Sign in

quail

@quail.wtf
126 followers 169 following 289 posts

A cursed quail. I think too much about databases.

PostsRepliesMedia
quail @quail.wtf · 14/02/2026
I've been scraping the metro transit GTFS feed for about 36 hours and it's kind of lovely to see the activity ebb and flow over the course of the day
020
quail @quail.wtf · 31/08/2025
would you believe that I've never done this before? (hello triangle with vulkan)
a multicolored triangle, the kind you would get from running through the basics of the vulkan tutorial
130
quail @quail.wtf · 31/05/2025
it really is this meme IMO. I think it takes seeing the outputs "live" for it to really click. Auto-instrumentation just doesn't know your domain.
190
quail @quail.wtf · 01/03/2025
well would you look at that, something's wrong with how lights work or something else with reflection, anyway.
000
quail @quail.wtf · 01/03/2025
Long time no outputs. I've been busy with other stuff. Lots of things to criticize in here - clearly I'm not handling transmission correctly (the sphere should be glass), and the reflection from the cube should be perfect and isn't. The teapot is way too bright, and the shadows are way too harsh.
100
quail @quail.wtf · 26/02/2025
Huge shout-out to @daisylovesart.bsky.social for this piece - it turned out amazing and is exactly what I wanted!
063
quail @quail.wtf · 01/02/2025
and we're back with more funky images. Currently trying to wrap my brain around the SimplePathIntegrator from pbrt - it's such a huge jump from a random walk integrator
120
quail @quail.wtf · 18/01/2025
Well clearly something is messed up I've changed the scene so it's entirely closed, which means rays will always bounce as many times as they can. I think this shows I'm weighting the results of bounces too highly.
120
quail @quail.wtf · 17/01/2025
honestly that looks better than I expected for 8 samples. BUT IT COULD BE BETTER
000
quail @quail.wtf · 17/01/2025
now that's a funky image
220
quail @quail.wtf · 15/01/2025
Sharing a work-in-progress image. Yes this looks like ass but it's 1 sample, 3 bounces, and I'm not even properly doing light sampling properly.
The bunny scene from previous posts. The image is very noisy due to the low sample rate
010
quail @quail.wtf · 14/01/2025
and now in color (and with correct aspect ratio calculations, in case anyone noticed :) )
the bunny scene, except now the bunny is green, the left wall is red, the right wall is green, and the background is very light turquoise.
120
quail @quail.wtf · 12/01/2025
Same scene, but now powered by BVH trees so I can render a pretty high-res image in not a lot of time.
The bunny scene, now in high res!
010
quail @quail.wtf · 09/01/2025
we are COOKING
the bunny scene, again, this time with shading that looks... maybe not realistic but it's at least pretty smooth
020
quail @quail.wtf · 09/01/2025
These normals are so normal
the bunny image, but it's multicolored because each pixel is the normal value at that point
000
quail @quail.wtf · 08/01/2025
lets go
a render of a bunny in a box. The bunny has a distinct shadow, and is slightly outlined by shadow
110
quail @quail.wtf · 05/01/2025
Does this count for genuary
A fucked up rendering of a cube. There are clearly many bugs with how it was rendered.
010
quail @quail.wtf · 31/12/2024
The fates have no power over me
An embroidery of the three fates
020
quail @quail.wtf · 27/12/2024
and all it cost was a huge performance regression (memcpy is NOT FREE kiddos)
a render of the stanford bunny with no obvious bugs
100
quail @quail.wtf · 23/12/2024
lmao and the annoying "wtf why is this happening" bugs start
an image of a red rabbit against a green background, except some of the scanlines are off for unknown reasons
100
quail @quail.wtf · 01/12/2024
lets go (see disclaimer below)
a screenshot of hyperfine comparing quail-olap to duckdb for the one billion rows challenge. The results have quail-olap 2.4 times faster than duckdb, or 674ms vs 1.6s
120
quail @quail.wtf · 27/11/2024
bsky
an image of bisky from hunter x hunter
110