There are 50 new simulations in the beta queue. These are very simple test workunits, but if everything goes well I will submit a whole batch to the long queue tomorrow. This will help test the new app.

This new app also features the smaller upload sizes, so it would be good if we can deploy it for all long tasks ASAP.

I already had about 26 beta work units with similar results from one computer. From what I saw of the received beta work units that resulted in errors, others had similar results.

The same here: I18R1-NATHAN_tstdhfr-0-1-RND4913_3. W7 64bit GTX560Ti CC 2.1.

????
Project has no tasks???

22.01.2013 22:23:23 | GPUGRID | Requesting new tasks for NVIDIA
22.01.2013 22:23:23 | GPUGRID | update requested by user
22.01.2013 22:23:25 | GPUGRID | Scheduler request completed: got 0 new tasks
22.01.2013 22:23:25 | GPUGRID | No tasks sent
22.01.2013 22:23:25 | GPUGRID | No tasks are available for ACEMD beta version
22.01.2013 22:23:25 | GPUGRID | No tasks are available for Long runs (8-12 hours on fastest card)
22.01.2013 22:23:25 | GPUGRID | Project has no tasks available

I have now selected ALL... ACEMD standard, beta and long but nothing is coming in :-(
____________
Member of Boinc Italy.

Yes, we are a little "dry" right now (slight shortage of tasks). Bear with us for the next 24 or 48 hours. We are trying to change to an updated application, which we are testing in the beta queue. As soon as the new app is installed, we will have plenty of new WUs to send.

As for the WUs in the beta queue, the first 50 seem to be erroring. We are going to cancel them and send new ones very soon, which should be fixed.

15x error, 1x success, out of daily quota.
Finished task: I1R1-NATHAN_tstdhfr3-4-10-RND6104_0 . Any other task than I1R failed as skgiven mentioned.

I just destroyed something like 20 of the new beta tasks by getting a computer error on them after only 2 seconds of starting them. I don't overclock my cards, and they are at as low as I can get them in terms of temperature, which is 70 C or less, so I don't know why they failed.

We're aware of the continued problems. We believe it is an issue with the application, but are checking the simulations as well. We'll update as soon as we know something.

You can see why we beta test every change!

The NATHAN_tstdhfr4 batch seems to be working fine on all of my hosts.

However, there's still no detailed info in the stderr output file about the GPU used to process the workunit.... (which would be very useful for figuring out the source of some errors)

The NATHAN_tstdhfr4 batch seems to be working fine on all of my hosts.

BTW I've received some Ann***_r*-TONI_AGGd8 workunits from the long queue, so there is some new batch also.

There was an issue with the simulations that has been corrected, and most seem to be finishing successfully now. We are having some other issues with the application that need to be addressed before we can deploy the app on Long. Sometime in the next few days, hopefully.

Recently received two Beta's. One ran successfully, the other did not.

Project has no tasks available

and i still wonder why everybody in Austria except me get work units and make up to half a million points per day O.o
____________
DSKAG Austria Research Team: http://www.research.dskag.at

Are the new workunits now with smaller upload files? Because the appversion didnt change so i wonder ;)
____________
DSKAG Austria Research Team: http://www.research.dskag.at

trypsin_lig_1_-NOELIA_RL_equ-0-1-RND1972_0 finished at 2202s @ W7 64bit GTX560Ti CC2.1/872MHz, driver 310.90. Low GPU load (begun at 40%, finished at 30%), low VRAM load (119 MB), high CPU load (0,7 CPU core of HT 3770K) => low credit.

getting around 36-39% usage on GTX 470 in win7
____________
XtremeSystems.org - #1 Team in GPUGrid

can not force project to send me beta wus, excluded everthing but beta , but nothings been sended.

must select "Run test applications?"

also, currently seeing 16-19% usage... not a good sign :(
____________
XtremeSystems.org - #1 Team in GPUGrid

must select "Run test applications?"

also, currently seeing 16-19% usage... not a good sign :(

also, currently seeing 16-19% usage... not a good sign :(

I am running a "acemdbeta version 648 (cuda42)" task called "trypsin_lig_1205_run3-NOELIA_RL2_equ-0-1-RND5309_0"
... and, monitoring GPU Load using GPU-Z
... the task appears to only be using 33% of my EVGA GeForce GTX 660 Ti FTW (clocked at 1045.2 MHz), using 313.95 (beta) drivers.

This means the application is under-utilizing the GPU, right?

right, u could use 3 wus at the same time with an "app_config.xml" file. And boinc version 7.0.44. Specified the app name = "acemdbeta".

... or is there a bug with the app that leads to poor GPU load?

counldn´t run any beta app right now.

Rantanplan, please try to post these kinds of issues in "Number Crunching" or "Server and website". Make sure you don't have specific location preferences set for your computer under "GPUGRID preferences" It is explained more here: http://boinc.berkeley.edu/wiki/Preferences#Location-specific_preferences


As for those of you noting the low GPU usage on these WUs (Noelia's WUs, to be specific), you are correct. They are experimental and for this reason we put them on the beta queue. We must run a few of them as preparation for many, many more simulations that will come later on the long or short queue (where they will no longer have the low % utilization issue). They shouldn't result in low credit, and typically we compensate heavily for beta WUs. I will look into that.

Thanks for the confirmation about low GPU usage not being a bug, Nate.

For future beta tests, for beta aspects that function differently than the final version but are still expected behavior, it might be wise to announce them up front. As it was, I (and other users) had no idea that low GPU usage was expected for this beta run.

Thanks,
Jacob

Pulling out of BETA testing as several Noelia ones on different machines are stalling. This one got to 20 hours before I could abort it!

These Noelia beta units are entirely too dependent on the CPU. I had to free up another CPU to get them to run faster. They also have a habit of starting out fast and then slow down toward the end. I observed GPU usage can be as high as 77% at the beginning of the task, and drop to 13% in the end.

beta units were working ok, albeit very low gpu usage, and then yesterday they started locking up my system. Had to do a hard reset. Now system locks up on boot when BOINC starts. Host id 125227, win7 x64 driver 310.70. Will try to update driver.
____________
XtremeSystems.org - #1 Team in GPUGrid

Seeing LOTS of errors on the Beta tasks right now.
They say:
ERROR: file mdioload.cpp line 207: Error reading parmtop file

Did I do something wrong? (I recently added the WUProp project)

Yesterday, I opted out of the beta testing because of the problems it was causing on my computers. Yet, I still kept getting beta WU's. I had to abort them, then set GPUGRID to "No new tasks". Forcing beta WU's on us is NOT COOL! You are jeopardizing the integrity of BOINC. Please let us know when it's SAFE to VOLUNTEER on this project again.
Regards (but disappointed),
Rick
____________

When I attempted to opt out of the GPUGRID Beta, I did:
"go to GPUGRID preferences,
Unselect "Run test applications?"
unselect "ACEMD beta"
and unselect "If no work for selected applications is available, accept work from other applications?""

I also did it for all my profiles:
Primary (default) preferences
Separate preferences for home
Separate preferences for work
Separate preferences for school

After I did all the above, why did I continue to get beta WU's?

Rick
____________

I have Short & Long projects selected for the default, and all 3 other profiles. Only the beta is unchected. This was the situation when I continued to get beta wu's and had to click "No new tasks" in the BOINC Client in order to keep my computers from freezing up and rebooting. I would love to resume GPUGRID work, but I'm concerned I'll still get more beta wu's, like I did yesterday. Is there a long lag between unchecking beta in the profile, and the scheduler getting the message? Rick
____________

Regarding Sponholz's inability to turn the betas off.. I haven't tried it myself so I can't verify that it works or does not work but I know this feature fails at 2 other projects and they claim it's a known bug in the server code. I have no idea if they are correct in saying it's a known bug, whether they have the feature misconfigured, whether they're using the same server code as GPUgrid or whatever.

The reason I have not tried turning beta off to test the feature is that I could never be certain that I am not not receiving betas simply because they have none to send, so it's pointless trying to test it.

All my Beta's fail except those ending in 1:

trypsin_lig_965_run2-NOELIA_RL2_equ-0-1-RND9201

Stderr output

<core_client_version>7.0.28</core_client_version>
<![CDATA[
<message>
process exited with code 98 (0x62, -158)
</message>
<stderr_txt>
ERROR: file mdioload.cpp line 207: Error reading parmtop file
21:06:01 (6125): called boinc_finish

</stderr_txt>
]]>

Thanks for the 'undebugable' server code bug tip. Might explain some of the scheduler quirks, or not.

Yesterday I got 3 Beta's and 2 errored out very quickly. By the third the system froze after a few seconds running the Beta and had to reboot to get any response. After the reboot I got the message that the graphic driver was restored after failing.
Windows Vista Ultimate x64, i7 12GB Ram, boinc 7.0.28, nVidia 285GTX driver 310.90

Long runs run fine on this machine, but take most of the time around 30 hours.
____________
Greetings from TJ

Yesterday I got 3 Beta's and 2 errored out very quickly. By the third the system froze after a few seconds running the Beta and had to reboot to get any response. After the reboot I got the message that the graphic driver was restored after failing.
Windows Vista Ultimate x64, i7 12GB Ram, boinc 7.0.28, nVidia 285GTX driver 310.90

Long runs run fine on this machine, but take most of the time around 30 hours.

Looks like the GPUGRID server is hung up, I've had this message in my Acctivity Log;

Sat 02 Feb 2013 08:27:17 AM MST | GPUGRID | Server can't open database

I was under the impression that fan speed shall be kept at max during crunching, and for me, everything I do.

My observation, on running the NOELIA_RC3 beta units, is they running faster and are less CPU dependent the NOELIA_RL2 units, but they still need work. On Windows XP, the gpu usage is about 60% to 80%, and by freeing another cpu, this increased by a few points. On Windows 7, the gpu usage is about 30% to 50%, with no noticeable increase is gpu usage when I free up a cpu. Also on Windows 7, gpu usage decreases as the unit crunches, though on XP, I didn't notice the drop.

I just finish crunching a I47R1-NATHAN_tstdhfr6-0-1-RND5218_0 on a Windows XP machine. This had gpu usage at 97%. See link:


http://www.gpugrid.net/result.php?resultid=6463597

My observation, on running the NOELIA_RC3 beta units, is they running faster and are less CPU dependent the NOELIA_RL2 units, but they still need work. On Windows XP, the gpu usage is about 60% to 80%, and by freeing another cpu, this increased by a few points. On Windows 7, the gpu usage is about 30% to 50%, with no noticeable increase is gpu usage when I free up a cpu. Also on Windows 7, gpu usage decreases as the unit crunches, though on XP, I didn't notice the drop.

I just finish crunching a I47R1-NATHAN_tstdhfr6-0-1-RND5218_0 on a Windows XP machine. This had gpu usage at 97%. See link:


http://www.gpugrid.net/result.php?resultid=6463597

The next thing is when the computers run these NOELIA_RC3 beta units......

6 days after changing all my profiles to NOT accept beta WU, they are still being forced on me. This is disgusting behavior, causing my machines to lock up and requiring a reboot. I've had to place a "No new tasks" embargo on GPUGRID until you get your act together. Moderator, please notify us when it's SAFE to resume volenteering for GPUGRID. Dissappointed, Rick
____________

Do you have set yes in the "run test apps" check-box? If you do, un-check it.

I've had the "ACEMD Beta" UNCHECKED for 6 days. I'm NOT using ANY manager. The only project I'm having unwanted beta WU's is from GPUGRID. However, I would really appreciate you letting me (us) know when the beta testing has stopped, so I can resume accepting ANY GPUGRID tasks. Thanks in advance, Rick
____________

I've had the "ACEMD Beta" UNCHECKED for 6 days.

I've had the "ACEMD Beta" UNCHECKED for 6 days.

Do you have set yes in the "run test apps" check-box? If you do, un-check it.

Retvari, you are my hero! I also feel real dumb, because I did not see the run test apps box above the other application choices. Thanks all of you for helping me get it right. I'll begin accepting WU's again, and contribute to the cause. Regards,(and embarassed) Rick
____________

Dagorath & Rick, I have nothing to lose or gain either, other than fun so it's fine by me and I'm quite prepared to 'up the ante', significantly!

It does go both ways, I wanted to run some Beta Apps on one of my machines, and it turns out for nearly two weeks I forgot to Check the "Send Test Applications " box. Hopefully I start getting some soon, but its my slowest machine, and since the only tasks have been the Huge Long Runs, it only searches for work every two days or so.
____________

Besides testing a new application, are we doing any other scientific work with these betas?

Actually yes, WUs in betaqueue right now are the first step of the simulations which will be sent to short queue. So this is already the real thing.

I have had a beta WU that completed without error:
6 Feb 2013 | 21:58:39 UTC 7 Feb 2013 | 2:17:47 UTC Completed and validated 12,202.02 2,548.72 13,500.00 ACEMD beta version v6.48 (cuda42).
It's about half time faster on my pc (i7, vista ultimate x64, GTX285) but uses approximately 800 seconds more CPU time. Temperature of the card was 60°C while non-beta it is 70-74°C.
____________
Greetings from TJ

I suspended a queued GPUGrid Beta this morning, allowed a running GPUGrid task to finished and then Resumed the GPUGrid Beta, trypsin_lig_127_1-NOELIA_RC3_equ-0-1-RND4222_0 (still running).
It started at 74% GPU utilization, rose to about 76% and then suddenly dropped to 66%. The power usage was about 65%, GPU temp 46°C with the fans fixed at 75%. Core clock was 1176MHz, which is normal. GPU utilization (W7x64) has subsequently dropped to around 55% and is fairly erratic; varies between 50 and 60% (two CPU threads free).

Hi,
we cannot test until the beta queue is cleared.

There is a problem with the new app and it has been difficult test it out.

gdf

I had particularly bad experience with a beta unit. Most bad WU simply give you computation error message when they crash, and you go on to the next WU, without any reboot or computer crash. This unit ran for a few seconds, froze up the computer, then blue screen, and the computer reboots. It did this few times, before I aborted the unit. It also cause another perfectly good WU to crash as well.

Here is the link:

http://www.gpugrid.net/workunit.php?wuid=4137081

I think this one belongs in the list too:

http://www.gpugrid.net/workunit.php?wuid=4138484

It gave me a BSOD after 7 seconds with

The problem seems to be caused by the following file: dxgkrnl.sys

STOP: 0x00000116 (0xfffffa801b5de010, 0xfffff88006dc4404, 0x0000000000000000,
0x000000000000000d)

That's from BlueScreenView: the original had a line about an NVIDIA driver crash and not restarting within the time allowed. BSV doesn't seem able to retrieve that information - I'll record it manually (and accurately) if it happens again.

Thanks for bringing it to our attention. There was an issue building a small number of the simulations, which our checks didn't catch before they were sent out. We have cancelled the work units that were crashing machines, but it is possible that there are others so let us know if it happens again. Crashing your machines is obviously the last thing we want to do. In the future we can avoid this with additional checks we'll be doing for this type of work unit.

nate

This WU: trypsin_lig_904_3-NOELIA_RC3_equ-0-1-RND0962, resultied in the nVidia driver to stop. However it recovered automatically without booting the system.
____________
Greetings from TJ

Had a repeat of my BSOD, though this time it seemed to be another project which triggered it.

The exact phrase on screen is:

"Attempt to reset the display driver and recover from timeout failed" (stop 116)

There was also a reference to nvlddkm.sys

Host is an i7 3770K with dual Gainward Phantom GTX 670 - driver 310.90 WHQL

Here is a beta unit that ran rather slowly.

trypsin_lig_491_run4-NOELIA_RL3_equ-0-1-RND5688

13 Feb 2013 | 11:57:24 UTC 13 Feb 2013 | 19:50:06 UTC Completed and validated 26,096.32 25,685.34 1,500.00 ACEMD beta version v6.48 (cuda42)


See link:

http://www.gpugrid.net/workunit.php?wuid=4142465