TFW 8-bit integers are too small, but 16-bit integers are too big, so you go with 12-bit integers… and it actually results in a nice perf win
-
Show this thread
-
Replying to @pcwalton
I've done 24 bit on AVR combined with hand-picked multiplication intrinsics. I'm assuming this is in some SIMD or GPU context?
1 reply 0 retweets 0 likes
Replying to @pepijndevos
Yeah, it’s for packing vertex buffers so that every piece of per-instance data is exactly 64 bits.
7:33 AM - 3 Jan 2019
0 replies
0 retweets
3 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.