Software rendering in 500 lines of bare C++

7/23/2026 at 3:05:54 PM

I went through this a few months ago in Rust. I wrote all the code by hand, no LLMs. Then I went ahead and added a small "game" on top, plus some special effects like pixelization shaders and chromatic aberration at the edge of a flashlight.

https://github.com/kshitijl/tinyrenderer-rs

if anyone is interested! The repo has lots and lots of in-progress screenshots so you can see the renderer come to life, plus all the hilarious visual bugs along the way.

I learned a lot! My biggest lesson, other than the specifics of how rendering works, was that modern CPUs are really fast: a single-threaded CPU renderer can definitely run an interactive 3D game with some fancy special effects.

by articulatepang

7/23/2026 at 4:39:16 PM

Why does it pull in wgpu if it's a software renderer?

by 0x1ceb00da

7/23/2026 at 4:53:07 PM

In this case wgpu is just providing the surface texture for the window that the software rendered pixels are drawn into.

by grovesNL

7/23/2026 at 6:57:20 PM

Softbuffer?

(I originally wrote this a few years ago but other people maintain it now)

by john01dav

7/23/2026 at 7:11:28 PM

I like softbuffer, but it owns its window and doesn't support as many events / integrations as winit.

by LoganDark

7/23/2026 at 11:23:08 PM

You might be confusing softbuffer for minifb here, as softbuffer doesn't do any window handling of its own. All window management and event handling happens through winit (or another crate providing a HasWindowHandle implementation).

by soxfox42

7/24/2026 at 12:08:21 AM

Oh, you're right! My bad

by LoganDark

7/23/2026 at 7:58:12 PM

Yes, exactly this -- it's a fast and convenient way for my code to just write the pixels into a spot in RAM (the CPU's RAM, not VRAM) and have those pixels end up on screen. On modern architectures this nearly always goes through a GPU, so even if you're not using the GPU to accelerate your 3D rendering/math, you gotta deal with it just to put pixels on a screen at the end of the day. So that's what wgpu does for me.

by articulatepang

7/23/2026 at 5:30:13 PM

You do often need a graphics context to create the surface that displays the framebuffer.

by LoganDark

7/23/2026 at 8:47:37 PM

Not necessarily.

Most desktop OS compositors provide a void** buffer that can be blitted to. Sure, this buffer will probably end up going through the compositor which does GPU hardware acceleration anyway, but it is definitely possible to shrink the application-side stack without relying on D2D/GLUT/GLEW/GLFW/SDL/WGPU in this case.

On Windows: CreateDIBSection and BitBlt

On macOS: CALayer and CATransaction commit

Not sure about Linux, too many stacks involved.

by delta_p_delta_x

7/23/2026 at 10:36:55 PM

I have done this on a number of platforms over the years. Going back to before createDIBSection existed (In fact I had something break when Windows provided an easy path and took away the method I had been using).

Linux has always been a weird kettle of fish, the thing that worked best for me was imlib2 which was originally part of enlightenment. It did what it was supposed to do, hid pixel format conversion without killing performance.

I'm not sure what would serve a similar function today,

If anyone knows of a library that takes a rectangle of memory, a pixel format and places it onscreen regardless of the display environment I'd like to hear about it.

The faster and more efficient, the better. More features than needed, worse.

by Lerc

7/24/2026 at 6:49:53 AM

Before those days, the solution would be to make use of WinG, the DirectX precursor, still introduced in Windows 3.0

https://www.gamedeveloper.com/programming/a-whirlwind-tour-o...

https://betawiki.net/wiki/WinG

by pjmlp

7/24/2026 at 6:38:59 AM

https://zserge.com/posts/fenster/ is what I used when I wanted "basic graphics" in a cross-platform way.

But of course that's X11 and much simpler.

by stevekemp

7/24/2026 at 8:50:18 AM

Ah, got me hooked readying the rest of the post with the "BGI and BASIC’s painting commands" reference. :)

Great overview of the whole approach you took there.

by pjmlp

7/24/2026 at 5:09:50 AM

> If anyone knows of a library that takes a rectangle of memory, a pixel format and places it onscreen regardless of the display environment I'd like to hear about it.

I've used SDL for this in the past.

by lelanthran

7/23/2026 at 9:09:15 PM

> Not sure about Linux, too many stacks involved.

Wayland: wl_shm and that’s it for the “stack”.

by mananaysiempre

7/23/2026 at 8:54:30 PM

Cross-platform support can be more difficult in that case, but yes it is possible.

by LoganDark

7/23/2026 at 8:00:00 PM

[dead]

by njjsjsys

7/24/2026 at 5:33:38 AM

[flagged]

by fridayis

7/24/2026 at 6:39:22 AM

It is a fashion thing, it is more relevant to discuss about a game engine does ECS, than what the game is all about.

Meanwhile, games like Minecraft, Balatro, Stardew Valley, Celeste, Geometry Wars,... blow up in adoption.

by pjmlp

7/24/2026 at 7:28:12 AM

[flagged]

by juuduahsvxyd

7/23/2026 at 3:42:05 PM

This resource, along with Mathematics for Computer Graphics by John Vince [1], was truly indispensable when I wrote my own software renderer [2]. This was long before LLMs, so the whole process took me at least a couple months - most of it trying to wrap my head around math behind computer graphics and tracking down C segmentation faults. Fun times.

[1]: https://www.amazon.co.uk/Mathematics-Computer-Graphics-John-...

[2]: https://github.com/nkanaev/tipsy

by nkanaev

7/24/2026 at 2:49:58 AM

Is the source of yours public anywhere? I'd like to take a look.

by DatCodeMania

7/24/2026 at 10:32:15 AM

How many hours a day over the couple months?

by mentos

7/23/2026 at 11:30:26 PM

Is the Foley/Van Dam book still a go to resource for this? It seems it was updated in 2013, but, honestly, I’m more familiar with the ‘82 edition that was dedicated to 2D.

Back in the day, it was The Book for computer graphics.

by whartung

7/24/2026 at 8:19:03 AM

I haven't looked at my copy in years. To me, the book is a peculiar encyclopedia with some historical importance.

I found the course note from this github to be a good refresher on the subject. While the repo source code style is distasteful to me and their old-school rasteriser in the course is too naive and unoptimal, it's still a better read than Foley I'd assume :)

by momocowcow

7/24/2026 at 6:52:13 AM

I also learnt with the second edition, I own the last one from 2013, it is alright.

The languages have evolved across editions, from Pascal, to C, to C and C++, and a bit of C# as well on the last one.

Naturally it misses on several new concepts, however I would assert it has quite valuable content.

by pjmlp

7/23/2026 at 3:00:53 PM

I wish we could have just one of these tutorials properly cover the concern of triangle clipping. This is the part that I struggle with the most in a software renderer. If you are going to be building a practical one, this is something you will eventually have to deal with, even for super basic scenes. Any time geometry intersects the view frustum you need to clip those triangles.

by bob1029

7/23/2026 at 3:16:43 PM

I have a whole chapter on that! https://gabrielgambetta.com/computer-graphics-from-scratch/1...

by ggambetta

7/23/2026 at 3:31:12 PM

Sorta unrelated but I really enjoy (I don't want to use the past tense as I often refer back to it) your article on Emulator Backed Remakes (https://gabrielgambetta.com/remakes.html). Thanks!

by aninteger

7/23/2026 at 3:48:02 PM

Thanks for saying that :) Have you built something based on that idea? I still haven't, but I'm thinking on throwing some Claude at it for the most mechanical parts.

by ggambetta

7/24/2026 at 3:29:33 AM

I bought and I am reading through your book as I build a sw rasterizer. On the chapter of clipping, it was not entirely clear to me how many planes you clip against and how many triangles result.

Presumably if you clip against all 6 planes, you get 3+6-2 = 7 triangles in the worst case, which is kind of annoying. However, if you're going to use the AAAB triangle raster method as described in this HN post, then it seems there's no need to clip against left/right/top/bottom planes of the frustum, since it is simpler to just clip the screen-space AABB of the triangle instead. So one just needs to clip against near/far. This is simpler and faster.

I was curious if you had given any thought on that, or if you plan on working on a second book with more details on these nuances.

by yunnpp

7/24/2026 at 9:12:15 AM

My book tries to be as accessible as possible; a high schooler with little to no knowledge of linear algebra should be able to understand all of it. So in every case I've chosen simplicity (code and/or conceptually) over performance, but without sacrificing correctness. Not always an easy balancing act!

I would not recommend anyone to write a production rasterizer using the algorithms in my book. This website seems more performance-focused (Bresenham is the obvious choice to draw lines, for example).

Specifically about clipping, you can ignore all the planes except the near plane, and everything will be fine. You'll try to draw more pixels outside the canvas but that's just slow, not problematic. The near plane does protect you from divisions by zero and from points with negative Z, which project upside down and mess things up.

by ggambetta

7/24/2026 at 4:10:27 PM

Yeah, the book has a nice balance, it's precisely why I picked it up. It allowed me to get something up and running quickly, then iterate over it.

by yunnpp

7/23/2026 at 3:26:22 PM

You only need to clip triangles is you're worried about attribute interpolation for very large triangles. There's two ways to handle this: (1) discard (fast but not a great user experience); or, (2) primitive synthesis. Just frustum clipping is enabled by point picking in the local tile. Primitive synthesis requires some FP kung fu; but, is easiest done in barycentric space against a reverse transformed clipping rectangle. This lets you carefully control clip rounding error using either doubles or (better) fixed point. Abrash likes to use integer fixed point, but that is historical — modern fixed point can be handled with careful control of the fp unit in the mantissa. The major issue is regenerating the Z and the 1/Z values for the new vertices of the synthesized primitives. Everything else should flow down the pipe naturally, assuming a deferred attribute synthesis rasterizer.

There are examples in the open source version of my rasterizer: OpenSWR.org.

by thechao

7/23/2026 at 3:38:07 PM

I'm clearly super old-school when it comes to rasterisation, and a lot of this has flown above my head. I have a ton of questions; I hope you can answer them.

> (1) discard (fast but not a great user experience)

What are we discarding here, and why is it fast but not a great user experience?

> (2) primitive synthesis

I assume this is retriangulating clipped triangles that are now no longer triangles?

> reverse transformed clipping rectangle

Which spaces does this reverse transformation map from and to? I assume the clipping rectangle here is the triangle's AABB in raster space (or as you say, barycentric space).

> integer fixed point, but that is historical — modern fixed point can be handled with careful control of the fp unit in the mantissa

So we are no longer doing 16.16 fixed point, but tweaking the FP representation itself?

> The major issue is regenerating the Z and the 1/Z values for the new vertices

Why is this a major issue?

> deferred attribute synthesis rasterizer

I assume this means attributes are perspective-correct interpolated in raster space.

by delta_p_delta_x

7/23/2026 at 4:08:44 PM

Sorry I was so terse! Let's ignore guard band, for now. Our rasterization surface is a rectangle. We don't need to "geometrically clip" a triangle to the rectangle's surface. Instead, we just walk the surface of the rectangle and ask two questions: (1) does the triangle capture this (sub)sample; and, (2) what's the interpolated value of the attributes at this (sub)sample. In practice, for software rasterizers, we're working on tiny subrectangles (the tile), e.g., 4x4 or 2x8, whatever. So, we can be a little "inefficient" with our walking with respect to the edge testing for interiority. (This is roughly how HW works at the 2x2 level, as well.) Because the (sub)sample is "pulling" the attribute, we don't need to geometrically "clip" the triangle to the tile: the tile's (x,y) subsamples do that "for free".

On the flip side, if a triangle is 'really big' we need a guard band to either reject or subdivide triangles. The first one is fairly cheap -- we're throwing away the triangle! -- the second one is a better user experience, but requires synthesizing primitives. (The worst case is that a single triangle becomes 5 triangles, I think.) Each of those triangles needs its Z and 1/Z calculated in fixed precision. The precision of that fixed precision (though) can be clamped to the local tile; so, even though the global precision might need to be 25.25 (or whatever), the tile-local precision is only 4.9 (or whatever), with an intermediate 24.24 that can be handled with a float-float patch-up. The computation should all occur in the triangle's barycentric space: that means you need the inverted barycentric mapping to invert the guard band into the triangle's barycentric space. You do that because it lets you control the fixed point calculations better. (You can leave off the inverted determinant multiplication until the last moment.)

When I say "deferred attribute synthesis" I mean that we don't calculate attributes in the vertex shader. Instead, we calculate the barycentric, Z, and 1/Z values and pass those along. When we fire up the tile walker for the triangle (in general we only need 1-3 tiles), we calculate the attributes "on the fly, as they're used" and then let the compiler do CSE to fold down the replicated constructions.

by thechao

7/24/2026 at 2:52:43 AM

Not OP, but thanks for the very detailed response.

by DatCodeMania

7/23/2026 at 3:54:28 PM

My own knowledge of GPU rasterization may be dated, but IIRC GPUs tend to rely on guard band clipping up to a guard band threshold before using geometric clipping. The guard band clipping involves rejecting 2D coarse rasterization blocks that are fully outside of the scissor rect. This is just a quick rectangle check, but the tradeoff is that a larger guard band means more GPU time lost in over-rasterization and potential higher precision requirements for rasterization values (which could be fixed point). Beyond the guard band, the triangles are clipped in floating point against the frustum clip planes.

by ack_complete

7/23/2026 at 4:27:11 PM

My renderer attempts always got stuck on the "should implement clipping" phase too, until I finally bit the bullet and managed to write a working one without much effort, independently "rediscovering" the Sutherland–Hodgman algorithm [1] as I found out later (googling it beforehand would've been cheating, of course).

The algorithm itself is fairly straightforward and intuitive, I think the biggest mental block is the weirdness of the projective space and working with homogeneous coordinates (actually the only frustum plane that you have to clip against in P₃(ℝ) is the front plane, the rest could be clipped after the perspective division, but no reason not to do it all at the same time while you're at it). The plane equations in the clip space are super simple, basically the six equations of the form ax + by + cz = w simplify to

  x = ±w 
  y = ±w 
  z = ±w.

Meaning, for example, that if the x coordinate of your vertex is greater than the w coordinate, that vertex is outside the right clipping plane. The Sutherland–Hodgman itself goes something like this:

  # Returns true if point is inside the half-space defined by plane
  def point_inside_plane(point, plane) -> bool:
    # single dot product, can be further simplified

  # Returns t such that the edge (p1, p2) intersects plane at lerp(t, p1, p2)
  def edge_intersect_plane(edge: (Point, Point), plane) -> float:
    # single dot product, can be further simplified

  # Given the vertices of a simple polygon and a plane,
  # returns the part of the polygon fully inside the plane 
  def clip_against_plane(poly: [Vertex], plane):
    let result: [Vertex] = []
    let [(v_1, v_2), (v_2, v_3), ..., (v_n, v_1)] = poly.edges()
    for each (v_i, v_j) of the edges:
      let i_inside = point_inside_plane(v_i, plane)
      let j_inside = point_inside_plane(n_j, plane)

      if i_inside and j_inside:
        # v_j will be pushed on the next iteration!
        result.push(v_i) 
      else if not i_inside and not j_inside:
        pass # Nothing to do!
      else:
        # One is inside, the other is not, we have to clip
        let t = edge_intersect_plane((v_i.pos, v_j.pos), plane)
        # Synthetize a new vertex straddling the plane
        let v_new = Vertex(
          pos = lerp(t, v_i.pos, v_j.pos),
          # For each vertex attribute
          attrib = lerp(t, v_i.attrib, v_j.attrib)
        )
        if i_inside:
          result.push(v_i); result.push(v_new) # discard v_j
        else:
          result.push(v_new); result.push(v_j) # discard v_i
      
   return result

Then you just call this for all the planes so that the output of one call becomes the input for the next call! The end result of this process is a convex polygon (of at most nine vertices for a triangle against six planes), which can be trivially triangulated. You can make the whole process faster by precomputing so-called outcodes which allow you to avoid clipping triangles known to be entirely outside at last one plane, or entirely inside every plane.

[1]: I. Sutherland and G. Hodgman. 1974. "Reentrant polygon clipping." Communications of the ACM, Volume 17, Issue. Available: https://dl.acm.org/doi/10.1145/360767.360802

by Sharlin

7/23/2026 at 8:11:53 PM

This is a great comment. As someone who is familiar with R^3 geometry but not projection or clipping, I feel like I could almost implement it from your comment alone.

Can you explain more about clipping before/after the divide by w operation? Is this about avoiding coordinates with w = 0? Otherwise all the geometric ideas seem to make sense in both the world/view-space frustum and the ±1 box.

by blt

7/23/2026 at 9:13:24 PM

Essentially, yes. The points of any edge that crosses the near plane escape to the infinity and if you naively just w-divide the endpoints, the image of the lineseg is not the points between the endpoints, but actually all the other points on the line that "wrap around" through infinity. The projective space not only contains all the points at infinity, it identifies each pair of infinity points "opposite" to each other (and then maps them to the w=0 plane), so lines are actually loops and between any two points there are two distinct line segments, an "internal" and "external" one. This is disconcerting to say the least.

Moreover, all geometry that is fully behind the camera will actually get projected in front of the camera into the distance by the w division because they have z<0, w<0 so the signs cancel out. I think clipping against the back plane would get rid of these, but it's still weird. And now my brain hurts.

Historically, of course, divisions were expensive and culling as much as possible before the perspective division was a performance question as well.

by Sharlin

7/23/2026 at 3:53:10 PM

Finally an engineering feat that's not built in Rust

by espetro

7/23/2026 at 4:23:02 PM

There was a software rendered game in vein of Tomb Raider 1 graphics for both DOS and Unix, but I can't remember its name. It was an exploration game, modern, a bit cyberpunkish, in 3D.

by anthk

7/24/2026 at 7:55:06 AM

There's basically no such thing as "bare C++" anymore. On any modern machine you are relying on piles of code. You can't just write to registers to edit vram and output video like on some 80s computer. It all happens on top of a thick stack of APIs, drivers and firmware.

by smolder

7/24/2026 at 12:54:17 AM

In a fit of 90s nostalgia I’ve also been going back to software rendering, though I’m doing a hybrid of 2D style CLUT banks with a more modern binned triangle and barycentric technique. Since I’m sticking to a fixed pipeline look, I’ve been amazed at just how many triangles one can push even with a fairly naive draw function.

by xgkickt

7/24/2026 at 2:51:02 PM

That's impressive. Building a renderer from scratch sounds like one of those projects that teaches you far more than following tutorials ever could. I'll definitely check out the screenshots.

by Zfyuchar

7/23/2026 at 2:46:54 PM

I jumped when the first image loaded on the page.. didn't expect that.

by t1234s

7/23/2026 at 3:00:15 PM

What image was that? The first image I get is a headshot.

Personally I’d have gone with a teapot as my demo but I don’t see anything shocking about the model the author chose.

by hnlmorg

7/23/2026 at 3:11:31 PM

> I jumped when the first image loaded on the page.. didn't expect that.

mbappe'?

by mito88

7/23/2026 at 3:35:14 PM

[dead]

by fleroviumna

7/24/2026 at 6:54:55 AM

I actually submitted my first PR against this repo in forever because I found a bug in how macOS handles OpenMP. Cool article and code.

by Wintermute21

7/23/2026 at 9:42:06 PM

on the topic of software rendering, im surprised how little gustavo pezzi's lectures get mentioned here on hacker news.

by atan2

7/24/2026 at 1:40:24 AM

For anyone wondering, Gustavo runs https://pikuma.com/ which has a host of lectures on a variety of topics from PlayStation 1 programming to maths to triangle rasterisation.

by bananaboy

7/24/2026 at 5:24:08 PM

That's right. :-)

by gustavopezzi

7/24/2026 at 5:23:11 PM

Hey! Gustavo here. Thanks for the mention. The 3D software renderer course is still one of my favorites, even though it was one of the first ones I've published. I recently re-recorded the chapter on perspective projection matrix.

by gustavopezzi

7/24/2026 at 12:06:59 AM

Gustavo's lectures are great! I work through his Compiler lectures now and I bought the 3D renderer course too.

by AlexeyBrin

7/24/2026 at 5:25:41 PM

Glad to hear that! All the topics are incredibly fun to play around with.

by gustavopezzi

7/23/2026 at 5:34:05 PM

How does this differ from POVRay?

by brcmthrowaway

7/23/2026 at 6:12:48 PM

POVray is a declarative language used to render raytraced scenes. This is a tutorial that demonstrates how the rasterization technique works from a programmatic perspective.

by 42days

7/23/2026 at 4:44:24 PM

This is a very good article.

by fithisux

7/23/2026 at 9:29:28 PM

Is something broken on mobile? It has code for writing a few pixels to a tga and then shows you how to clone their repo and then shows some photos... There is.. no real article..?

Edit: Ah: there's a quiet hamburger menubar. The page really should have forward/next/toc links.

by layla5alive

7/24/2026 at 6:45:12 AM

[flagged]

by jis7gdhshs

7/24/2026 at 6:55:51 AM

For whatever reason you planted this on your head, there are people that can provide professional help.

by pjmlp

7/24/2026 at 7:02:24 AM

[dead]

by uiiqhsgxg

7/24/2026 at 7:06:53 AM

Again, seek professional help.

by pjmlp

7/24/2026 at 7:17:32 AM

[dead]

by uisodnvjjsjsh

7/23/2026 at 3:15:56 PM

"Comments

An error occurred: API rate limit already exceeded "

by chrisjj

7/23/2026 at 6:50:19 PM

Ok I'm following along but how do I view the tga files that this program produces? Windows can't open them.

by 0x1ceb00da

7/24/2026 at 12:10:43 AM

You can replace the tga with bmp which is natively supported by Windows and not difficult to implement (if you prefer to use a C library look for stb libs).

by AlexeyBrin

7/23/2026 at 6:52:56 PM

Irfanview?

by mhoela

7/24/2026 at 6:54:43 AM

Now that is an application I don't see in a long time, it used to be part of my default set of stuff to install on a new Windows PC.

by pjmlp

7/23/2026 at 4:11:35 PM

I thought this would be something new but it’s just ssloy’s tinyrenderer. Article should have a date since it’s old as dirt

by uncivilized

7/23/2026 at 6:40:58 PM

Just because you've seen something before doesn't mean it isn't new for some readers! This is my first time seeing it

by mootdentures

7/23/2026 at 10:01:44 PM

They just said it should have a date and it should.

by CyberDildonics

7/23/2026 at 4:55:33 PM

I've been wanting to write a series of articles exactly like this for a long time. Good that someone else has done it now =D

by Sharlin