Private
Public Access
docs(twitter): add 1597798161665253376 corpus (NOTimothyLottes explicit packed 16-bit GLSL + FSR1)
7-post thread with @rianflo, Nov 2022. Confirms NOTimothyLottes wrote CAS/FSR1/etc GLSL versions with packed 16-bit math by hand. 'Explicit packed 16-bit works on AMD VK Vega and up. Up to 30% improvement on ALU bound stuff. Lots of occupancy wins.' Constants all packed and aliased as UINT (no conversion overheads). FSR1 source has 'F' (32-bit) and 'H' and 'Hx2' (packed 16-bit) function variants. Sibling to the Nov 2022 GPU tip-line thread (1588906002212323328).
This commit is contained in:
@@ -0,0 +1,46 @@
|
||||
---
|
||||
title: "@rianflo Explicit packed 16-bit code. FSR1 example: https://github.com/GPUOpen-E"
|
||||
author: "NOTimothyLottes"
|
||||
handle: "@NOTimothyLottes"
|
||||
post_url: "https://x.com/NOTimothyLottes/status/1597798161665253376"
|
||||
post_id: "1597798161665253376"
|
||||
timestamp: "2022-11-30 03:42:22"
|
||||
post_count: 7
|
||||
reply_count: 0
|
||||
repost_count: 0
|
||||
like_count: 1
|
||||
view_count: 0
|
||||
---
|
||||
|
||||
# @NOTimothyLottes — @rianflo Explicit packed 16-bit code. FSR1 example: https://github.com/GPUOpen-E
|
||||
|
||||
## Post 1 (2022-11-29 18:38:24)
|
||||
|
||||
"The road to 16-bit floats GPU is paved with our blood"
|
||||
:-/
|
||||
|
||||
https://www.yosoygames.com.ar/wp/2022/01/the-road-to-16-bit-floats-gpu-is-paved-with-our-blood/
|
||||
|
||||
## Post 2 (2022-11-29 22:19:53) — reply to Post 1
|
||||
|
||||
@rianflo Explicit packed 16-bit works on AMD Vulkan Vega and up. I typically get up to 30% improvement on ALU bound stuff. Lots of occupancy wins. I don't use {HLSL, RenderDoc, Reflection, RADV, or VS/PS}. All constants are packed and aliased as UINT, so no coversion overheads.
|
||||
|
||||
## Post 3 (2022-11-29 22:24:24) — reply to Post 2
|
||||
|
||||
@NOTimothyLottes Oh I know the benefits. Just no simple clear way to write it in GLSL for vulkan.
|
||||
|
||||
## Post 4 (2022-11-29 22:26:03) — reply to Post 3
|
||||
|
||||
@rianflo Sure there is. CAS/FSR1/etc all shipped with fantastic GLSL versions using 16-bit packed math (I wrote those), all which at the time got fantastic code generation using AMD's drivers.
|
||||
|
||||
## Post 5 (2022-11-29 22:32:10) — reply to Post 4
|
||||
|
||||
@NOTimothyLottes What GLSL extension did you use?
|
||||
|
||||
## Post 6 (2022-11-29 22:33:35) — reply to Post 5
|
||||
|
||||
@NOTimothyLottes Oh wait, you're saying you wrote the fp16 math manually?
|
||||
|
||||
## Post 7 (2022-11-30 03:42:22) — reply to Post 6
|
||||
|
||||
@rianflo Explicit packed 16-bit code. FSR1 example: https://github.com/GPUOpen-Effects/FidelityFX-FSR/blob/master/ffx-fsr/ffx_fsr1.h - There are different 'F' (32-bit) and 'H' and 'Hx2' (packed 16-bit) functions.
|
||||
@@ -0,0 +1,118 @@
|
||||
{
|
||||
"root_post_id": "1597798161665253376",
|
||||
"posts": [
|
||||
{
|
||||
"post_id": "1597661267845865474",
|
||||
"author": "Florian Hoenig",
|
||||
"handle": "rianflo",
|
||||
"text": "\"The road to 16-bit floats GPU is paved with our blood\"\n:-/\n\nhttps://www.yosoygames.com.ar/wp/2022/01/the-road-to-16-bit-floats-gpu-is-paved-with-our-blood/",
|
||||
"timestamp": "2022-11-29 18:38:24",
|
||||
"media_urls": [],
|
||||
"reply_to_id": null,
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 1,
|
||||
"repost_count": 0,
|
||||
"like_count": 4,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597717009369735169",
|
||||
"author": "NOTimothyLottes",
|
||||
"handle": "NOTimothyLottes",
|
||||
"text": "@rianflo Explicit packed 16-bit works on AMD Vulkan Vega and up. I typically get up to 30% improvement on ALU bound stuff. Lots of occupancy wins. I don't use {HLSL, RenderDoc, Reflection, RADV, or VS/PS}. All constants are packed and aliased as UINT, so no coversion overheads.",
|
||||
"timestamp": "2022-11-29 22:19:53",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597661267845865474",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 1,
|
||||
"repost_count": 0,
|
||||
"like_count": 1,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597718146663690240",
|
||||
"author": "Florian Hoenig",
|
||||
"handle": "rianflo",
|
||||
"text": "@NOTimothyLottes Oh I know the benefits. Just no simple clear way to write it in GLSL for vulkan.",
|
||||
"timestamp": "2022-11-29 22:24:24",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597717009369735169",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 1,
|
||||
"repost_count": 0,
|
||||
"like_count": 0,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597718560511385600",
|
||||
"author": "NOTimothyLottes",
|
||||
"handle": "NOTimothyLottes",
|
||||
"text": "@rianflo Sure there is. CAS/FSR1/etc all shipped with fantastic GLSL versions using 16-bit packed math (I wrote those), all which at the time got fantastic code generation using AMD's drivers.",
|
||||
"timestamp": "2022-11-29 22:26:03",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597718146663690240",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 1,
|
||||
"repost_count": 0,
|
||||
"like_count": 0,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597720097753542656",
|
||||
"author": "Florian Hoenig",
|
||||
"handle": "rianflo",
|
||||
"text": "@NOTimothyLottes What GLSL extension did you use?",
|
||||
"timestamp": "2022-11-29 22:32:10",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597718560511385600",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 2,
|
||||
"repost_count": 0,
|
||||
"like_count": 0,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597720454000541696",
|
||||
"author": "Florian Hoenig",
|
||||
"handle": "rianflo",
|
||||
"text": "@NOTimothyLottes Oh wait, you're saying you wrote the fp16 math manually?",
|
||||
"timestamp": "2022-11-29 22:33:35",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597720097753542656",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 1,
|
||||
"repost_count": 0,
|
||||
"like_count": 0,
|
||||
"view_count": 0
|
||||
}
|
||||
},
|
||||
{
|
||||
"post_id": "1597798161665253376",
|
||||
"author": "NOTimothyLottes",
|
||||
"handle": "NOTimothyLottes",
|
||||
"text": "@rianflo Explicit packed 16-bit code. FSR1 example: https://github.com/GPUOpen-Effects/FidelityFX-FSR/blob/master/ffx-fsr/ffx_fsr1.h - There are different 'F' (32-bit) and 'H' and 'Hx2' (packed 16-bit) functions.",
|
||||
"timestamp": "2022-11-30 03:42:22",
|
||||
"media_urls": [],
|
||||
"reply_to_id": "1597720454000541696",
|
||||
"quote_of_id": null,
|
||||
"metrics": {
|
||||
"reply_count": 0,
|
||||
"repost_count": 0,
|
||||
"like_count": 1,
|
||||
"view_count": 0
|
||||
}
|
||||
}
|
||||
],
|
||||
"source_url": "https://x.com/NOTimothyLottes/status/1597798161665253376"
|
||||
}
|
||||
Reference in New Issue
Block a user