MultiBeam application for ATi GPUs released

Message boards : Number crunching : MultiBeam application for ATi GPUs released
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 11 · Next

AuthorMessage
Saaby900T

Send message
Joined: 24 Dec 10
Posts: 76
Credit: 4,971,171
RAC: 0
United States
Message 1075519 - Posted: 9 Feb 2011, 15:01:32 UTC - in response to Message 1075514.  

I've got my 6970 onboard now. No seti work yet though. I've ran some Primegrid and Milkyway on it. I've noticed a major improvement on times with both. MW as usual runs the GPU very hot and forced me to run the Fan much higher and with much more noise.


Keep us/me posted I am looking into getting 2 of these cards to replace/Grow my Farm.
ID: 1075519 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1075526 - Posted: 9 Feb 2011, 15:34:33 UTC - in response to Message 1075514.  
Last modified: 9 Feb 2011, 15:38:45 UTC

Hi testers, stupid question (again), can't find the HD5 version of rev.177
for MB, or is there (only) this version.
Installed latest ATI drivers, (forgot/can't find the number)11.1 (?) and get OpenCL support, according to GPUz 0.51 .

Took out the 4850, bad compatabillity and heat problems, so gonna test with
1 HD5870, first, when working stable, add a HD5770, which is just lying around.
ID: 1075526 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34258
Credit: 79,922,639
RAC: 80
Germany
Message 1075531 - Posted: 9 Feb 2011, 15:46:10 UTC

There are 3 files in the package.

2 apps and the CL file.
Look at its names, one of them is the HD5 version.



With each crime and every kindness we birth our future.
ID: 1075531 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1075537 - Posted: 9 Feb 2011, 15:53:43 UTC - in response to Message 1075531.  

Thanks, hadn't opened the 'package', sorry, my (stupid) bad...


ID: 1075537 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1075577 - Posted: 9 Feb 2011, 18:46:28 UTC - in response to Message 1075537.  

Another 'smart question', should I leave the Optimized SSSE3x entries,
in the app_info.xml, and (if) the 6.03 'SSSE3-executables?
Don't know, what's going to happen if a VLAR, gets cru(n)ched, by ATI's 6.10?
Although, several people stated, ATI's do like VLAR's, I'm stunned, how
a 'personal' downtime,(head/eye-pain) affects my memory.....?! (Makes me feel real stupid :-/ )

Since I do have a few MB's, don't want to crash them, ohh boy,
having a 'hard time', putting an xml-file together..........
(Better disconnect from the Project, although it's down)
Lucky, the Forums are still working.

But now (I hope) it's gonna work, an hour rest or so, some coffee, should do the 'trick' ;).




ID: 1075577 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1075603 - Posted: 9 Feb 2011, 20:23:05 UTC - in response to Message 1075577.  

Another 'smart question', should I leave the Optimized SSSE3x entries,
in the app_info.xml, and (if) the 6.03 'SSSE3-executables?
Don't know, what's going to happen if a VLAR, gets cru(n)ched, by ATI's 6.10?
Although, several people stated, ATI's do like VLAR's, I'm stunned, how
a 'personal' downtime,(head/eye-pain) affects my memory.....?! (Makes me feel real stupid :-/ )

Since I do have a few MB's, don't want to crash them, ohh boy,
having a 'hard time', putting an xml-file together..........
(Better disconnect from the Project, although it's down)
Lucky, the Forums are still working.

But now (I hope) it's gonna work, an hour rest or so, some coffee, should do the 'trick' ;).


Leave the SSSE3 entries in the app_info, don't worry about the running the VLAR's on the GPU, while they do take longer, it's not as bad as running them on a Nvidia GPU,

Claggy
ID: 1075603 · Report as offensive
aad

Send message
Joined: 3 Apr 99
Posts: 101
Credit: 204,131,099
RAC: 26
Netherlands
Message 1075657 - Posted: 9 Feb 2011, 22:37:38 UTC - in response to Message 1075577.  
Last modified: 9 Feb 2011, 22:47:42 UTC

ohh boy,having a 'hard time', putting an xml-file together..........


I just used the unified installer (0.37) and then added the part that Raistmer wrote here http://lunatics.kwsn.net/15-optimized-applications-release-news/ati-opencl-multibeam-application-released-rev177.msg35371.html;topicseen in the xml-file.
Copied the files from the download to the right dir. and started Boinc again.

I only tried two jobs together on my HD6970 and It crunched just fine.

On Milkyway there was no gain in doin'2 jobs together. (it was worse)
ID: 1075657 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1075676 - Posted: 9 Feb 2011, 23:25:44 UTC - in response to Message 1075577.  
Last modified: 9 Feb 2011, 23:27:42 UTC

Amazing, how fast you 'forget', better unfamiliar, with software 'coding', I've mostly learned the basics of electricity, electronics, most things about computers were from were you worked.
But I figured it out, I just have to 'see' the logic, otherwise.....

Unfortunatly, most of the SERVERS [As of 9 Feb 2011 | 23:00:05 UTC]are
DOWN or DIABLED
Server status
Program Host Status
BOINC master database carolyn Running
BOINC replica database jocelyn Running
SETI@home science database oscar Running
Astropulse science database marvin Running
data-driven web pages thinman Running
scheduling server bane Disabled
scheduler process bane Disabled
download server 1 anakin Running
download server 2 vader Running
upload server bruno Running, but no chance UPLoading, cause [i]Sceduler is Disabled, strange downloads are a No Go, too

Hopefully, someone has enough time, tomorrow, it's UTC+9 hours, Berkeley, or 8,
New York is +6UTC(?), is there a clue as to what happened, excatly?
Except one of the Storage Servers, with 24 drives, had difficulties, BOOTing from those disks, which appears, randomly choosen and numbered...
Not funny, if you have to deal with such a Terra Byte Storage.
For now I stick too SATA 2 or 3, when available.
Although, some situations, I had to put 3500 photos on a USB stick, appeared to have choosen an 1.1 point, it took 40 minutes, (with or without BOINC, though),
I checked, cause of the low 'speed'.
ID: 1075676 · Report as offensive
Profile Fred J. Verster
Volunteer tester
Avatar

Send message
Joined: 21 Apr 04
Posts: 3252
Credit: 31,903,643
RAC: 0
Netherlands
Message 1075814 - Posted: 10 Feb 2011, 15:19:02 UTC - in response to Message 1075676.  

Microsoft Windows XP [versie 5.1.2600]
(C) Copyright 1985-2001 Microsoft Corp.

C:\Documents and Settings\Gebruiker>CLinfo
Number of platforms: 1
Platform Profile: FULL_PROFILE
Platform Version: OpenCL 1.1 ATI-Stream-v2.3 (451)
Platform Name: ATI Stream
Platform Vendor: Advanced Micro Devices, Inc.
Platform Extensions: cl_khr_icd cl_amd_event_callback cl_amd_offline_devices


Platform Name: ATI Stream
Number of devices: 2
Device Type: CL_DEVICE_TYPE_GPU
Device ID: 4098
Max compute units: 20
Max work items dimensions: 3
Max work items[0]: 256
Max work items[1]: 256
Max work items[2]: 256
Max work group size: 256
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 0
Native vector width char: 0
Native vector width short: 0
Native vector width int: 0
Native vector width long: 0
Native vector width float: 0
Native vector width double: 0
Max clock frequency: 837Mhz
Address bits: 32
Max memory allocation: 134217728
Image support: Yes
Max number of images read arguments: 128
Max number of images write arguments: 8
Max image 2D width: 8192
Max image 2D height: 8192
Max image 3D width: 2048
Max image 3D height: 2048
Max image 3D depth: 2048
Max samplers within kernel: 16
Max size of kernel argument: 1024
Alignment (bits) of base address: 32768
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: No
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: Yes
Cache type: None
Cache line size: 0
Cache size: 0
Global memory size: 536870912
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Scratchpad
Local memory size: 32768
Kernel Preferred work group size multiple: 64
Error correction support: 0
Unified memory for Host and Device: 0
Profiling timer resolution: 1
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: No
Queue properties:
Out-of-Order: No
Profiling : Yes
Platform ID: 00CFA40C
Name: Cypress
Vendor: Advanced Micro Devices, Inc.
Driver version: CAL 1.4.900
Profile: FULL_PROFILE
Version: OpenCL 1.1 ATI-Stream-v2.3 (451)
Extensions: cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomi
cs cl_khr_local_int32_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_amd_device_attribute_query cl_amd_printf cl_amd
_media_ops cl_amd_popcnt


Device Type: CL_DEVICE_TYPE_CPU
Device ID: 4098
Max compute units: 4
Max work items dimensions: 3
Max work items[0]: 1024
Max work items[1]: 1024
Max work items[2]: 1024
Max work group size: 1024
Preferred vector width char: 16
Preferred vector width short: 8
Preferred vector width int: 4
Preferred vector width long: 2
Preferred vector width float: 4
Preferred vector width double: 0
Native vector width char: 16
Native vector width short: 8
Native vector width int: 4
Native vector width long: 2
Native vector width float: 4
Native vector width double: 0
Max clock frequency: 2405Mhz
Address bits: 32
Max memory allocation: 536870912
Image support: No
Max size of kernel argument: 4096
Alignment (bits) of base address: 1024
Minimum alignment (bytes) for any datatype: 128
Single precision floating point capability
Denorms: Yes
Quiet NaNs: Yes
Round to nearest even: Yes
Round to zero: Yes
Round to +ve and infinity: Yes
IEEE754-2008 fused multiply-add: No
Cache type: Read/Write
Cache line size: 64
Cache size: 32768
Global memory size: 1073741824
Constant buffer size: 65536
Max number of constant args: 8
Local memory type: Global
Local memory size: 32768
Kernel Preferred work group size multiple: 1
Error correction support: 0
Unified memory for Host and Device: 1
Profiling timer resolution: 0
Device endianess: Little
Available: Yes
Compiler available: Yes
Execution capabilities:
Execute OpenCL kernels: Yes
Execute native function: Yes
Queue properties:
Out-of-Order: No
Profiling : Yes
Platform ID: 00CFA40C
Name: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz
Vendor: GenuineIntel
Driver version: 2.0
Profile: FULL_PROFILE
Version: OpenCL 1.1 ATI-Stream-v2.3 (451)
Extensions: cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomi
cs cl_khr_local_int32_extended_atomics cl_khr_byte_addressable_store cl_khr_gl_sharing cl_ext_device_fission cl_amd_device_attribute_query cl_amd_media_ops cl_a
md_popcnt cl_amd_printf

CLInfo, above, looks OK?
But refuses to use the 5870 GPU, check the app_info.xml file, again.

ID: 1075814 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1075848 - Posted: 10 Feb 2011, 17:34:47 UTC - in response to Message 1075814.  

what errors are you getting? does it work on MW or another project? Have you added the following app_info to your current app_info?




<app>
<name>setiathome_enhanced</name>
</app>
<file_info>
<name>MB_6.10_win_SSE3_ATI_HD5_r177.exe</name>
<executable/>
</file_info>
<file_info>
<name>MultiBeam_Kernels.cl</name>
<executable/>
</file_info>
<app_version>
<app_name>setiathome_enhanced</app_name>
<version_num>610</version_num>
<platform>windows_intelx86</platform>
<avg_ncpus>0.05</avg_ncpus>
<max_ncpus>0.05</max_ncpus>
<plan_class>ati13ati</plan_class>
<cmdline>-period_iterations_num 2 -instances_per_device 2</cmdline>
<flops>20987654321</flops>
<file_ref>
<file_name>MB_6.10_win_SSE3_ATI_HD5_r177.exe</file_name>
<main_program/>
</file_ref>
<file_ref>
<file_name>MultiBeam_Kernels.cl</file_name>
<copy_file/>
</file_ref>
<coproc>
<type>ATI</type>
<count>0.5</count>
</coproc>
</app_version>



In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1075848 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1075876 - Posted: 10 Feb 2011, 18:50:58 UTC - in response to Message 1075814.  
Last modified: 10 Feb 2011, 18:53:30 UTC

Your ATI host only has three CPU MB tasks, how are you expecting r177 to work if you're got no ATI MB tasks and the project is down?

Claggy
ID: 1075876 · Report as offensive
aad

Send message
Joined: 3 Apr 99
Posts: 101
Credit: 204,131,099
RAC: 26
Netherlands
Message 1076269 - Posted: 11 Feb 2011, 22:24:27 UTC - in response to Message 1075876.  

Your ATI host only has three CPU MB tasks, how are you expecting r177 to work if you're got no ATI MB tasks and the project is down?

Claggy


A rescheduler....?
http://www.efmer.eu/forum_tt/index.php?topic=428.0
ID: 1076269 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34258
Credit: 79,922,639
RAC: 80
Germany
Message 1076276 - Posted: 11 Feb 2011, 22:38:09 UTC - in response to Message 1076269.  

Your ATI host only has three CPU MB tasks, how are you expecting r177 to work if you're got no ATI MB tasks and the project is down?

Claggy


A rescheduler....?
http://www.efmer.eu/forum_tt/index.php?topic=428.0


Yep, i´m glad to have it.
I still have 1100 units for my CPU to crunch.
I reschedule 100 to GPU each day to feed my GPU.
Works fine.



With each crime and every kindness we birth our future.
ID: 1076276 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1076295 - Posted: 11 Feb 2011, 23:35:16 UTC - in response to Message 1076269.  

Your ATI host only has three CPU MB tasks, how are you expecting r177 to work if you're got no ATI MB tasks and the project is down?

Claggy


A rescheduler....?
http://www.efmer.eu/forum_tt/index.php?topic=428.0


I've seen no evidence of Fred using it on his other hosts, he's unlikely to even have those three tasks to crunch since all his recent CPU tasks have been crunched and reported within a few hours,
So 6 days on you really think those 3 tasks are still uncrunched by the CPU?

Claggy
ID: 1076295 · Report as offensive
aad

Send message
Joined: 3 Apr 99
Posts: 101
Credit: 204,131,099
RAC: 26
Netherlands
Message 1076504 - Posted: 12 Feb 2011, 18:56:36 UTC - in response to Message 1076295.  

Your ATI host only has three CPU MB tasks, how are you expecting r177 to work if you're got no ATI MB tasks and the project is down?

Claggy


A rescheduler....?
http://www.efmer.eu/forum_tt/index.php?topic=428.0


I've seen no evidence of Fred using it on his other hosts, he's unlikely to even have those three tasks to crunch since all his recent CPU tasks have been crunched and reported within a few hours,
So 6 days on you really think those 3 tasks are still uncrunched by the CPU?

Claggy


I do not need any evidence.
It's not my problem, just literally answering your question!

Aad
ID: 1076504 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1078007 - Posted: 16 Feb 2011, 19:45:25 UTC - in response to Message 1076504.  

comparison of 2 similar WU's both ran 2 at a time
the first is from my 6970 running the non HD5 r177 Angle range 0.411

[url] http://setiathome.berkeley.edu/result.php?resultid=1802106239[/url]

the second from my 5850 running the HD5 r277

angle range 0.42

http://setiathome.berkeley.edu/result.php?resultid=1801866388

this WU was ran on my 6970 but I believe is wasn't run 2 at a time

[/url]http://setiathome.berkeley.edu/result.php?resultid=1802106233[url]

It looks like the 6970 is much faster than the 5850 even when slowed down by using the 4XXX r177 app. I mention this because it hasn't seemed to be all that much faster on other projects. Perhaps like the HD5 app the other projects havent gotten the kernel working correctly with the Card[/url]


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1078007 · Report as offensive
Profile Gundolf Jahn

Send message
Joined: 19 Sep 00
Posts: 3184
Credit: 446,358
RAC: 0
Germany
Message 1078023 - Posted: 16 Feb 2011, 20:04:02 UTC - in response to Message 1078007.  

http://setiathome.berkeley.edu/result.php?resultid=1802106239

http://setiathome.berkeley.edu/result.php?resultid=1801866388

http://setiathome.berkeley.edu/result.php?resultid=1802106233

There mustn't be any blanks within a URL (or around it) and it's easier to highlight the URL and then click the URL button once than to rely on the toggle function of the button. ;-)

Gruß,
Gundolf
ID: 1078023 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1078055 - Posted: 16 Feb 2011, 21:13:18 UTC - in response to Message 1078023.  

http://setiathome.berkeley.edu/result.php?resultid=1802106239

http://setiathome.berkeley.edu/result.php?resultid=1801866388

http://setiathome.berkeley.edu/result.php?resultid=1802106233

There mustn't be any blanks within a URL (or around it) and it's easier to highlight the URL and then click the URL button once than to rely on the toggle function of the button. ;-)

Gruß,
Gundolf

actually its a bad case of dyslexia notice the / placement DOH


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1078055 · Report as offensive
baron_iv
Volunteer tester
Avatar

Send message
Joined: 4 Nov 02
Posts: 109
Credit: 104,905,241
RAC: 0
United States
Message 1078137 - Posted: 16 Feb 2011, 23:46:42 UTC

Just so everyone knows, ATI released a new driver for their GPUs yesterday, version 11.2. So far, I have had ZERO driver restarts when running the multibeam application, even when I am running 2 tasks per GPU on the 6950. This is very good news!!

I have switched to the NON-HD5 version too, because I was having trouble with the HD5 version.

It also works flawlessly with the Astropulse application from lunatics.

Let the ATI/AMD fun begin! :)

http://sites.amd.com/us/game/downloads/Pages/downloads.aspx

You can choose your OS version there.

I am running on Winblowz 7 x64. Just be sure to uninstall your old drivers and SDK before you install the new drivers/sdk or you will have issues.

Thanks to Raistmer for the app and the help getting it set up properly.
-baron_iv
Proud member of:
GPU Users Group
ID: 1078137 · Report as offensive
Saaby900T

Send message
Joined: 24 Dec 10
Posts: 76
Credit: 4,971,171
RAC: 0
United States
Message 1078166 - Posted: 17 Feb 2011, 1:00:37 UTC - in response to Message 1078137.  

Just so everyone knows, ATI released a new driver for their GPUs yesterday, version 11.2. So far, I have had ZERO driver restarts when running the multibeam application, even when I am running 2 tasks per GPU on the 6950. This is very good news!!

I have switched to the NON-HD5 version too, because I was having trouble with the HD5 version.

It also works flawlessly with the Astropulse application from lunatics.

Let the ATI/AMD fun begin! :)

http://sites.amd.com/us/game/downloads/Pages/downloads.aspx

You can choose your OS version there.

I am running on Winblowz 7 x64. Just be sure to uninstall your old drivers and SDK before you install the new drivers/sdk or you will have issues.

Thanks to Raistmer for the app and the help getting it set up properly.


Good to know that the New drivers seem to be helping.
ID: 1078166 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · 6 · 7 · 8 . . . 11 · Next

Message boards : Number crunching : MultiBeam application for ATi GPUs released


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.