Welcome to MilkyWay@home

Posts by Starfire

1) Message boards : News : New Separation Runs (Message 53558)
Posted 6 Mar 2012 by Starfire
Post:
Your Catalyst ist to new for the 4850. Downgrade to 11.9!

The newer driver doesn't incluedd something that the 4850 need to work.

See alos:

http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=2457&nowrap=true#52751

regardes Franz


That was in relation to the old app (0.82). CAL was broken for the HD4000 GPUs in Catalyst 11.11 - 12.1.

The new OpenCL-app (1.02) works for me (HD4870) with both Catalyst 12.1 and 12.2 (pre-cert). If I remember correctly OpenCL was broken in Catalyst 11.10 - 11.12.
2) Message boards : Number crunching : No ATI work with BOINC 7.0.18 (Message 53528)
Posted 3 Mar 2012 by Starfire
Post:
Version 7.?.? is still in Beta testing, meaning not all projects have embraced it yet.


That I'm aware of. I expected troubles before I made the switch ;) Although I thought they would lie with projects with old server versions like WCG *shrug*
I worked around the problem with Milkyway for myself by creating an appropriate app_info file to resume work fetch with 7.0.18.

I only mentioned it because a public release may not be that far off according to the 7.0.18 announcement on the alpha e-mail-list:

We believe we are really close to a public release, please report test
results to http://boinc.berkeley.edu/alpha as soon as possible.
3) Message boards : Number crunching : No ATI work with BOINC 7.0.18 (Message 53520)
Posted 2 Mar 2012 by Starfire
Post:
Hi,

today I installed Virtual Box and upgraded BOINC to 7.0.18 from 6.12.43 to test the new stuff. Since then I'm not getting any ATI tasks from Milkyway even through BOINC is requesting lots of them:

3/2/2012 9:11:27 PM | Milkyway@Home | [sched_op] Starting scheduler request
3/2/2012 9:11:27 PM | Milkyway@Home | Sending scheduler request: To fetch work.
3/2/2012 9:11:27 PM | Milkyway@Home | Reporting 1 completed tasks, requesting new tasks for ATI
3/2/2012 9:11:27 PM | Milkyway@Home | [sched_op] CPU work request: 0.00 seconds; 0.00 devices
3/2/2012 9:11:27 PM | Milkyway@Home | [sched_op] ATI work request: 77779.06 seconds; 0.00 devices
3/2/2012 9:11:29 PM | Milkyway@Home | Scheduler request completed: got 0 new tasks
3/2/2012 9:11:29 PM | Milkyway@Home | [sched_op] Server version 700
3/2/2012 9:11:29 PM | Milkyway@Home | No tasks sent
3/2/2012 9:11:29 PM | Milkyway@Home | Tasks for CPU are available, but your preferences are set to not accept them
3/2/2012 9:11:29 PM | Milkyway@Home | Project requested delay of 61 seconds
3/2/2012 9:11:29 PM | Milkyway@Home | [sched_op] handle_scheduler_reply(): got ack for task ps_separation_15_2s_20per_1_54219_1
3/2/2012 9:11:29 PM | Milkyway@Home | [sched_op] Deferring communication for 1 min 1 sec
3/2/2012 9:11:29 PM | Milkyway@Home | [sched_op] Reason: requested by project


The following is the coproc-section from the corresponding scheduler request:

 <coprocs>
<coproc_ati>
   <count>1</count>
   <name>ATI Radeon HD 4700/4800 (RV740/RV770)</name>
   <available_ram>1040187392.000000</available_ram>
   <have_cal>1</have_cal>
   <have_opencl>1</have_opencl>
   <req_secs>77779.061815</req_secs>
   <req_instances>0.000000</req_instances>
   <estimated_delay>0.000000</estimated_delay>
   <peak_flops>2400000000000.000000</peak_flops>
   <CALVersion>1.4.1703</CALVersion>
   <target>5</target>
   <localRAM>1024</localRAM>
   <uncachedRemoteRAM>2047</uncachedRemoteRAM>
   <cachedRemoteRAM>2047</cachedRemoteRAM>
   <engineClock>750</engineClock>
   <memoryClock>900</memoryClock>
   <wavefrontSize>64</wavefrontSize>
   <numberOfSIMD>10</numberOfSIMD>
   <doublePrecision>1</doublePrecision>
   <pitch_alignment>256</pitch_alignment>
   <surface_alignment>4096</surface_alignment>
   <maxResource1DWidth>8192</maxResource1DWidth>
   <maxResource2DWidth>8192</maxResource2DWidth>
   <maxResource2DHeight>8192</maxResource2DHeight>
    <atirt_detected/>
   <coproc_opencl>
      <name>ATI RV770</name>
      <vendor>Advanced Micro Devices, Inc.</vendor>
      <vendor_id>4098</vendor_id>
      <available>1</available>
      <half_fp_config>0</half_fp_config>
      <single_fp_config>62</single_fp_config>
      <double_fp_config>63</double_fp_config>
      <endian_little>1</endian_little>
      <execution_capabilities>1</execution_capabilities>
      <extensions>cl_khr_fp64 cl_amd_fp64 cl_khr_gl_sharing cl_amd_device_attribute_query cl_khr_d3d10_sharing </extensions>
      <global_mem_size>1073741824</global_mem_size>
      <local_mem_size>16384</local_mem_size>
      <max_clock_frequency>750</max_clock_frequency>
      <max_compute_units>10</max_compute_units>
      <opencl_platform_version>OpenCL 1.1 AMD-APP (898.1)</opencl_platform_version>
      <opencl_device_version>OpenCL 1.0 AMD-APP (898.1)</opencl_device_version>
      <opencl_driver_version>CAL 1.4.1703</opencl_driver_version>
   </coproc_opencl>
</coproc_ati>
    </coprocs>
4) Message boards : News : Separation updated to 1.00 (Message 53114)
Posted 12 Feb 2012 by Starfire
Post:
For some reason the frequency setting in the M@H preferences is ignored. The GPU runs @ 99% no matter what the setting. I tried a target frequency in the app-info file with separation 1.02 and it caused the GPU frequency and power draw to be all over the place. Had to switch back to 0.82 to get things running smoothly again. Only thing I can figure is it must be some kind of a hardware thing on my setup.


Strange

I'm running now without app_info and the default frequency of 60Hz -> GPU load 82-85%. Before I removed the app_info I was running with "--gpu-target-frequency 40". I left it in when I switched from 0.82 to 1.00. According to the results it was used and run times where lower than now (with 60Hz).
Maybe the varying GPUs react differently to the setting.

Edit: A quick test changing the web preferences to 40Hz: result
The setting change was recognized by the app -> GPU load 90-91%.
5) Message boards : News : Separation updated to 1.00 (Message 53094)
Posted 11 Feb 2012 by Starfire
Post:
What command would be added to the the app-info file to throttle back GPU usage? It's currently at 99% and I'd like to pull it back to 80-85%.
Win7 64 bit. Radeon 5830.


With the new app you don't need the app_info-file to control GPU load - it's now only necessary if you want to run multiple instances on one GPU. If you want to use it anyway the command line parameter apparently wasn't changed from the old app:
--gpu-target-frequency XX

Raising XX lessens the GPU load and increases the run time (default is now 60).

With the new app it's easier controlled in the Milkyway@home preferences: Frequency (in Hz) that should try to complete individual work chunks
6) Message boards : Number crunching : New app ? (Message 53016)
Posted 10 Feb 2012 by Starfire
Post:
Look inside here.
1.02 is an update to 1.00 - it didn't get it's own news item (like 1.00).
7) Message boards : News : Separation updated to 1.00 (Message 53012)
Posted 10 Feb 2012 by Starfire
Post:
Right -- but it did for me -- two different systems -- and one not all that much different from yours -- Phenom II 925, Win 7 64 bit, 4850 and BOINC 7.0.08.

FYI, 12.1 drivers don't break Collatz - I tried them.
Equipment
AMD Phenom 9850 4x, Radeon HD5830 on Win 7 PRO 64



The important difference is the GPU. CAL (which is used by Collatz) is only broken since 11.11 for HD4XXX-GPUs. Newer GPUs (like HD5XXXX) are working.

So far the 1.00 OpenCL app has worked well for my HD4870 using Catalyst 12.1 :)
Thank you very much Matt.
8) Message boards : News : Separation updated to 1.00 (Message 52937)
Posted 9 Feb 2012 by Starfire
Post:
R700 stuff (4770 and 4850, 4870) support OpenCL but just barely. There are various limitations and they're missing features nbody uses.


So this means 4XXX users can't upgrade to the new app and are stuck with 0.82 and Cat 11.10? (since CAL is broken in later versions)

9) Message boards : News : Raised GPU work limits (Message 52770)
Posted 1 Feb 2012 by Starfire
Post:
By the way, they're all running just fine on the latest AMD catalyst version 12.1.


Just on a side note: CAL is still broken in 12.1 for HD4xxx cards (and has been since 11.11) - looks like AMD really doesn't care.
10) Message boards : Number crunching : Upgraded ATI-drivers to 11.11 All WUs exits with computanional error (Message 51970)
Posted 14 Dec 2011 by Starfire
Post:
OP: I'm using CCC 11.11 on a HD5850 - so far everything's working fine. Maybe this isn't affecting all AMD cards?


Errors with Cat 11.11 have shown up on Collatz too. Apparently (at least) some HD4xxx cards are effected - never cards seem to be OK. Unfortunately I tested Collatz with the new Cat 11.12 and the problem is still occurring for me (HD4600) :(
11) Message boards : Number crunching : Bad WUs (Message 48746)
Posted 13 May 2011 by Starfire
Post:
A few from me (same error as mentioned above):

de_separation_10_3s_free_2_453478_1305305955_1
de_separation_10_3s_free_2_462196_1305307281_0
de_separation_10_3s_free_2_420164_1305301086_1


Error example:
<core_client_version>6.12.26</core_client_version>
<![CDATA[
<message>
Unzul´┐Żssige Funktion. (0x1) - exit code 1 (0x1)
</message>
<stderr_txt>
<search_application> milkywayathome_client separation 0.62 Windows x86 double CAL++ </search_application>
Found 1 CAL devices
Chose device 0

Device target:         CAL_TARGET_770
Revision:              2
CAL Version:           1.4.1385
Engine clock:          750 Mhz
Memory clock:          900 Mhz
GPU RAM:               1024
Wavefront size:        64
Double precision:      CAL_TRUE
Compute shader:        CAL_TRUE
Number SIMD:           10
Number shader engines: 1
Pitch alignment:       256
Surface alignment:     4096
Max size 2D:           { 8192, 8192 }

Estimated iteration time 284.281667 ms
Target frequency 40.000000 Hz, polling mode 1, using responsiveness factor of 1.000000
Dividing into 16 chunks
Integration range: { nu_steps = 640, mu_steps = 1600, r_steps = 1400 }
Using { 1, 16 } chunk(s) of size { 1400, 100 }
Integration time = 206.497489 s, average per iteration = 322.652326 ms
Integral 0 time = 208.028925 s
Estimated iteration time 71.070417 ms
Target frequency 40.000000 Hz, polling mode 1, using responsiveness factor of 1.000000
Dividing into 2 chunks
Integration range: { nu_steps = 640, mu_steps = 400, r_steps = 1400 }
Using { 1, 2 } chunk(s) of size { 1400, 200 }
Integration time = 51.262623 s, average per iteration = 80.097848 ms
Integral 1 time = 51.766143 s
Likelihood time = 8.690186 s
Non-finite result
Failed to calculate likelihood
<background_integral> 0.001134262006206 </background_integral>
<stream_integral>  24.680924661083985  1331.077010865039700  -0.317058930291751 </stream_integral>
<background_likelihood> -3.043218251293072 </background_likelihood>
<stream_only_likelihood>  -10.208126603949683  -8.771491003836179  -1.#IND00000000000 </stream_only_likelihood>
<search_likelihood> -1.#IND00000000000 </search_likelihood>
19:26:59 (3724): called boinc_finish

</stderr_txt>
]]>
12) Message boards : News : ATI application updated to 0.60 (Message 47958)
Posted 17 Apr 2011 by Starfire
Post:
The latest version (0.62) is running very smooth for me (Win7_64 Catalyst 11.3) - thanks for getting it to work this fast :)
Yesterday I noticed a validate error - unfortunately the task is already gone from the account overview.
Run time for the task was within normal boundaries compared to my other tasks. In the result the stderr was empty. If I remember it looked something like this:

<core_client_version>6.12.22</core_client_version>
<![CDATA[

]]>


It wasn't much more than that - so the whole output was somehow missing. I only noticed it at this one task, but with completed tasks being removed very quickly it's hard to be sure.

Yes I just noticed one invalid the same:
Stderr output

<core_client_version>6.12.22</core_client_version>
<![CDATA[
<stderr_txt>

</stderr_txt>
]]>

Hope I haven't been getting many of these. They clear quickly so could have had a few. Win 7 64, HD 5970, Cat 11.4.



Just noticed another one:
7320292
<core_client_version>6.12.22</core_client_version>
<![CDATA[
<stderr_txt>

</stderr_txt>
]]>
[/url]
13) Message boards : News : ATI application updated to 0.60 (Message 47853)
Posted 14 Apr 2011 by Starfire
Post:
The latest version (0.62) is running very smooth for me (Win7_64 Catalyst 11.3) - thanks for getting it to work this fast :)
Yesterday I noticed a validate error - unfortunately the task is already gone from the account overview.
Run time for the task was within normal boundaries compared to my other tasks. In the result the stderr was empty. If I remember it looked something like this:

<core_client_version>6.12.22</core_client_version>
<![CDATA[

]]>


It wasn't much more than that - so the whole output was somehow missing. I only noticed it at this one task, but with completed tasks being removed very quickly it's hard to be sure.
14) Message boards : News : ATI application updated to 0.60 (Message 47734)
Posted 12 Apr 2011 by Starfire
Post:
Looking very good here but even after updating the app_info file, I can't get it to work. I'm doing something wrong. Any ideas?


If your talking about the frequency parameter - it wasn't working in 0.60 but in 0.62 it is fixed.

PS: Your second computer errors out all tasks complaining about missing milkyway_0.60_windows_intelx86__ati14.exe. When using app_info.xml file you need to put it in the project folder manually.
15) Message boards : News : ATI application updated to 0.60 (Message 47733)
Posted 12 Apr 2011 by Starfire
Post:
Should be fixed now with 0.62


Thumbs up. The 40 Hz are there now :)
16) Message boards : News : ATI application updated to 0.60 (Message 47721)
Posted 12 Apr 2011 by Starfire
Post:
I now had a similar task like Milchband. It was de_separation_10_3s_fix10_1_463621_1302635252_1. While it was running the UI was really sluggish and the Target frequency in the result is strange. Currently another de_separation_10 is running and the UI is a lot smoother.

<core_client_version>6.12.22</core_client_version>
<![CDATA[
<stderr_txt>
<search_application> milkywayathome_client separation 0.60 Windows x86 double CAL++ </search_application>
Found 1 CAL devices
Chose device 0

Device target:         CAL_TARGET_770
Revision:              2
CAL Version:           1.4.1332
Engine clock:          750 Mhz
Memory clock:          900 Mhz
GPU RAM:               1024
Wavefront size:        64
Double precision:      CAL_TRUE
Compute shader:        CAL_TRUE
Number SIMD:           10
Number shader engines: 1
Pitch alignment:       256
Surface alignment:     4096
Max size 2D:           { 8192, 8192 }

Estimated iteration time 284.281667 ms
Target frequency 5735634568500407000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.000000 ms, polling mode 1, using responsiveness factor of 1.000000
Dividing into 1 chunks
Integration range: { nu_steps = 640, mu_steps = 1600, r_steps = 1400 }
Using { 1, 1 } chunk(s) of size { 1400, 1600 }
Integration time = 199.189885 s, average per iteration = 311.234195 ms
Integral 0 time = 200.606518 s
Estimated iteration time 71.070417 ms
Target frequency 5735634568500407000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000.000000 ms, polling mode 1, using responsiveness factor of 1.000000
Dividing into 1 chunks
Integration range: { nu_steps = 640, mu_steps = 400, r_steps = 1400 }
Using { 1, 1 } chunk(s) of size { 1400, 400 }
Integration time = 53.654794 s, average per iteration = 83.835615 ms
Integral 1 time = 54.111614 s
Likelihood time = 5.941989 s
<background_integral> 0.000331750494935 </background_integral>
<stream_integral>  109.619376970424070  1951.519873483356200  266.117270577347310 </stream_integral>
<background_likelihood> -3.051034779156822 </background_likelihood>
<stream_only_likelihood>  -4.832203828758965  -4.439834432194514  -4.548461163928033 </stream_only_likelihood>
<search_likelihood> -2.824976493982051 </search_likelihood>
22:26:05 (5984): called boinc_finish

</stderr_txt>



Edit:
Target frequency looks normal for the second task (even through the setting to 40 seems to be ignored).
Target frequency 30.000000 ms, polling mode 1, using responsiveness factor of 1.000000
17) Message boards : News : ATI application updated to 0.60 (Message 47719)
Posted 12 Apr 2011 by Starfire
Post:
The --gpu-target-frequency <number> is similar to the -f flag the old one had. The default is 30 hz.


How should this be set. I tried

<cmdline>--gpu-target-frequency 40</cmdline>


in the app_info.xml. I verified that --gpu-target-frequency 40 is passed to the app using Process Explorer.
However in the finished results I find "Target frequency 30.000000 ms, polling mode 1, using responsiveness factor of 1.000000".

On the other side I can confirm that GPU load is now high (98% up) with full load on all CPU cores. It's way faster than 0.59 :)
18) Message boards : News : Old ATI applications deprecated (Message 47266)
Posted 9 Apr 2011 by Starfire
Post:
Is the optimized 0.23 app currently hosted on Arkayn's site one of the deprecated ones, or is it still good?


It's one of the apps that has been deprecated.
19) Message boards : News : Old ATI applications deprecated (Message 47255)
Posted 9 Apr 2011 by Starfire
Post:
Will the new app also have the ability to configure the load on the GPU (either by command line or by options on the webpage)?
The reason I'm asking is because with the old one and default settings the UI (Win7) is realllllyyyyy sluggish. So I've used the Gipsel's command line switches to lower the load. If the new app doesn't have this ability it would be a show stopper for me, because I really need to use the system while the app is running.

+1 I would really like to know this one as well...
I can add it easily. I'll be doing an update for all of the separation applications soon, and I'll include that.


Thank you - that's really good news :)
20) Message boards : News : Old ATI applications deprecated (Message 47203)
Posted 9 Apr 2011 by Starfire
Post:
Will the new app also have the ability to configure the load on the GPU (either by command line or by options on the webpage)?
The reason I'm asking is because with the old one and default settings the UI (Win7) is realllllyyyyy sluggish. So I've used the Gipsel's command line switches to lower the load. If the new app doesn't have this ability it would be a show stopper for me, because I really need to use the system while the app is running.


Next 20

©2019 Astroinformatics Group