No one can read that error. So if it's the same as before, no need to repost. If it's a new one, please repost.

But let's try something different: 3DMark, download, install, run the Firestrike Test/Demo.
See if it survives that. Any artifcating you see, tell us about it.

You can also install Whocrashed and have it run over your dump files to tell you what crashed your computer.

Some observations:
1. There have been no changes between these versions with return to how GPU tasks suspend/resume.
2. You also changed your videocard drivers. It's much more likely you fixed your problem doing that.
3. BOINC by itself does nothing with the GPUs. It merely detects what kind of GPU is in the system, according to the drivers that you installed and what the capabilities are (CUDA & OpenCL mainly). It does not put load on them, which will usually cause the BSODs.
4. 7.7.2 was mostly a concept release to see if the developer in question could pack everything together, build a working client out of it and if the installer he built worked in the different Windows versions (which it doesn't).
5. In itself any 7.7 version will never become a recommended version, because of its numbering. Major.odd.revision means it's an unstable alpha/beta version. Major.even.revision means it's a stable release candidate/recommended version. So any next recommended version will always be a 7.8

Please, could you stop upping this topic? The Prefsremodel page has been around since 5 years ago. It's not one that will be done quickly, and until someone finds someone else to do these things, it's probably never going to be added all.

So, if you truly want what's on that page added to BOINC, best find a buddy programmer with lots of time on his hands. Continuing to up this topic or asking every 2 weeks if it's done yet isn't going to be helpful.

Whenever http://boinc.berkeley.edu/trac/wiki/PrefsRemodel can be implemented, be it fully or partially. Depends on if programmers for it can be found.

At the moment the three main BOINC developers are no longer paid by BOINC, so they've fanned out to other paid jobs that they primarily work for to be able to live their lives and pay the rent. It all is an open source thing, we'll have to wait until someone capable says they want to take it up.

I've just sent this to the BOINC Development email list:

The past couple of days we've had a person try to suspend individual GPUs in https://boinc.berkeley.edu/dev/forum_thread.php?id=10698
When he games, he wants his Nvidia GPU to suspend and his AMD GPU to continue.

I'm thinking this could be solved by adding a couple of exclusive_gpu switches, <exclusive_amdgpu_app>, <exclusive_nvidiagpu_app> and <exclusive_intelgpu_app>, where when these are set and the exclusion program is in memory all of the AMD/Nvidia/Intel GPUs get suspended, but those not specifically mentioned continue.

Is that a doable thing?
Is that something we'd want to add? Or is that thought to be too fiddly?

I'm thinking these new switches can be used for finer control than just <exclusive_app> and <exlcusive_gpu_app>

28-12-15 16:08:41 | | OpenCL: AMD/ATI GPU 0: Hawaii (driver version 1912.5 (VM), device version OpenCL 2.0 AMD-APP (1912.5), 4096MB, 4096MB available, 3200 GFLOPS peak)

@xixou hi, what model AMD GPU ?