Last night I crunched a bunch of Wus in 1 hour 20 minutes and see the dnetc-gpu-1.3 file has the processor priority at 9, so I cannot do anything.

There's definitely something wacky going on - thats similar to Dan's problem in the OP - and he has been through nothing short of heroic efforts to try and find the issue, so far to no avail. He has 5870s, you are reporting it on 5850s, and I dont get the issue on twin 5970s.

This one is baffling.

Try the app_info file - even for single GPUs [just put 1 as the number of GPUs in the app_info] - and see if explicit instructions from app_info overcomes the problem

EDIT: Just had a thought, bit wacky but ..... ,I did have an issue with it being set to 9 when the option first came out. Many reported it being quicker, but I found it slower, so I always had it set to 4, which was the old default. On change to V1.3, I went in and changed mine from the new default of 9 back to 4. Its a long shot .... but .... I did have slow down issues at 9 wa y back when it first came in, although nothing like reported at the OP - try it at 4 not 9 and see what happens - bit of a wild card, but you never know, and easy to check.

Regards
Zy

Last night I crunched a bunch of Wus in 1 hour 20 minutes and see the dnetc-gpu-1.3 file has the processor priority at 9, so I cannot do anything.

There's definitely something wacky going on - thats similar to Dan's problem in the OP - and he has been through nothing short of heroic efforts to try and find the issue, so far to no avail. He has 5870s, you are reporting it on 5850s, and I dont get the issue on twin 5970s.

This one is baffling.

Try the app_info file - even for single GPUs [just put 1 as the number of GPUs in the app_info] - and see if explicit instructions from app_info overcomes the problem

EDIT: Just had a thought, bit wacky but ..... ,I did have an issue with it being set to 9 when the option first came out. Many reported it being quicker, but I found it slower, so I always had it set to 4, which was the old default. On change to V1.3, I went in and changed mine from the new default of 9 back to 4. Its a long shot .... but .... I did have slow down issues at 9 wa y back when it first came in, although nothing like reported at the OP - try it at 4 not 9 and see what happens - bit of a wild card, but you never know, and easy to check.

Regards
Zy

I changed back to prio 4 and the process changed from 38 minutes to 14 minutes for a WU.

B

---

Whoa .... thats set the cat amongst the pigeons ...... is there anyone else can reproduce that ?

Regards
Zy

Last night I crunched a bunch of Wus in 1 hour 20 minutes and see the dnetc-gpu-1.3 file has the processor priority at 9, so I cannot do anything.

There's definitely something wacky going on - thats similar to Dan's problem in the OP - and he has been through nothing short of heroic efforts to try and find the issue, so far to no avail. He has 5870s, you are reporting it on 5850s, and I dont get the issue on twin 5970s.

This one is baffling.

Try the app_info file - even for single GPUs [just put 1 as the number of GPUs in the app_info] - and see if explicit instructions from app_info overcomes the problem

EDIT: Just had a thought, bit wacky but ..... ,I did have an issue with it being set to 9 when the option first came out. Many reported it being quicker, but I found it slower, so I always had it set to 4, which was the old default. On change to V1.3, I went in and changed mine from the new default of 9 back to 4. Its a long shot .... but .... I did have slow down issues at 9 wa y back when it first came in, although nothing like reported at the OP - try it at 4 not 9 and see what happens - bit of a wild card, but you never know, and easy to check.

Regards
Zy

Sorry, I thought I posted this morning that I've fixed my problem and am back to 20 minute runs. I just needed to set the core in the 1.3 preferences to #3. Funny I didn't see that earlier, has it been there all the time.

Thanks for the help.

Dan

Pretty well seals it I reckon, the default of 9 needs revisiting, its not looking to be a good option as a default value.

Regards
Zy

Pretty well seals it I reckon, the default of 9 needs revisiting, its not looking to be a good option as a default value.

Regards
Zy

I am getting normal run times, but normal WU credit (~1780).

I have reset the count to 4 and will let things run overnight to see if things return to normal.

"processor -usage priority from 4 to 9" in the dnetc-gpu-1.2 file

I have been having the same prblm w/my 5870's, Lg wu's? fragmentation? 32 min wu times. Ran thru CCC drivers, setteled on 11.9 for both. Where is this file? or do I need to try the app_info? On the Moo preferance page I changed ATI core to 3, but I dont think thats whats being talked about?

---

As before - project reset then change processor usage to 4 from 9.

Overnight the results on the HD5850 are run times @ 1200 seconds (was 2,300) and credit 1750, was @6.5K to 7.0K.

Weider and weirder.

My HD5970, on the same settings (ATI 11.3 CCC) running absolutely normally if a bit slow (crunch times ~1,150 seconds, credit ~7.x)

I have been having the same prblm w/my 5870's, Lg wu's? fragmentation? 32 min wu times. Ran thru CCC drivers, setteled on 11.9 for both. Where is this file? or do I need to try the app_info? On the Moo preferance page I changed ATI core to 3, but I dont think thats whats being talked about?

Change Moo preferences back to what they were - if forgotten, change to defaults (values are shown there).

What is being talked about is a value inside the project ini file in the Project Directory under "Program Data/Boinc/Projects/ .... moo

In there you will see a file called "dnetc-gpu-1.3ini"

Look just under halfway down and you will see


[processor-usage]
priority=9

If you want to test to see if a difference, change 9 to a 4. I have done this in the past, and its being reset to 9, seems not possible to stay at 4, not sure .... anyway, test it at 4 see what happens to RAC. Keep an eye on it to see if it resets itself to 9.

The purpose of the ini file is to present specific generic graphics card options to moo WU, that one inside the file you are changing, specifies the level of priority between GPU & CPU. It was 4 in the past, there is now a suspicion 9 is too much and 4 may have been ok.

Caution needed on rushed deductions, need evidence, but the more try the 9 to 4 change, the more solid the deduction. So at present its a suspicion only, need more to see if it makes a difference.

Regards
Zy

Overnight the results on the HD5850 are run times @ 1200 seconds (was 2,300) and credit 1750, was @6.5K to 7.0K.

Weider and weirder.

Have a look at file sizes, I cant check, PCs are hidden - particularly the numbers of Stat Unit Groups etc. Chances are you are getting the smaller WUs because you reset the Project, it will eventually reset back to bigger ones when the adaptive mechanism it uses gets used to the fact you have a card capable of bigger WUs.

Regards
Zy

Overnight the results on the HD5850 are run times @ 1200 seconds (was 2,300) and credit 1750, was @6.5K to 7.0K.

Weider and weirder.

Have a look at file sizes, I cant check, PCs are hidden - particularly the numbers of Stat Unit Groups etc. Chances are you are getting the smaller WUs because you reset the Project, it will eventually reset back to bigger ones when the adaptive mechanism it uses gets used to the fact you have a card capable of bigger WUs.

Regards
Zy

That's what I thought Zydor.

I looked at my processor reset from 9 to 4 last night and see both are still set to 4.

I will let the the small WUs run and they should reset in a day or so. But, the HD5970 is knocking them out on processor usage 4 as normal.

Overnight the results on the HD5850 are run times @ 1200 seconds (was 2,300) and credit 1750, was @6.5K to 7.0K.

Weider and weirder.

My experience so far

I looked at the run time I’m getting with a say”192” task on a HD6950 running 2 WU together using the app-info.
It varies between 625 & more than double 1435 seconds. All of then gave 1766 credits
This makes it impossible to fine-tune a box by looking at the run times
.
The priority change from 4 to 9 should affect only those running the CPU hungry ATI 11.7 up if the box is running CPU apps on all the cores.
I’m running 11.4 it shows a peak usage of 0.1 CPU x 2 in task manager.
I did try 11.8 & the 11.9 but with zero performance increase. All I got was the CPU usage jumping to 1 causing the GPU to wait for CPU time.

I read somewhere that this bug should be fixed when they release 12.1

I have found monitoring and logging GPU Load using GPU-Z, log to file, for Moo to be highly effective in troubleshooting run times, especially since I cannot always watch my crunchers.

During fragmentation of the dnet you will see very long times in which the GPU Load drops drastically and hovers around 55% for 5-10 minutes of each WU.

I solved this by doubling up the WU's using app_info.xml , this kept the GPU load at 99% and performed 4 WU's at a time and while I don't have the percentages persay, typical run time of one WU during fragmentation was 36-38 minutes, and running 4 WU's @ a time would be from 56-109 minutes.

I have noticed over the past few days that the fragmentation is nearly ceased on several of my systems and reduced those back to 1 WU per GPU coproc. With that said there is one of my systems that is actually dual 5870's still running 2 WUs per GPU coproc and the Credit coming from that system is amazingly high.

Overnight the results on the HD5850 are run times @ 1200 seconds (was 2,300) and credit 1750, was @6.5K to 7.0K.

Weider and weirder.

Have a look at file sizes, I cant check, PCs are hidden - particularly the numbers of Stat Unit Groups etc. Chances are you are getting the smaller WUs because you reset the Project, it will eventually reset back to bigger ones when the adaptive mechanism it uses gets used to the fact you have a card capable of bigger WUs.

Regards
Zy

I have nothing but 5870 and 5770 gpu's, in some cases two 5770's in one pc, and it seems almost all are getting the short units right now. I used to be at about an hour for the regular units on a 5770 now I am doing units in about 15 minutes. Some of my pc's have had the gpu's moved around ,while others have not, still even the ones left untouched are all getting the short units.

I changed back to prio 4 and the process changed from 38 minutes to 14 minutes for a WU.

Those of you who found changing priority back to 4 helped, are you on a Windows or Linux system?

-w

I changed back to prio 4 and the process changed from 38 minutes to 14 minutes for a WU.

Those of you who found changing priority back to 4 helped, are you on a Windows or Linux system?

-w

Windows (XP 32 bit in my case)