My GTX 260 has always gotten 2 WUs at a time. Yesterday I got 6 in total. Is this normal? Has been on GPUGrid since about the 8th.

It seams it is 4 task per gpu or 2 per cpu core...?!

---

POLISH NATIONAL TEAM - Join! Crunch! Win!

WHY am I getting so many WUs? I set the prefs to 1 day and aborted 4 WUs. Sometime later it gets 4 more?

We have done a server update few days ago, but not changed any other setting.
Is anyone receiving more WUs?

gdf

We have done a server update few days ago, but not changed any other setting.
Is anyone receiving more WUs?

gdf

I added a new machine with 190.62 drivers and BOINC v6.10.0. It was up to 5 WUs a couple days ago and 3 last night.
I think it might be related to the new v2.3 app test.

i did recieve like 5 wu at once, but have only recieved 2 since but that of course, i hit no new task after the first 2 are sent.

i'll let it naturally try later. and report the results.

I received 6 workunits today, all with a severe underestimate of the usual time they take to complete, after a power failure apparantly lost all of the workunits already in the queue on my 64-bit computer. Are they actually shorter workunits, in which case this is reasonable, or did this computer also lose something needed for an accurate estimate of how fast the workunits will run?

i've got 4 wu's for my one card on a quad.

I've got GPU-Grid set to 1/4th the combined ressource share of the cpu projects and have a cache setting of a few hours (so BOINC caches enough work for about 1 day) and am not getting more WUs than previously.

MrS

---

Scanning for our furry friends since Jan 2002

Thanks for replies!

I set resource share to 25 and "Maintain enough work for an additional" .5 days. Hit update and in messages it says I have reached max of 6 tasks.

I hate to keep aborting tasks so will wait and see if it downloads more when it finishes the current task.

Thanks

2009-08-26 07:48:15 GPUGRID Message from server: (reached limit of 6 tasks)

---

POLISH NATIONAL TEAM - Join! Crunch! Win!

I received 6 workunits today, all with a severe underestimate of the usual time they take to complete, after a power failure apparantly lost all of the workunits already in the queue on my 64-bit computer. Are they actually shorter workunits, in which case this is reasonable, or did this computer also lose something needed for an accurate estimate of how fast the workunits will run?

I got the message about a limit of 6 workunits also, and have disabled downloading any more GPUGRID workunits until I see how accurate the apparantly rather low expected time is for the ones I already have.

Nothing I do seems to make a difference. It is maintaining 5 in queue one working. I set it to no more tasks until this queue clears as I am now no longer getting bonus points.

Which boinc version do you use?

gdf

6.6.36 on Vista 64, 190.38 drivers, GTX 260 216.

Of the six workunits I downloaded at once, all with an estimated time of a little over 2 hours, at least two reached a computation error after about 11 hours, and one was aborted by the project.The last two now have estimated total time back near normal, so I have enabled getting new workunits again.

My computer apparantly erased its copy of these workunits during a power failure, or possibly during my driver upgrade to 190.38.

1152064
1152000
1139877
1137835

I'll let you decide if it's better to send another copy of them, or reassign them to someone else.

Did it again. Had 2 Redundant results so was down to one on GPU, enabled Get new tasks and it got 5.

I've tried setting different settings and nothing makes a difference. I just got 6 more.

A person on our team with a GTX 260 has stated it is happening to him. I asked him to post it here,

Yes, as PG said, I am having this problem now too. I'm getting 6 tasks at a time, and am not consitent enough to abort WUs. I'm running Bionic 6.5.0 on Win7 64bit on a GTX 260 216 Driver 190.37. Any fix?