How to get 2 Stock ATI AP tasks run on your 4670 at the same time, Sorta.

Message boards : Number crunching : How to get 2 Stock ATI AP tasks run on your 4670 at the same time, Sorta.
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · Next

AuthorMessage
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1302993 - Posted: 7 Nov 2012, 5:43:24 UTC
Last modified: 7 Nov 2012, 5:56:35 UTC

It can be done using the stock SETI app, but, a few things have to go just right. First, complain about not getting any ATI APs for your 4670 here. Someone will hear you and fix that for you. Then, when you look at the 6 days worth of APs on your machine, you think, maybe if I put the old ATI 3650 in the other video slot it would speed things up at least a little. It helps if you are unaware that a 3650 will probably not work, at this point. That's all it takes. Here is the proof. Note how there is only one Open CL device listed here, and the 3650 is being called a 2600. There are 2 AP tasks running though;
06-Nov-2012 06:04:22 [---] ATI GPU 0: ATI RV730 (CAL version 1.4.1664, 1024MB, 1001MB available, 960 GFLOPS peak)
06-Nov-2012 06:04:22 [---] ATI GPU 1: ATI Radeon HD 2600 (RV630) (CAL version 1.4.1664, 1024MB, 1015MB available, 360 GFLOPS peak)
06-Nov-2012 06:04:22 [---] OpenCL: ATI GPU 0: ATI RV730 (driver version CAL 1.4.1664, device version OpenCL 1.0 AMD-APP (851.4), 1024MB, 1001MB available)
06-Nov-2012 06:04:22 [---] Config: use all coprocessors
Initialization completed
06-Nov-2012 06:04:23 [SETI@home] Restarting task 09se12ad.18647.3339.140733193388047.10.133_0 using setiathome_enhanced version 603 in slot 1
06-Nov-2012 06:04:23 [SETI@home] Restarting task ap_11se12aa_B1_P1_00307_20121105_27495.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 0
06-Nov-2012 06:04:23 [SETI@home] Starting task ap_10se12ad_B5_P0_00279_20121105_27432.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 2

At first, all seemed well. I thought both cards were being used. As time passed, it became apparent that wasn't the case. The 3650 was idle and the 4670 was running both tasks. It was taking at least twice as long, and the screen lag was pretty bad. As was the temperature on the 4670. I started trying to correct the matter. Upgrading to BOINC .36 seemed to work, but, things went downhill at that point.
06-Nov-2012 16:30:04 [---] Version change (7.0.28 -> 7.0.36)
06-Nov-2012 16:30:04 [---] Starting BOINC client version 7.0.36 for windows_intelx86
06-Nov-2012 16:30:37 [SETI@home] Restarting task ap_11se12aa_B1_P1_00307_20121105_27495.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 0
06-Nov-2012 16:30:37 [SETI@home] Restarting task ap_10se12ad_B5_P0_00279_20121105_27432.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 2
06-Nov-2012 16:30:37 [SETI@home] Restarting task 10se12ab.4549.24607.140733193388038.10.193_0 using setiathome_enhanced version 603 in slot 1
06-Nov-2012 16:30:39 [SETI@home] Task ap_10se12ad_B5_P0_00279_20121105_27432.wu_1 exited with zero status but no 'finished' file
06-Nov-2012 16:30:39 [SETI@home] If this happens repeatedly you may need to reset the project.
06-Nov-2012 16:30:39 [SETI@home] Starting task ap_11se12aa_B2_P0_00127_20121105_30034.wu_0 using astropulse_v6 version 604 (opencl_ati_100) in slot 3
06-Nov-2012 16:30:40 [SETI@home] Task ap_11se12aa_B2_P0_00127_20121105_30034.wu_0 exited with zero status but no 'finished' file...

Yep, it trashed everyone of the scheduled APs, keeping only the 2 it had been working on all day, it finished those. There went all my tasks...again. So, I guess a 3650 just won't work with the stock app in SETI? It's listed as an Open CL card in GPU-Z, with 1024 mb of ram. I know it would be slow, but, it's faster in the machine than the box. Oh, 2 at a time on a 4670 doesn't work very well, I don't recommend it.
:)
ID: 1302993 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303009 - Posted: 7 Nov 2012, 6:42:07 UTC - in response to Message 1302993.  
Last modified: 7 Nov 2012, 6:56:02 UTC

Just as I posted this, the APs are being downloaded again to the computer. As they are finished downloading, they change to 'Waiting to Run' instead of 'Ready to start'. It did that the last time and then all gave that error and were deleted when the running task was finished and tried to start a new AP. The 3650 is still in the machine but not being used. The 4670 is working on ONE AP at present and will take another 4 hours to finish. I have a feeling it will run down the list and delete all the APs in about 4 hours. All the drivers have been removed and reinstalled since the incident, the machine restarted a number of times...
ID: 1303009 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303013 - Posted: 7 Nov 2012, 7:12:03 UTC
Last modified: 7 Nov 2012, 7:44:48 UTC

Here's the last time it started;
06-Nov-2012 20:43:50 [---] Starting BOINC client version 7.0.36 for windows_intelx86
06-Nov-2012 20:43:50 [---] log flags: file_xfer, sched_ops, task
06-Nov-2012 20:43:50 [---] Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
06-Nov-2012 20:43:50 [---] Data directory: C:\Documents and Settings\All Users\Application Data\BOINC
06-Nov-2012 20:43:50 [---] Running under account Tom
06-Nov-2012 20:43:50 [---] Processor: 2 GenuineIntel Pentium(R) Dual-Core CPU E5200 @ 2.50GHz [Family 6 Model 23 Stepping 6]
06-Nov-2012 20:43:50 [---] Processor: 2.00 MB cache
06-Nov-2012 20:43:50 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 lm tm2 pbe
06-Nov-2012 20:43:50 [---] OS: Microsoft Windows XP: Professional x86 Edition, Service Pack 3, (05.01.2600.00)
06-Nov-2012 20:43:50 [---] Memory: 2.00 GB physical, 3.85 GB virtual
06-Nov-2012 20:43:50 [---] Disk: 97.75 GB total, 44.53 GB free
06-Nov-2012 20:43:50 [---] Local time is UTC -5 hours
06-Nov-2012 20:43:50 [---] ATI GPU 0: ATI Radeon HD 4600 series (R730) (CAL version 1.4.1664, 1024MB, 1002MB available, 960 GFLOPS peak)
06-Nov-2012 20:43:50 [---] ATI GPU 1: ATI Radeon HD 2600 (RV630) (CAL version 1.4.1664, 1024MB, 1015MB available, 348 GFLOPS peak)
06-Nov-2012 20:43:50 [---] OpenCL: ATI GPU 0: ATI Radeon HD 4600 series (R730) (driver version CAL 1.4.1664, device version OpenCL 1.0 AMD-APP (851.4), 1024MB, 1002MB available)
06-Nov-2012 20:43:50 [---] Config: use all coprocessors
06-Nov-2012 20:43:50 [SETI@home] URL http://setiathome.berkeley.edu/; Computer ID 6796475; resource share 100
06-Nov-2012 20:43:50 [SETI@home] General prefs: from SETI@home (last modified 13-Oct-2012 21:46:00)
06-Nov-2012 20:43:50 [SETI@home] Host location: none
06-Nov-2012 20:43:50 [SETI@home] General prefs: using your defaults
06-Nov-2012 20:43:50 [---] Reading preferences override file
06-Nov-2012 20:43:50 [---] Preferences:
06-Nov-2012 20:43:50 [---] max memory usage when active: 1228.25MB
06-Nov-2012 20:43:50 [---] max memory usage when idle: 1842.38MB
06-Nov-2012 20:43:50 [---] max disk usage: 20.00GB
06-Nov-2012 20:43:50 [---] max CPUs used: 1
06-Nov-2012 20:43:50 [---] suspend work if non-BOINC CPU load exceeds 50 %
06-Nov-2012 20:43:50 [---] (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
06-Nov-2012 20:43:50 [---] Not using a proxy
Initialization completed
06-Nov-2012 20:43:51 [SETI@home] Restarting task ap_10se12ad_B5_P0_00279_20121105_27432.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 2
06-Nov-2012 20:43:51 [SETI@home] Restarting task 10se12ab.4549.24607.140733193388038.10.193_0 using setiathome_enhanced version 603 in slot 1

Now some strange stuff going on;
07-Nov-2012 01:59:11 [SETI@home] Task ap_01se12aa_B3_P1_00014_20121106_23962.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 01:59:11 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:11 [SETI@home] Restarting task ap_02se12ab_B0_P1_00240_20121106_02688.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 12
07-Nov-2012 02:00:12 [SETI@home] Task ap_02se12ab_B0_P1_00240_20121106_02688.wu_0 exited with zero status but no 'finished' file
07-Nov-2012 02:00:12 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:12 [SETI@home] Restarting task ap_02se12aa_B1_P0_00296_20121106_02852.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 3
07-Nov-2012 02:00:13 [SETI@home] Task ap_02se12aa_B1_P0_00296_20121106_02852.wu_0 exited with zero status but no 'finished' file
07-Nov-2012 02:00:13 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:13 [SETI@home] Restarting task ap_01se12aa_B3_P0_00339_20121106_23416.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 4
07-Nov-2012 02:00:14 [SETI@home] Task ap_01se12aa_B3_P0_00339_20121106_23416.wu_0 exited with zero status but no 'finished' file
07-Nov-2012 02:00:14 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:14 [SETI@home] Restarting task ap_11se12ac_B5_P0_00219_20121106_28068.wu_2 using astropulse_v6 version 604 (ati_opencl_100) in slot 2
07-Nov-2012 02:00:15 [SETI@home] Task ap_11se12ac_B5_P0_00219_20121106_28068.wu_2 exited with zero status but no 'finished' file
07-Nov-2012 02:00:15 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:16 [SETI@home] Restarting task ap_02se12aa_B1_P0_00305_20121106_02852.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 6
07-Nov-2012 02:00:17 [SETI@home] Task ap_02se12aa_B1_P0_00305_20121106_02852.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:17 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:17 [SETI@home] Restarting task ap_01se12aa_B2_P1_00356_20121106_23192.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 5
07-Nov-2012 02:00:18 [SETI@home] Task ap_01se12aa_B2_P1_00356_20121106_23192.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:18 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:18 [SETI@home] Restarting task ap_02se12aa_B1_P0_00293_20121106_02852.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 7
07-Nov-2012 02:00:19 [SETI@home] Task ap_02se12aa_B1_P0_00293_20121106_02852.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:19 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:19 [SETI@home] Restarting task ap_03se12aa_B0_P0_00012_20121106_23964.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 10
07-Nov-2012 02:00:20 [SETI@home] Task ap_03se12aa_B0_P0_00012_20121106_23964.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:20 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:20 [SETI@home] Restarting task ap_01se12aa_B3_P1_00015_20121106_23962.wu_0 using astropulse_v6 version 604 (ati_opencl_100) in slot 11
07-Nov-2012 02:00:21 [SETI@home] Task ap_01se12aa_B3_P1_00015_20121106_23962.wu_0 exited with zero status but no 'finished' file
07-Nov-2012 02:00:21 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:21 [SETI@home] Restarting task ap_02se12aa_B1_P0_00349_20121106_02852.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 8
07-Nov-2012 02:00:22 [SETI@home] Task ap_02se12aa_B1_P0_00349_20121106_02852.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:22 [SETI@home] If this happens repeatedly you may need to reset the project.
07-Nov-2012 02:00:22 [SETI@home] Restarting task ap_01se12aa_B3_P1_00014_20121106_23962.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 9
07-Nov-2012 02:00:23 [SETI@home] Task ap_01se12aa_B3_P1_00014_20121106_23962.wu_1 exited with zero status but no 'finished' file
07-Nov-2012 02:00:23 [SETI@home] If this happens repeatedly you may need to reset the project.

SLOT 12?

My best guess is it's trying to start another task using the 3650 that it thinks is a 2600 and failing. It is just repeating the process over and over. I'm going to shut it down and remove the 3650. There are 12 slots on the machine now. There should be 3.
ID: 1303013 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303025 - Posted: 7 Nov 2012, 9:02:08 UTC

It looks like that might have solved it. After I removed the 3650 it downloaded 5 more APs and all those are listed as 'Ready to start' with nothing strange in the log. I still have 6 APs listed as 'Waiting to run', two finally gave up and changed to 'Computation error' before I shut down the machine to remove the card. Speaking of 'Computation error', is there some known problem with the AMD HD6850 and Astropulse 6.04? MOST of the APs run through the 6850 work all the way to the end and then end with 'Computation error'. It's a fairly new card and doesn't do 'Games'. I bought it because it also works in OSX out of the box. It's basically a straight Windows card and is giving the errors in XPsp3.
ID: 1303025 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34250
Credit: 79,922,639
RAC: 80
Germany
Message 1303032 - Posted: 7 Nov 2012, 9:34:39 UTC

Hello TBar.

First of all HD 4670 only has OpenCL beta level support.
The 3650 wont work at all.

There is no issue with 6850.

Makes sure you removed old drivers completely.



With each crime and every kindness we birth our future.
ID: 1303032 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1303036 - Posted: 7 Nov 2012, 9:50:17 UTC - in response to Message 1302993.  
Last modified: 7 Nov 2012, 9:54:42 UTC

For OpenCL support on ATI/AMD GPUs they must be HD4*** series or later, sometimes ATI/AMD have tweaked a GPU specks a tadge, then released it under a new Series number, but it's still basically an ealier model and reported by Boinc as such.

Eric when he devised the planclasses for OpenCL apps thoght of a work around so users of Boinc 6 would still get OpenCL work, first he did a (opencl_ati_100) plan class,
only Boinc 7 users will get work from this planclass, and Boinc 7 should only start the same number of Wu to OpenCL GPUs on your system,
second Eric did a (ati_opencl_100) planclass, this planclass is sent to both Boinc 6 and Boinc 7 reporting an ATI CAL device, using the assumption that a recent enough CAL device, with recent enough drivers should have OpenCL support,
but this falls down in a few instances when OpenCL support isn't available, best try and not mix OpenCL capable and non-OpenCL capable ATI/AMD GPUs in the same host, otherwise Boinc will just start what it thinks is an CAL app (but isn't) on all the CAL devices it has.

Claggy
ID: 1303036 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303044 - Posted: 7 Nov 2012, 10:19:17 UTC - in response to Message 1303032.  

I'm getting mostly errors with this machine. It's a 2008 MacPro with the standard NV8800, which is working fine, and a new AMD6850 running XP http://setiathome.berkeley.edu/results.php?hostid=6797524&offset=0&show_names=0&state=0&appid=12 Only a couple are working. The other machine with the 4670 isn't having any problems, except with the ATI3650, which is gone now. I installed the AMD drivers again after installing the NV8800 a few days ago. The 8800 had been in a box since I got the 6850. This is the first time I've run them together. The NV8800 is maxed out while the 6850 is only running at around 70%. I've just gotten 4 more errors in a row to add to that list.

Something isn't right...
ID: 1303044 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303243 - Posted: 7 Nov 2012, 19:27:33 UTC

I'm still getting mostly errors with the HD 6850. I removed the option to use all GPUs to see if it would still use both cards. It still used both cards but I got two successes in a row. Then all errors again. I've noticed when it gets to about 95% finished the time remaining starts hanging and jumping every 40 secs or so. Then the error appears. I'm going to try removing/reinstalling the AMD drivers again and see what happens. Any thoughts?
ID: 1303243 · Report as offensive
Profile skildude
Avatar

Send message
Joined: 4 Oct 00
Posts: 9541
Credit: 50,759,529
RAC: 60
Yemen
Message 1303250 - Posted: 7 Nov 2012, 19:41:44 UTC

4XXX series ATI cards are like the Nvidia 2XX series. Neither should have more than 1 WU per GPU. They both run much more efficiently with 1 WU at a time


In a rich man's house there is no place to spit but his face.
Diogenes Of Sinope
ID: 1303250 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303280 - Posted: 7 Nov 2012, 21:05:44 UTC - in response to Message 1303243.  
Last modified: 7 Nov 2012, 21:52:25 UTC

I'm still getting mostly errors with the HD 6850. I removed the option to use all GPUs to see if it would still use both cards. It still used both cards but I got two successes in a row. Then all errors again. I've noticed when it gets to about 95% finished the time remaining starts hanging and jumping every 40 secs or so. Then the error appears. I'm going to try removing/reinstalling the AMD drivers again and see what happens. Any thoughts?

Removed drivers, cleaned with Driver Sweeper, Installed AMD 12.1 without the HDMI driver, and my last AP just Erred again. It got right to the end, the Remaining Estimate started hanging, then the error.

My other machine with the 4670 just downloaded another 6 APs and is loaded up again. I'm not receiving any for the 6850 machine. I guess the errors don't help. The only non-standard software is Apple Bootcamp & VMware Fusion, but I'm booted to Windows and it doesn't seem to bother the NV8800. Is there an easy way to suspend just the NV8800 without losing all the Cuda tasks on the machine? I can just try the 6850 by itself the next time I receive APs.

I just received APs for the 6850, what can I try next?

Can someone look at one of the long errors and see whats going on right at the end? I'm not sure about the really short ones, obviously another type of error.
http://setiathome.berkeley.edu/results.php?hostid=6797524&offset=0&show_names=0&state=0&appid=12
ID: 1303280 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303296 - Posted: 7 Nov 2012, 22:05:30 UTC
Last modified: 7 Nov 2012, 22:21:48 UTC

After 51 minutes, another error. Should I just give up running APs on this machine? The card has to stay in this machine, it's one of the few that also works in OSX.
http://setiathome.berkeley.edu/results.php?hostid=6797524&offset=0&show_names=0&state=0&appid=12

This is the closest match to my card. The connectors are the same but it only has one fan. Seems they don't have my exact card anymore. It's a HD-685X-ZCFC
http://www.newegg.com/Product/Product.aspx?Item=N82E16814150515
ID: 1303296 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303316 - Posted: 7 Nov 2012, 22:55:48 UTC

Another 51 minutes, and another error.
ID: 1303316 · Report as offensive
Claggy
Volunteer tester

Send message
Joined: 5 Jul 99
Posts: 4654
Credit: 47,537,079
RAC: 4
United Kingdom
Message 1303322 - Posted: 7 Nov 2012, 23:08:26 UTC - in response to Message 1303316.  
Last modified: 7 Nov 2012, 23:09:00 UTC

Another 51 minutes, and another error.

Sounds like Maximum Time Exceeded Errors, try using Fred's BoincRescheduler, and use the protect against -177 errors setting, see if that makes a difference,

Then try adding -unroll 10 -ffa_block 6144 -ffa_block_fetch 1536 to your ap_cmdline_6.04_windows_intelx86__opencl_ati.txt file, that should speed the app up.

Claggy
ID: 1303322 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303327 - Posted: 7 Nov 2012, 23:19:12 UTC - in response to Message 1303322.  
Last modified: 7 Nov 2012, 23:48:29 UTC

It's doing the work. Right now the card is running at 70% GPU load, 70C Temp, 67% fan speed. I'm thinking about changing the Slots with the NV8800 which is in Slot 1. The 6850 is the only one with a fan though. The NV8800 has a large radiator. The way it is now the heat rises to the 6850 then gets pushed out of the machine. I think all changing slots will do is make the NV8800 run hotter though. It's staying around 68C at 95% load right now.

The last one just finished. It ran for 00:50:38 using around 70% of the GPU then threw the 'Computation Error'. I'll change the slots and see what my last 2 APs do.
ID: 1303327 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303406 - Posted: 8 Nov 2012, 4:08:56 UTC
Last modified: 8 Nov 2012, 4:48:26 UTC

Well, I just went through a lot of work changing the cards around and completely reinstalling the drivers. The only apparent result is the NV8800 is running about 5C cooler. I had to place the 8800 in slot 3, which is a 4x PCIe because it would block the 6850's fan if I placed it in the 16x slot 2. The 6850 is in the 16x slot 1 and still giving errors. The AMD 6850 ran for 00:49:49 at around 70% GPU load then ended with 'Computation Error'. I did receive 5 more APs, so, I have about 7 hours worth to work on. The NV8800 is working about the same as before.

I'm out of ideas...

Last run was 00:51:09 at 70% and NO ERROR. Whoa, I got some points for that one...
:)
ID: 1303406 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303425 - Posted: 8 Nov 2012, 5:57:33 UTC

The last run was a little slower, I saw a lot of GPU load around 50-60%. The 6850 ran for 01:04:18 and ended with 'Computation Error'. There isn't any mention of any error in the log;
08-Nov-2012 00:37:18 [SETI@home] Starting task 10se12ad.27589.21335.140733193388046.10.237_0 using setiathome_enhanced version 603 in slot 2
08-Nov-2012 00:37:20 [SETI@home] Started upload of 11se12aa.27752.21744.140733193388039.10.197_1_0
08-Nov-2012 00:37:32 [SETI@home] Finished upload of 11se12aa.27752.21744.140733193388039.10.197_1_0
08-Nov-2012 00:50:24 [SETI@home] Computation for task ap_04se12ab_B1_P0_00117_20121107_09192.wu_0 finished
08-Nov-2012 00:50:24 [SETI@home] Starting task ap_03se12aa_B6_P1_00291_20121107_28015.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 1
08-Nov-2012 00:50:26 [SETI@home] Started upload of ap_04se12ab_B1_P0_00117_20121107_09192.wu_0_0
08-Nov-2012 00:50:41 [SETI@home] Finished upload of ap_04se12ab_B1_P0_00117_20121107_09192.wu_0_0
ID: 1303425 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303447 - Posted: 8 Nov 2012, 6:48:04 UTC

Another Computation Error after running for 00:51:44 at around 70% GPU load. No mention of an error.
08-Nov-2012 01:23:39 [SETI@home] Started upload of 10se12ad.24569.19699.140733193388035.10.162_0_0
08-Nov-2012 01:23:44 [SETI@home] Finished upload of 10se12ad.24569.19699.140733193388035.10.162_0_0
08-Nov-2012 01:42:10 [SETI@home] Computation for task ap_03se12aa_B6_P1_00291_20121107_28015.wu_1 finished
08-Nov-2012 01:42:10 [SETI@home] Starting task ap_04se12aa_B5_P0_00036_20121107_08710.wu_0 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 01:42:12 [SETI@home] Started upload of ap_03se12aa_B6_P1_00291_20121107_28015.wu_1_0
08-Nov-2012 01:42:22 [SETI@home] Finished upload of ap_03se12aa_B6_P1_00291_20121107_28015.wu_1_0

Looks pretty bad... http://setiathome.berkeley.edu/results.php?hostid=6797524&offset=0&show_names=0&state=0&appid=12
ID: 1303447 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34250
Credit: 79,922,639
RAC: 80
Germany
Message 1303465 - Posted: 8 Nov 2012, 8:57:26 UTC


What brand is this card ?

You could install MSI Afterburner and reduce core and memory speed.
Lets see if this helps.

Any firewall or virus program interfering maybe ?
Make sure you excluded Boinc from scanning those.



With each crime and every kindness we birth our future.
ID: 1303465 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303476 - Posted: 8 Nov 2012, 10:10:41 UTC - in response to Message 1303465.  

The card seems to be working just fine with everything except AstroPulse. I've been using it for about 6 months mostly in OSX 10.6.8. I've made numerous screen recordings with QuickTime while running all sorts of other programs. I've never had any trouble with it. There is a link above to the card at Newegg. AMD Overdrive allows for custom settings but I've never had any need to change the default settings. The last AstroPulse run was successful, it seems like 1 success for every 4 'Computation Errors'. Right now it's running at Default settings and showing 70 C at 60-70% Load. I'm happy with 70% Load while running AstroPulse.
ID: 1303476 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1303599 - Posted: 8 Nov 2012, 17:42:14 UTC
Last modified: 8 Nov 2012, 18:35:19 UTC

The last nine have all ended with errors. Nothing in the log;
08-Nov-2012 08:28:48 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 08:30:58 [SETI@home] Computation for task ap_04se12aa_B6_P1_00135_20121107_14692.wu_1 finished
08-Nov-2012 08:30:58 [SETI@home] Starting task ap_04se12ab_B3_P1_00226_20121107_05551.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 1
08-Nov-2012 08:31:00 [SETI@home] Started upload of ap_04se12aa_B6_P1_00135_20121107_14692.wu_1_0
08-Nov-2012 08:31:03 [SETI@home] Computation for task ap_04se12ab_B3_P1_00226_20121107_05551.wu_1 finished
08-Nov-2012 08:31:03 [SETI@home] Starting task ap_04se12ab_B4_P1_00081_20121107_21259.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 1
08-Nov-2012 08:31:06 [SETI@home] Started upload of ap_04se12ab_B3_P1_00226_20121107_05551.wu_1_0
08-Nov-2012 08:36:07 [SETI@home] Temporarily failed upload of ap_04se12aa_B6_P1_00135_20121107_14692.wu_1_0: transient HTTP error
08-Nov-2012 08:36:07 [SETI@home] Backing off 2 min 48 sec on upload of ap_04se12aa_B6_P1_00135_20121107_14692.wu_1_0
08-Nov-2012 08:36:08 [SETI@home] Started upload of 14fe12aa.10068.15302.8.10.118_2_0
08-Nov-2012 08:36:10 [---] Project communication failed: attempting access to reference site
08-Nov-2012 08:36:11 [SETI@home] Finished upload of 14fe12aa.10068.15302.8.10.118_2_0
08-Nov-2012 08:36:12 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 08:36:15 [SETI@home] Temporarily failed upload of ap_04se12ab_B3_P1_00226_20121107_05551.wu_1_0: transient HTTP error
08-Nov-2012 08:36:15 [SETI@home] Backing off 3 min 42 sec on upload of ap_04se12ab_B3_P1_00226_20121107_05551.wu_1_0
08-Nov-2012 08:36:18 [---] Project communication failed: attempting access to reference site
08-Nov-2012 08:36:19 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 08:38:55 [SETI@home] Started upload of ap_04se12aa_B6_P1_00135_20121107_14692.wu_1_0
08-Nov-2012 08:39:04 [SETI@home] Finished upload of ap_04se12aa_B6_P1_00135_20121107_14692.wu_1_0
08-Nov-2012 08:39:58 [SETI@home] Started upload of ap_04se12ab_B3_P1_00226_20121107_05551.wu_1_0
08-Nov-2012 08:40:05 [SETI@home] Finished upload of ap_04se12ab_B3_P1_00226_20121107_05551.wu_1_0
08-Nov-2012 08:52:05 [SETI@home] Sending scheduler request: To fetch work.
08-Nov-2012 08:52:05 [SETI@home] Reporting 8 completed tasks
08-Nov-2012 08:52:05 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI
08-Nov-2012 08:57:13 [SETI@home] Scheduler request failed: Timeout was reached
08-Nov-2012 08:57:17 [---] Project communication failed: attempting access to reference site
08-Nov-2012 08:57:19 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 09:07:08 [SETI@home] Sending scheduler request: To fetch work.
08-Nov-2012 09:07:08 [SETI@home] Reporting 8 completed tasks
08-Nov-2012 09:07:08 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI
08-Nov-2012 09:12:20 [SETI@home] Scheduler request failed: Timeout was reached
08-Nov-2012 09:12:23 [---] Project communication failed: attempting access to reference site
08-Nov-2012 09:12:25 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 09:14:29 [SETI@home] Computation for task 11se12ad.3504.10701.140733193388040.10.31_1 finished
08-Nov-2012 09:14:29 [SETI@home] Starting task 11se12ab.4954.13973.140733193388041.10.168_1 using setiathome_enhanced version 609 (cuda23) in slot 3
08-Nov-2012 09:14:32 [SETI@home] Started upload of 11se12ad.3504.10701.140733193388040.10.31_1_0
08-Nov-2012 09:14:50 [SETI@home] Finished upload of 11se12ad.3504.10701.140733193388040.10.31_1_0
08-Nov-2012 09:21:39 [SETI@home] Computation for task ap_04se12ab_B4_P1_00081_20121107_21259.wu_1 finished
08-Nov-2012 09:21:39 [SETI@home] Starting task ap_04se12ab_B3_P1_00227_20121107_05551.wu_1 using astropulse_v6 version 604 (ati_opencl_100) in slot 1
08-Nov-2012 09:21:41 [SETI@home] Started upload of ap_04se12ab_B4_P1_00081_20121107_21259.wu_1_0
08-Nov-2012 09:21:44 [SETI@home] Computation for task ap_04se12ab_B3_P1_00227_20121107_05551.wu_1 finished
08-Nov-2012 09:21:44 [SETI@home] Starting task ap_08se12aa_B6_P1_00119_20121108_30584.wu_0 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 09:21:46 [SETI@home] Finished upload of ap_04se12ab_B4_P1_00081_20121107_21259.wu_1_0
08-Nov-2012 09:21:46 [SETI@home] Started upload of ap_04se12ab_B3_P1_00227_20121107_05551.wu_1_0
08-Nov-2012 09:22:20 [SETI@home] Finished upload of ap_04se12ab_B3_P1_00227_20121107_05551.wu_1_0
08-Nov-2012 10:00:42 [SETI@home] Sending scheduler request: To fetch work.
08-Nov-2012 10:00:42 [SETI@home] Reporting 11 completed tasks
08-Nov-2012 10:00:42 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI
08-Nov-2012 10:01:14 [SETI@home] Computation for task 11se12ab.4954.13973.140733193388041.10.168_1 finished
08-Nov-2012 10:01:14 [SETI@home] Starting task 11se12ac.10920.326462.140733193388038.10.136_0 using setiathome_enhanced version 609 (cuda23) in slot 3
08-Nov-2012 10:01:16 [SETI@home] Started upload of 11se12ab.4954.13973.140733193388041.10.168_1_0
08-Nov-2012 10:01:20 [SETI@home] Finished upload of 11se12ab.4954.13973.140733193388041.10.168_1_0
08-Nov-2012 10:05:49 [SETI@home] Scheduler request failed: Timeout was reached
08-Nov-2012 10:05:52 [---] Project communication failed: attempting access to reference site
08-Nov-2012 10:05:54 [---] Internet access OK - project servers may be temporarily down.
08-Nov-2012 10:12:04 [SETI@home] Computation for task ap_08se12aa_B6_P1_00119_20121108_30584.wu_0 finished
08-Nov-2012 10:12:04 [SETI@home] Starting task ap_24au12aa_B1_P0_00139_20121108_29014.wu_0 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 10:12:07 [SETI@home] Started upload of ap_08se12aa_B6_P1_00119_20121108_30584.wu_0_0
08-Nov-2012 10:12:43 [SETI@home] Finished upload of ap_08se12aa_B6_P1_00119_20121108_30584.wu_0_0
08-Nov-2012 10:47:29 [SETI@home] Computation for task 11se12ac.10920.326462.140733193388038.10.136_0 finished
08-Nov-2012 10:47:29 [SETI@home] Starting task 11se12ab.3998.22971.140733193388040.10.184_0 using setiathome_enhanced version 609 (cuda23) in slot 3
08-Nov-2012 10:47:31 [SETI@home] Started upload of 11se12ac.10920.326462.140733193388038.10.136_0_0
08-Nov-2012 10:47:41 [SETI@home] Finished upload of 11se12ac.10920.326462.140733193388038.10.136_0_0
08-Nov-2012 11:04:39 [SETI@home] Computation for task ap_24au12aa_B1_P0_00139_20121108_29014.wu_0 finished
08-Nov-2012 11:04:39 [SETI@home] Starting task ap_24au12aa_B1_P0_00138_20121108_29014.wu_1 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 11:04:42 [SETI@home] Started upload of ap_24au12aa_B1_P0_00139_20121108_29014.wu_0_0
08-Nov-2012 11:04:51 [SETI@home] Finished upload of ap_24au12aa_B1_P0_00139_20121108_29014.wu_0_0
08-Nov-2012 11:08:48 [SETI@home] Computation for task 11se12aa.27752.23789.140733193388039.10.250_0 finished
08-Nov-2012 11:08:48 [SETI@home] Starting task 11se12ab.3998.15200.140733193388040.10.203_1 using setiathome_enhanced version 603 in slot 0
08-Nov-2012 11:08:50 [SETI@home] Started upload of 11se12aa.27752.23789.140733193388039.10.250_0_0
08-Nov-2012 11:08:57 [SETI@home] Finished upload of 11se12aa.27752.23789.140733193388039.10.250_0_0
08-Nov-2012 11:35:08 [SETI@home] Computation for task 11se12ab.3998.22971.140733193388040.10.184_0 finished
08-Nov-2012 11:35:08 [SETI@home] Starting task 11se12ab.18828.885.140733193388043.10.189_0 using setiathome_enhanced version 609 (cuda23) in slot 3
08-Nov-2012 11:35:11 [SETI@home] Started upload of 11se12ab.3998.22971.140733193388040.10.184_0_0
08-Nov-2012 11:35:24 [SETI@home] Finished upload of 11se12ab.3998.22971.140733193388040.10.184_0_0
08-Nov-2012 12:00:58 [SETI@home] Computation for task ap_24au12aa_B1_P0_00138_20121108_29014.wu_1 finished
08-Nov-2012 12:00:58 [SETI@home] Starting task ap_21au12aa_B0_P0_00340_20121108_18839.wu_1 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 12:01:00 [SETI@home] Started upload of ap_24au12aa_B1_P0_00138_20121108_29014.wu_1_0
08-Nov-2012 12:01:26 [SETI@home] Finished upload of ap_24au12aa_B1_P0_00138_20121108_29014.wu_1_0
08-Nov-2012 12:11:12 [SETI@home] update requested by user
08-Nov-2012 12:11:17 [SETI@home] Sending scheduler request: Requested by user.
08-Nov-2012 12:11:17 [SETI@home] Reporting 18 completed tasks
08-Nov-2012 12:11:17 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI

This was the last reboot;

08-Nov-2012 06:30:48 [---] Starting BOINC client version 7.0.36 for windows_intelx86
08-Nov-2012 06:30:48 [---] log flags: file_xfer, sched_ops, task
08-Nov-2012 06:30:48 [---] Libraries: libcurl/7.25.0 OpenSSL/1.0.1 zlib/1.2.6
08-Nov-2012 06:30:48 [---] Data directory: C:\Documents and Settings\All Users\Application Data\BOINC
08-Nov-2012 06:30:48 [---] Running under account Tom
08-Nov-2012 06:30:48 [---] Processor: 3 GenuineIntel Intel(R) Xeon(R) CPU E5462 @ 2.80GHz [Family 6 Model 23 Stepping 6]
08-Nov-2012 06:30:48 [---] Processor: 6.00 MB cache
08-Nov-2012 06:30:48 [---] Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss htt tm pni ssse3 cx16 sse4_1 nx lm vmx tm2 dca pbe
08-Nov-2012 06:30:48 [---] OS: Microsoft Windows XP: Professional x86 Edition, Service Pack 3, (05.01.2600.00)
08-Nov-2012 06:30:48 [---] Memory: 1.99 GB physical, 3.83 GB virtual
08-Nov-2012 06:30:48 [---] Disk: 217.90 GB total, 182.87 GB free
08-Nov-2012 06:30:48 [---] Local time is UTC -5 hours
08-Nov-2012 06:30:48 [---] NVIDIA GPU 0: GeForce 8800 GT (driver version 306.81, CUDA version 5.0, compute capability 1.1, 512MB, 467MB available, 504 GFLOPS peak)
08-Nov-2012 06:30:48 [---] ATI GPU 0: AMD Radeon HD 6800 series (Barts) (CAL version 1.4.1664, 1024MB, 999MB available, 2976 GFLOPS peak)
08-Nov-2012 06:30:48 [---] OpenCL: NVIDIA GPU 0: GeForce 8800 GT (driver version 306.81, device version OpenCL 1.0 CUDA, 512MB, 467MB available)
08-Nov-2012 06:30:48 [---] OpenCL: ATI GPU 0: AMD Radeon HD 6800 series (Barts) (driver version CAL 1.4.1664, device version OpenCL 1.1 AMD-APP (851.4), 1024MB, 999MB available)
08-Nov-2012 06:30:48 [---] Config: simulate 3 CPUs
08-Nov-2012 06:30:48 [SETI@home] URL http://setiathome.berkeley.edu/; Computer ID 6797524; resource share 100
08-Nov-2012 06:30:48 [SETI@home] General prefs: from SETI@home (last modified 13-Oct-2012 21:46:00)
08-Nov-2012 06:30:48 [SETI@home] Host location: none
08-Nov-2012 06:30:48 [SETI@home] General prefs: using your defaults
08-Nov-2012 06:30:48 [---] Reading preferences override file
08-Nov-2012 06:30:48 [---] Preferences:
08-Nov-2012 06:30:48 [---] max memory usage when active: 1222.82MB
08-Nov-2012 06:30:48 [---] max memory usage when idle: 1936.14MB
08-Nov-2012 06:30:48 [---] max disk usage: 30.00GB
08-Nov-2012 06:30:48 [---] max CPUs used: 2
08-Nov-2012 06:30:48 [---] suspend work if non-BOINC CPU load exceeds 50 %
08-Nov-2012 06:30:48 [---] (to change preferences, visit the web site of an attached project, or select Preferences in the Manager)
08-Nov-2012 06:30:48 [---] Not using a proxy
Initialization completed
08-Nov-2012 06:30:52 [SETI@home] Restarting task 10se12ad.27589.21335.140733193388046.10.237_0 using setiathome_enhanced version 603 in slot 2
08-Nov-2012 06:30:52 [SETI@home] Restarting task 11se12aa.27752.23789.140733193388039.10.250_0 using setiathome_enhanced version 603 in slot 0
08-Nov-2012 06:30:52 [SETI@home] Restarting task ap_04se12ab_B3_P0_00121_20121107_32186.wu_0 using astropulse_v6 version 604 (opencl_ati_100) in slot 1
08-Nov-2012 06:30:52 [SETI@home] Restarting task 10se12ad.24569.19699.140733193388035.10.158_1 using setiathome_enhanced version 608 (cuda) in slot 3
08-Nov-2012 06:30:52 [SETI@home] Sending scheduler request: To fetch work.
08-Nov-2012 06:30:52 [SETI@home] Reporting 10 completed tasks
08-Nov-2012 06:30:52 [SETI@home] Requesting new tasks for CPU and NVIDIA and ATI

Is there some Benchmarking software for XPsp3 I could run that might reveal something? It just finished the last AP, the card is idle now. I just went back and checked, the card has been installed in this machine since mid-April. All the videos on this site were done using the card in OSX using QT screen recording http://www.youtube.com/user/TBar1984?feature=mhee
ID: 1303599 · Report as offensive
1 · 2 · 3 · Next

Message boards : Number crunching : How to get 2 Stock ATI AP tasks run on your 4670 at the same time, Sorta.


 
©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.