Display driver crashing causing virtual machines to mess up.

[Simonzi]

First, here's the specs of my machine

i7 2700k
Gigabyte P67A-UD7
AMD 7970
16GB G.Skill Ram
256GB Crucial M4
3x Dell U2412M's in Eyefinity

I have 5 VM's set up, each with 1 processor\2GB of ram. Randomly, all 3 screens will go blank, and I'll get the error on my desktop (not inside a VM), that my display driver has stopped working and recovered. Which is great that it recovered, but the VM's don't recover. I have to restart them all. Which sucks, if it happens 20 minutes after I leave for work, or go to bed. I'm using the 12.6 Catalyst drivers, which I just did a fresh install of yesterday, uninstalling the old ones with DriverSweeper as well as Catalyst Uninstaller. I was going to update to the 12.7 Beta drivers, but the link was dead on AMD's website, so I could not download them. This is getting really frustrating, and am looking for any help and suggestions.

[Assistend]

check your GPU temp, seems to be either overloaded or overheating

[psychoangler]

i have the same problem currently, i tryed various drivers for my ati card and i cant get it to work properly.

temperatures are 50°C card is not used at all (as expected) i tryed to set clock timings for the card down, no help, it just feels like ati drivers are too bad :/

i have ordered a gd 560gtx to test it out which should arrive tomorrow

[Ledoocat]

This is something I could never stop happening, although it was irregular.

My only solution was, with 4 VM's, 2 processors and 1gb RAM. This reduced the amount of crashes.

[boredevil]

suffered from that problem for over a year. but with nividia

updated everything, tried the tdr fir, tried setting energyplan to full power...
this seems to be an error when switching power states.
in the end i had to use msi afterburner. "unlock voltage control". set it up, so your card always runs with full vcore even across energy plan changes. no more freezes since then.

a friend of mine also had the same problem. but for him, lowering gpu/memory clock just a little bit did the trick.

maybe you want to try one of the above

[Simonzi]

Yeah, the card doesn't really run hot at all. I upgraded from a GTX480, and that thing could bake a pizza.

I'll try some of the suggestions in this thread. It's just overly annoying. Some times it will go 10 hours without error. Other times, half an hour.

[turbobee]

I'm also experiencing this problem on my EVGA 580 Ultra Classified. The card itself doesn't seem to be under much load, plenty of vram, temps around 65-70c, gpu-z reports 15% load. I'm trying what boredevil suggested or at least similar to what he suggested. I changed my nvidia control panel settings from adaptive to performance as per the instructions of an EVGA tech support rep for keeping core voltage the same at all times. I'm hoping this works since it would be a waste to only be able to run a couple vm's at a time. Will post again once I have some results.

[siklon]

Having same crashing issues with my dual GTX295. I changed some settings in Nvidia control panel to maximum performance, we'll see if that fixes it.

[Markn12]

I have the same issues with a ATI 6850, and Nvidia GTX 480.

[siklon]

Oh god it crashed again, real annoying...

[Markn12]

It isn't a driver issue guys. Its a GPU Load issue if you want to be able to run say 3 VMs with 2-3 bots each get a 7850 or better. I've used 3 Video Cards in the last month 6850, GTX 480 and 7850 results are stunning.

6850 4 bots+DB GPU Load 75% + entire time
GTX 480 5 bots+DB GPU Load 85% + entire time
7850 6 bots+DB GP Load 28-34% entire time.

I have not had a driver fail issue since switching to the 7850 in VMware. This is using innerspace on the VM btw with isboxer shrinking the windows to 100x50 with an inactive window at 800,600 that i can bring up one of the smaller ones and send it back to home region when im done looking at loot.

[Simonzi]

It isn't a driver issue guys. Its a GPU Load issue if you want to be able to run say 3 VMs with 2-3 bots each get a 7850 or better. I've used 3 Video Cards in the last month 6850, GTX 480 and 7850 results are stunning.

6850 4 bots+DB GPU Load 75% + entire time
GTX 480 5 bots+DB GPU Load 85% + entire time
7850 6 bots+DB GP Load 28-34% entire time.

I have not had a driver fail issue since switching to the 7850 in VMware. This is using innerspace on the VM btw with isboxer shrinking the windows to 100x50 with an inactive window at 800,600 that i can bring up one of the smaller ones and send it back to home region when im done looking at loot.

I have a 7970, that doesn't go past 60% GPU load with 5 bots, so I don't really think this applies to me. I don't have them running with ISBoxer though, they are all just running at 1920x1080.

[psychoangler]

how could it be a gpu load isse -.- my gpu doesnt go over 20% load when running the vms. im pretty sure it has to do something with vm ware, if i use my regular pc i can start 15 diablo 3 windows at once and it doesnt matter, the gpu goes to 80 - 90% but it doesnt crash. pretty sure vmware 3d accelleration buggin out

[Simonzi]

Are there any alternatives to VMWare?

[yellowspark]

Are there any alternatives to VMWare?

Try Sandboxie