Making a 3D modeler in C in a week

https://danielchasehooper.com/posts/shapeup/

by jasim • 2 years ago

680 187 2 years ago

dhooper5• 2 years ago

Thanks for the share!

netule2• 2 years ago

I agree entirely with the author on the limitations of Raylib. I'm currently working on a tower-defense style game that I started in Raylib, but I'm running into many of the same limitations (and more). Things such as toggling fullscreen not working consistently across platforms, not being able to enumerate screen modes, toggling rendering features at runtime, saving compiled shaders etc., etc. Having said that, I appreciate Ray's work on this library and will continue to sponsor him. Raylib is great for quickly banging out a prototype, but not much beyond that unless you're okay with living with severe limitations.

Lesson learned, for sure, but I'm too far into the development to swap all of the Raylib stuff out for SDL (or something else) now.

rwbt• 2 years ago

Raylib is easy to get started but once the project gets a little complex it bites back. SDL on the other hand takes more time to setup everything but scales extremely well as the project gets bigger and bigger. Also, SDL is exceptionally well written code.

visil• 2 years ago

And an exceptionally well written documentation, too! One of the first big-ish projects of mine was a raytracer I wrote in C with SDL.

sgt5• 2 years ago

This made me want to look at raylib. It comes with some cute examples that run using WebAssembly: https://www.raylib.com/examples.html

One thing that's always bothered me about Wasm and browser 3d/2d graphics is that I often find minor issues such as scrolling. Look at the example called "Background scrolling & parallax" here: https://www.raylib.com/examples.html

I've tested on several devices and it's definitely not smooth scrolling, unless there's something wrong with my eyes.

How can 2D smooth scrolling not be a solved problem in 2024?

dekhn• 2 years ago

the answer to your last question is "inner platform effect" and "second system effect".

modeless• 2 years ago

In that sample the foreground scrolls perfectly smoothly for me, but the background looks jittery. This indicates to me that it's not a platform issue at all. That sample is just doing something weird with the background.

flohofwoe2• 2 years ago

Because it's a surprisingly tricky topic on modern operating systems [1], and even trickier in web browsers (smooth scrolling was actually much easier to achieve on hard-realtime systems like the 8- and 16-bit home computers of the 80's and early 90s).

TL;DR: if you base your animation- or scroll-speed on the 'raw' measured time between two frames, you'll get micro-stutter because it's pretty much impossible to obtain a non-jittery frame duration on modern operating systems or web browsers, all you can do is try to remove the noise via filtering, or 'align' your measured frame duration with the display refresh interval, which on some platforms cannot be queried.

In web browsers the most important problem is that you can't measure the exact frame duration (which is fallout from Spectre/Meltdown), or obtain a precise 'presentation timestamp', or even query the display refresh frequency.

Even in the native OS APIs that provide a presentation timestamp (like DXGI on Windows or CVDisplayLink on macOS) that timestamp has considerable jitter and has not much to do with the actual presentation time when the frame becomes visible to the user.

And as soon as you base your animation timings on such a jittery timestamp you'll get micro-stutter (the easiest way to get smooth animation is actually to assume a fixed frame duration, but then your animation speed will be tied to the display refresh rate).

It's often possible to eliminate the timing jitter with 'noise removal' filters or just tracking an average over the last couple dozen frames, but those then may behave funny in situations where the frame duration changes drastically (such as moving a window between displays with different refresh rate, or when rendering stops and then resumes because the window or browser tab is fully obscured and then becomes visible again).

PS: Raylib's frame pacing code on the web is also a bit on the crude side [2].

...e.g. it just sleeps for 16 milliseconds, and relies on ASYNCIFY to enable a traditional render loop in browsers. It would actually be better to use a frame callback via requestAnimationFrame (or the Emscripten wrapper function emscripten_request_animation_frame), but this means giving up the cross-platform 'own the game loop' application model. Not that requestAnimationFrame alone solves any of the above mentioned time jitter problems though.

[1] https://medium.com/@alen.ladavac/the-elusive-frame-timing-16...

[2[ https://github.com/raysan5/raylib/blob/f1007554a0a8145060797...

oersted3• 2 years ago

Quick appreciation for the detail that Raylib is named after the creator's name Ray and not ray-tracing, fun.

Things Unexpectedly Named After People: https://notes.rolandcrosby.com/posts/unexpectedly-eponymous/

airstrike• 2 years ago

Such a good list.. worth a submission of its own IMHO

TrainedMonkey• 2 years ago

How do you know Ray was not named after ray-tracing?

pests• 2 years ago

IIRC it defines some common words too like all the color names and uses a lot of names that should be prefixed. Good otherwise.

diath3• 2 years ago

At least they're all-caps, but as somebody that writes C++ and uses Raylib, I just wrapped it in a namespace in my project that I include, like so (note that cstdio must be included before raylib if you're using it from C++):

    #pragma once

    #include <cstdio>

    namespace raylib {
        #include <raylib.h>
    }

vsuperpower20202• 2 years ago

Raylib has a lot of issues that are never going to be fixed, but I wouldn't blame fullscreen on it. Fullscreen is just absolutely unusably FUBARed on windows and has been for decades. It's probably the same for other platforms. The modern strategy is to just do borderless windowed and pretend true fullscreen doesn't exist.

netule2• 2 years ago

True, it’s just one of the many issues that came to mind while writing this.

My solution to the issue is also full screen borderless combined with resolution scaling.

spacechild1• 2 years ago

just do borderless windowed and pretend true fullscreen doesn't exist.

Hah, good to know I am not the only one.

keyle• 2 years ago

I do the same, I do wonder however, if there are performance issue at the OS level, from running (on a 4K screen) 4K borderless vs. 4K fullscreen.

Like, the whole giving the application maximum priority would work the same?

spookie• 2 years ago

At least on Xorg it isn't an issue, but to ensure cross-compatibility your solution trumps all. I cannot comment on other platforms.

duxup• 2 years ago

I have no knowledge about such things but…

The modern strategy is to just do borderless windowed and pretend true fullscreen doesn't exist.

This explains a lot as to why I experienced some interesting inconsistency going full screen in different applications in Windows ;)

diath3• 2 years ago

Kind of share the same feeling, started a project about 2 months ago and chose to use Raylib, and while the basic stuff is really simple to get going, the more you use it, the more random minor inconveniences you run into, but at this point I've invested too much into this project to back out of using Raylib. My biggest issue with it right now is the font handling and text rendering, I think I'll have to switch from TTF fonts to pre-baked bitmap fonts for it (which will suck for localization later). The biggest two features I'm missing after switching from Love2D is being able to render multi-color text (with Raylib, you have to manually split up the text into chunks based on color markup, then apply width offset, then call the draw function for each chunk, while also taking into account things like linebreaks and such), not to mention that it seems to tank the FPS a lot when you try to draw a lot of text on the screen (perhaps the draw call batching is broken for text?) and being able to easily chop up textures and make them wrap or tile, Raylib used to have tiled texture draw function in the past but for some reason they removed it.

abnercoimbre• 2 years ago

I'm the organizer for the conference [0] mentioned in TFA.

We had a professional UI/UX designer react to ShapeUp [1], and one of the things she commented on was the font being hard to visually parse.

I laughed a little when the author yelled "raylib!" to make sure blame was assigned appropriately XD. I'm currently the top GitHub sponsor for raylib, so there's no hate, but I wish he changed some of his defaults.

[0] https://handmadecities.com/seattle

[1] https://vimeo.com/887532756/2972a82e55#t=49m58s (timestamped)

edflsafoiewq• 2 years ago

Why did you switch from Love2D?

nextaccountic• 2 years ago

I agree entirely with the author on the limitations of Raylib.

Wow this is kind of insane. About this

Raylib doesn’t do basic parameter validation, by design. This function segfaults when dataSize is null: (...)

The developer answered this

For most of the raylib functions is up to the user to validate the inputs, if raylib should consider all possible bad-use scenarios it would require reviewing most of the library functions and it will increase source-code complexity.

kragen5• 2 years ago

in c and c++ this is not just normal but almost unavoidable. if your function takes a t* and someone casts a random int to a t* and pass it in, it is gonna segfault. no possible way to validate it, though in theory you could open /proc/self/maps and iterate through it to catch the segfault cases, or install a segfault handler

lelanthran11• 2 years ago

I think the word 'insane' is going to far to describe the behaviour of the specified function.

It returns an array of bytes. If you, the programmer, wrote a line that called that function, on the very next line you are going to try to use the array, realise that you don't know the length, and realise that the `NULL` that passed in on the line above is probably the output for the length!

In order to actually write a call with `NULL` for the dataSize argument, the programmer needs to be clueless about how to write a for loop.

So, no, I can't easily see a situation where a programmer accidentally uses a `dataSize` parameter of `NULL`, because that would mean they don't know that arrays in C have no length information, which is C 101.

jakubtomsu• 2 years ago

I switched from raylib to Sokol some time ago and it's the best. Super simple and single header C code, no dependencies, cross platform code and shaders, etc etc. I've shipped a steam game and I plan to continue using Sokol in the future.

user4326783• 2 years ago

Do you mind sharing a link or game name? Despite all the advice out there picking up a game engine, I too am pursuing a goal of making a game from scratch using simple parts of C++ and SFML for now, but was also looking at SDL and Sokol. What was your experience? Are you a solo dev? Was it worth it?

parasti• 2 years ago

There's something really powerful about taking the tools that you know very well and just making something cool with them. Really enjoyed this writeup, thanks.

xixixao• 2 years ago

Super impresssive for getting this done in a week. Being able to make pretty demo models definitely helps too! :)

sandwichukulele• 2 years ago

is the source code available? I looked through the blog post and linked videos but could not find a github repo or anything similar

TruthWillHurt• 2 years ago

Or just being able to save/load creations would be nice :)

dhooper5• 2 years ago

yeah I never implemented saving/loading for the web. Thats one example of how raylib doesn't totally abstract the underlying platform for you.

dhooper5• 2 years ago

I just made it public: https://github.com/danielchasehooper/ShapeUp-public

jbritton• 2 years ago

How did you create / obtain the example shapes? Is there a standard format your code parses?

SoKamil• 2 years ago

The Shapes are kept in a statically allocated array [...] Can’t fail to allocate, can’t be leaked, no fluff. Lovely. The 100 shape limit wasn’t limiting in practice. With very little time to optimize the renderer, the framerate would drop before you even got to 100 shapes.

That's the best example of avoiding premature optimization I've seen in a while.

Narishma• 2 years ago

I think this is the opposite. It's avoiding premature abstraction/generalization.

bruce343434• 2 years ago

But finding abstractions and generalizations optimizes scratching that itch inside our heads

HumblyTossed• 2 years ago

Best example of those who do vs. those who sit around bickering over HOW to do.

ffitch• 2 years ago

The project is 2024 lines of C

got to appreciate the effort to make the irony possible : )

poopicus• 2 years ago

As someone who has difficulty in detecting irony, could you explain the irony in this statement?

booleandilemma• 2 years ago

2024 is the current year and it's the same as the number of lines of code. I don't think describing it as ironic is correct though.

enumjorge• 2 years ago

The other response is correct that this is not ironic. Roughly speaking, irony is when something happens that is the opposite of what you'd expect. A firefighter's home burning down is ironic. Sometimes irony is related to unfortunate or funny coincidences/timing, and it's easy to confuse the two. Alanis's song Ironic famously has a lot of examples of this. Rain on your wedding day--is that ironic? Maybe? You certainly hope there is no rain on your wedding day, but I don't think there's an expectation that there won't be rain. Now if your parents decided to get a divorce on your wedding day, I think that's ironic.

But the parent commenter dilutes the definition further. A project with 2024 lines of code in 2024 is just an amusing coincidence. There's no reason why you'd expect a project in 2024 to not have 2024 lines of code.

runevault• 2 years ago

Super interesting post, and appreciate him talking about the various decisions like his handling of memory (and the issues he ran into with raylib). As someone who's finally diving into part 2 of crafting interpreters (and using it to refresh myself on C) being reminded of what C does well is great.

fallingsquirrel• 2 years ago

I really love the live demonstrations in the video. Forget building the app, I couldn't even produce that video in a week if I tried.

dhooper5• 2 years ago

The video took me longer to make than the app! I don't know how youtubers do it so regularly.

an_aparallel• 2 years ago

They have a team :)

swiftcoder• 2 years ago

That's some impressive development speed. Really enjoyed the explainer video too!

emmanueloga_• 2 years ago

What is a stablished 3d modeler that uses the same kind of modeling as this one?

Lichtso• 2 years ago

Many 3D content creation tools such as Blender [0] have SDF [1] (e.g. metaballs in Blender) and CSG [2] (e.g. boolean modifier in Blender) features. But these are rarely used as they can only define volumes, not surfaces. And we are usually interested in surfaces for assigning textures and materials. Thus, polygons / meshes and curves / splines dominate the industry.

[0] https://www.blender.org/ [1]: https://en.wikipedia.org/wiki/Signed_distance_function [2]: https://en.wikipedia.org/wiki/Constructive_solid_geometry

antirez3• 2 years ago

TinkerCAD.

starmole• 2 years ago

Adobe/Substance3D Modeler

dantondwa• 2 years ago

The two best SDF modelers in existence are MagicaCSG[1] and Adobe Substance Modeler[2]. There are also a few others, like Womp, but those two are the most feature-complete. Blender is also adding them as part of geometry nodes, and there is also an add-on that is working on adding SDF for hard-surface.

[1] https://ephtracy.github.io/index.html?page=magicacsg & https://www.patreon.com/magicavoxel

[2] https://www.adobe.com/products/substance3d/apps/modeler.html

joeld422• 2 years ago

Blender Geometry nodes supports SDF modelling. It's just not a widely known or used technique. But it's super powerful.

tcfunk• 2 years ago

Plasticity (https://www.plasticity.xyz/) is a new-ish one that looks neat, but I wish it offered a lower-price entry point.

RamiAwar• 2 years ago

Amazing write up, thanks! Really enjoyed it, miss working on C/C++ apps from scratch and having full control

jasonjmcghee• 2 years ago

This looks like such a fun jam - wish I'd have known about it!

When's the next one?

bvisness• 2 years ago

The next Wheel Reinvention Jam will be in September! We're just finalizing our plans for that jam and our smaller Visibility Jam in July. If you're interested in participating, then join the Handmade Network Discord server (link on our home page at https://handmade.network/).

antirez3• 2 years ago

I hope somebody will continue this project. It's a few months away to be a serious alternative to Blender / FreeCAD for certain use cases, with a much softer learning curve.

turtledragonfly• 2 years ago

EDIT aha, the program already supports exporting to a mesh via marching cubes; see the youtube video on the site. I hadn't realized that (:

----

Be aware that since it fundamentally works with SDFs, it is a somewhat different modeling experience (and stores different data) than traditional meshes with triangles, verts, etc.

Transforming it from SDFs into meshes could be done with marching cubes or similar, but you'd likely need to "clean up" such data afterwards in a Blender-style app anyway.

SDFs are great though, if your renderer is SDF-based, too (most are not).

[sorry if you knew this already, wasn't sure]

antirez3• 2 years ago

Yep the export is already done! I used TinkerCAD a ton for things ways more complex than it should be used to, so even when I use more advanced CADs at this point I tend to think in SDF terms. For many things it's faster and more natural than extruding, rotating, ... But the fact is, the engine behind TinkerCAD is quite good, but there is little interest for AutoDesk to compete with its own Fusion360, so TinkerCAD is left forever as a children / beginners tool, without the more advanced stuff that could implement.

johnnyanmac3• 2 years ago

SDFs are great though, if your renderer is SDF-based, too (most are not).

Thanks for reminding me that one of the best SDF renderers in the industry is stuck behind Sony, deemed a failure because Sony didn't want to publish it to PC (where all the dev scene is).

adastra222• 2 years ago

Blender maybe, but not CAD work unfortunately.

nineteen9995• 2 years ago

Not even Blender. Not even close.

No disrespect to the author, it's impressive, but Blender gives you extreme control over mesh topology and this doesn't.

moarinfoszszz• 2 years ago

Check out Dune3D and Salome-Platform

OskarS3• 2 years ago

You should check out MagicaCSG, which is a more sophisticated version (though still free!) of this: https://ephtracy.github.io/index.html?page=magicacsg#ss-caro...

captainhorst• 2 years ago

There's also https://womp.com as an SDF modeler

fsloth• 2 years ago

If you like SDF:s Womp is pretty nice for starting out. Tinkercad is another pretty good beginner “cad”.

rationalfaith• 2 years ago

Good stuff! As a c/c++ coder maintaining his game engine (for commercial and hobby purposes), this is always good to see!

freecodyx• 2 years ago

I sometimes think, that c is all we need

yazzku• 2 years ago

I write C all day and enjoy every second of it. I assume there are many like myself. There's just nothing new to make noise about, and that's a good thing, other than small quality-of-life improvements in the standard.

lionkor• 2 years ago

you dont need that comma fyi

While I fully support liking one language, it wouldnt feel right for me to not mention the benefits other languages bring, such as GC for large dev teams of lower skilled devs, languages with builtin unit tests, languages with templating or less bad macros, etc.

ngcc_hk2• 2 years ago

Really agreed with his assertion of c. More to his “ Its syntax doesn’t hide complex operations. It’s simple enough that I don’t have to constantly look things up”, and further if you need to look up something about c, it is very easy and very informative. Simple and old lang has its benefits.

syphiant• 2 years ago

Impossible! Wow! This is absolutely mind blowing.

ederamen• 2 years ago

Just started using Raylib, bummed to hear about the limitations!

As a novice C programmer, the simplicity and immediacy of results opened my eyes to how C can feel as productive as higher level languages with robust standard libs.

lelanthran11• 2 years ago

As a novice C programmer, the simplicity and immediacy of results opened my eyes to how C can feel as productive as higher level languages with robust standard libs.

TBH, once you have halfway-good libraries for dealing with `char *` strings as-is, dynamic arrays and hashmaps, you are not going to be much slowed-down using C than using a higher-level language.

You even get much stronger isolation guarantees than most other high-level languages, while getting much more compatibility[1] with any other language you may wish to interface to: https://www.lelanthran.com/chap9/content.html

[1] I did a little Go project, and it annoyed me slightly when I wanted to do performant FFI. For Go, I think the situation has improved since I last checked, though.

neonsunset6• 2 years ago

You can trivially replace Go with C# in order to get almost zero-cost FFI (you can make it fully zero-cost with additional effort but even the baseline is better than the alternatives, hell you can statically link other .lib/.a's into AOT compiled .NET binaries).

naasking9• 2 years ago

TBH, once you have halfway-good libraries for dealing with `char *` strings as-is, dynamic arrays and hashmaps, you are not going to be much slowed-down using C than using a higher-level language.

That can't possibly be true. Not having to even think about object lifecycles and ownership because all memory is GC'd saves a lot of time all by itself, not even getting into debugging issues when you get it wrong.

Jach• 2 years ago

Really cool and the video was great too. Indie devs should probably consider doing this sort of thing more often, building these simple tools in service of a game rather than just using the industry standard tool. It can be a great way to add artistic character as well as rigorously enforce some limitations if that's what you're after.

koushik• 2 years ago

This looks cool, after 3 years in financial technology industry working on c/c++ projects I’m in process of revisiting textbooks and relearning computer science fundamentals! Added raylib to my ‘explore’ list

I love this idea of reinventing wheel with such explicit goal (even if it sounds counterintuitive to some), we can rethink initial assumptions, best case scenario we come out with better implementations and paradigms than existing standards, worst case scenario you learn internals of tools and techniques you use daily!

gorkermann• 2 years ago

To get a look at SDF rendering in a game, check out the blue clouds on the ground in Solar Ash:

https://youtu.be/HqQpYSQDIZQ?si=vMKplmGJIGvUn_LT

drtgh2• 2 years ago

Off-Topic:

Glad to see for the first time a WebAssembly interface where the text does not look blurry. I repeat, it is the first time.

Extending this to programs and some operating systems (such as Windows), in the past few years, there has been a pervasive issue with the text rasterization methods that have become a common trend and default setting.

Unfortunately, users often do not have the option of turning off anti-aliasing to get sharp text, and in the rare cases where this option is available, the interface (menus, etc.) still uses anti-aliasing.

themerone• 2 years ago

The crisp text is unimpressive. The typeface has no curves and there is no smoothing. That text will be crisp and blocky at any resolution.

drtgh2• 2 years ago

The crisp text maybe it is unimpressive in the sense that is being used an graphic library targeted to games, so is using an homologous to pixel fonts. But it is not the norm.

I'm simply increasing awareness among the development community about the problems with text rasterization in applications and OS interfaces. I'm glad not to see blurry text.

flohofwoe2• 2 years ago

The crisp text is unimpressive.

Not if you actually tried to achieve that in practice across all browsers and display configurations ;)

For instance Safari had for the longest time a hardwired (e.g. not fixable via CSS) linear filter applied when a WebGL canvas had to be upscaled. It's only been fixed last year or so.

Also unfiltered upscaling done breaks down when fractional scaling outside the browser happens, the result will generally look horrible and there is no good way to fix it. The only workaround is to make hide the scaling artifacts behind a filter, which then makes everything look blurry.

Browsers running on top of Wayland most likely still have all those issues.

TachyonicBytes• 2 years ago

Why is there a link between WebAssembly and rasterization? Genuinely curious, it seems interesting

kewp• 2 years ago

I wish he had tried instead to do the faster subset of typescript, that is a pet peeve of mine and I've love to see how it would be done!

cnity• 2 years ago

AssemblyScript?

rkagerer• 2 years ago

You could certainly make it harder on yourself by malloc-ing each Shape individually and storing those pointers in a dynamic array. Using a language like C# ... would force that allocation structure.

What's stopping you from using a fixed array of structs in C#, just as the author has done in C?

kronovecta• 2 years ago

Nothing. In C# it's not unusual to use struct arrays in this way.

OskarS3• 2 years ago

Probably confusing it with Java, where the only value types are the basic ones (unless you store each individual coord in a different array).

neonsunset6• 2 years ago

"Using a language like C#, Javascript, or Python would force that allocation structure."

No. C# structs are C structs. Shape[], Span<Shape> or Shape* would not be an array of pointers. Any of the following would work:

    var fromHeap = new Shape[100];
    var fromStack = (stackalloc Shape[100]);
    var fromPool = ArrayPool<Shape>.Shared.Rent(100);
    var fromMalloc = (Shape*)NativeMemory.Alloc(sizeof(Shape) * 100);

kuon• 2 years ago

I really like this kind of projects. I still like the low level of C. Now I work with rust a lot and elixir/erlang but I often miss the simplicity and explicitness of C. For this, I use zig a lot too. It is a very nice improvement over C while keeping a lot of its philosophy.

nesarkvechnep• 2 years ago

Erlang is a pretty simple language too.

anthk• 2 years ago

Can I compile it with GNUStep?

naasking9• 2 years ago

Nice, I was considering a project like this myself. Signed distance fields are awesome. Everyone into modelling should check out ImplicitCAD based on SDFs.

JabavuAdams• 2 years ago

This is really great! Nice work, and thanks for sharing.

_ddbw• 2 years ago

getting tired of the underhanded shilling for c and procedural style programming. the people doing this still pretend they are the underdogs but their point of view is over saturated.

TacticalCoder• 2 years ago

I don't know why you got that feeling, TFA says this:

Some view C as a language so simple and raw that you’ll spend all your time working around the language’s lack of built in data structures, and fixing pointer bugs. The truth is that C’s simplicity is a strength. It compiles quickly. Its syntax doesn’t hide complex operations. It’s simple enough that I don’t have to constantly look things up. And I can easily compile it to both native and web assembly. While C has its share of quirks, I avoid them by habits developed over 22 years of use.

Which is nothing unreasonable.

I code mostly in a Lisp dialect but I don't take offense at, say, the Linux kernel being mostly C.

gromneer• 2 years ago

i am not assessing the truthfulness, the reasonableness, the correctness, or the veracity of the claims made in the article in any way. i simply do not care because the claims are purely subjective, untestable, unmeasurable. it is an opinion piece and one that is dulled out ad nauseam in the programming world today. enough to be annoying now. one opinion for another.

layer8• 2 years ago

I’m getting tired of people not using the Shift key.

CamperBob2• 2 years ago

Other sites beckon.

ttul• 2 years ago

Long ago, I worked on the operating system for a desk phone. With only 64K of RAM, there was no dynamic memory management whatsoever. We made heavy use of static variables and let the compiler figure out how to allocate everything at compile time.

It’s easy to forget that many applications probably don’t need dynamic memory management at all. You can often get away with allocating a few fixed size buffers and just handling the edge cases nicely when those buffers are full.

And in such a context, C is indeed a whole lot safer. No memory leaks. Your only concern is buffer overflows, which can be managed through careful use of sizeof when all of your variables are statically allocated. I’m not saying Rust and Go aren’t great options these days, but humble old C still works and doesn’t have to be nightmarishly complex.

Making a 3D modeler in C in a week

Related Stories