Anyone noticing work not currently being issued to GPUs?

For the first time in the last 24~hours my 3850 cache has just dried up, but the Penny CPUs are going OK?

OK, not sure but panic over? Just D/Led 4 more, but not up to cache size now considered normal.

The CPU was getting red in the messages tab of BOINC Manager on "upto 6 CPU limit" and no work available.

---

Go away, I was asleep

I haven't seen the 6 per cpu limit on my quady with the GPUs since the torrent began, but I think that is more of a BOINC issue and not a MW issue as the machines that I have cpu only crunching are seeing the 6 per cpu limit. BOINC doesn't understand MW GPU crunching and so treats it like a CPU application and so it's a little fubar. You really have to set the resource share just right if you are trying to do both cpu and gpu work.

Yes you do, but the old GPU cache seems to have filled up again. So, the loss of work was temporary and very much local.

---

Go away, I was asleep

Amyone having problems downloading MW work to their GPU?

One of my rigs has just completed the cache of MW work, and is sitting there, unusually, not downloading new work. The other rigs - mixed CPU and GPU - seem to be OK.

I say unusually as for the last 20 hours it seems to have maintained a good cache, which filled back up when it was down to crunching the last 3 WUs.

Sorry to post to my own question

It took nearly an hour (2 x 8 FreeHAL WUs completed), but the new work seems to be flowing now.

---

Go away, I was asleep

I don't get it ...

http://milkyway.cs.rpi.edu/milkyway/results.php?hostid=65070

This guy is getting 12 WUs to process whereas I'm only getting 2 !

What settings is he using in app_info.xml ?

Or should I try using the same client as him 6.6.20 ?

---

Amyone having problems downloading MW work to their GPU?

One of my rigs has just completed the cache of MW work, and is sitting there, unusually, not downloading new work. The other rigs - mixed CPU and GPU - seem to be OK.

I had this quite recently, just days ago, two of my rigs just sat there with a column of "Ready to report", but did nothing else. "Update" got things going again, and I still check from time-to-time that my rigs aren't taking unauthorized tea breaks.

---

No worries... I've got it sorted...

Slip back to 6.6.20 and modified app_info, now the GPUs are fully loaded...

Next problem is that there's a bit too much heat on the motherboard ;)

---

I got home yesterday arvo to find a bunch of 'uploaded but not reported' WUs.
Clicked Update, '41 WUs reported'.
So I've restarted my update script, but changed it to run only every 15mins or so, so as to reduce the backlog.

---

Cheers,

PeterV

.