[Bug 94133] relatevly low performance rv740

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 1797 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

            Bug ID: 94133
           Summary: relatevly low performance rv740
           Product: Mesa
           Version: unspecified
          Hardware: Other
                OS: All
            Status: NEW
          Severity: normal
          Priority: medium
         Component: Drivers/Gallium/r600
          Assignee: dri-devel@lists.freedesktop.org
          Reporter: roman.elshin@gmail.com
        QA Contact: dri-devel@lists.freedesktop.org

Created attachment 121737
  --> https://bugs.freedesktop.org/attachment.cgi?id=121737&action=edit
Xorg.log

I just installed ubuntu 15.10 (with oibaf and commendsarnex ppa to see how good 
is my radeon hd4770 (rv740) works with fresh graphics stack).
Unigine Valley-1.0 (native) and MafiaII Game (wine with Gallium Nine) was used 
to see video performance differences against windows7.

Linux Unigine Valley Benchmark 1.0:
FPS:    8.5
Score:    356
Min FPS:4.2
Max FPS:13.8

Windows7 Unigine Valley Benchmark 1.0 (dx11 render, as opengl wasn`t able to
run):
FPS:    15.9
Score:    666
Min FPS:9.2
Max FPS:29.5

Linux Mafia2 Game (internal benchmark) under wine with Gallium Nine, dri3):
22.2fps

Windows7 Mafia2 Game(internal benchmark) under wine with Gallium Nine, dri3):
42.4fps

It works (without visible artifacts) and it is great!, but it is almost twice
as slow against windows. There are some benchmarks on phoronix
(http://www.phoronix.com/scan.php?page=article&item=radeon_1404_win81&num=3)
for more recent and powerfull r600g hw, and there was no such big differences
there.
So is it r700 classs hw works slow with gallium in general, or it something
wrong with just rv740 support? (or something else ...?)

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 3294 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 293 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

--- Comment #1 from Roman Elshin <roman.elshin@gmail.com> ---
Created attachment 121738
  --> https://bugs.freedesktop.org/attachment.cgi?id=121738&action=edit
dmesg

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 1128 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 507 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

Alex Deucher <alexdeucher@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |enhancement

--- Comment #2 from Alex Deucher <alexdeucher@gmail.com> ---
The open source Linux driver has not been as optimized as the windows driver.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 1741 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 372 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

--- Comment #3 from Roman Elshin <roman.elshin@gmail.com> ---
yes, but my quastion: is it optimized for rv740 not less than for other chips
supported by r600g (for some reasons, bugs, etc..)? performance difference at
first sight seems too big.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 1104 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 784 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

--- Comment #4 from John Bridgman <john.bridgman@amd.com> ---
I think the answer is "not knowingly". Optimizations tend to be done for an
entire generation (or a few generations), not for specific chips. 

That said, if a specific chip is observed to be an outlier (performance quite a
bit lower or higher than expected) then that specific discrepancy will probably
get some attention.

I don't understand your Mafia2 numbers - the Linux version wouldn't be running
under Nine, would it ? If the Linux version were running under OpenGL, then
this would look like a fairly typical "DX renderer in the app gets a lot more
work than the GL renderer" scenario.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 1536 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 328 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

--- Comment #5 from Roman Elshin <roman.elshin@gmail.com> ---
I will try to find different card from r600g driver for compare at the same
conditions.
Mafia2 works winth Nine :), without it - 15.8 fps.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 1060 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

[Bug 94133] relatevly low performance rv740

[bugzilla-daemon]

[-- Attachment #1.1: Type: text/plain, Size: 852 bytes --]

https://bugs.freedesktop.org/show_bug.cgi?id=94133

Roman Elshin <roman.elshin@gmail.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |WONTFIX

--- Comment #6 from Roman Elshin <roman.elshin@gmail.com> ---
HD4850 results:
Linux Unigine Valley Benchmark 1.0:
FPS:    10.2
Score:    428
Min FPS:4.2
Max FPS:19.3

Windows7 Unigine Valley Benchmark 1.0 (dx11 render):
FPS:    16.9
Score:    709
Min FPS:10.3
Max FPS:33.1

Linux Mafia2 Game (internal benchmark) under wine with Gallium Nine, dri3):
25fps

Windows7 Mafia2 Game(internal benchmark): 42.7fps
Not much differences.

-- 
You are receiving this mail because:
You are the assignee for the bug.

[-- Attachment #1.2: Type: text/html, Size: 2273 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel