My bad. I thought everyone knew that using the app_info would cause you to lose your CPU tasks. You can run CPU tasks with GPU tasks when using the app_info it just requires more code be added to the app-info file for each of the CPU projects you want to run.

That annoying display driver crap is Windows fault. What happens is the new video stack in Windows Vista/Seven runs seperate from the kernel so that video driver restarts are possible without a full reboot. Thus preventing many video related BSOD's. However as a part of this if the driver doesn't respond to windows fast enough windows automatically restarts the driver and throws an error when actually nothing could be wrong. It is really annoying on some systems. Some systems have higher internal latency (can't remember what that is called right now) that cause this to trip. There is a registry fix for this that I tried and it worked for me.
Copy and paste the following into notepad and save it as a .reg file and run that to add the following to the registry. Copy and paste all of it. I named mine repair.reg

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Watchdog]
"DisableBugCheck"="1"

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Control\Watchdog\Display]
"EaRecovery"="0


Right click on the file and open with registry editor. It will ask if you want to add this to the registry, select yes. That fixed the display crashes for me. As always, there are dangers when working in the registry so try this at your own risk.

I can't take credit for this fix. You can find the thread about it here XtremeSystems

This has nothing to do with power of the other machines....When your wingmans set a cache of, as example, 3 days, you have to wait up to 3 days until they get calculated on the wingmens machine. Nothing to care about...

FWIW your cache settings will be ignored for any setting that would give you more than 100 WUs per GPU. That's the limit.

If you forget to include an application in your app_info.xml, all work for the unlisted applications will error-out. It's always been this way, since optimized applications for SETI@home was made available back in 2005.

But while here at WCG users must make their own app_info.xml or try a file posted to the forums, for other BOINC-projects the app_info.xml is normally downloaded together with the application, either optimized or beta, and these app_info.xml often includes info of older application-versions and so on, to make sure no work currently on client will error-out.

HT doesn't really make you able to run two tasks with one core, there is only one execution unit in your core.

The processor is a FX-8120 @ 4Ghz is plenty for this many tasks, but WCG is limited to running 9 work units when I was trying to run 16 to fully saturate my processor and further eliminate droops.

How are you not following? The GPU isnt being used when the CPU is working, and the CPU isnt being used when the GPU is crunching. Open MSI Afterburner Hardware Monitor and monitor your GPU usage, you'll find there is a huge gap where the GPU is doing nothing, and monitor task manager to see a huge gap where the CPU is doing nothing, so much potential lost. If you run one or two work units. The more work units you run, the less gaps you will see, this will keep both your GPU, CPU, and RAM running at the best they can. A single 7970 can complete the GPU portion of a work unit in around 17-18 seconds, so we are trying to reduce the amount of CPU time by increasing the number of work units beyond the max cores you have. The more real cores you have the better.

Also, there is no lag on this machine, I've been watching Hi10 10 bit anime on this machine while doing this, but recently just installed Synergy so I can move my mouse off the side of my screen and my mouse cursor pops up on my Ubuntu machine so I am doing all of my x264 encoding and anime watching there and give HCC extra CPU cycles.