/v/ - Video Games

it's fucking video games, baby

New Reply
Files Max 5 files32MB total
[New Reply]


[Hide] (78.5KB, 462x496) Reverse
>#8/agdg/ via irc.rizon.net
>https://matrix.to/#/+agdg:matrix.org via matrix programs
>Dev resources: http://8agdg.wikidot.com/resources
>Wiki: http://8agdg.wikidot.com/
>Previous: >>147361
[Hide] (21.4KB, 202x204) Reverse
Any ludum fags this weekend? This is the second time in a row, 3/4 of my latest gamejams that I would consider a complete garbage fire failure and I am not happy about it.
Replies: >>166198 >>167303
I didn't even know it was happening.
Replies: >>166207
Well you're not missing out because the theme is "Every 10 seconds" and it sucks dick.
Replies: >>166212
It depends a lot on your skills but I think the best way of doing what you want is to either make GZDoom/eduke32 mod (examples: https://wiki.eduke32.com/wiki/EDuke32_Modifications) or make a game based on ioquake3 engine. Or you can just use ((( Unity )))?
I guess I'll just keeping working on my existing projects.
[Hide] (406.6KB, 1597x869) Reverse
[Hide] (1MB, 1593x871) Reverse
[Hide] (1.1MB, 1595x871) Reverse
[Hide] (1.2MB, 1598x871) Reverse
[Hide] (1.1MB, 1596x869) Reverse
Progress. I'm not much of an artist so most of the props in my game are going to be an eclectic mix of whatever I find for free online.
GDDs are for absolute tools.
Replies: >>166310
Spoiler File
(471.5KB, 460x482, 00:06)
people who don't use GDDs are absolute tools
Game Design Document. It's where you write down, in great detail, how you want the game to be. Think of it like an engineering blueprint. There are likely to be several iterations as you discover what does and doesn't work.
[Hide] (82.7KB, 958x796) Reverse
More like a dry, professional, technical document. Though if you want to create a folder full of weird, autistic, queer bullshit and call that your GDD, you can do that. Pretty sure that's how Cruelty Squad was developed.
Clear out ayys from wrecked spaceships and stations with bros, buy new equipment or abilities with salvage money.
[Hide] (247.5KB, 768x576, 00:20)
Anyone know of a way to make menu navigation less painful?
Nature of this game unfortunately relies on submenu upon submenu due to being a "spellcrafting" type game, and I don't think just implementing hotkeys are going to save me.
[Hide] (116.9KB, 768x576) Reverse
For a start, don't erase the previous menu layers. There's no reason for it, and it's easier to go across the screen with your cursor (and do add hotkeys, there's no reason not to) as you go into each category once you know what you're doing and in addition to actually saving clicks going back to a single layer back it prevents both the 'the whole screen layout changed and I need to readjust' problem as well as lets you know what you're working on at a given time.
Crude mockup like so, obviously you'd add the hotkeys as a little letter in the bottom right corner of each box
Replies: >>166432
I originally had the menu configured like that, but I had worries that it'd result in far too much "clutter" on the screen when it gets past 5 columns
Replies: >>166435
The biiig problem with too many columns is just pushing the text/input box at the end too far to the right. Even 5 columns is pushing it, since not only are you going corner to corner but you're putting the stuff you have to actually read/look at all the way to the right.
Do you have an example of 5+ columns?
Replies: >>166444
What about a node based spell crafting system instead?
Replies: >>166444
[Hide] (47.1KB, 968x673) Reverse
I thought I had a case where I reached 5 columns at one point, but I was mistaken. I'll go back to using this type of menu until I find something "better"
please digit wizard spare me from the transparent and same-as-background menuboxes
[Hide] (192.4KB, 1619x507) Reverse
Yeah so instead of having a million menus with submenus you would drag and drop nodes with different attributes then connect them to create your spell.
[Hide] (85KB, 1920x1080) Reverse
[Hide] (65.8KB, 1920x1080) Reverse
[Hide] (179.1KB, 1920x1080) Reverse
As I understand it, the player has a preset list of spell behaviors (aura and missile) with set attributes of (missile speed,  etc.). Attached to any behavior is an effect (physical, body, magical) that also have set attributes. What the menus need to facilitate is rapidly choosing behavior and effects, and rapidly editing the attributes of each.

My proposal is as follows. Keep the first two layers of menu's, but rather than have spell type, spell properties, and preset spells, you instead have a "new spell" button, a spellbook (with ALL preset spells) and a quick access list of 5 presets.

In menu 1, when you click the "new spell" button, your mouse disappears and the + circle in the middle shows up, along with the text. Right click moves you up one layer in the menu, left click moves you down one layer. You don't have a mouse, instead the entirety of this central menu moves as you move your mouse. You move your mouse left, the menu moves to the right. Selecting makes the current and all lower layers shrink in size and blur out, showing that they're still there but otherwise getting out of the way.

The circles in the current menu layer grow larger with respect to their distance from the center. When the player clicks, the game checks if any circles are above a threshold size, and if they are, the largest circle is selected. The selected circle moves to the center of the screen, the past layers fade further into the background, and the next layer in the menu appears.

The final layer is different. A central circle appears with numbers showing the current value of all the properties, plus the total cost. Selecting a bubble reveals a UI element that signifies which way to move the mouse to change the value. In the Menu3 image, after selecting the size bubble, the player would move the mouse left and right to adjust the slider. The new value would appear real time in the central circle, and the description text would appear underneath the bubble menu. Left click accepts the current value, right click cancels any changes.

True/false values could work by having two check boxes side by side. As the player moves their mouse left or right, one checkbox would grow larger and the other shrink. This trend has been thoughtfully established by this point that it should be obvious to the player which one is selected when the player clicks.

The one big disadvantage I see is that you have to descend the tree to get back to spell properties. Sure, it's only a few left clicks, but that risks going too far back and canceling out of the entire spell creation process. Now that I think about it, there should probably be some protection (like an "are you sure?" menu) that keeps you from accidentally hitting left click one too many times and leaving the new-spell menu.

I'm showing the menu's here with letters, but icons representing the properties should be used in the final version.
Replies: >>167787
[Hide] (532.8KB, 1197x772) Reverse
>trying to extract textures from an xbox 360 game
>some post at some random forum
>"you can try using the le fluffie tool"
>let's check that out then
>open up the le fluffie whatever that is
>see this
Replies: >>166499
furfags and the hacking community intersect quite a bit. if the tool works though that's all that matters.
Replies: >>166500
[Hide] (43.8KB, 637x637) Reverse
Problem is it may not even do what I want it to do.
I had already extracted the contents of the game's iso with another tool and was checking out whether you can read any of the packages with any tool.
This le fluffie nigger can't read the .xbp, .xen files that the program spewed out, so might as well execute the furfag on the spot.
If you just want to rip textures on anything, you could try RenderDoc. It's not straightforward, but it should do the job.
Replies: >>166503
Spoiler File
(1.5MB, 1024x1024) Reverse
Finally got the texture using the workaround >>166501 suggested.
What I don't understand is why the program hadn't worked the first time around.
[Hide] (88.7KB, 1200x675) Reverse
Got around to using the PS4 unpacker from previosu thread and managed to snatch the Raoh and Toki models from Hokuto ga Gotoku.
Is there any way I can open a Blender 2.8+ file with Blender 2.79? Or at the very least export both the model and the skeleton, either together or separated.
Replies: >>166601 >>166614
[Hide] (2.4MB, 495x379) Reverse
write a five thoundred word document on why you're or you're nottoers makering you're gamer this very momentoes matoeable
Replies: >>166635 >>166747
Try if you can export the model as collada (dae) and then import it.
Replies: >>166617
[Hide] (39.1KB, 625x469) Reverse
oh yee gennie gunna dee another moonman dooman 1488 matoeable please enjoy your self above all
Replies: >>166627 >>166635
Did you try to just open it? Blender has some weird format that allows it to be pretty backward and forward compatible at the same time. You might need some touch ups with materials and new features, but the models itself should work.
Replies: >>166617
[Hide] (75.2KB, 493x515) Reverse
I'm accostumed to exporting in .obj which doesn't support skeletons. .dae does support rigging data right?
>did you try to just open it
I did. It was too stupid of an idea for it to work.
What should have been done was to export the model in another format, though I got a brain fart when I realized I couldn't use the .obj extension.
>you might need some touch ups with materials
UV maps apparently work and the textures were converted alongside the model. I know I'm going to have transparencies with the models, though that's for another post once I get time to illustrate my point.
big bnuy
draw a space marine
[Hide] (841.7KB, 1503x844) Reverse
[Hide] (844.8KB, 1517x848) Reverse
I've been messing around with the Toki model. Should be good other than the transparency issue surrounding all textures using the alpha channel.
It's not too clear in the 1st picture but you can see through the model where the bits of texture using the alpha channel should be. The 2nd pictures shows how the texture should display in Object Mode yet doesn't.
Is there a workaround with this using the cycles render or similar? I have been trying to search for this exact same problem and while replicated, none of the solutions are actually useful.
Replies: >>166694 >>166726
[Hide] (14.7KB, 221x221) Reverse
>0 replies
Come on, Blender is a bitch. Surely someone's had the same issue as I had right?
Replies: >>166697
>no replies in 4.5 hours
Next time try something other than a dead obscure imageboard if you want a fast reply.
I don't know much about blender, but you could go to Material Properties and later to Viewport Display The Sphere Icon on the right to Settings and mess with the two options of "Blend Mode" and "Shadow Mode" it should fix the transparency problem.
Replies: >>166778
[Hide] (1.4MB, 360x238) Reverse
Nice numbers but no.
[Hide] (25.5KB, 308x448) Reverse
>its just a quirk of blender to display transparency
The problem carries over to other game engines that's the problem.
>try going Alt+N 
This is to recalculate normals right? If so I don't know how that'll help. It didn't.
>you could go to Viewport Display
Note this is an older version of Blender and Viewport Display isn't an option here. I did find the Blend Mode option in 2.79, though it isn't changing much.
Wow, what an informative and helpful post! This post right here converted me from Blender newbie to super class ultra mega wizard 3D modeler that can make 3 gorillion photorealistic 3D assets in uder a minute.
>technically that's true for basically this whole site
3 or 4 nigger rapefugees shitting all boards don't equal to the entirety of the site. Proceed to kill yourself immediately.
[Hide] (294.2KB, 640x360, 00:04)
>"now that the dust has settled"
Stop derailing the AGDG thread and go back to cuckchan.
>How hard could it be to make a 3D stealth game anyways?
Very difficult.
[Hide] (146KB, 329x277) Reverse
Sounds like fuckin' korean morrowind
[Hide] (11.3KB, 320x288) Reverse
have sex
>There's a nigger there.
<There's a nigger there!
>Really? even a shit one?
Enough with the abstract and hypothetical. You're caught in an endless cycle of "This is so easy, why isn't anyone doing this? This is so easy, why isn't anyone doing this? This is so easy, why isn't anyone doing this?". Stop talking about how easy it is, install Unity, and make a 3d stealth game. If I'm wrong and it's actually as easy as you say it is, then I look forward to playing your game.

The tagline of AGDG is "just like, make game" for a reason. Not because it's descriptive, but because that's what you *should* be doing.
>no talking about videogames on /v/
<half the posts are gone
I'm beginning to understand why anons say it's pointless to post itt
My bad, didn't noticed that he was posting from tor and deleted from ip. Continue as usual.
Replies: >>167009
[Hide] (5.8MB, 962x520, 00:29)
>he pushed the mark button
>he fucked up the capcode
Replies: >>167025
I'm extremely sorry, will not happen again.Mark deleted from the whole board instead of just a thread, not trying to make excuses but also don't want people to forget about what actually happened.
[Hide] (390.9KB, 500x280) Reverse
The incompetence is off the chart.
You can ensure that it won't happen again by streaming your suicide
any other option is chickening out
You are the Blackest Gorilla Nigger I have EVER seen.
>this wont happen again i swear you guys!
once is one too many, enough is enough with your rulecuckery
[Hide] (351.5KB, 1374x596) Reverse
>bunch of niggers chimping out in aggydaggy
Fuck you faggots post progress. I'm currently working on porting hexasphere to Unreal Engine for my hexbased planet building game.
We should totally have our own game jam! just think of the potential for talent to emerge!
Replies: >>167098
>keeps acting like a nigger
I'm already working on the logo as we speak.
[Hide] (2.3MB, 360x200) Reverse
I want to bring MGO to PC/Linux, but how should I do it? How can i bring it to PC without losing its charm? Any ideas for what weapons/camos/items to include into the game? I'm all ears as I want to make this game for (You) and not pander to reddit or cuckchan or Eden.

First things first I need 2 factions with unique equipment for both sides then "hero" or "villain" characters that have special abilities like shooting around corners with revolvers. US vs RU seems too generic so I expect something wild like rhodesia vs NASAs SWAT team.
Replies: >>167222
[Hide] (851.1KB, 667x936) Reverse
I recommend getting a prototype working before you worry too much about the skin that those mechanics are going to wear. At least get two cylinders in a room together sliding around and updating over a network. What you've demonstrated thus far is that you're an idea guy that can't come up with his own ideas.
I've analyzed a vast variety of classics, indies and other well regarded games. These seem to be the most common traits among them.

1) Really low system requirements. Making your game a resource hog limits your audience right off the gate. 

2) Ease of "mod"-ability: name the 10 most longevous games. Odds are they have a mod scene, or at the very least some kind of user generated content.

3 Strong multiplayer support: this one is very genre dependant, but where it works, it encompasses the entire market.

With this in mind, I'm better off making an engine from scratch or is there an engine that can lend itself well to these three traits?
the classics weren't necessarily "low system requirements", it was more just now computers are so far beyond what they used to that even today's toasters are dozens of times more powerful than the most powerful consumer grade hardware of the 90s, some of them had some high requirements for the time.
Moddability is nice, but I think some genres are more just meant to attract people who sink ungodly number of hours into them specifically and encourage such things.
See: sandbox, sim, and autism sim genres
Multiplayer support just goes without saying for certain genres, so yeah.

However, those criteria/traits just favor sims/sandboxes with simplistic looking graphics, whereas in terms of genres, its very different between each other what makes a game "quality" or not.
Replies: >>167296
Perhaps "low system requirements" was a misnomer. Maximizing resource efficiency may be a better fit. Gods knows optimization is a lost art today
Replies: >>167300
For #1 I would say "performant". That is, it runs well on the targeted hardware. This includes frame-rate, bugs, and so on for anyone who isn't actively trying to break the game.
Modding support can improve a game's longevity, but I don't think it's required to be a classic. Tetris, for example, is a classic with many different versions. But I wouldn't exactly say it has a modding scene.
Strong multiplayer support doesn't make something a classic. It's the bare minimum preventing a multiplayer game from dying.

1 and 3 are necessary but not sufficient. 2 can increase longevity and is nice to have, but I don't think it's necessary for something to be considered a classic.

In terms of engine choice, it has far more to do with the developer than with the engine. If you have to ask "should I make a game engine from scratch" you probably shouldn't.

Lack of optimization isn't the problem. You can write acceptable code that isn't optimized, especially on modern hardware. The problem with modern software is that people are actively writing bad code.
Replies: >>167304
You need a producer bro, otherwise you are at the mercy of whatever random person put together the team. Usually they are too busy arting or coding to keep the project scope down to minimum and on schedule
Replies: >>167312
Well, that's the zeroth law of optimization.

0) Do not write bad code.
Replies: >>167310
Depending on your definition of low resource usage, you can make a low resource usage game in Unity or Godot for sure, even Unreal. Use the profiling tools to optimize bottlenecks, limit draw calls, use object pools instead of dynamic allocation at runtime, consider patterns like ECS for CPU cache opimitzation, multi-threaded animation, LODs, etc. Quake source ports, Build engine, and GZDoom are also open source. Can also check out libs like raylib that are pretty lightweight. But usually games with high resource usage are either unoptimized, lacking data structure/algorithm knowledge in how they were implemented, or are high fidelity 3D simulations. Given your goals, you might find the books Data Oriented Design and Game Programming Patterns interesting.
[Hide] (25.5KB, 474x257) Reverse
How low can you go when it comes to games of today? could you make a shooter like perfect dark or a platformer like mario64 under 20mb? I really want to make a fun game thats really low file size and low system requirements as to get as many people playing as possible but is there anything special i need to do to achieve sub 100mb total file size?
Replies: >>167312 >>167315
I still consider not-writing-bad-code and optimization to be two different things. Real optimization to me means stuff like code motion, loop unrolling, in-lining, running through all the available algorithms to see which one works best, that kind of stuff.
For example, take memory optimization vs not writing bad memory code. If you're going "I should reduce the number of variables being passed to this function to 6 because x86 reserves 6 registers for passing to functions before it starts writing to the stack" then you're optimizing (and your an idiot if you do that without a profiler telling you it's a good idea). If you're going "I should cache this stuff I use 1000's of times a second instead of malloc and freeing it 1000's of times a second" then you're not writing bad code. If you fix the latter you aren't "optimizing", you're fixing bad code.

Though I'm being more pedantic than anything. I'll probably still say "I'm doing an optimization pass" when really I'm just fixing my dumb mistakes that impact performance. The point I want to make is that the problem with modern code isn't a lack of "I'm going to put this value stored in an array into a temporary variable because I know without that the compiler won't optimize as much as it can due to memory aliasing" and more a lack of "I probably shouldn't use the standard library here because of how many hidden memory allocations there will be".
[Hide] (53.1KB, 1000x1000) Reverse
It's mostly on me not shooting down the other people's shitty ideas.

<100 mb
For what purpose? Disk space is literally free.
Replies: >>167321
If you want a game with a low file size then the "secret" is compression. How much are you willing to sacrifice to get a lower file size? The more you are willing to sacrifice, the lower the file size.
You should be compressing your audio files and textures regardless, though if you sacrifice detail and/or quality you can get surprisingly small file sizes. Low poly models are essentially a form of compression performed with artistic intent rather than with an algorithm.
That being said, there are correct and incorrect ways to use compression. You should research whatever compression you're using to know what kind of content it works best for. Jpeg is optimized for photographs and png is better flat-color images. Games often have their own special compression algorithms optimized for textures, like BPTC.

If you want to go really crazy, you can look at .kkreiger. It's a 3d FPS that fits in 96 kilobytes. Lots of very fancy compression tricks were used, though because so much effort was put into reducing the file size it suffers in other areas, such as portability. The demoscene is full of fancy compression. Mod tracker music is probably one of the most prevalent.
Replies: >>167321
>for what purpose
So poojeets and Argentinians can play, I also like the idea of a game taking up as little space possible to be extremely portable and can be downloaded easily within a short amount of time. who the fuck want spunkgargleweewees 100s of GB file sizes that eat your storage space and do absolutely nothing, 100mb of RAW gameplay is something to strive for as if classic games like duke3D or goldeneye can be made under 100mb then modern games under 100mb should be even better! imagine a goldeneye tier game with slightly higher polycount on everything and sharper textures i'd fucking cum buckets at the thought of it. the lower the file size means the higher the chance you can play the game on a smart fridge.

i was just watching a documentary on .kkreiger it was still confusing to figure out how and why they did their different compression and other wizardry to decrease the file size but in the end .kkreiger still needed high specs for the time to run whereas i want both low spec and low file size.
Yeah, .kkreiger sacrificed a lot for file size, including performance. Everything else I said about compression still applies to low-spec systems though. Good luck with your game.
[Hide] (388.1KB, 1200x830) Reverse
[Hide] (455.4KB, 1200x830) Reverse
[Hide] (393.2KB, 1200x830) Reverse
[Hide] (542.5KB, 1920x1029) Reverse
[Hide] (2.3MB, 1920x1080) Reverse
did some progress on the model converter for gravity rush 2:
- skins are now exported (joints, weights and inverse bind matrices) making skeletons available
- LODs can now be filtered by their index
- character faces are now correctly filtered when requested
only thing missing is reversing the PBR parameters (roughness, specular, etc...), but i think these might be defined elsewhere.

i also created a tool to convert from GLTF back to the game's GFX format.
i planned to replace the player's model with some other funny/cool model, but ended up converting the model's meshes only.
i could come back to this and convert skins back so the models could get animated, but i'd rather work on my game or attempt to "emulate" the bloodborne box's games

source code is available at https://gitgud.io/glue_sniffer_420/gr2-gfx-tools
Obligatory nude mod when?
Replies: >>167354
>still no WW2 singleplayer FPS where you play as the good guys
Do we REALLY have to make it ourselves?
Spoiler File
(26.8KB, 514x410, 00:00)
[Hide] (25.8KB, 150x188) Reverse
Ayo what fucking year is this
Replies: >>167364
Very cool.
[Hide] (809.6KB, 960x540, 00:11)
>you'll be able to play GR 1 and 2 with naked Kat.
Very nice.
these games never went away, COD and its contenders still pump out regular shitfests that act like glorified wack-a-moles, which is why we need NEW FPS games that are not shit and also not 100s of GB in size
[Hide] (80.1KB, 884x576) Reverse
say that I have several entities on screen, each of which are reporting their position to the player when they enter a certain mode. what is the most efficient way of having each report their straight-line distance to the player?
Replies: >>167575 >>167576
there was some game I heard that was under 5MB and was procedurally generated FPS. can't remember much else about it except that it had Quake/Unreal aesthetics. I think Druaga1 played it on one of his videos but I have no idea which one it was.
What type of game could do well for a gamenight?
Replies: >>167574
the one you make yourself
You already know both positions, distance between those is literally a basic math function. Only real optimization you can do there is not doing a square root and instead comparing against the square of a value, which is fine as long as your distances are never very large.
If you're asking how to calculate the distance between two points, stop trying to make games and go learn what a vector is first.
Create an array of positions.
When an enemy is spawned, it gets the next available index in that array, iterating a counter when it's added.
When an enemy is de-spawned, the last enemy in the array update's its index to that enemy's index, and the counter is decremented. If the last enemy was killed, then you just need to decrement the counter.
An enemy writes it's position to the associated index as often as you want.
The player may then iterate over that array any time they need to get the enemies' positions.
Replies: >>167650
>Mark deleted from the whole board instead of just a thread
Yes, after being encouraged to do so during said time.
[Hide] (1.6MB, 296x240, 01:37)
Replies: >>167693
heres an idea. why dont we have a gamejam where you have to make a feature rich 3D game but keep it under 1gb in size?
i'll make the logo
Pretty easy to do this even with unreal and wise. Main thing to watch out for is massive textures and only packaging the assets needed
[Hide] (480.7KB, 4931x2088) Reverse
>Feature Rich
>Game Jam
Do you understand the concept of a game jam?
1 GB is too big to be an interesting challenge, especially considering game jams already produce small games. Set the limit to 1 mb.

What's the joke? The video is just explaining a basic localization and mapping algorithm. :^)

Here's an image that hopefully explains what I was talking about better.
You have two options. You can store just the array of pointers to the enemies, or you can do what is shown in the image and have two parallel arrays. The two parallel arrays area easier (and probably faster) to iterate through. The single enemy array works better if you want to store the enemy positions within the Enemy data structure.
Replies: >>167698 >>170330
[Hide] (23.9KB, 474x474) Reverse
OK, what if we did something like you have 10mb to create an hour long FPS game OR focus on a decent multiplayer mode, can be any theme or setting

OR we could do something like 15mb to create an old school survival horror game that takes at least an hour to beat if you know where everything is, just like the original RE, RE2 was able to fit onto an n64 cart so it is possible to do on 15mb if you cut some corners here and there
Replies: >>167715
It can be difficult to judge if a game is an hour-long game. Let's say the developer rely's on difficulty to extend the length of the game to the point they finish it in about an hour. Then someone who's really good may finish in 30 minutes and complain that the game wasn't actually an hour long. I'd say instead of giving a length requirement, extend the length of the jam to something like 2 weeks.
Making the limit an actual hardware limit rather than an arbitrary number sounds fun. The N64 cartridge can hold up to 64 MB like RE2, though there were smaller carts that only held 4 MB such as Dr Mario and Automobili Lamborghini. In the middle you had 32 MB games like OoT. I feel like 4MB is more game jam size, so that's where I would cast my vote. You're not getting a game like RE2 out of a game jam.

To formalize the rules for an N64 jam:
>4 MB maximum. Distributing a zip file containing the game is fine, but the unzipped game must be 4MB or less. Any further decompression must happen automatically after starting the game.
>Can't use more than 10 buttons, a D-pad (arrow keys/wasd), and a single analog stick/mouse
>If any texture filtering is used, it must be bilinear
>If there is any multiplayer at all, you must also include couch co-op
Limits like ram usage or texture cache size I feel would get in the way more than be fun limitations, especially for a game jam, so we ignore them. Bonus points if you do hold yourself to those limits though.
Wouldn't it make much more sense to say the game has to run on an actual N64? I know, that would increase the difficulty of the jam significantly, as you can't use many modern tools that way, and it's an unknown platform for most developers, but what you're doing now is also almost arbitrary limits.
Replies: >>167721 >>167728
This is a retarded idea and I'm surprised you suggested it without thinking about the implications of such a post.
>I know, that would increase the difficulty of the jam significantly
No, it would make the jam impossible. You would make a game jam that nobody would participate in.
The limitations I give are not remotely arbitrary. I started from the N64 and removed limitations that would be horrendously difficult or unfun to deal with. The juggle of fitting a game on 4 MB seems fun. Limiting screen resolution is trivial. Thinking about juggling the RSP unit, microcode, RDRAM latency, and 4k texture cache during a game jam makes me want to die.
I haven't checked N64 personally, but a lot of old consoles have active homebrew community who still develop games. Are there may people who are not willing to deal with it? Sure, but it still does not make it impossible.
Replies: >>167741 >>167822
I'm saying impossible in the practical sense, not the theoretical sense. None of the people who can develop for the N64 would participate in the game jam. Everyone who would participate in the game jam couldn't/wouldn't if they had to make it run on the N64. The fact that the ven diagram is two disconnected circles is what makes it impossible. I want to make a game jam with fun limitations that people (in this thread, because apparently you need me to specify that) can actually participate in.

I can't figure out if you are actually retarded, if you're trying to make a point that *any* limitations for the game jam are arbitrary and dumb, or if you hear "N64 game jam" and your autism cannot stand that it be anything other than writing a game that can literally run on an N64.
In my opinion, a GBA jam would be much better, as it's much easier to program for and there are already tutorials on how to do it: https://www.coranac.com/tonc/text/toc.htm
And since 2d games are generally easier to develop for than 3d games.
Replies: >>167762 >>167990
[Hide] (57.5KB, 552x520) Reverse
Oh wow. Another game jam idea that will produce zero games.
The game jam theme and rules will not restrict platform. Eligible games can be made for any platform.
If you want to develop a game for GBA, more power to you. But I'd rather our first game jam not end in utter failure because nobody wanted to do engine dev for a handheld old enough to drink.
Thank you for your insight and effort spent on the demonstrational images. I wasn't expecting this modern type of UI to not only be thematically "appropriate" to the whole Wizardry and spellcrafting, but also highly functional.
Replies: >>167850
What is the N64 homebrew scene like? I know there are game boy jams. There are tools that make it easier to dev GB games though, like GUIs and C APIs
yeah lets learn some lua sounds like fun
what the fuck
Replies: >>167831
>Anon is such a baby scientific notation scares him
Replies: >>167836
telepathy fail
[Hide] (13.2KB, 474x355) Reverse
mite b kewl, and sounds like a fun idea for a game jam. forgive my retardedness but I have no idea what game jams are about anyways, i thought it was just grab an idea then make some fags do it in X amount of time. An N64 jam sounds pretty lit and i would participate in it tbh
>dude make it run on an actual n64 lmao
that would be nice but I think that makes the bar of entry too high for most people here to enter.

All in all i think that this would be perfect for a gamejam on here and we should make the logo starting sometime soon either RIGHT NOW or maybe later
Replies: >>167850
Happy to help.

>i thought it was just grab an idea then make some fags do it in X amount of time.
Pretty much yeah. But you can't make anyone do anything with these sorts of online collaborations. People have to want to participate. Hence why I think the barrier to entry shouldn't be very high.
Replies: >>167853
[Hide] (2.2MB, 400x300, 00:34)
>barrier of entry shouldn't be high.
Then lets do it! n64 gamejam sounds cool and I would make a thing for it if it happened.
>people must WANT to participate
yeah i get it but then we just need to figure out what type of gamejam people would want? like a survival horror jam, bitches LOVE zombies!
Replies: >>167872
I think the existing restrictions on file size and screen resolution are enough, no need to apply additional genre constraints.
[Hide] (3.9MB, 1280x720, 00:16)
I'll just post this here. Haven't worked on this in 3 months. Neither the time nor the motivation I have.
Maybe I'll pick it up one day.
Replies: >>167928 >>168037
Meh I'm going to work on a GZDoom mod for now because at least I can use existing maps to play it on, since I don't enjoy making maps. If there was a good Sven Co-op map importer for Godot I think I'd would use rather that but there ain't one. 

So in the past few days I spend debugging and working on a barebones shop menu system, it took me a while to code it because writing the fucking position per hand is a pain in the ass as GZDoom doesn't come with any sort of GUI designer. 

And as for weapons I have implemented only 4 so far yet, more to come. Which is 1 shotgun, 2 SMG and 1 rocket launcher. 

Money in this mod is permanent so feel free to hoard as much money as you like.
Replies: >>167953
>fov constantly changing 
This is THE most reddit thing you could put into a game except for anthony burchs writting
[Hide] (470.1KB, 715x718) Reverse
[Hide] (24.5KB, 201x279) Reverse
Are you going to start shit just because or what? This thread barely got progress posting for several days by now.
>discord tranny
Do you got a proof that I'm a tranny?
Replies: >>167949
>replying seriously to retards
>on tor of all things
you never had a chance
Replies: >>167954
Pretty good gunplay. Is this going to be a total conversion or something?
Replies: >>167954
A chance with what? Finishing this project? I only have to do weapons now, add items, improve the scripts a bit and other stuff then this mod will be pretty much complete as I don't do anything experimental this time. My other Godot project failed because there is no good map editor that I can use, Trenchbroom is not suited and I don't like Blender controls for making maps with, it does stupid shit like selecting polygons when I want to move the camera which is annoying, Trenchbroom doesn't do that but its not well suited for outdoor maps. 

Thanks. And no it's not going to be a TC. Just a class and weapon replacement mod.
>Game jams
Why not make FreeDOS game jam? I think it's much easier to develop for it. It even supports multiple programming languages well, at least ASM, C and Lua? And probably Pascal, too... FreeDOS even comes with a package manager (the packages are just zip files: http://wiki.freedos.org/wiki/index.php/Package). Packages can be distributed over Internet since FreeDOS has networking. There are even C programming tutorials on FreeDOS YT channel.

But if there is going to be N64 or GBA game jam, be sure to actually post instructions for writing and building programs for these platforms! For example, for a GB/C game jam I would link Pandocs (https://gbdev.io/pandocs/), GBDev wiki and RGBDS.

But his game looks actually good!
Replies: >>167994
<paying to give some CIA front your IP address and what BLACKED porn you torrent

sounds cool to actually develop for a system but and this is a big butt! you'd have to make it niggeretard proof in order to get anyone to submit a game.

a simple 3D or 2.5D game in DOS does sound good but I doubt anyone here knows what the fuck to do with it to make a game, making a game for DOS is like making a PS1 game only less lame and less hardware restrictions alone in the dark and wipeout were made for DOS and they're good games, give us 2 weeks to try and make something in DOS or at least give us a base to use in DOS
[Hide] (237.8KB, 894x1024) Reverse
[Hide] (225.6KB, 800x433) Reverse
[Hide] (130.5KB, 735x545) Reverse
I've got a design question. The game I'm working on is a game about the colonization of a fantasy North America, and I plan to have three modes of play: an action mode, where you and your party go into battle, kick ass, take loot, and repeat; a battle mode, where you have control of a small unit and fight against your enemies; and a kingdom mode, where you play as a lord, governor, or chief, and defeat the other factions to end up on top. It would basically be Mount and Blade mixed with Morrowind and Arcanum. 
The question is, how do I integrate these systems into the greater whole? I can always copy M&B's way, but I want to consider some other options, like an isometric view that can seamlessly zoom in and out from action, battle, and kingdom mode with a single scroll. If I were to choose the latter option, what would be the most optimal way to implement it?
Replies: >>168012 >>168146
make the logo first, the rest comes naturally
Replies: >>168013
ill make the logo, you market the game!
Replies: >>168020
sure i'll make posts on furaffinity and eka's portal
[Hide] (92KB, 683x681) Reverse
Sometimes the non-progress posts on this /agdg/ are so bad that I wonder if someone is trying to kill it out of spite. It's not outside the realm of possibility, considering how zzzchan had autists going around and demanding /agdg/'s removal on random threads for a while. I even looked at Markchan's /agdg/ for comparison and, as suspected, they don't have this garbage clogging up the thread.
Bitch 8/agdg/ had an entire furry gzdoom game called Memoirs of Magic. We gave the dev a hard time anytime he posted anything sufficiently heinous (like gay as shit character designs, NOT using the wrong linux distro or chat programs) and we did it without trashing our own thread. Meanwhile I look here and not only is OP shitting up the thread, he's so oblivious he can't even use a working link to the Matrix room.
Are the gun model and animations your work? Those aren't half-bad.
Replies: >>168047 >>169255
[Hide] (15.5KB, 256x234) Reverse
>"someone" using off-site meta drama to D&C shill in the dubs thread
>gets told to fuck off
>"someone" starts doing the same thing in the aggydag thread
<someone is trying to kill it out of spite
You don't say.
What about an NES challenge where it runs on actual NES hardware? Since 2D games are easier and because it only has 2 KB of ram, there's less variables to worry about.
Replies: >>168112
<what about a challenge where everyone has to write assembly code
Did you really have to even ask?
If you want anyone to actually make a game you have to set your limits at a fixed palette and some sprite and screen size restrictions, not actual hardware.
Replies: >>168113
To be fair, NESmaker is easy to use for basic games and runs on real hardware.
Replies: >>168115
Well unless you have a trustworthy source to pirate it from without getting dolpin porn, no one's going to shell out 36 dollaridoos to enter an anonymous gamejam. Not to mention how many of us are Linuxfags, and from a cursory gance it seems NESmaker doesn't work under WINE.
Replies: >>168116
[Hide] (10.3KB, 1313x949) Reverse
Here, I got it from an anon in a QTDDTOT thread.
Check if it works.
Replies: >>168150
[Hide] (19.6MB, 1280x720, 01:41)
How much direct control do you want to have over your units in battle?
if you are fine with not issuing direct orders you might be able to integrate action and battle into a semi-unified system,something along the lines of Kingom Under Fire's troop mechanics.

Strategy view to battle can be a zoom in or out,yes.
Depending on how you want the strategy layer to function it can either be a different simulation entirely (which will require some switching/synch logic) or can also be folded into a single world where issuing orders actually causes the battle mode units to act in world physically. (which allows seamless integration but might end up being a performance hog or unneeded complexity)
Replies: >>168166
Works on my machine!
>How much direct control do you want to have over your units in battle?
I want to be able to issue orders to my units: hold position, go to position, form ranks, fire, charge, and retreat, while also fighting alongside them. Other than that, the AI should be autonomous enough to do what it needs to without player interference. 
>if you are fine with not issuing direct orders you might be able to integrate action and battle into a semi-unified system
I could. The only real difference between action and battle modes would be the scale and logistics, since action would be you and your companions, versus battle which is you and your company of men. 
>Strategy view to battle can be a zoom in or out,yes. Depending on how you want the strategy layer to function it can either be a different simulation entirely (which will require some switching/synch logic) or can also be folded into a single world where issuing orders actually causes the battle mode units to act in world physically. (which allows seamless integration but might end up being a performance hog or unneeded complexity)
Good point. I'm using Unity so it's bound to be a resource hog if the player can zoom in and out all the time. I'm not sure whether the strategic mode should be an overworld or something like an interactive "desk" where you interact with the world from your manor, with a map and spreadsheets. That'd probably be gay for the official release, though, but it could work as a prototype.
[Hide] (43.5KB, 500x478) Reverse
My ideas are too strong for you, ideafag
int x = 1;
int y = 1;
Console.WriteLine(x + y % 2 == 0);
N64 jam starting. >>168376
Push all columns to the left with the first/oldest column falling off of the screen when there are too many. Lets players see all of the options without crowding the screen too much. Alternatively, limit columns to the upper half of the window and give the player the ability to scroll them. Leaves the bottom half open the spell information.
[Hide] (129.5KB, 437x434) Reverse
i got started on a game loader for the ps quadruple.
it can only load a simple homebrew sample so far but it's a start to implement whatever else is needed to load other stuff.
i'm looking towards uploading the source code once i get it to load some other samples.
here's a log of what the loader output, since there's nothing else to show:

possible language: perl, relevance: 15
Loaded 37328 bytes from ./samples/hello_world_c/eboot.bin
phdr[0]: type 0x1 off: 0x4000 vaddr: 0x0 paddr: 0x0 filesz: 0x440 memsz: 0x440 align: 0x4000
phdr[1]: type 0x61000010 off: 0x8000 vaddr: 0x4000 paddr: 0x4000 filesz: 0x8038 memsz: 0xc000 align: 0x4000
phdr[2]: type 0x1 off: 0x14000 vaddr: 0x10000 paddr: 0x10000 filesz: 0x78 memsz: 0x138 align: 0x4000
phdr[3]: type 0x61000001 off: 0x14000 vaddr: 0x10000 paddr: 0x10000 filesz: 0x50 memsz: 0x50 align: 0x8
phdr[4]: type 0x2 off: 0x158f8 vaddr: 0x0 paddr: 0x0 filesz: 0x1c0 memsz: 0x1c0 align: 0x8
phdr[5]: type 0x3 off: 0x4000 vaddr: 0x0 paddr: 0x0 filesz: 0x15 memsz: 0x15 align: 0x1
phdr[6]: type 0x7 off: 0x0 vaddr: 0x0 paddr: 0x0 filesz: 0x0 memsz: 0x0 align: 0x1
phdr[7]: type 0x61000000 off: 0x15608 vaddr: 0x0 paddr: 0x0 filesz: 0x4b0 memsz: 0x0 align: 0x10
DT_SCE_HASH: 0x2c0
DT_SCE_RELA: 0x1d0
DT_INIT: 0x7c
DT_FINI: 0x7c
warn: unused tag type 0x15 at index 16 ptr 0x0
warn: unused tag type 0x16 at index 17 ptr 0x0
warn: unused tag type 0x1e at index 18 ptr 0x4
warn: unused tag type 0x1 at index 19 ptr 0x1
DT_SCE_NEEDED_MODULE: 0x101010000000f
warn: unused tag type 0x61000015 at index 21 ptr 0x100010000000f
warn: unused tag type 0x61000019 at index 22 ptr 0x1000000000009
warn: unused tag type 0x61000007 at index 23 ptr 0x0
warn: unused tag type 0x61000009 at index 24 ptr 0x27
warn: unused tag type 0x6100000d at index 25 ptr 0x10100000019
warn: unused tag type 0x61000011 at index 26 ptr 0x0
virtsize: 0x10138
alignedvirtsize: 0x11000
resolved symbol[2] to libkernel/1jfXLRVzisc
resolved symbol[3] to libkernel/wuCroIGjt2g
resolved symbol[4] to libkernel/6Z83sYWFlA8
resolved symbol[5] to libkernel/ku7D4q1Y9PI
eboot image base: 0x7f486ab8d000
gonna exec
sceKernelUSleep: ms 2000000
_exit: status 0
[Hide] (193.5KB, 1200x739) Reverse
low iq posts are not welcome here
Replies: >>169258
[Hide] (8.6KB, 303x149) Reverse
[Hide] (27.9KB, 400x300) Reverse
nevermind I found it
>paste in draft of a report I was to email to a coworker
I knew it
Replies: >>169267 >>169276
>Testing different 5G-coronavirus conspiracy theorists using posts from one of their Facebook groups, I got IQs ranging from 75 to 89.
Ok, this got me thinking. Let's try some text from Robert W Malone
>musk & twitter: 131
>giving kids covid boosters: 110
Let's try boriquagato
>canceled by twitter: 146
>some rant about indigenous "science": 143
>pfizer lying again: 119
Yeah, the algorithm is not rigged, Not at all.
[Hide] (945KB, 155x173) Reverse
This IQ estimator probably tests a line of text by correct sentence structure, and the grade level of words used. Put in a part of some peer reviewed article about anything, some other line of broken or poorly written text, and then some quotes of Hitler and see what happens.
Replies: >>169285
[Hide] (72.9KB, 655x491) Reverse
[Hide] (114.8KB, 647x495) Reverse
Here's your post compared to a random /cow/ post I found. Most posts on /cow/ didn't have enough info to estimate IQ for some reason.
Replies: >>169283
What does putting in 1000 niggers give?
Replies: >>169306
And regardless, now that people know this is a metric by which they will be judged they will formulate their sentences and test them specifically in an attempt to game the system and appear more intelligence than they are.
Replies: >>169288
[Hide] (47.7KB, 649x495) Reverse
>and appear more intelligence than they are.
I don't think this estimator does a good job.
Replies: >>169290 >>169292
[Hide] (41.2KB, 854x523) Reverse
It works for me.
>Fixing the typo brings it down to 113 IQ
I tried that it doesn't work.
[Hide] (328.3KB, 1280x960) Reverse
Replies: >>169651 >>169655
Did the THX criteria ever leak? Was it always a brand recognition scam or was there professional rigor involved for video games? It actually mattered in theaters, or at least it did in the 20th century.
[Hide] (60.3KB, 600x800) Reverse
of course
[Hide] (46.9KB, 1288x752) Reverse
i got more progress done on my ps4 program loader. done enough to display a simple homebrew graphics example to screen.
some corners were cut when implementing some of the needed APIs - the virtual memory allocator API is just malloc'ing memory for sake of simplicity, the event queue API is incomplete and the rendering process occurs in the CPU.
pic rel is the resulting render
Replies: >>170161
How long did it take to render? Fractint could do a Mandelbrot in about a second on a 486.
Replies: >>170286
[Hide] (497.3KB, 610x508) Reverse
without resorting to any timers, it takes a quarter of a second or less on a 5600x.
the sample code is from https://github.com/OpenOrbis/OpenOrbis-PS4-Toolchain/blob/master/samples/graphics/graphics/main.cpp , which is based off https://github.com/olekssy/ican-c/blob/master/math/mandelbrot.c
[Hide] (224KB, 640x360, 00:03)
got it working, not sure if this helped or not but i think i did your implementation more or less, one longass string later
selectedScanEntity = to_local(get_parent().get_node(pv.scannableEntities[pv.scannableEntitiesDistances.find((pv.scannableEntitiesDistances.min()))]).global_position)
Replies: >>170338
Glad it worked out for you. As long as it works and you understand it well enough to debug it.
Personally I would break it up into multiple lines. It won't make the program run any faster or slower but it will be easier to read.
[Hide] (9.5MB, 480x360, 03:04)
getting to the point where i should probably start working on a menu/backend system for muh gaym. post cool vidya main menus and what makes them cool.
can someone give me a quick rundown on licenses? whats the difference between GNUgpl and MIT,etc?
is there a way I can sell my games yet still have them be open source but not get fucked over by the jews?
Replies: >>170807
GPL is poorly written but it basically means anyone can do anything with your code as long as they publish the source somewhere. Incidentally, it explicitly allows charging for source code access, so users interested in fixing a bug on their own or making a port could be kiked out of it that way.

MIT allows even greater freedom, only requiring attribution of the code base. The Godot engine chose this license to encourage commercial adoption.

The Creative Commons series is like a bitfield of license flags you can tick on and off, for such things as noncommercial use or source republishing. The CC0 variant is very nearly public domain, useful for countries that do not recognize public domain.
Replies: >>170813 >>170820
>Incidentally, it explicitly allows charging for source code access, so users interested in fixing a bug on their own or making a port could be kiked out of it that way.
Conversely, anyone trying this can get foiled instantly by someone paying the cost and then redistributing the fixes for free. It's why corporate exploitation of free software is always focused on either making the software impossible to use (so you need to pay for support, e.g. all forms of Linux since about 2009) or making the software impossible for goyim to build without proprietary code or being a shabbos goy with access to corporate resources.
Replies: >>170820
So there is no way of not getting buttfucked by jews on this one? damn!
Replies: >>170821
Creative Commons-BY-NC-SA will require all derived works to credit the original, publish source, and be non-commercial.
Replies: >>170825
That doesn't cover patent rights though, so some asshole can bribe enough clerks to get an invalid software patent issued and lock you out of selling your own code that way. Really your choices are WTFPL (when your code has no value to anyone whatsoever and therefore you don't care what some gay area jew does with it) and AGPLv3+NIGGER.
Help! I am retarded!
anyone know how I can create a half life style inventory system? I need to know how to have multiple slots occupy the same keybind and have specific weapons in their respective slots.
Replies: >>170886 >>170904
You can have arrays inside other arrays.

If you complain that this answer is too vague, ask a more specific question. At what point have you gotten stuck? Do you know how to get user input? Do you know how to create inventory items that the player can swap in and out? Do you know the data-structure to store the weapons in? What engine are you using? What language? What have you tried up until this point? How far have you gotten into figuring this out?
Replies: >>170890
OK so lets say we have object A in one inventory slot and we need object B to be in a different slot but using the same keybind to chose it. how exactly does that work? what do i use to cycle through items? I know that if on button pressed gets the first item, but how does it work when I press it again? do i need to write an if else statement or something to say if X is pressed twice then select the second option then if options are out then return to first option?

Also for items in specific slots does it go like if item is X then slot A can only pick up item X whilst slot B can swap between either item Y or Item Z.

I dont need to know in depth how to do it just how to think the way in which to do it, like how you can use HP bars and animation trickery to implement a stun grenade and stun effect into your game
Replies: >>170897
>OK so lets say we have object A in one inventory slot and we need object B to be in a different slot but using the same keybind to chose it. how exactly does that work? what do i use to cycle through items? I know that if on button pressed gets the first item, but how does it work when I press it again?
Store the index of the currently selected item in the selected slot. Every time you push the key associated with that slot, increment the index and get the next item in the slot. When you reach the end, go back to zero. If you push a different key, then the index is reset to 0.

>Also for items in specific slots does it go like if item is X then slot A can only pick up item X whilst slot B can swap between either item Y or Item Z.
I'll be honest, I don't totally understand what you're saying here.
The way half-life does it, each weapon has a specific slot. Not just a column, but a specific row in that column. If I were to guess, I'd say that they have the arrays allocated up-front and all the gun data exists in the player's inventory, there's just a boolean flag to determine if the gun has been "picked up", and will be accessible to the player. In the above example, rather than simply going to the next row it also checks if the gun in the next row has been picked up, and if not it keeps going. Probably with some check to determine of the column is empty before it gets stuck in an infinite loop searching for what isn't there.
Replies: >>170899 >>170901
Thanks for the info, i'll try to see if I can get it implemented one day soon.
Replies: >>170903
Also I mean for the second question to apply to something like L4D or CS where you can only pick up rifles in slot A and pistols in slot B but cannot have 2 rifles or 2 different pistols
Replies: >>170903 >>171482
Happy to help.

Ah, I see. In that case it wouldn't make sense to have every gun already in the player's inventory and turn them "on" as needed. You would swap the guns and all their data in and out of a given slot.
half-life's inventory system sucks, too many weapons and too few tbuttons
inventory in those games is just a pointer array, every copy of a weapon is just a pointer to the same weapon in the data list, without type checking its literally just void *inventory[] if you want weapon type checking then just make a type in the weapon list
struct weapon
struct inventory
	struct weapon *rifle,*pistol,*etc[].
eg. picked up weapon; weapon->type is rifle; if inventory->rifle is empty then set to &weapon
Replies: >>171486 >>171493
[Hide] (1.6MB, 1920x922) Reverse
>meanwhile in unreal
>all of these tutorial videos take 20+ minutes just to tell people what a vector is
>doesn't go over shit like Lerps and breaking pins
>come across documentation on it buried under political hitpieces and chinkspam
>everyone seems to only care about shilling half baked products onto the marketplace

>start on making game
>make weapons and player character programming from scratch in less than a week
>still havent rigged playermodel because tutorials are dogshit redditors  being "buy my book" shills on the topic
>retards on cuckchan insist on using store bought assets 
>inb4 "that's the default mannequin" 

Why do you faggots wonder why games are so shit?
All those retards care about is taking credit for other people's work. Its no wonder the indie scene is full of shovelware garbage with an attitude like that as they drove out anyone who actually made anything worth playing.
>Why do you faggots wonder why games are so shit?
I don't.
Cool, now can someone tell me how to achieve the effect that happens in games such as goldeneye007 or red orchestra osfront where the aiming is off center inside a deadzone?
I did watch a tutorial or two on it one in unreal and one in unity, the unreal one takes forever to get to the point and the unity one says to just copy paste a script into the game which is fucking useless

This, every single time its always some redditor taking credit for doing basic bitch store bought asset flips, you know like those
It grinds my gears to no end, if these fagdevs actually made a game with their own code and assets it wouldnt be an issue.
>make weapons and player character programming from scratch in less than a week
Any tips for where to start in godot for characters and weapons? I really don't want to just copy some code i found then stitch together a flimsy game based on shit i know nothing about
>indie scene is shovelware garbage
maybe if faggots like >us started to make more games then the world would become a better place? I want to start making my own games solely because of all the shit games flooding the market both AAA and indie, even if it is just "X game but with Y"
>Any tips for where to start in godot for characters and weapons?
First you copy, then you dissect, then you expand and improve.
Replies: >>171508
>Cool, now can someone tell me how to achieve the effect that happens in games such as goldeneye007 or red orchestra osfront where the aiming is off center inside a deadzone?
Like autoaim? Simplest method is you use a cone-cast in front of the player and lerp the aim raytrace towards the closest enemy. That is, it's easy if whatever you're using has cone-casting/cone-tracing pre-implemented for you. Pretty sure Unity does, for instance. Also often people use a separate collider for heads for this to prioritize headshots if you happen to aim near there.
Another method is to render enemies IDs to a separate texture, much like a depth filter, and do math on how much of a certain enemy fills the deadzone area. This can also help in determining how much you should correct towards the enemy, so just barely having them within your deadzone aims towards them only a little.

If you meant more like bullet spread, just a random offset to the aiming vector. You can use good ol' SOHCATOA to figure out how much of a max angle you want to get a specific amount of distance at a specific distance. There's ways to go more into autistic realism details about distribution but for most games no one's gonna give a shit.
Replies: >>171512
I don't know as I have never used Godot. Eventually what the tutorials were saying sunk in and I figured out how to do some things. Unreal tutorials are very inconsistent and like most games today, have no respect for the end user's time. They all just shill stealing stuff off the internet instead of anything requiring effort while doomfagging about how all is lost when it isn't.

Store bought and free assets are not bad in of themselves as placeholders for your own assets and in a job you will be using other people's assets from your studio as a given. Those "developers" are just asset flipping. 

An odd phenomena I found is that a ton of older games do use some of the same sounds. Like how the Tie fighter laser cannon sound is used as the firing sound in Megaman ZX for the first weapon you get and also in few other older games. 

Basically this.
I just want to know how to make a continuous laserbeam weapon in Unreal. Whenever I'd ask they would act like I was speaking in tongues and all tutorials just feature AK's as if any creativity whatsoever would be like throwing water on a vampire to the video makers.
Replies: >>171519
no like free aim or just the ability to aim around the screen without the screen actually moving, think of it as your aim is bound to a box and you can move freely within the box without moving the screen and only when you hit the boxs boundary then it moves the screen.

all I want to know how to do is make an inventory system for a 3D game, all these fags use 2D instead or dont explain things properly
[Hide] (484.5KB, 1664x893) Reverse
I don't know how to do weapon animations or vfx in game and blender makes me want to scream with how often it will just randomly freeze on simple meshes.  Have all the different parts of various equipment and player characters made, but there are no tutorials on how to add more than one animation to the skeletal mesh. 

All the tutorials are by fagdevs who just sperg endlessly about their personal life and taking shit off the internet.  

This is a doodle of a laser I made in about half an hour.
Replies: >>171605 >>171607
Nice rifle except for the clashing blue/red colors, just make all the lights the same color to get a better overall effect, also what engine are you using?
Replies: >>171610
The Action Editor is what you're looking for in blender. It's accessible through the Dope Sheet window. Once you add an action (animation) you can click "Push Down" and add a new one. I use this when exporting to gltf.
same anon from >>171486

It might look better if the lights were all one color and I added more detail around the grip. Still using Unreal and still stuck on setting up character animations like the different poses for holding weapons. 

Problem is I don't have any custom skeletal meshes of my own with animations added and the tutorials I've encountered haven't been helpful. Though I did just find some answers upon looking again when it comes to multiple animations added to the same skeletal rig.  Will make some test animations and try to export them to Unreal.
Replies: >>171612
> I don't have any custom skeletal meshes of my own with animations added 
then make some, its not hard it just takes time, quite tedious
>multiple animations on the same rig
you mean like reloading a weapon whilst moving so your legs have an animation playing as well as your arms?
Replies: >>171613 >>171645
I just mean basic stuff like reloading, firing, and ADS. Push down didn't actually work back when I tried it a while ago. It will probably work now though that I have some idea of what I am doing. 

>you mean like reloading a weapon whilst moving so your legs have an animation playing as well as your arms? 

Assumed that is what the animation state machine is for. To blend those together with moving and crouching.
[Hide] (70.2KB, 350x399) Reverse
got more stuff done on my ps4 program loader.
this time only non-visual related stuff was touched: added support for tracking virtual memory allocations, threads and mutexes, added support for offscreen rendering so graphics can be automatically tested, and reorganized some other code.
moving on i'll have to work on loading more modules other than the game's main executable, especially the libc that comes packaged with most (if not all) games, so i won't face to implement libcxx crap like _ZNSt6locale16_GetgloballocaleEv and friends (ps4's libc bundles a standard cplusplus library), which are many. will have to work on TLS support aswell.
pic unrelated
Replies: >>171705
[Hide] (54.2KB, 640x360, 00:02)
[Hide] (36.9KB, 640x360, 00:02)
[Hide] (110.5KB, 640x360, 00:03)
Weapon rigs themselves were fairly easy. Thanks for the encouragement. They don't have half of the tedium that characters do with IK and everything. Still have to finish sculpting one of them and sculpting tools hate me.
Replies: >>171991
Why are you making a PS4 loader when there's nothing to load with it?
Replies: >>172942
Does anyone have the image with the whole "before you make your own game try"?
Replies: >>171933
the one thats just some popular 8bit games?
Replies: >>171948
No i meant what you need or reccomended to do be dire making your own game i think there was a pong clone and a 2d platformer. It was simple and such.
I am having trouble setting up animations in game. 

The weapon will disappear completely when an animation is played but when the bones are selected, the animation is still playing properly. I did fix any other issues and have the animations ported in the project.
any guides on making a game look like a PSX title?
Replies: >>171999
Low poly models + unfiltered/nearest neighbour textures are the first thing you'll need, but you shouldn't need a guide on that. If you can't make good low poly models you should go learn how to model properly, not try gimmicks.
Then to really make it look like PSX, you'll need two specific things: floating point errors on your vertex positions (basically just round their position to a certain level of precision in a shader), and you need affine texture mapping instead of perspective. That can be as easy as putting a keyword in the shader if the rendering framework of your engine/that you're using supports it, but not all of them do. Here's the first result for it in a search engine, it explains how to do it manually too https://danielilett.com/2021-11-06-tut5-21-ps1-affine-textures/

Then you just render the output to a specific size, but the PSX supported various modes from 256×224 (or 256×448 interlaced) up to 640x480 (interlaced).
[Hide] (108KB, 800x800) Reverse
[Hide] (11.2KB, 800x800) Reverse
[Hide] (20.4KB, 800x800) Reverse
[Hide] (20.7KB, 800x800) Reverse
You can play around with the different types of noise, but for this example we'll use Difference Clouds. If you want the texture to tile then this is also the point where you would want to do that.

Choose a number of levels that works for you. Probably something between 3 and 7. I used 5.

Duplicate the layer, rotate, then change the blend mode of the top-most layer to multiply, then merge the layers
Another good option is "Luma darken only"

Pick a camo-like color, like dark green. If there's a specific color pallet you want, then you could use the Color Select tool to apply it one color at a time.

Ta-Da! Different kinds of noise would create different patterns , and if you want a specific camo look, like Rhodesian Camo, it may mean more layers, more steps, and more work.
Replies: >>172946
[Hide] (1.3MB, 1024x768) Reverse
this week i progressed with my ps4 loader/compatibility layer by:
- implementing loading of shared libraries, and using its exports when available
- added half assed flexible memory API support
- wrote pthread-like condition variables and read/write locks related functions, along with some time related functions
- created tests for the above functionality
- removed a feature where you could override any function/object export - this was used by tests and has been replaced with checking the exit status of the test programs

now there's some more functions that a game's libc is asking for, and i'm working on getting them done.
as a side note i should look into cross compatibility between OSes, as there's some code using functions such as mmap directly and i don't think windows exports them.

for fun
Replies: >>174240
That is actually super cool and ingenious, is this how they make real camo?
Replies: >>172966 >>172969
No idea if this is how they make real camo patterns. I just made it up. I'm familiar with procedural generation techniques so I knew generally the sort of the path to take but for the most part I played around until I figured out something that looked halfway decent without being too much work.
>is this how they make real camo?
No, but its close enough, they take the % of foliage within the country then mix and match it with various plants to create a pattern and shapes, or at least thats what they did for jeitai (JGSDF) and a few other patterns
[Hide] (65.8KB, 1200x632) Reverse
Replies: >>173700 >>174237
[Hide] (552.1KB, 1600x873) Reverse
How do anons get to contribute to some other anon's project? How do you coordinate?
Replies: >>173803 >>173804
you don't, everyone here is usually one man army so you cant do shit to help unless someone specifically asks for it.
If you meant like how to get into a group project then your shit outta luck cuz no nigga here is doing any groups projects because they always end in shit, see anton and coolpecker for example
Replies: >>173830
Say what you have to offer and if anyone needs your services they can present their project to you so you can decide if you want to contribute. Further organization can be done via Email, IRC, Matrix, or just talking in the thread. Expect short-term collaborations.
[Hide] (104.1KB, 364x725) Reverse
How does one practice/study 3D modelling ? 
I made this a few hours ago, its peeping eye from Castlevania, but it does not feel like I'm improving, or even learning something, I'm not even sure if this looks good or bad.
I will never not be pissed with how A&C died. They had everything going for them, they weren't getting hung up too badly on assets unlike a lot of group projects, and then it seems like the whole thing exploded from (for lack of a better term) bad group politics. At least 7SU2 was done in by inexperience and RPG Maker being an awful experience for anything that isn't one-man-army.
[Hide] (16.8KB, 400x396) Reverse
>I'm not even sure if this looks good or bad
Are you sure on that? That looks undoubtably good to me.
What most games do to improve looks on models is to properly texture your model with UV maps and such.
I truly believe your model would look very competent if you spent some effort on extracting the UV maps then doing some magic on GIMP.
Replies: >>175346
>how do you improve
simple really
>get reference image
>model it
>see how you can get in enough detail but make it lower poly
>repeat for other reference images
Its all about low poly baby, mostly because games will run like shit if you got higher poly models taking up most of the screen.
Replies: >>173888
you are such a retarded useless piece of shit nigger, his question had nothing to do with "low poly" and even if he was trying model for a game console from 2001 it's already "low poly".  You're also completely wrong on performance, no modern hardware is going to have trouble rendering thousands of those, and it has nothing to do with polygons but draw calls and fill rates. You're such a stupid nigger and it makes me so mad that people like you have taken over this board and made AGDG into this parody of game development. 

yes I HATE it when my star trek critter takes up the entire screen and causes the game to lag because it has over 1000 triangles.
I would if someone would tell me how to fix the fucking  “imported bone transform is not the same as exported bone transform” error in unreal when importing a skeletal mesh. My animations won’t work until I fix it and no Ive already tried applying all transforms and got as far as my mesh appearing in the animations but not doing anything while the error is persistent.
Replies: >>174503
>nogames4 loader

Stop being a useless  nigger and find a way to let us play custom unreal projects on the jailbroken console.
Replies: >>174642
[Hide] (7.9KB, 330x290) Reverse
Still have no idea how I'm supposed to come up with enemy ideas. I thought this would be easy but apparently I'm a terrible ideafag.
[Hide] (384.6KB, 630x420) Reverse
>design/plan thing
>make thing
>after finishing thing, reflect upon what you did right or wrong
>WHY did you do it right or wrong?
If you want criticism or feedback on your model, you should compare it to source material like concept art or bestiary entries.  I think the thing that makes your model difficult to compare to a real enemy, 2D or 3D, is that the eye stem isn't curved like it would actually appear in gameplay.  Can you bend the eye stem on your model?  If not, then that's an issue.
Replies: >>175346
I solved the problem and lmported animations successfully.
[Hide] (5.8KB, 320x133) Reverse
got a couple things done last week on my ps4 compatibility layer - most if not all of it uninteresting:
- added support for calling DT_INIT and DT_FINI functions on shared libraries (executables take care of this by themselves before and after calling main)
- dropped embedded libc functions in favor using the native shared library countepart
- created custom pthread structures for compatibility with any code that accesses any of its private members (libc does this)
- simplified tests code and made them load shared libraries
i'm looking towards working on a GNM (ps4's rendering API) driver, but i want to work with smaller sample programs, like a simple triangle drawing program, instead of more complicated game programs to ease development.

(you) can technically already do this by using any of the console's leaked SDKs, but you're likely to run into issues when trying to distribute any of the resulting products without a contract with sony.
alternatively, there's work being done for an open SDK replacement at https://github.com/OpenOrbis/OpenOrbis-PS4-Toolchain , though it's not complete:
one of its missing features is a reimplementation of the GNM API, a layer between a program and the GNM driver, which would allow homebrew games have accelerated graphics. i may try to implement a very small set of this in order to get one of the simple drawing programs i've desired above.
[Hide] (95.4KB, 293x683) Reverse
[Hide] (94.4KB, 232x647) Reverse
I remade the peeping eye model, I tried a different design this time, more spikes, there is also a loop around the eyelid for animation purposes.
I'm not satisfied with the stinger and tail, they look bad, maybe its the silhouette, there are spikes at the top and bottom of each segment, and the stinger looks like a boot.

Thanks, that is what I'm going to to next, extract the UV maps and paint over it.

It is possible to bend the tail, I just need to add bones to the segments.
Replies: >>175456
[Hide] (10.5KB, 500x500) Reverse
I can tell there is something odd with the way the sadows display on the model.
Can you show us the tail's geometry on edit mode? Try to clean up the model first (remove doubles, deelete loose geometry, etc).
I've read that a practical way to model in 3D in CY+7 is to first create the base of your model using sculpting tools, then use retopology tools to create the mesh on top of the sculpture.
I have never modeled anything complex in 3D and I don't know if what I've typed before applies to older game developing environments, though the method sounds novel enough to me to at least warrant a mention.
Replies: >>175464 >>175469
>I can tell there is something odd with the way the sadows display on the model.
It's a screenshot from a modeling program. It's not going to have real shadows.
Replies: >>175469 >>175474
[Hide] (125.5KB, 415x406) Reverse
[Hide] (173.2KB, 532x507) Reverse
[Hide] (244.3KB, 592x557) Reverse
[Hide] (80.6KB, 341x643) Reverse
[Hide] (77.9KB, 575x549) Reverse
This is my second attempt at generating a texture, just the head for now, The first attempt I tried to create a generic texture, which it would allow me to change the color with the shader, I wanted the shadows to have a dark purple ting rather than pure black, and the highlights to have an orange ting, it worked, but then I tried to export to Godot, and I forgot  Godot cannot convert Blender shaders, I will have to study shaders at some point, I'm avoiding shaders for a while now.
The second attempt, I just painted the texture over the model, I used the painting tools available in blender, this one I can easily export to godot compared to my first attempt.

Now I have question, I know I can create texture maps that allows me to change colors, emission maps, and whatever I want in run time using shaders, but, is it resource intensive compared to extracting textures from models using a 3D suite, or it depends, on the shader, on how many objects using the shader, how optimized the shader is, and the engine ? 

I checked mesh, and orientation of the face were in the wrong direction.

> is to first create the base of your model using sculpting tools, then use retopology tools to create the mesh on top of the sculpture.
Its something that I wanted to try, create a detailed high poly model, and extract a texture map and normal map, there is also a feature in blender called Dynamic Paint, https://docs.blender.org/manual/en/latest/physics/dynamic_paint/introduction.html, this video explains how to use to generate texture maps https://yewtu.be/watch?v=HUdxpSXei2Y
>I checked mesh, and orientation of the face were in the wrong direction.
Oh shit.
I'm retarded.
In that mode, it displays the shadows based on normals. The shadows on the body segments being upside down implied a serious issue, which turned out to be inverted faces. And if he switched to Viewport Shading it'd totally have "real shadows" as well. Why say dumb shit if you don't know what you're talking about?

Looks pretty nice, especially the front half's texture. You might have add specific topology on your remesh so those wringle textures deform in a nice way when the eye opens though.
As for your question on how to study this, you can look at free tutorial shit on jewtube, buy or find courses on some trackers like cgpeers (not that they ever open registration), and in general just practice a lot.
Replies: >>175483 >>175565
I didn't notice the inverted normals on the the stalk and I thought he was  talking about something else. If I'd known someone would have been so pedantic I would have specified that it won't have real shadows when using the Workbench Engine (Solid Viewport Shading mode) to render the view. In hindsight I should have realized he was talking about the shading of the model and not the shadows.
>Why say dumb shit if you don't know what you're talking about?
Because sometimes I make a mistake. Happens to the best of us. For example, in your post you talk about "Viewport Shading" as if it were a specific render mode and not the name of a setting that has four distinct render modes, some of which render shadows and some of which don't. If there's a 3d viewport then there's some sort of viewport shading going on, even if it's in wireframe mode. So switching "to Viewport Shading" does not necessarily mean that there will be real shadows in the same way that switching a texture "to Color" doesn't necessarily mean there will be yellow. I think you meant to say if he switched his Viewport Shading to Rendered or Material Preview he'd have real shadows, right?
[Hide] (131.3KB, 1024x1024) Reverse
[Hide] (1.8MB, 2048x1024) Reverse
[Hide] (3.4MB, 1536x1792) Reverse
[Hide] (1.6MB, 1410x1200) Reverse
[Hide] (2.8MB, 320x180, 01:39)
That looks pretty good, I concur with >>175474.
I don't know much about shaders, and I can't navigate the cycles render without a tutorial at hand, so I can't tell you about how intensive Blender shaders or materials are.
Looking at other games, in case you want to add more detail into the model, since you already have a diffuse map I'd say you took some time into making both a normal map to add depth to the wrinkles of the eye and perhaps a specular map to draw the slimy surface on the eye itself.
As far as I'm aware most games tend to use a combination of diffuse, normal and specular map alongside good lightning (read. carefully placed and designed light sources) to render the graphics on screen. However, you can tell some more primitive games tried to cram most of the shading detail already on the diffuse map. I'm not very knowledgeable at the moment and my experience is next to null, so hopefully some anon can correct me on what I've said.
>Its something that I wanted to try
In case you want to try this out, I've heard Maya is the king of sculpting and retopology tools, so you may want to give it a pirate. This faggot gives you a light tour on what can be done in Maya in terms of retopology: https://yewtu.be/watch?v=xpDWta5O3n8
There's this video that's been posted on plw/animu/ that showcases a simple workflow on an anime head using this technique albeit it's hard to tell since it plays lightning fast and the resolution is ant-sized. Whoever posted webm related also said he knew someone skilled in 3D. Perhaps considering asking for help there too.
Can you send out some invitations?
[Hide] (37.2KB, 1920x1080) Reverse
ast week i worked on getting a bare minimum GNM sample to help progressing my ps4 compatibility layer.
i reverse engineered enough functions to get the GPU to draw a triangle in a real console (see pic rel) - functions to create a framebuffer, a command buffer, write some commands into the command buffer, etc.

while the sample is very simple, it should help with developing the compat layer to:
- allocate GPU memory, which may be visible to the CPU
- parse and process commands from a command buffer
- setup a rendering pipeline for the command buffer (i think)
- convert PSSL shaders to SPIRV
this week i'll try to get the first 3 points implemented
Replies: >>175760 >>175855
[Hide] (244.2KB, 1805x425) Reverse
you're gonna make it
Are enginedevs black triangle devs?
Replies: >>175850
Most of them don't even make it to the black triangle part
Very cool, are you going to use the compat layer for?
[Hide] (274.1KB, 448x797) Reverse
[Hide] (440.9KB, 618x792) Reverse
[Hide] (359.8KB, 822x656) Reverse
[Hide] (131.1KB, 257x644) Reverse
[Hide] (397.9KB, 578x791) Reverse
Finally, I've finished the peeping eye, fully textured, but I'm not going to rig the model, nor I will animate it, maybe another time, next I'm going to try to model the slime, from Aria of Sorrow.
Checked and miring, great work anon! post model pls?
Replies: >>176589 >>176692
Looks cool. Does the eye glow in the dark?
Replies: >>176589
Checked.  Great work.
[Hide] (10.7KB, 464x188) Reverse
rate my spider
It looks like a deformed potato.
Spider legs don't just stick out the bottom.
But spiders have cool hydraulic legs it looks like a diglett got digested
[Hide] (40.1KB, 277x275) Reverse
[Hide] (227.1KB, 690x374) Reverse
[Hide] (411.8KB, 700x700) Reverse
[Hide] (151KB, 510x363) Reverse
[Hide] (336KB, 757x571) Reverse
Before the peeping eye I tried to create the slime from Aria of Sorrow, because I thought it was an easy thing to model and render, but it was not, mesh was easy, the problem was the material, this is my second attempt. I really like the result, especially the dark core, I use the reflection normal.z to create the core, however there a few problems, I think the shader is heavy, the bottom of the  slime does not look good, the bubbles don't look good, and I need to convert the Blender shader to Godot, and I'm not sure if it's possible to create this on GLES3.

Can do.

Yes, it glows in the dark.
Replies: >>176634 >>176692
[Hide] (526.6KB, 1600x1600) Reverse
Spoiler File
(16.4KB, 188x284) Reverse
That's not bad either, but are you sure you can port those Blender materials into a game engine?
By the way, what shaders did you use to render the slime translucent?
In case you know a bit of Blender, can you help me out on what attaching numbers like this to the end of an object's name does?
Replies: >>176692
[Hide] (34.9KB, 418x433) Reverse
[Hide] (21.8KB, 419x294) Reverse
>you sure you can port those Blender materials into a game engine?
No, it is something that I need to research more, but I think it can wait, I have other things that I want to finish before moving to shaders. 

>By the way, what shaders did you use to render the slime translucent?
Principled BSDF, you have to enable screen space reflection on the Render properties, then, you have to configure the material settings on the Material Properties.

>attaching numbers like this to the end of an object's name does?
No idea, maybe something related to scripts, or naming convention. 

[Hide] (393KB, 640x640, 00:10)
I keep getting more fucking ideas for my never-ever fetishistic eroge but I can't draw well enough to make the porn the way I want it to be and I'm too autistic to hire a drawfag. What the fuck do I do?
Replies: >>176873
Make a text-only game
Replies: >>176883
>want to make game about my disgusting fetishes
>can't draw
>also can't program
>don't want to pay someone to do it
what do, agdg?
Replies: >>176883
[Hide] (971.1KB, 1080x1067) Reverse
I actually have this idea for an autistic modular text game written down, but I simply don't feel like working on that right now. Plus I don't have anything written down for the text part of the text game. 
Are you autistic? 
If you're not autistic learning to draw should be relatively easy, and if you're incompetent AI art is a possibility. Programming is harder but with game engines and python it's really easy to shit up a disgusting program that's still "acceptable" by normalfag standards. 
If you're autistic programming is really easy but learning to draw is hell. I have no idea what the fuck to do.
Replies: >>176892 >>176952
>I actually have this idea for an autistic modular text game written down, but I simply don't feel like working on that right now. Plus I don't have anything written down for the text part of the text game. 
Are you me?
When I finish my first game and get gud enough at drawing I'm going to make a cooking mama eroge. The first levels are going to be cooking on mama's naked body, then you're going to breed mama and then you're going to fuck your daughter while mama sucks your balls.
> I have no idea what the fuck to do.
Start with courses.
[Hide] (12.5KB, 1288x752) Reverse
managed to get the PS4 GPU triangle sample working on the PC compat layer.
it's now got very basic parsing of GNM draw command buffers so they can be turned into Vulkan command buffers.
some of the stuff got implemented in a rush - it's using a hand written version of the original PSSL shaders instead of translating them, and the rendering pipeline isn't very flexible.
pic rel is the result - the background is black because the framebuffer is being cleared before draw, unlike under the PS4 where there's no hardware clear.
now i'm looking towards making the rendering pipeline more dynamic, and perhaps caching it too. may also look into translating PSSL shaders to SPIRV.
Replies: >>177081
Good job anon, are you going to put it into a git at some point?
Replies: >>177911
[Hide] (29.5KB, 680x659) Reverse
Will I be disqualified from the epic enginedev club if my game has OpenGL 4.2 as the minimum requirement?
no, even if you made a pacman clone on UE 4 you'd still be ahead of 80% of the people in agdg who just whine and have no gaem
Replies: >>177312
Just, like, make game.
Actually I don't think I even need to do that.
glTexStorage3D is from 4.2, but seems like glTexImage3D can do literally the same thing and it's much older.
Replies: >>177248
Lol no, I suppose there are some bounds but directly using an actual graphics API squarely places you in enginedev territory no matter what version it is.
Replies: >>177276
If you want to put in too much effort, like a real enginedev, you could always add multiple rendering paths depending on the highest supported  OpenGL version detected. And Vulkan, can't half-ass this kind of stuff of course.
Replies: >>177276
It's enginedev but it's more cool if your engine supports very old toasters.

It's hard enough wrapping my head around a single renderer. This feature is specially important because as far as I can tell it lets me select textures in a shader, which changes everything. I'll never need to switch textures or split objects based on what texture they use.
Even my 6 years old laptop with integrated intel haswell graphics (which is a 2013 technology) can do OpenGL 4.6 (which is probably the last version of OpenGL anyway, all the cool kids use vulkan these days). Unless your graphics is very simple (but in that case you wouldn't need 4.2), your game would probably have massive performance problems on HW too old to run GL4.2 anyway.
Useful site: https://opengl.gpuinfo.org/
Also this: >>177226
[Hide] (6KB, 512x89) Reverse
I went through Steam hardware survey, and apparently at least 0.58% of people are stuck on OpenGL 4.0 or lower.

However, there's a big 11% portion of people that don't have their GPU listed, I assume it's a hodgepodge of random old GPU models in very low percentages.
0.58% sounds low, but steam has something like 100 million active users, so 0.58% would still increase your audience by 580,000 players. Those players are the ones most desperate for games because they probably can't play any bloated Unreal Engine asset flip games no matter how low they're priced.
If anyone here is mathlet and is having trouble adding nodes in Godot's GraphEdit relative to the zoom, divide the offset by the zoom, and if you want to add a node to a specific position, you also divide the position by the zoom.
pos = GraphEdit.scroll_offset / GraphEdit.zoom
pos += target_position / GraphEdit.zoom
[Hide] (12KB, 1288x752) Reverse
this week on my ps4 compatibility layer, git log shows the following progress:
- added MD5 and Murmur3-32 implementations - these are used to make hashes out of shader code
- implemented a graphics pipeline cache - it compiles a new pipeline whenever a shader or pipeline state is changed
- created a 'context' graphics object that keeps track of a command buffer, the current graphics pipeline and pipeline state
- made a GPU resource manager - it currently tracks RenderTargets and DrawCommandBuffers
- DrawCommandBuffer execution is now asynchronous
- misc GPU fixes

this new 'context' object should help creating functions that mimic GNM commands 1to1 as possible.
pic rel is the same sample as before, but now its RenderTarget data is uploaded to the GPU before any work, so the background color is now correct.
also I need to read about pipeline barriers, since the code pasted to transition image layouts is showing some issues.

sure: https://gitgud.io/glue_sniffer_420/orbis-loader
please note that this has only been tested under linux (with musl and glibc distros), tests are currently uncompilable and source code for test eboots is missing.
[Hide] (262.6KB, 871x827) Reverse
[Hide] (68.1KB, 854x817) Reverse
[Hide] (76KB, 629x619) Reverse
I was trying to import a facemask I was creating on Maya when suddenly, as I was softening the edges, this bizarre shader appeared.
How in this clear hell am I supposed to fix this? It's not the normals or the doubles. Basically nothing from the cleanup menu solves this.
Is there a way to revert the shader to its original look, either in Blender or somewhere else?
[Hide] (33.4KB, 409x482) Reverse
[Hide] (181.6KB, 803x781) Reverse
Finally solved it.
Properties Window > Object Data > Geometry Data then Clear Custom Split Normals.
The shading is far from perfect but at tleast it's better than what it used to look like.
>fucked up solid view
more like fucked up solid wew
Replies: >>178422
[Hide] (83.5KB, 662x679) Reverse
Finally finished the model. I should get to extracting UV maps and texturing now.
Maya is hard and the tools are strange shut your trap.
What do you mean fucked up? This is aesthetic as hell anon, reminds me of metal mario. I wish I knew how to apply this effect.
Replies: >>178445
[Hide] (74KB, 280x373) Reverse
Jesus Christ what in the god damn
[Hide] (314.6KB, 996x891) Reverse
[Hide] (283.6KB, 1078x936) Reverse
[Hide] (516.1KB, 1255x953) Reverse
[Hide] (337.5KB, 1108x955) Reverse
Apparently my mesh was a shit and Maya had trouble making sense of it in the cleanup option. At the very least it tried to do something.
>this is aesthetic as hell anon
What do you mean, easthetically pleasing? This is what I picture the damned sould of dead children to look like.
Here, I've made a mockup texture for you, seeing that you like it so much.
>reminds me of metal mario
Nigger that's what specular maps are for, reflecting an image over your solid.
You do now want to fuck up your model just because you want to see some pretty grey gradients on it.
[Hide] (114KB, 1024x764) Reverse
[Hide] (744.2KB, 850x1275) Reverse
[Hide] (599.1KB, 1108x955) Reverse
>tfw booba
Thumbnail looks like a rabbit girl, especially around the right side mouth. Buckteeth, small nose, even has the puffed upperlip like she would have whiskers.
[Hide] (208.3KB, 571x542) Reverse
[Hide] (17.3KB, 220x245) Reverse
[Hide] (9.2KB, 217x225) Reverse
>they're still going on about it
Get already over it /v/, it didn't look that bad.
[Hide] (1.6MB, 2619x1966) Reverse
juicy lips
[Hide] (35.4KB, 640x373) Reverse
this week i've been working on a PS4 GNM shader parser and disassembler. so far it's capable of decoding shader bytecode and reading metadata of a very simple sample.
i plan on using this to translate the GNM instructions to SPIRV, perhaps for a SPIRV to GNM translator in the future, or even more insane, a GLSL or HLSL to GNM compiler. i should also publish its code sometime next week.

here's a disassembly example:
$ ./pssldisasm -f ./vex_vv.sb
Read 316 bytes from vex_vv.sb
SB Header:
Version: 0.4 (revision 48939)
Association hash: 0x8fdc17db 0x0
Shader type: Vertex Shader
Code type: ISA
Uses SRT: false
Compiler type: orbis-wave
Code size: 188
Vertex variant: Vertex

GNM Header:
Magic: 0x72646853
Version: 7.2
Shader type: Vertex shader
Header size (DWORDs): 10
Auxiliary data: 0
Target GPU mode: Base and NEO modes
Shader size: 132
Is using SRT: false
Num input usage slots: 0
Embedded constant buffer (DQWORDS): 0
Scratch size per thread (DWORDS): 0
Num input semantics: 0
Num export semantics: 0
GS mode or num input semantics CS: 0
Fetch control: 0

Code length 92 at offset 92
00000000: s_mov_b32 vcc_hi, #0x0000000c // 0.000000 = 12
00000008: v_cmp_eq_u32 vcc, 1, v0
0000000c: v_mov_b32 v1, #0xbecccccd // -0.400000 = -1093874483
00000014: v_cndmask_b32 v2, 0.5, 0, vcc
0000001c: v_cmp_eq_u32 s[0:1], 0, v0
00000024: v_mov_b32 v0, #0x3f19999a // 0.600000 = 1058642330
0000002c: v_cndmask_b32 v3, v1, v0
00000030: v_cndmask_b32 v0, v2, -0.5, s[0:1]
00000038: v_cndmask_b32 v1, v3, v1, s[0:1]
00000040: v_mov_b32 v2, 0
00000044: v_mov_b32 v3, 1.0
00000048: exp pos0, v0, v1, v2, v3 done
00000050: exp param0, v2, v2, v2, v2
00000058: s_endpgm

Code section:
ShaderBinary version: 7
Is PSSL/CG: true
Is source cached: false
Shader type: Vertex Shader (VS)
Shader code length: 92
Num input usage slots: 0
Has Shader Resource Tables (SRT): false
Has SRT used valid info: false
Has Extended Usage Info: false
Association hash: 0x8fdc17db 0x00000000
CRC32: 0x14ebc4a4
Replies: >>179195 >>179400
How do I achieve the pretty gradient without a spectacular map and/or fucking up my model? A lot of old CGI games/films had such gradients and they looked nice.
What's the end goal of this?
Replies: >>179197
Why does vtx. input assembly and vertex shader look the same?
Replies: >>179401
Because whoever draw this doesn't know how a model looks like before applying skeletal animation
[Hide] (1.8MB, 1920x1043) Reverse
[Hide] (1.8MB, 1920x1043) Reverse
[Hide] (1.7MB, 1920x1043) Reverse
I tried to recreate the shader from ArcSys game in Godot, its a poor man's version of the original thing, but I'm happy with the results.
It also gave me a goo idea where I should focus my studies.
Have you watched the GDC video for ArcSys? Maybe it might help you replicate the style: https://invidious.snopyta.org/watch?v=yhGjCzxJV3E (go to around 16:50)
>first thumbnail
I legit thought it shows a pregnant girl until I opened the full size image.
[Hide] (1.5MB, 640x360, 00:35)
got some progress done with the PS4 GNM/GCN shader to SPIRV translator:
- wrote a dynamic SPIRV assembler - like https://github.com/ReinUsesLisp/sirit but in C
- started work on converting GCN instructions to SPIRV
the translator still needs some work before converting instructions, like setting up general purpose and hardware registers in SPIRV.

in a more fun task, i managed to give all dusty rewards in Gravity Rush 2 to the character (online only rewards that are no longer available), see webm related.
there's potential in modding the series' games, but their platforms are likely an obstacle
Replies: >>180774 >>180990
That looks sweet.  Do you have any examples of using it in motion?
Any alternatives for atom?
There's VSCodium if you really want to use your hyperbloated browser as a text editor. But I'd recommend checking out some real editor, like emacs or vi.
Community-driven continuation of Atom. I personally like VIM but if you really want to stick to Atom this is gonna be your best bet I think. Wouldn't be surprised if it's gonna be overrun by trannies soon enough though. It's already got a CoC.
are you the fpps4 developer?
Replies: >>182263
>Atom is officially discontinued
>Any alternatives for atom?
- VSCode (but it contains ((( telemetry ))) If you want VSCode but without propitiatory telemetry code, check out VSCodium)
- Kate or Notepad++ (p.s. Gedit sucks)
- NeoVim (tutorial: open nvim and type :Tutor)
- GNU Emacs (tutorial: open emacs and type C-h t or use the menu-bar: Help --> Emacs Tutorial)
  If you want vi keybindings, use Doom Emacs or Spacemacs (or just install evil-mode)
  There are also some video tutorials:
There is also lightweight config file generator for GNU Emacs:
(But don't use it for 3rd-party GNU Emacs distributions (like Doom Emacs)).
For easy customization, try M-x customize in emacs. For finding packages, use M-x list-package 
"M-x" has Tab completion. And if GNU Emacs is too slow for you, use it in daemon mode (run emacs as server and connect to it using emacsclient) or enable native compilation (nativecomp) of elisp code using libgcc jit.

If choose to use GNU Emacs, I highly recommend you install the GUI version (preferably the Xaw/Athena Widgets version or Motif version if you can. Look for a package called emacs-lucid or emacs-x11.) but using the gvim GUI doesn't make any sense because it doesn't add any functionality to Vim (which is why NeoVim got rid of it). I recommend you try both NeoVim and GNU Emacs and see which one you prefer. But if you want to have editor that's closest to Atom, I recommend VS{Code,Codium}. Look at this page to learn the difference between VSCode/VSCodium: https://vscodium.com/#why
>"Microsoft’s vscode source code is open source (MIT-licensed), but the product available for download (Visual Studio Code) is licensed under this not-FLOSS license and contains telemetry/tracking."
thinking about c or c++ to make game?
minecraft is in java but i dont want it to be sllow
Replies: >>180988 >>180990
You should use C#, it's a lot like Java but less verbose and follows the principles of POOD more rigorously (practical object-oriented design). It has all the features of Microsoft's WIndows platform, and is used in the work place so you can get a job after your game fails to make you more than minimum wage. It essentially invented LINQ and Lambdas by popularizing them in such a well designed language.
You are doing some real interesting work anon, nice.

c or c++ are for enginedev-ing; Between the two I think it's personal choice. Try to find a problem that'll take you two days to solve and do it in both languages to get an idea for which of the two you prefer.
[Hide] (11.4MB, 1920x1080, 00:38)
Is the rigid body 3d physic scuffed for anybody else too in Godot 3.5.1? I can't test it out in Godot 4.0 Beta 9 because there is way too many damn script errors which prevents me from quickly testing it. This sucks because I can actually make prefabs with Blender 3.4 now since the Blender devs finally unfucked the polygon selections in edit mode.
Replies: >>181045
Are you using Godot or Bullet Physics ?
The changelog for 3.5.1 has a regression for Bullet Physics.
>Physics: Hack a hot fix for Bullet's collision margin regression
Replies: >>181046
The physics being used in the video is Godot Physics. I'd would try using Bullet physics but it seems it doesn't like it when I scaled the terrain transform which causes issues where projectiles don't collide with the terrain at all.
[Hide] (144.8KB, 1862x967) Reverse
rate my cave
[Hide] (47.3KB, 800x518) Reverse
Looks like a functional tutorial level, though I couldn't distinguish it from any indie game I've played in the last seven or eight years.
Looks like a cave.
[Hide] (189.1KB, 1862x967) Reverse
My first thought is can I get up there without any movement items in rabi-ribi. Probably no, reverse walljump + wallkick only goes 6 blocks high, and I think this is higher, but it's not standard rabi-ribi blocks high so dunno. Maybe if you have a fairy enemy there, item hover + damage boost + lots of luck, but that's getting into the really masochist category. Unless of course you have to go from right to left, because that's trivial and why the fuck do you waste my time with trivial shit like this.
Replies: >>182090
Yes, it's a basic tutorial level that teaches you ledgegrabbing and walljumping.
>Where'd everyone go? Did you all move to 8chan's /agdg/?
8chan /agdg/ is not much anymore active than this thread.
[Hide] (1.2MB, 800x600, 00:25)
[Hide] (786.4KB, 800x600, 00:16)
[Hide] (556.3KB, 800x600, 00:12)
>Finally have an excuse to add new mechanics
Today was a good day.

If Fugue or her ghosts get outside of their "chain range", the non-moving members will be forcibly moved to within range. Fugue will be damaged for every tile either her or a ghost are forced to move.

Next, I think I'm going to create an ability that will let Berserker throw Fugue.
Replies: >>182123 >>182317
post art of fugue
Replies: >>182317
I made a game but it was fucking garbage for various reasons, so I never posted it (and still won't).
[Hide] (58.1KB, 474x553) Reverse
really it seemed to me that this place had more actual discussion/progress...
I've started yet another project instead of working on any of my existing projects
This time a spite- er "fan" project.
Replies: >>182268 >>182293
[Hide] (14.7KB, 1442x105) Reverse
Why'd you delete your own post?  Seemed fine enough to me.
[Hide] (6.8KB, 1280x720) Reverse
finally got my PSSL to SPIR-V shader translator to work with two very, very basic shader samples. pic related
besides getting the translator parts working, i had to add code in the GCN disassembler to calculate the number of DWORDs used in a register per opcode, along with checking if a GPR operand is a scalar or vector correctly, along with other small bug fixes.
the next step is getting this integrated with the PS4 compatibility layer so shaders are compiled on demand.
here's the tool's source code if anyone is interested: https://gitgud.io/glue_sniffer_420/freepssl

have some disassembly of a translated sample too:
; Version: 1.6
; Generator: Khronos; 0
; Bound: 47
; Schema: 0
               OpCapability Shader
         %38 = OpExtInstImport "GLSL.std.450"
               OpMemoryModel Logical GLSL450
               OpEntryPoint Fragment %main "main" %sgpr %vgpr %vcc_lo %vcc_hi %vcc_z %mrt_color0
               OpExecutionMode %main OriginUpperLeft

               ; Debug Information
               OpName %sgpr "sgpr"  ; id %5
               OpName %vgpr "vgpr"  ; id %9
               OpName %vcc_lo "vcc_lo"  ; id %13
               OpName %vcc_hi "vcc_hi"  ; id %14
               OpName %vcc_z "vcc_z"  ; id %15
               OpName %exec_lo "exec_lo"  ; id %16
               OpName %exec_hi "exec_hi"  ; id %17
               OpName %exec_z "exec_z"  ; id %18
               OpName %mrt_color0 "mrt_color0"  ; id %33
               OpName %main "main"  ; id %21

               ; Annotations
               OpDecorate %mrt_color0 Location 0

               ; Types, variables and constants
       %uint = OpTypeInt 32 0
     %uint_1 = OpConstant %uint 1
%_arr_uint_uint_1 = OpTypeArray %uint %uint_1
%_ptr_Private__arr_uint_uint_1 = OpTypePointer Private %_arr_uint_uint_1
     %uint_2 = OpConstant %uint 2
%_arr_uint_uint_2 = OpTypeArray %uint %uint_2
%_ptr_Private__arr_uint_uint_2 = OpTypePointer Private %_arr_uint_uint_2
%_ptr_Private_uint = OpTypePointer Private %uint
       %bool = OpTypeBool
%_ptr_Private_bool = OpTypePointer Private %bool
       %void = OpTypeVoid
         %20 = OpTypeFunction %void
     %uint_5 = OpConstant %uint 5
%uint_993211596 = OpConstant %uint 993211596
     %uint_0 = OpConstant %uint 0
%uint_1006648115 = OpConstant %uint 1006648115
   %uint_1_0 = OpConstant %uint 1
      %float = OpTypeFloat 32
    %v4float = OpTypeVector %float 4
%_ptr_Output_v4float = OpTypePointer Output %v4float
    %v2float = OpTypeVector %float 2
       %sgpr = OpVariable %_ptr_Private__arr_uint_uint_1 Private
       %vgpr = OpVariable %_ptr_Private__arr_uint_uint_2 Private
     %vcc_lo = OpVariable %_ptr_Private_uint Private
     %vcc_hi = OpVariable %_ptr_Private_uint Private
      %vcc_z = OpVariable %_ptr_Private_bool Private
    %exec_lo = OpVariable %_ptr_Private_uint Private
    %exec_hi = OpVariable %_ptr_Private_uint Private
     %exec_z = OpVariable %_ptr_Private_bool Private
 %mrt_color0 = OpVariable %_ptr_Output_v4float Output

               ; Function main
       %main = OpFunction %void None %20
         %22 = OpLabel
               OpStore %vcc_hi %uint_5
         %26 = OpAccessChain %_ptr_Private_uint %vgpr %uint_0
               OpStore %26 %uint_993211596
         %29 = OpAccessChain %_ptr_Private_uint %vgpr %uint_1_0
               OpStore %29 %uint_1006648115
         %35 = OpAccessChain %_ptr_Private_uint %vgpr %uint_0
         %36 = OpLoad %uint %35
         %37 = OpExtInst %v2float %38 UnpackHalf2x16 %36
         %39 = OpCompositeExtract %float %37 0
         %40 = OpCompositeExtract %float %37 1
         %41 = OpAccessChain %_ptr_Private_uint %vgpr %uint_1_0
         %42 = OpLoad %uint %41
         %43 = OpExtInst %v2float %38 UnpackHalf2x16 %42
         %44 = OpCompositeExtract %float %43 0
         %45 = OpCompositeExtract %float %43 1
         %46 = OpCompositeConstruct %v4float %39 %40 %44 %45
               OpStore %mrt_color0 %46

i can't read pascal :^)
[Hide] (212.9KB, 1779x950) Reverse
I'm working on a tool suite to speed up game dev, I developed an api that I allows me to create and modify state machines, characters and other things and store the data in a json file, similar to how CK and other games work. The api works, but, after a while I lost track of the files, so I decide to create a tool to manage it, however, its not perfect, good example is the state machine, uses the default graph edit tree, and its a horrible node to work with. 
I'm a few hours away from finishing it, its great.
cute blob
[Hide] (917.7KB, 1920x628, 00:22)
[Hide] (39.1KB, 284x227) Reverse
At ten million youtube views, this video may be the most widely-seen thing made by an 8chan /agdg/ anon. Not Red Sky, not Speebot, not Beezlebox, Aethyr, or Dorf RTS. This.
fuck videogames they're degenerate
please draw her pregnant
Replies: >>182303
[Hide] (916.3KB, 612x2067) Reverse
[Hide] (555.1KB, 400x300, 00:45)
Spoiler File
(1.5KB, 200x300) Reverse
Berserker can now throw Fugue. Working on the mechanics for gave me the idea for a cohesive mechanic across all three characters. In short, some abilities can only be used if certain members of the trio are within a certain distance. Right now I'm going with they need to be adjacent to the user with passive abilities planned to increase this range.

>post art of fugue
Spoiler related, anon.
Instead of having Fugue get damaged when the chain moves it would be more fun to have her and the ghosts pull each other as they move. The result would add a little more depth to the game by allowing her to circumvent limited movement points, and also have strong characters like Teagan be able to throw them to. Might be cool.
Replies: >>182324
[Hide] (822.1KB, 400x300, 01:08)
[Hide] (121.2KB, 400x300, 00:11)
Forgot to mention, Berserker now as a pull ability which will taunt enemies he pulls.
Also, if it is impossible for a ghost to get into range, Fugue will immediately die, which will cause the ghosts to immediately die as well.
Of note, Fugue does not have a death sprite, so Alto's is used as a default death sprite.

>Fugue get damaged when the chain moves
I could have an ability maybe at level 8 of the Dysfunction path that negates damage dealt to Fugue from pulling/being pulled by the chain. That way the player can throw/push/pull to their heart's content without killing Fugue. I think I will keep the automatic death if it is impossible to get back into range, though.

Actually, thinking on it a bit more, pushing and pulling units is pretty fun, so I'm going to remove damage from Teagan's push if the target is an ally. That way the player will be able to push allies around without damaging them.
Thank you for the idea, anon.
People are addicted to memes instead of enjoying quality kino content.
I gotta say I heckled at "megachonker" it was missing the downpitched voice and the echo effect to make it like an unreal tournament kill streak
Any good tutorials on VR out there?
Replies: >>182720
>Any good tutorials on VR out there?
It should be easy enough to google "[Engine of Choice] VR Tutorial" and just go on from there. If you haven't chosen an engine yet, try looking up tutorials for different engines and see which one has a workflow you would be more comfortable with.

If you want to make your own engine, you can still probably find tutorials with your language of choice, though I would ask you to keep in mind you will most likely never make game if you are going to program an engine first.

One final update, I've added a bunch of stuff that doesn't warrant a video. This appears to be all the programming that needs to be done and after this it's back to "data entry."
Coward now pulls out a bow before firing an arrow, exactly like the pirates do.
The Path menu can now only show nodes up to your level, meaning a character needs to be level 10 before all nodes are visible.
Path menu now plays sounds when adding or removing a point from a Path.
Fugue has been given a downed sprite.
The AI has been updated so it is now aware of the element of damage it deals on a per-frame-per-tile basis, meaning if an attack deals multiple damage types on a variety of tiles, it will be fully aware of which tiles will take what damage. This will prevent things like Appassionato inadvertently healing one of the ghosts with darkness damage.
Hitting the action button on the victory screen will now skip to the end of counting xp instead of immediately closing the victory screen.
Victory screen now shows "Level Up" text when a character levels up and plays a sound.

Some rewrites that I got done so far:
The torture scene is now cut short. As soon as Fortissimo is handed the knife, Teagan will state that she is done listening and the screen will fade back in on Teagan and Alto walking away, performing the same conversation they previously did after listening to the whole scene.
The "Outside" scene with Joseph and Arcturus is now different. I didn't like the original version because it felt too much like Arcturus was written as a strawman, and I didn't like that. Now after Alto interrupts the two, Arcturus offers an olive branch, which Joseph refuses. This prompts Arcturus to challenge Joseph to a duel, which then leads to the same outcome.
Additionally, something I completely forgot is that Teagan outright tells Joseph that he was supposed to die, but Joseph never actually brings this up to Alto. As such I want Joseph to ask Alto about it, but I have no idea where to put the conversation. I'm thinking of putting it in the scene where Joseph and Alto are heading back from Polyphony to Minuet Forest.
Replies: >>183063
What's the best way to implement a Metroid Prime-like scan system? The idea is that you can individual enemies/landmarks/computers/other points of interest and view specific stats, background, strengths, and weaknesses on them. Should I keep each scan entry with the entity itself (in Godot)? Or should there be a database for each type of scan (one for enemies, one for landmarks, etc.) to keep things simplified?
Replies: >>183055 >>183058
I'd say keep it in a database, because in the long run that will be easier to maintain. If you ever want to switch languages you can always just switch out the database for a German or Japanese version.
Replies: >>183081
IIRC in Prime scans are basically point entities containing a pointer to the actual scan data. That seems like a reasonable way to do things. If you were targeting old hardware instead of using Godot it might make sense to store minor scans (compared to Prime these would be the ones that don't add to your logbook) with the entity instead of forcing another trip to memory.
Replies: >>183067 >>183081
>throw ability
>can potentially throw fugue to someplace where the chain damage will kill her
>teamkilling has no consequence as long as the mission ends in the same turn
Excellent, dood, now you just need to add an ability that makes Fugue explode when she dies.

Aren't you going to make the segments of Fugue's chains snap to grid squares and restrict the turn angle between squares to multiples of 90 degrees?
>If you were targeting old hardware instead of using Godot it might make sense to store minor scans (compared to Prime these would be the ones that don't add to your logbook) with the entity instead of forcing another trip to memory.
Even for a computer from the 90's, one trip to memory every once in a while for some text doesn't really matter all that much.
For example, let's say we have a scan that is 1 kb in size, which is about 200 words. If you assume a computer with a bus speed of 10 µs and a bus width of 64 bits (which is comparable to the original Pentium) it would take 128 clock cycles, 1,280 µs, or 1.28 milliseconds to transfer all the data from memory. At 60 fps, a frame is about 16.67 milliseconds. If you were doing this every frame, or multiple times a frame, then there's a significant problem. But these scans are accessed once every couple of seconds at the most. Hide the memory access with a few frames of animation and call it a day. Or acknowledge that if you're targeting a P5 Pentium, you have bigger optimization problems ahead of you.

Fiddling with cache coherency is better saved for bigger amounts of data, or data accessed more frequently.
Replies: >>183068
[Hide] (2MB, 370x319) Reverse
that's a lotta math
Replies: >>183070
Knowing me I still probably messed up at some point.
[Hide] (577KB, 624x352, 00:13)
A few other things I'd like to keep track of:
>locking onto an entity while it is being scanned
Right now the player's scan target is the entity they are closest to. If they are scanning something but another entity moves closer to it, how would I lock/unlock the array keeping track of scannable entities safely so that something doesn't go wrong?
>scan progress
This should probably be linked with foreign keys into the main enemy database, so for example, enemy ID X currently is 45% scanned, enemy ID Y is 100% scanned, and enemy ID Z is 0% scanned. Is there a way I could bundle all this up into a single savefile, including current player stats, items, and completion percentages?
Quickly realizing that my professor passing me in database classes out of pity may be catching up with me. But at least it's not Oracle. How easy is it to corrupt a database anyways?
Replies: >>183084 >>183085
>How easy is it to corrupt a database anyways?
Depends on your level of access lol.

>locking onto an entity while it is being scanned
It sounds like you want two things.
1. Scan the thing they are looking at.
2. Direct the player's aim toward that item.
There are a number of ways to determine what the player is looking at.
The most straightforward when using a game engine with an already-written physics engine for it to perform a ray-cast from the center of the screen out to some maximum scan distance, and then get the result. All scanable objects will be on a physics layer that only belongs to scanable objects in order to optimize things a bit.
Another option is to find all scanable objects around the player camera by some means (AABB, physics collider sphere centered around the player camera, going through the list of all available points and doing sqrt(a^2+b^2) ). Once you have all nearby points, you perform a dot product between the normalized versions of a vector coming straight out of the player's camera, and a vector going from the camera to each point. The dot product between two normalized vectors can essentially tell you how far apart these two vectors are pointing. 1 means they're the same, 0 means their perpendicular, and -1 means they're pointing in the exact opposite direction. For determining what point the player is looking at, you simply perform a cutoff that prevents the player from scanning objects too far away from the center of the screen, then pick the point that returns a value closest to 1.

To direct the players aim toward that point, you can use a bit of control theory. Control theory can get pretty complicated ("the missile knows where it is because it knows where it isn't"), but in short you get the "error", the difference between where you are and where you want to be, and you apply some change to where you are to get closer to where you want to be. The simplest would be to make where you are equal to where you want to be, snapping to the point, but that could feel pretty bad. It would be better to apply a change in state depending on the size of the error. A bigger error results in a bigger change. This would feel more like a gradual force pulling the player to the point, rather than the player simply snapping to the point.
[Hide] (141.3KB, 1340x1032) Reverse
>scan progress
You're right that paying attention in your databases class would make this trivial. A database has rows and columns. Each row is an entry, and each column is a values that entry stores. It's common for a database to have an index or other unique ID associated with each entry. This ID stays the same, while all other values can change. Hence why you could, theoretically, switch an English and French version of the database and have it work perfectly. I'm not going to worry too much about how to get a database up and running, since I'm sure there are already tools and tutorials out that for whatever you're working in.

For your purposes, each entry would include the ID of the scan, the text of the scan, and the percentage completed. In the game data you don't save the percentage (unless some things start at a percentage other than 0%) and when saving, you would only need to save the ID and the percentage since the text remains the same.

This may be overkill for your purposes, but if you want different text depending on the percentage, then you would have to include that in some way in either the entry or in additional rows. For example, you could include some metadata in the scan text. For example, let's say text between tags like <50></50> display when the scan level is at or above 50%, and <-50></-50> displays text when the scan level is below 50. If we take the following plain text entry in the database.

>The green bug is an interesting insect.<-25> You still don't know much about it.</-25><25> It spits a deadly poison that will confuse and confound those hit by it.</25><100> It's weak to fire and acid.</100>

At 1% to 24%, it will display
>The green bug is an interesting insect. You still don't know much about it.
At 25% to 99% it will display
>The green bug is an interesting insect. It spits a deadly poison that will confuse and confound those hit by it.
And at 100% it will display
>The green bug is an interesting insect. It spits a deadly poison that will confuse and confound those hit by it. It's weak to fire and acid.

Though don't do any of this meta data stuff until you have the rest of the database working, and only if you need it.
Replies: >>183213
[Hide] (12.4KB, 255x235) Reverse
Anyone here like me? Sometimes i would sob uncontrollably because i couldn't bring myself to work on my games due to laziness? Like I know that i should sit on my desk coding away, but instead i just keep lying on my bed crying my eyes out. Its so frustrating bros.
Sounds like you need some depression meds and/or a good beating.
[Hide] (354.3KB, 904x1080) Reverse
Replies: >>183139
Recommend me some good pills.
Im not kidding.
Replies: >>183141 >>183170
enlightened neet pill
>depression meds
[Hide] (300KB, 1140x641) Reverse
>depression meds
>Eastern Europe
>Recommend me some good pills.
Go to a shrink, retard, and not the pill-giving kind. Your problems are likely behavioral and can be fixed with some guidance. if that doesn't work then you might need to go to an actual pill-giving shrink
[Hide] (467.6KB, 1038x725) Reverse
Just need 2 more buildingz for level 1.
Any ideas?
Dr. Pavel's General Practice
Big Guy Bane's Flight Instructors
2 Dollars for a box of milk? What a rip off!
Replies: >>183177 >>183179
it's special milk though.
Jill's Sandwiches.
Replies: >>183276
that's a bargain in today's obliterated wreckage of an economy thanks obama
The scan percentage is throwing me here, since that value is going to update very quickly (float updating several times per tick as long as the player is actively scanning something), and I don't believe databases are meant to have a constant write of something like that. In that case, should I store the current scan percentage value of something as a local variable, and then when the player stops scanning immediately write that value to the corresponding entry in the DB?
Replies: >>183216
Correct, though it doesn't even have to be immediately after they stop scanning. You could keep the value in RAM and only write it to disk when the game is saved.
[Hide] (423.9KB, 705x651) Reverse
Starting to seem more and more like the DBs should just be read only, and I should keep track of what has been scanned/collected/read in a simple dictionary (one for each type) in the player's save file. It will be a really big dictionary for each type but probably more efficient that constant writes to a database, especially when you bring multiple save files into play.
Replies: >>183231
[Hide] (330KB, 1268x1341) Reverse
You are correct. The stuff you don't save and the stuff you save should be in separate files. I realize, reading back my initial description of the database, I wasn't exactly clear about this.

I will say that working on player saving too early is usually a waste of time, since you will inevitably have to re-write parts of it as your game progresses and changes. However, if you don't have the experience to know what to look for then you do probably want to add saving early on just so you don't code yourself into a corner. I'd say start out writing some saving code, and once you feel you have a good hang of it, hold off on saving everything until you've finished the interesting parts of the game.
[Hide] (494.1KB, 1086x777) Reverse
[Hide] (232.2KB, 1280x1024) Reverse
Replies: >>183304 >>183319
[Hide] (200.7KB, 476x268, 00:06)
>want to talk in agdg matrix room
<matrix forgot to renew their cert today
of course
Replies: >>183360
Good, your little circlejerk faux-IRC prevents you from doing actual game dev work. Plus it makes you look like a complete faggot.
Replies: >>183363
That's a bit harsh.
[Hide] (2.1MB, 320x240, 00:35)
One of these days, I'll have the Zen discipline to not be so easily distracted.
Today however, is not that day.
Replies: >>183819
[Hide] (48.6KB, 278x520) Reverse
are you feeling too challenged by coding, or not challenged enough?
or are you like me, someone who seeks a constant stream of dopamine to alleviate boredom, only working on the project out of a sense of dutycrunch.
[Hide] (242.1KB, 1322x1009) Reverse
Is there anyone who has used uniform buffers in OpenGL? I can't get them to work and I don't know why. I've looked everything over 100 times and read several tutorials and re-read my old code where this shit worked, but for some reason it's just not working now.

It seems like the uniform buffer isn't connected to the shader or something? No matter what I upload into it, the values always seem to be 0.
Also if anyone has ideas what I could even try in order to find the problem, that would help too, because I have no idea what to do now except give up.
Replies: >>183537 >>183859
[Hide] (96KB, 780x780) Reverse
you could try using renderdoc to check the if the correct buffer is used and it's contents, and to debug the values used by the shader
Replies: >>183554
It's downloading at like 10kbps, I guess I'll try it tomorrow.

I'm looking at my other program where uniform buffers work, but no matter how I look at it it's the same fucking code. It's as if some unrelated thing in-between everything else is making it work.
[Hide] (95.6KB, 1024x1024) Reverse
got the shader translator working with my PS4 compatibility layer - it translates and compiles a new shader, then stores it in a hash map so it's not compiled every command submission.
also fixed its tests and added offscreen rendering when using the GPU system. i need to review the GPU and the video system soon since they feel awkward to work with.
next week I'd like to implement mapping the vertex shader output to the fragment shader input, which will also need to be implemented in my GNM library so i have a sample program to work with
Replies: >>183679
Good stuff as always. Your Makefile will be most helpful to me to be honest since I find make to be borderline incomprehensible
I can't check right now due to gitgud's maintanence but are you using openGL or Vulkan?
Replies: >>183682
[Hide] (29.8KB, 377x200) Reverse
these are not pointers, you are not writing anything to the layout by writing to those variable
[Hide] (65.6KB, 366x401) Reverse
i'm using Vulkan, i think it would be painful to use openGL with these type of projects.
>Your Makefile will be most helpful to me to be honest since I find make to be borderline incomprehensible
they're based of https://github.com/rofl0r/modern-autotools without autoconf since i don't have a use for it by now.
also >gitgud is down AGAIN
Replies: >>183771
I see, thanks for the resource, I will give that a read and hopefully finally get a grip on make. Why is openGL painful for this? Is it just that Vulkan allows you to construct/translate the calls more easily?
Replies: >>183782
[Hide] (23.8KB, 400x300) Reverse
i assume OpenGL resources are harder to setup and manage due to the API having a global state and its implicity.
maybe it's not that bad and this is just my Vulkan bias/better understanding speaking
[Hide] (1.2MB, 512x384, 00:16)
[Hide] (735.5KB, 512x384, 00:22)
i fucking love adjusting hitboxes
At least this is more understandable to me than the psuedo 3d i'll have to implement
Replies: >>183864
[Hide] (37.1KB, 700x430) Reverse
One thing that I see here is that in your draw call, you don't actually bind your uniform block to index 0. Similar to how you need to bind a texture to GL_TEXTURE0 before draw, you also need to bind your uniform block to an index. (Which in your case is zero, as you've set at line 16-17).
glBindBuffer doesn't bind an index, you need to use glBindBufferBase instead. Though if you don't use any other uniform blocks in your code, its state from line 18 should be preserved (but obviously you want to use more than one uniform block in your code, so you should add a proper bind call either way).

tl;dr at line 43, replace with "glBindBufferBase(GL_UNIFORM_BUFFER, 0, ubby);"

pic is from https://learnopengl.com/Advanced-OpenGL/Advanced-GLSL
Is vattletoads getting some progress?
I can't access the slack anymore.
Replies: >>183868 >>183898
>making progress on anything other than cardiovascular disease

All they needed to do was remake Super Battletoads (the arcade game) and actually finish it but they went full ideafag instead.
Replies: >>183899 >>184035
[Hide] (1.1MB, 720x420, 00:19)
Dead in the water as far as I'm aware. Only reason I'm using their TOTALLY COOL ORIGINAL CHARACTER WHO DEFINITELY FITS THE AESTHETICS, STYLE, AND TONE of the BattleToads series is purely out of spite.
I honestly wish I was making this in Godot such that there's easier access for other anons to fuck around in it, and for more connective tissue to /v/attletoads, but that doesn't seem feasible for the current scope of this project's deadline (demo day).
[Hide] (104.7KB, 1134x383) Reverse
[Hide] (38.7KB, 253x216) Reverse
I think real issue was lack of competent cohesion.
Took months for someone to get a prototype in godot engine, months of discussion, etc.
Making art assets and game documents is relatively easy compared to plugging them into an engine, especially if multiple peeps should be working on it.
[Hide] (12.7MB, 1066x918, 109:44:03)
wasted my time trying to make an engine, turns out you have zero access to hardware on a modern os without hacking or root, you have to use their shitty driver either from the kernel or wm even on linux 
still got it working on a terminal though, switched whats supposed to be pixels with characters and the real fb with a character buffer that gets printed, surprised it worked
no wonder no one makes games on linux fucking aids, you cant even get keyboard input ( as scancodes not fucking characters ) without root
Replies: >>183921 >>183922
Don't reinvent the wheel, build your engine on top of a media library like SDL or SFML.
>webm length 109:44:03
What the fuck?
Replies: >>183997
Dude you're supposed to use driver APIs. Mesa/libvulkan/SDL for graphics, libalsa/Cubeb/SDL for sound, SDL/libusb/whatever the fuck you do to access the evdev API for input. I believe Allegro can do most of this for you.
Replies: >>183997
i know the point was to not use any libraries otherwise i wouldnt have bothered, i only need to access the shared memory of the devices nothing else, i thought since linux has fucking /dev  you could just map those devices to memory but theyre all fucking fake its not pointing to the actual hardware memory just some pseudo shit the kernel sets up to trick you, i can use ioctl calls but i have to kill all the drivers so thats not a solution
of course theres always voodoo mode on x86 to just bypass the kernel and use the bios but this is ridiculous, i literally just want a memory address or the port number on the memory bus i dont need an entire library just for basic i/o
Replies: >>184008 >>184010
The problem with your approach is that it won't work if there's more than 1 application trying to access that device. It also won't work, unless you plan to implement the driver for all past, present, and future intel and amd and nvidia cards out there in your app. The days when drivers were backward compatible on a HW level (sound blaster compatible sound cards and the likes) are over.
If you want low level tinkering like that, run your program in dosbox or find some pre-6th gen console (I think they started to weed out direct HW access from consoles around that time) and play with that. (Yes, what you want is not possible even on consoles which are designed for games and where you don't have to worry about different HW.)
>use the bios
Even my 10 old PC doesn't have that (I mean, I could enable CSM in the UEFI settings, and fuck up VGA passthrough in the process and bring in a lot of useless baggage, but why?).
>an entire library just for basic i/o
I wouldn't call a shader compiler, memory manager, a shit that handles render pipeline, etc. as "basic i/o". Heck, many devices don't even work until you upload a firmware.
[Hide] (13.1KB, 225x225) Reverse
Oh, I see. You're a lying nigger who can't even fizzbuzz. Otherwise you would know to program for GC/Wii.
[Hide] (224.9KB, 530x480) Reverse
I fucking hate wasting time on art asset R&D when I don't have a functional game yet, but adding code feels like playing with a towering pillar of cards or something.
Replies: >>184016
[Hide] (1.7MB, 512x384, 00:14)
Trying to have the camera rotate when the character moves up or down, but make it appear as though the character remains on the same Y axis from the POV of the camera.
Replies: >>184021 >>184035
[Hide] (2.1MB, 512x384, 00:32)
Not what I wanted, but I guess unstable washing machine seizures are cool too,.
Replies: >>184035
[Hide] (21KB, 500x500) Reverse
What are you attempting to do here, anon?
What's with the spinning circles? Are you trying to do some parallax scrolling here? Or you simply want some spinning swastikas in the background?
>remake Super Battletoads
They didn't even need to do that, a ROM hack of the original game of the original game would have sufficed.
The NES game had a lot more variety sprinkled in and both the platforming sections and the racing sections, while kind of bullshit, are very fun to master and go back to.
Compare that to Battletoads Arcade, where most of the joy comes from the fantastic spritework and special effects. A NES Battletoads ROM hack or inspired game would have been much easier to get assets for.
Replies: >>184039 >>184043
[Hide] (75KB, 449x663) Reverse
something like image related, where the "real positions" of the characters are translated to fighty game camera like positions.
Sorta like the camera in a 3d fighter, if that makes sense.
Replies: >>184048
Battletoads 1 probably isn't a good base for a hack as it's a buggy mess on the inside, like pretty much every Rare NES game. Just look up any speedrun of an NES Battletoads game. I specifically called out Super Battletoads because it's a basic beat-em-up that even an aggydag could crank out an engine for relatively quickly using Godot as a base, blatantly unfinished, but has plenty of unused assets in the ROM to work from and preview screenshots to act as a guide for recreating the rest: https://tcrf.net/Battletoads_(Arcade)
Replies: >>184048
Are GLSL shaders a security hazard? Like if I let people download player-made shaders, can someone upload a malicious shader that fucks people over? Crashing the game is fine.
Replies: >>184049
[Hide] (110.3KB, 243x271) Reverse
>the characters are translated to fighty game camera like positions
Do you mean how the arena moves in Tekken when you sidestep? If you're planning adding that to a beat-em-up I do not approve.
>Battletoads 1 probably isn't a good base for a hack
Yes, you're right. Even on a normal playthrough you can catch up to the terrain as it's generating.
Still, the game's a solid base, not necessarily for a ROM hack but a remake from the ground-up. The gravity's well programmed for a NES game and the movement is addicting.
It's a shame Battletoads vs Double Dragon on the NES ditched the platforming when it could have added to the original.
Replies: >>184118
In theory, it should be safe, shaders don't have access to anything you don't explicitly give access to. In practice, however, graphics drivers are a buggy mess, and they don't really care about security issues, so there might be known security issues. You should look into what browsers to do to validate GLSL shaders, but that might be overkill...
[Hide] (1.4MB, 512x384, 00:22)
Christ, I hate how close I can taste the desired result with retarded Bruteforcing, yet how far it still is. I wish I had actually paid attention to calculus/trig. For all the normalfags that go "Yeah, like I'll ever use this after I graduate," calc and trig are actually pretty fucking important in this field for more "advanced" shit.
Replies: >>184240
[Hide] (45KB, 290x390) Reverse
[Hide] (682.3KB, 640x480, 00:15)
[Hide] (20.4KB, 212x159) Reverse
i'm going to scream, i should've seen it 3 days ago as soon as I saw these 2 fucking values.
Fucking 3 days of frustration, and pointless implementation of a pointless goofball feature for nothing.
oh well, at least the frustration lead to the creation of more art assets or whatever.
[Hide] (12.9KB, 1288x752) Reverse
[Hide] (37.8KB, 1920x1080) Reverse
managed to map vertex shader outputs to pixel shader inputs in my PS4 compatibility layer this week - interpolation's working too, though the wrong colors are being output.
it needs some fixes, like using the same variable type for VS outputs and PS inputs. i should also look into fixing whatever the Vulkan validation layer keeps complaining about.
later on I'd like to get started on setting up resources like buffers and textures to shaders
Replies: >>184341
You're doing God's work, anon.
Someone design my asset loader for me.
Replies: >>184628
The scheme I used is:
>use any common file format (ini, json, xml, whatever) for a "list" file
>file just contains a list of assets to load, and what index value they bind to
>upon game startup, load&parse file, load assets listed and story them into array at given index, if index is already used, override previous asset (hint: this is your future modding support)
>create a header file somewhere which contains definitions that resolve into indexes, so that in your code you use these definitions instead of just a magic number (improves code readability, makes it easy to tell where a file is used since you can ctrl+f your repository for the definition)
Replies: >>184643
My main issue is with updating in-game data when files are modified. Some files need to be processed first, for example bitmap fonts need to find the character boundaries from the image.

I need a system that can update my font data structure when the image it used is updated, so I need some way to link the image file into a font resource or detect what resources used the image or something. There's more complicated resources but fonts are the simplest example.
Replies: >>184725
[Hide] (249.7KB, 400x300, 00:19)
[Hide] (393.7KB, 400x300, 00:32)
[Hide] (260.8KB, 400x300, 00:21)
[Hide] (1.3MB, 400x300, 01:53)
Been over at /tg/ getting a lot of advice for mechanics and abilities. I've created a couple new mechanics, but I unfortunately was not able to continue making new mechanics and write a continuation to the story, so I've started moving the story forward again. As such, I've created the beginnings of a simple stealth minigame that will help further the story.

Right now, it will show the line of sight and for debugging purposes, will turn red when the player is spotted.
Replies: >>184722
[Hide] (21.6KB, 158x476) Reverse
Looking forward to destroying all enemies of the Cantabiles with incense
Replies: >>184727 >>184803
>bitmap fonts
Depending on the image format, you may be able to store metadata (e.g. character dimensions) in the image file itself, so when you load a font bitmap image you automatically know its properties... For example the TGA format always ignores bytes written after the pixel data, so you can dump information there. I think BMP has this as well, and PNG directly supports metadata chunks.
[Hide] (7.5MB, 1024x768, 00:55)
Well, I've gotten... somewhere close to what I wanted. This is likely going to fucking suck when I try to get scaling working to complete the faux 3d illusion.
Either that or I look to Sonic Battle for inspiration.
Replies: >>184728 >>184738
Looks promising.  You've come a long way in a pretty short time.
Replies: >>184738
[Hide] (8.6MB, 1024x768, 00:22)
thank you. I hope to have a relatively polished product by Demonstration Day
Z-Sorting implemented (messily)
Debating getting this implemented into the Combat branch immediately or trying to tackle perspective.
Replies: >>185282
>Looking forward to destroying all enemies of the Cantabiles with incense
Thank you, anon. Likewise I look forward to what you'll have available for demo day.
[Hide] (6.4MB, 640x480, 00:19)
Well, this sure went disastrously.
Replies: >>185283
[Hide] (3.8MB, 640x480, 00:18)
nevermind, i'm just clumsy and impatient.
Replies: >>185591
[Hide] (38.4KB, 1920x1080) Reverse
last week i did more code maintenance than actual progress on my PS4 compat layer:
- fixed the issue with wrong triangle colors being output - turns out the program was using a UNorm format instead of SRGB;
- moved a lot of GNM definitions and functions to a new library named libgnm (https://gitgud.io/glue_sniffer_420/libgnm ) - the compat layer, the disassembler/SPIRV translator and PS4 programs can use it;
- solved issues reported by the Vulkan validation layer;
- cleaned up and commited source code for the test programs used in the layer;
- submitted a very small patch to OpenOrbis upstream.
i also started work on getting textures to work but haven't got anything to show so far.
maybe i should group all the PS4 tools i made for into one project.
Replies: >>185589 >>185592
>me at Saturday afternoon
>I'll play a little bit before working on my game
>how does this fucking pc-98 thing work
>found some tool to extract the files from the floppy image
>what the heck is this image format
>open ghidra, throw in the com file
<why am I looking at x86 real mode code
<how the fuck does this pc98 thing work
>find a scanned pdf of a pc98 technical manual
>many characters are unreadable. it doesn't help that nips have fucking 3000+ characters, good lock figuring out what's written there
>anyway using that and dosbox-x source, mostly figure out what this shit does
>hey, those 8-color images are actually 125-color images with lame dithering
<what if I set pc98 to 256-color mode instead?...
>640x400=250 kbyte, of course that doesn't fit into the already pretty cramped 20 bit address space, so you have top use banking if you don't want to rewrite your whole code in 32-bit protected mode
>unless unreal mode is a thing
>black magic, I like it
>hack, hack, hack...
>getting more acquainted with x86 segmented memory, instruction encoding and dosbox's horrible debugger than I wanted
PSA: set core = normal if you want to use dosbox's debugger
>also figured out most of the game's weird VM, if I can call it a VM at all (probably turing complete. you have variables, conditions, and I think you can make loops if you really want. it won't be fast though)
>finally it works, except the game also used some lio routines to draw some shit
>from dosbox source it looks like it's 8-color only
>rewrite those codes in x86 assembly. fortunately it only needs a few basic rectangle drawing functions
>a few small details are still missing, but it looks working
And I still haven't played that game I mean, it looks like a basic VN. And it's probably a horrible one, given that I pretty much randomly selected it from my pc98 dump directory. Nor did I do any work on my game. And my weekend is over. I'm the worst.
Replies: >>185436
seems like a productive weekend to me.
Replies: >>185445
[Hide] (8.9KB, 640x400) Reverse
[Hide] (7.3KB, 640x400) Reverse
Well, here's a before/after image (black textbox is after) after I fixed my color calculations to do them in linear RGB so my non-dithered image doesn't look 50% darker. One thing's for sure, the after image doesn't look that jarring after a 3x upscale.
Not sure what I'm gonna do now, Maybe try to play the game? There are just too many moonrunes..
Replies: >>185446 >>185580
Seems like the most reasonable option, since you went through the trouble.
Or just moving on to your project or something.
[Hide] (93.3KB, 1190x1056) Reverse
>figured out how to save/load basic savefiles
Replies: >>185580 >>185592
Congrats to both of you. Every little amount of progress counts.
[Hide] (58.6KB, 928x619) Reverse
Hello Mr. Big Game Dev.

Since you've made a Gravity Rush model importer yourself I'd like you to commend me on how to fix a model porting script that's on Blender 2.79.
Apparently there's a plugin on this version of Blender that wasn't properly fixed until much later on Blender 2.8+, and shit fucking sucks since Blender 2.8+ changes the way you move around entirely.
What should I do in order for the plugin to work? Consider I need to import a .psk file that's a skeletal mesh first and then a .psa file with all animations.
For instance, here are both scripts attached to this post, both the working one and the messed up one.

Sank yu, I rove yu big taim.
Replies: >>185604
[Hide] (13.8MB, 640x480, 00:46)
I've gotten some very basic AI working. I guess I should probably add some limiters or something.
I don't even know what you're doing but it seems important
[Hide] (320KB, 401x446) Reverse
I couldn't manage to download the scripts you attached, so I looked up a PSK model importer and found https://github.com/Befzz/blender3d_import_psk_psa , and I'm assuming this is the source for both scripts.
have you tried any of the forks listed in https://github.com/Befzz/blender3d_import_psk_psa/issues/88 ?
if they don't work for you, how about diff'ing the 2.79 and 2.80 scripts, and try applying any meaningful change into 2.79.
either way I don't have any Blender scripting experience so I'm unsure what else to do
Replies: >>185606
[Hide] (62.3KB, 772x905) Reverse
[Hide] (57.9KB, 788x807) Reverse
[Hide] (61.3KB, 735x868) Reverse
[Hide] (58.7KB, 655x926) Reverse
An update here, the plugin has always worked and the fix is fairly simple.
Turns out, the plugin's default option is to invert the root bone to the skeleton. For the models to work it's needed to check the Don't invert root bone option on the import.
Really, I should have just read the man page from the beginning.
Replies: >>185612
>default behavior is to do something retarded
Replies: >>185613
Coordinate systems (in 3D) are something that no two tools can agree on, ever.
[Hide] (17.5KB, 317x316) Reverse
>finally put together the .psk file to use for Unreal
>assemble the System file to load the model
>start the game
>model doesn't load in-game
>check the game's log file
>mesh could not be loaded
>check the System file in search for typos
>System file has no typos
>the mesh simply doesn't load
>check on Blender just in case
>select non manifold geometry
>edges selected up the ass
The porting process was fairly easy at the end of the day so I can afford to do it without feeling too robbed.
How do I solve the non-manifold geometry without losing the UV maps? Mesh is some hot mess I imported from [REDACTED] that leaks everywhere.
Here's a link just in case someone wants to lend me a hand: https://anonfiles.com/P3F4V3T5y7/done_7z
Replies: >>185742
>finally editied out all non-manifold geometry from the model
Holy shit that was some sweatshop hard work right there. Hopefully that was the issue as to why the model wasn't loading in-game, will report back soon enough.
[Hide] (871.2KB, 976x1280) Reverse
>it still can't manage to load the model's mesh
Well I'll be damned.
I'll try a simpler model some other day to see if it's Blender that shat the bed here.
[New Reply]
430 replies | 238 files | 186 UIDs
Show Post Actions



Select the solid/filled icons
- news - rules - faq -
jschan 0.10.2