rpi_logo
ATI application (v0.57) should be available now
ATI application (v0.57) should be available now
log in

Advanced search

Message boards : News : ATI application (v0.57) should be available now

Previous · 1 · 2 · 3 · 4 · 5
Author Message
Patrick
Send message
Joined: 30 Aug 10
Posts: 3
Credit: 18,752,125
RAC: 0

Message 47474 - Posted: 10 Apr 2011, 13:05:40 UTC

Running the new 0.57 WU now and ~95% lead to computation errors. I'm running on Win 7 64 with an ati 5830 and catalyst 11.3.

So far I've observed that the only units to work are those with _free in the file name all of the _fix ones go to computation error after 4-5 s.

Michael
Send message
Joined: 14 Nov 10
Posts: 13
Credit: 3,616,915
RAC: 0

Message 47481 - Posted: 10 Apr 2011, 13:26:26 UTC - in response to Message 47474.
Last modified: 10 Apr 2011, 13:28:08 UTC

Running the new 0.57 WU now and ~95% lead to computation errors. I'm running on Win 7 64 with an ati 5830 and catalyst 11.3.

So far I've observed that the only units to work are those with _free in the file name all of the _fix ones go to computation error after 4-5 s.


i got the opposite scenario ... all tasks with _free in the filename lead to computation error and all with _fix are working fine

win 7, boinc 60.10.60 x64, catalyst 11.3., gainward hd 4870 gs 1024MB

Chris S
Avatar
Send message
Joined: 20 Sep 08
Posts: 1387
Credit: 186,689,786
RAC: 9

Message 47487 - Posted: 10 Apr 2011, 13:55:48 UTC

I think that the new 0.57's are very touchy as to what they will run on, in terms of the boinc version, CCC version, the card itself, and maye the OS as well. The past setup was a lot more tolerant.

____________
Don't drink water, that's the stuff that rusts pipes

Profile Beyond
Send message
Joined: 15 Jul 08
Posts: 383
Credit: 501,817,790
RAC: 0

Message 47491 - Posted: 10 Apr 2011, 14:15:58 UTC - in response to Message 47466.

- I used "b-1" for the polling behaviour with 0.23. This allowed for optimal GPU performance without loosing a full cpu core -> increased CPU throughput (and satisfaction ;)

How to solve this? Well, get the CPU-utilization sorted out. And could we use 0.57 with an app_info and configure it the same way we used to with 0.23?

MrS

Seconded. I also used b-1 in the commandline on all machines and it allowed high GPU utilization coupled with very little hit on the CPU.

Jesse Viviano
Send message
Joined: 4 Feb 11
Posts: 82
Credit: 38,118,736
RAC: 16,959

Message 47497 - Posted: 10 Apr 2011, 15:13:52 UTC

I have two errors I found last night:

One of them is that if something suspends a task running on the ATI GPU client just as it begins, it will cause a computation error when the task is resumed.

The other one is that if the ATI GPU client tries to chew on one of the de_separation_17_3s_fix_4_* series work units (which seem to error out on any client including CPU, Nvidia GPU, and ATI GPU clients), the GPU client sometimes fails to detect the computation error immediately and gets stuck at 0% forever until someone suspends and then resumes the task.

PDelgado
Send message
Joined: 16 Nov 10
Posts: 17
Credit: 3,138,711
RAC: 0

Message 47501 - Posted: 10 Apr 2011, 16:07:27 UTC - in response to Message 47379.

New twist. This system does primarily Seti, but after the first few MW WU's processed normally, everyone after that starts immediately in high priority mode. Could be my version of Boinc Manager, could need a reboot, etc. Can't start troubleshooting right now so I'll just set NNT till I have more time.

After setting NNT for a while then resuming it displays the same behavior. The first few tasks run normally, then every MW task starts running in panic mode as soon as it is downloaded, even though the scheduling priority shows -16229.

I tried upgrading to the latest beta of Boinc Manager, but it shows the same behavior. Anybody else running multiple ATi projects having this problem?

Just glad I don't work at the RPI computer lab this week :-))

zzuupp
Send message
Joined: 6 Apr 10
Posts: 2
Credit: 15,846,668
RAC: 235

Message 47502 - Posted: 10 Apr 2011, 16:18:48 UTC

I'm getting WUs and they are successfully completing.
However, they are taking longer & using a lot more CPU.

Using Win7 32bit
Boinc 6.10.18
Catalyst 10.11 on an ATI 4850

Brickhead
Avatar
Send message
Joined: 20 Mar 08
Posts: 108
Credit: 2,540,684,670
RAC: 144,628

Message 47518 - Posted: 10 Apr 2011, 20:36:33 UTC - in response to Message 47419.

WUs all error out with Catalyst 10.7 XP32. Since AMD broke "classic" OpenGL in 10.8 on and can't be bothered to repair it, updating is not an option on this particular machine.

It seems they have in 11.3, old OpenGL games work after a clean install. I stand corrected.

And of course, MW works too now, although I need to find a replacement for the app_info setting to chop each task into 100 tiny bits per second and allow display updates to sneak in between them, I can't even see what I'm typing as the app behaves now (it's like using a terminal server through a link with 5 sec latency).
____________

aad
Send message
Joined: 30 Mar 09
Posts: 59
Credit: 385,868,906
RAC: 148,392

Message 47524 - Posted: 10 Apr 2011, 22:50:14 UTC

Using;
Win7 64 on a 1090T with 8 gig
Boinc 6.10.58
Catalist v11.4 on a HD6970

Notice the 'bump' in GPU-load in the middle of the 'fix' wu from around 60% to 100%.
This is while doing Seti on all 6 CPU cores.
MW takes 8/9% CPU and sometimes even 17% (whole core)

No problems or freeses yet.
When using 'priority' set to high, there are some little freeses but no crashes.

Previous · 1 · 2 · 3 · 4 · 5
Post to thread

Message boards : News : ATI application (v0.57) should be available now


Main page · Your account · Message boards


Copyright © 2018 AstroInformatics Group