@ENDESGA

Programmer @NightdiveStudio :::. C, Vulkan :::. Math-obsessed https://nitter.cf/t.co/aIaJr2hqGD :::. Pixel Artist :::. EDG32 palette :::. deity of dodecahedra

 HEPTAVERSE
patreon.com/endesga
Joined June 2011
  • Tweets54K
  • Following294
  • Followers22K
7.5KMedia
GPUs are overrated 😏 (this takes about 0.1ms per frame on my i9 9900K... it's fast.)
End@ENDESGA
Sep 26
The idea I had was to use some form of Bresenham's line algorithm to draw lines for the edges (two at a time) so that the rows can be filled very quickly as the edges are traced. Performance seems to be extremely good, which is pretty cool
22
16
3
539
64,357
A gentle reminder that I’m just enjoying my weekend, please don’t take what I post so seriously. Yes I know a GPU can do this faster, that’s not the point 🫩
12
1,488
Single-threaded of course, it just heavily relies on auto-vectorization to blit pixels as fast as the CPU possibly can. Rendering a whole lot of cubes would require complex culling, but it's not impossible. The cool thing is that this .exe is compatible with every PC, and is 43KB
3
1
27
1,962
The idea I had was to use some form of Bresenham's line algorithm to draw lines for the edges (two at a time) so that the rows can be filled very quickly as the edges are traced. Performance seems to be extremely good, which is pretty cool
End@ENDESGA
Sep 25
I wanted to use a C rasterization technique that combined everything I've learned over the past many years. With GCC's auto-vectorization I knew I'd be able to render extremely efficiently without the GPU With just whole colors you can do some really cool optimizations...
3
9
3
180
73,093
I wanted to use a C rasterization technique that combined everything I've learned over the past many years. With GCC's auto-vectorization I knew I'd be able to render extremely efficiently without the GPU With just whole colors you can do some really cool optimizations...
2
2
3
208
21,278
"hey why have you gotten so obsessed with C and software rendering over the past few years?"
8
29
2
528
30,534
In my C code I've gotten a little obsessed with "automated" bitfields, after discovering that you can use the auto-count trick in an enum to determine how many bits are needed to hold the values. Is there a better way to do this? I'm curious if anyone else has explored this...
13
5
276
19,961
this weekend I was playing with some old code from last year the idea was an emphasis on rain and growth, but just like last year I couldn't think of any worthwhile game mechanics... so it'll probably sit in my folder for another year 😅 what would you expect from such tiny game?
9
3
96
3,698
the air smells sweet #pixelart
4
17
272
6,119
unfinished remnants under an ancient tree #pixelart
1
6
162
4,147
some quick typography cool down this evening. sorta y2k, inspired by the Planet Kosmos font, but in sharp 45 degree angles~
1
1
19
1,313
I unironically think this. With today's CPUs, if we programmed efficiently with low-level code there's a lot of headroom. You'll be surprised how quickly your CPU can render a complex scene to a window. With SIMD and other optimizations it's quite impressive what you can do
David Amador 🐙@DJ_Link
Jun 25
Let’s ditch GPUs and go back to software render
66
82
7
2,368
95,744
To clarify further: The reason I have this strong opinion is because you can't actually "just use" the GPU. You HAVE to use some library because AMD/Nvidia want to control how it's used. It sucks. I'll change my mind if someone gives me a GPU that I can throw function-pointers at
1
30
1,353
Wasn't expecting this many replies. I must add that I personally prefer low-res/fidelity graphics/games/tools/etc. Like of course you can't render Elden Ring in 4K 60fps with Software Rendering. I'm more talking Notepad, 720p games, tools, that sorta thing. It's just cool to me
4
146
6,623
Rewatched Backrooms yesterday before it's no longer in cinemas, but this time I binged Kane's youtube videos right beforehand to refresh my memory, along with the behind-the-scenes... and that made the experience so much better. What an amazing film, I love it so much
1
1
1,394
Interesting, these latest upgrades to pep I'm testing seem to compress better and faster than all formats (even max-JPEG XL). How does one advertise that their format is the best pixel art compression format on the planet without sounding egotistical? lmao Maybe a YouTube video??
3
19
2,762
5-minute 32x32 Apple Timelapse. was trying to get the subtle pinks and oranges in, was hard to do fast haha #pixelart
1
2
89
3,408
these timelapses use my HEPT32 palette you can find in Aseprite!
6
1,225
While testing I stumbled upon a way to match palette size to the right model, which skips the "test every model" code - and I found it compresses low-res (<=16 colors) even better in almost half the time! The 256-color images take a slight penalty, but I think this is worth it 🤔
End@ENDESGA
Jun 11
On a whim I emailed my old Calculus professor last week (he's the one who told me about Markov Chains), showing him my pep code and the ideas behind it. He just replied and suggested I look into "Context-Mixing" and "Matt Mahoney's PAQ" I now know what I'll be doing this weekend
5
1,664
On a whim I emailed my old Calculus professor last week (he's the one who told me about Markov Chains), showing him my pep code and the ideas behind it. He just replied and suggested I look into "Context-Mixing" and "Matt Mahoney's PAQ" I now know what I'll be doing this weekend
End@ENDESGA
Jun 7
Turns out it's only effective for small-palette images. The 64+ color images actually got worse, so I had to make it multi-pass; checking which 2D-tapping model compresses the best. Slightly slower, but <=16-color images compress a LOT, especially bitmaps. Tool link below:
2
2
1
58
5,677
After a bit of research it seems to be bigger-O than PPM; a lot more to compute and predict, so it'll probably be a lot slower decoding. I'm not sure I want to make it slower for a more-than-likely minimal size decrease... I could make it an option but I wanted pep to be minimal!
1
568
Load more