blender3.3 HIP performance down 30% on Windows #100891

New Issue

周建峰 · 2022-09-07T23:07:37+02:00

周建峰 commented

2022-09-07 23:07:37 +02:00

System Information
Operating system:windows10 21h2
Graphics card:rx6800xt

Blender Version：3.3 Long-term support version
Broken: Version ：3.2 has no problems
Worked: no
Short description of error3.2
3.2 version of the classroom scene only needs 32s, 3.3lts need 44s
Standard classroom, no need for me to provide it.

**System Information** Operating system:windows10 21h2 Graphics card:rx6800xt **Blender Version**：3.3 Long-term support version Broken: Version ：3.2 has no problems Worked: no **Short description of error**3.2 3.2 version of the classroom scene only needs 32s, 3.3lts need 44s Standard classroom, no need for me to provide it. ![3.3lts.png](https://archive.blender.org/developer/F13465926/3.3lts.png) ![3.2.jpeg](https://archive.blender.org/developer/F13465925/3.2.jpeg)

周建峰 commented

2022-09-07 23:07:37 +02:00

Added subscriber: @2905710881

blender-admin commented

2022-09-07 23:07:37 +02:00

#101151 was marked as duplicate of this issue

周建峰 commented

2022-09-08 00:02:02 +02:00

![3.3lts.png](https://archive.blender.org/developer/F13466009/3.3lts.png)

Germano Cavalcante commented

2022-09-08 22:17:52 +02:00

Added subscriber: @mano-wii

Germano Cavalcante commented

2022-09-08 22:17:52 +02:00

It must be a problem only with AMD GPUs as I didn't notice any difference using an RTX 3060.

Operating system: Windows-10-10.0.22000-SP0 64 Bits
Graphics card: NVIDIA GeForce RTX 3060 Laptop GPU/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 516.94

It must be a problem only with AMD GPUs as I didn't notice any difference using an RTX 3060. --- Operating system: Windows-10-10.0.22000-SP0 64 Bits Graphics card: NVIDIA GeForce RTX 3060 Laptop GPU/PCIe/SSE2 NVIDIA Corporation 4.5.0 NVIDIA 516.94

Omar Emara changed title from ~~blender3.3 hpi performance down 30%~~ to blender3.3 HIP performance down 30%

2022-09-13 13:59:07 +02:00

Omar Emara commented

2022-09-13 14:11:46 +02:00

Added subscribers: @BrianSavery, @OmarEmaraDev

Omar Emara commented

2022-09-13 14:11:46 +02:00

Changed status from 'Needs Triage' to: 'Needs Developer To Reproduce'

Omar Emara commented

2022-09-13 14:11:46 +02:00

@BrianSavery Are you aware about any slow downs in 3.3? It seem Open Data queries seems to confirm some slow down, see:

@BrianSavery Are you aware about any slow downs in 3.3? It seem Open Data queries seems to confirm some slow down, see: - https://opendata.blender.org/benchmarks/query/?device_name=AMD%20Radeon%20RX%206800%20XT&compute_type=HIP&blender_version=3.3.0 - https://opendata.blender.org/benchmarks/query/?device_name=AMD%20Radeon%20RX%206800%20XT&compute_type=HIP&blender_version=3.2.0

Omar Emara changed title from ~~blender3.3 HIP performance down 30%~~ to blender3.3 HIP performance down 30% on Windows

2022-09-13 14:12:00 +02:00

Brian Savery (AMD) commented

2022-09-13 15:15:05 +02:00

Added subscriber: @Sayak-Biswas

Brian Savery (AMD) commented

2022-09-13 15:15:05 +02:00

@Sayak-Biswas is taking a look.

Sayak Biswas commented

2022-09-13 21:30:05 +02:00

Added subscriber: @sayakbiswas

Sayak Biswas commented

2022-09-13 21:30:05 +02:00

@2905710881 Are those two numbers on the same driver or different?

周建峰 commented

2022-09-15 04:12:56 +02:00

In #100891#1416621, @sayakbiswas wrote:
@2905710881 Are those two numbers on the same driver or different?

Similarly, I tested several drivers with the same result, including 7.1,8.1,8.2,5.2 22q2, each time using Display Driver Uninstaller to uninstall

> In #100891#1416621, @sayakbiswas wrote: > @2905710881 Are those two numbers on the same driver or different? Similarly, I tested several drivers with the same result, including 7.1,8.1,8.2,5.2 22q2, each time using Display Driver Uninstaller to uninstall

周建峰 commented

2022-09-16 09:59:28 +02:00

In #100891#1416621, @sayakbiswas wrote:
@2905710881 Are those two numbers on the same driver or different?

I purposely downloaded version 3.22 to compare the run, the current driver is 22Q2 (I tested multiple versions of the driver, each time using the duu tool to uninstall)

> In #100891#1416621, @sayakbiswas wrote: > @2905710881 Are those two numbers on the same driver or different? I purposely downloaded version 3.22 to compare the run, the current driver is 22Q2 (I tested multiple versions of the driver, each time using the duu tool to uninstall) ![T@ZY%KNTF)4QVC%~70~L_4N.png](https://archive.blender.org/developer/F13498556/T_ZY_KNTF_4QVC__70_L_4N.png)

Jesse Yurkovich commented

2022-09-17 23:56:10 +02:00

Added subscriber: @Pionpouage

Jesse Yurkovich commented

2022-09-17 23:56:43 +02:00

Added subscriber: @deadpin

Jesse Yurkovich commented

2022-09-17 23:56:43 +02:00

Additional data from the merged report:

Benchmark using 3.2.1:
https://opendata.blender.org/benchmarks/4b575db1-e191-401a-898d-e4e33b3820a8/

Benchmark using 3.3.0:
https://opendata.blender.org/benchmarks/28b80b4a-fb76-4af8-8b8c-c4599b01c6e2/

Additional data from the merged report: Benchmark using 3.2.1: https://opendata.blender.org/benchmarks/4b575db1-e191-401a-898d-e4e33b3820a8/ Benchmark using 3.3.0: https://opendata.blender.org/benchmarks/28b80b4a-fb76-4af8-8b8c-c4599b01c6e2/

Sayak Biswas commented

2022-09-20 00:42:00 +02:00

Okay it seems this commit is causing the performance differential: 023eb2ea7c
On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse?

Okay it seems this commit is causing the performance differential: https://github.com/blender/blender/commit/023eb2ea7c16a00272f83d564145e28aeb9ed2b7 On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658349504dc98c693c8c46c30e9a60c44) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse?

周建峰 commented

2022-09-20 09:08:35 +02:00

In #100891#1419724, @Sayak-Biswas wrote:
Okay it seems this commit is causing the performance differential: 023eb2ea7c
On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse?

https://devtalk.blender.org/t/cycles-amd-hip-device-feedback/21400/405
There is a tested linux without performance degradation problem

> In #100891#1419724, @Sayak-Biswas wrote: > Okay it seems this commit is causing the performance differential: https://github.com/blender/blender/commit/023eb2ea7c16a00272f83d564145e28aeb9ed2b7 > On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658349504dc98c693c8c46c30e9a60c44) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse? https://devtalk.blender.org/t/cycles-amd-hip-device-feedback/21400/405 There is a tested linux without performance degradation problem

nathan commented

2022-09-23 18:23:54 +02:00

In #100891#1419724, @Sayak-Biswas wrote:
Okay it seems this commit is causing the performance differential: 023eb2ea7c
On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse?

I can also confirm this behavior with my RX 6800XT.

With version "3.3.0 Alpha, branch: master, commit date: 2022-07-25 22:17, hash: 4cf6524731", the scene takes 44s to render.
With version "3.3.0 Alpha, branch: master, commit date: 2022-07-24 19:33, hash: 6db059e3d7", the build just before the commit in question, the scene takes 33s to render.

> In #100891#1419724, @Sayak-Biswas wrote: > Okay it seems this commit is causing the performance differential: https://github.com/blender/blender/commit/023eb2ea7c16a00272f83d564145e28aeb9ed2b7 > On my W6800, the classroom scene renders in about 59s on a build from this commit. However the previous commit (d567785658349504dc98c693c8c46c30e9a60c44) the scene renders in ~44s. I'm not entirely sure yet which change is exactly causing the issue, maybe the new transform_inverse? I can also confirm this behavior with my RX 6800XT. With version "3.3.0 Alpha, branch: master, commit date: 2022-07-25 22:17, hash: `4cf6524731`", the scene takes 44s to render. With version "3.3.0 Alpha, branch: master, commit date: 2022-07-24 19:33, hash: `6db059e3d7`", the build just before the commit in question, the scene takes 33s to render.

blender-admin commented

2022-10-03 17:48:03 +02:00

This issue was referenced by d20be55c1a

This issue was referenced by d20be55c1a9fc9f503a507fe47ace6ef5d70573e

Brecht Van Lommel commented

2022-10-03 17:58:23 +02:00

Added subscriber: @brecht

Brecht Van Lommel commented

2022-10-03 17:58:23 +02:00

I committed a change that may help but I don't really expect it to.

@Sayak-Biswas are you looking into this? The way I suggest to approach it is to effectively bisect that commit, step by step undo changes by that commit until you hit the change(s) that caused the performance difference.

Like, revert changes in ray_triangle_intersect, revert changes to use cross(), revert division by det to be done once in advance, etc.

I committed a change that may help but I don't really expect it to. @Sayak-Biswas are you looking into this? The way I suggest to approach it is to effectively bisect that commit, step by step undo changes by that commit until you hit the change(s) that caused the performance difference. Like, revert changes in ray_triangle_intersect, revert changes to use cross(), revert division by det to be done once in advance, etc.

Brecht Van Lommel commented

2022-10-03 18:00:07 +02:00

Changed status from 'Needs Developer To Reproduce' to: 'Confirmed'

Sayak Biswas commented

2022-10-03 21:26:01 +02:00

@brecht thanks, will try the change. We have the compiler team looking into the issue as well.

周建峰 commented

2022-10-14 14:09:03 +02:00

In #100891#1427026, @Sayak-Biswas wrote:
@brecht thanks, will try the change. We have the compiler team looking into the issue as well.

Can you restore the performance when 3.4 is released?

> In #100891#1427026, @Sayak-Biswas wrote: > @brecht thanks, will try the change. We have the compiler team looking into the issue as well. Can you restore the performance when 3.4 is released?

Brecht Van Lommel commented

2022-11-09 15:56:37 +01:00

The current status is that we expect to get a new HIP compiler version in the coming weeks that will fix this regression. This should then fix the issue in time for the 3.4 release, and also 3.3.2 or 3.3.3 LTS depending on the timing.