I just checked yesterday's results. First I subtracted hcc's cpu time and results from the total and calculated the average result from everything except hcc. I got around 5 hour per result. Then I took a random date from prior to gpu and I got 3.19 hours per result or an increase of about 60%. That could explain why my fastest 6 core computers are only doing about 70% of what they were doing prior to gpu. It would not explain why I am only getting about 65% of real time for cpu time for my worst two computers and only 85% for all of them. In other words my two worst computers one of which has a working gpu are wasting around 35% of their electrical cost since they seem to be producing nothing during 35% of their time.

I have already pointed out an error on wcg's part. I am still running a clean energy unit. The only way I could get a unit besides hcc is if there were no hcc units to give. I would think that clean energy would be the last to be given so am I to assume that there is a lack of units in all the rest of the projects thus forcing wcg to give me a clean energy unit? I am still in 404th position in cpu time(down by about 20 since gpu) and 432 in points(down by around 60 since gpu) and 542 in results returned(down by around 38). Here is a link to a story from Nividia http://hothardware.com/News/NVIDIA-CEO-Unveil...lized-Rendering-Systems//. As anyone can see the gigaflops per watt are increasing a lot in the near future. The time when the cost of generating the results will be less than the cost of the bandwidth to transmit and receive them is coming soon. I need to maximize my effort.

I'm seeing "No" shortage of work. It "All" depends on what you've set your system to run. CEP2 does take alot to run..Hince the reason you've got to opt-in (to run more than 1 at a time). I've got 47 cores running and average 44dys CPU time daily. If your not (seeing similar) it's in your settings and/or the amount of time your system is crunching.

The problem more than likely is on your end and not WCG's.

Please help everyone help you and post the requested/required info. So everyone can troubleshoot your problem.

To not get CEP2 tasks with "If ..." ticked, a known "hole" on the ToDo list is, to set the allowed disk space for BOINC to a lower value. Each time a CEP2 is send, the server checks if there's 2GB free for BOINC, if not CEP2 is not send.

It's easy to compute how much space that needs. When saving the local prefs, the log prints how much is free to use. Lower the "use at most xx GB diskspace" value so that the printed value in the log drops below 2GB, say 1.8GB. Far as I know, there's a limit on "In Progress" buffering to include HCC. which when maxed easily fits in 1GB [But have not checked the actual HCC storage use, so it will have to be a user hands-on to find out the correct value]

Hope this one helps those who have a problem processing the occasional CEP2 slipping through.

I guess I can not complain any more since I am again going up in rank in cpu time. I am now 402, up from 404 so working with my preferences has helped. Thanks everyone.