Chowbow,

Do read the link Sekerob provides above. In the next 2 days you will likely see your ratio go up. Within 3 days I suspect you'll see 99% valid. If something were wrong with your computer then you would likely be seeing more "errors" and fewer "pending". Give it a day or 2 before adjusting things on your end.

I ran the queue down to nothing and accepted new tasks. When I opened it, it wanted 216000 seconds of work. One workunit was downloaded and it said my computer was overcommitted again. That has been happening for a couple of days.

I reset the WCG project and it just sat there sulking and didn't do anything.

I downloaded a workunit from ralph@home. That must have got WCG/BOINC's attention and it thought it should do something. It asked for 325000 seconds of work and immediately downloaded 16 workunits. <shrug> I only had ralph set to 0.1% of resources but a 3 hour cutover. FAAH was shorter, so it downloaded the program quickly and started running. HPF2 is downloading the program again since I reset it.

It's not a punishment, it's a safety precaution. It's designed to stop sending work to people who never return valid results because their computer isn't working properly.

Don't worry, BOINC will get back to business as usual soon enough.

My system is not overclocked, I refuse as it decreases component life, and my memory is brand new, and thoroughly scanned with memtest and no errors, so my PC is not at fault, it has to have something to do with the tasks themselves, my CPU is an Athlon64 4000+ so its not a rubbish one. Its only been happening the last few days, so it could just be a teething thing with the new hpf2 projects.

same guess: Dagorath's message got in while I was formatting this longer one.

We're just trying to eliminate common things that have caused errors in the past.

How much memory is on these systems? I have no less than 512Mb on my own 2.4MHz AMD systems, and am not having problems with either HPF2 or FAAH. The HPF2 are even running faster than the 5 hour FAAH, completing in about 3.5 hours.

One system is overclocked a bit and two are not. The new overclocked Venice core even runs a lot cooler than the systems that are not.

I can't reproduce this problem myself, as I have had no bad workunits here.

If you still have some HPF2 version 5.06 around, they have had some freezing problems. The HPF2 version 5.07 should automatically be sent for the next download. I aborted some of the 5.06 myself, but that causes problems with quotas. I'm back to normal activity now, with about 16 workunits on each computer, with 2.5 days set in BOINC.

It is also possible to set your overall project profile to FAAH or HPF2 if you are having a problem with one or the other. From the messages it sounds like you are using BOINC. Projects in the UD interface are set in the device profiles.

These projects usually run better on fast AMD machines anyway. Please keep trying.