Ok So I upgraded my GPU i think the rest has to go...
So i just purchased a MSI Radeon HD 6870 Hawk edition got it for 175$ on craigslist and love it! Though i am having some freezing and it is recovering from errors is this my cpu or maybe Mobo PSU? i have a feeling its my Biostar Mobo or Rosewill 550watt psu the Cpu is a phenom 2 925 amd quad core. I want to eventually crossfire another Hawk will my cpu bottleneck? I really want to get an i5 2500k anyway so im thinking that processor and Dual hawk 6870 will that work out well?
What exact PSU model is that? Rosewills are bad in general. If yours came with the case, you should go ahead and upgrade to a non-free PSU.
Whether or not you get a CPU bottleneck depends on the game. Nearly all games can run just fine on a Phenom II x4 that's overclocked. Have you tried overclocking yet?
I feel like an i5-2500K isn't the best buy right now when you have no guarantee that the CPU is your issue and the Ivy Bridge refresh is right around the corner. Could you describe the problem in more detail? We should figure out what's causing your errors and freezing before you go spending a bunch of money.
While i play WOW the game will freeze then the screen will go black my monitor will flip out losing DVI signal from the GPU, then the GPU it will recover and come back, only went blue screen once. Yes i am overclocking stable at 3.4ghz atm. i just got my GPU and downloaded the drivers from radeon should i maybe download MSI drivers? thank you for your time
So I can assume everything ran perfectly prior to hooking up the GPU?
Based off of what you said, it sounds like your driver is crashing. That can be caused by numerous GPU failures such as a bad overclock (too high) or maybe not enough power? A driver update may fix this. Radeon's website should have you covered and you probably won't benefit from downloading MSI's drivers.
Something you may find very useful though is MSI's Afterburner and Kombustor tools. Download and install those (Google to find them). See if the Kombuster (essentially Furmark) benchmark/burn-in tool crashes your driver. See if the crashing occurs in non-WOW games. You can use Afterburner to reduce your clock settings and see if the crashes still occur.
There's a possibility your graphics card is not stable at it's current speed and needs to be underclocked (essentially meaning your card is somewhat broken).
Once again, we still need your Power Supply Unit (PSU) information. We can use that to tell you whether your card is getting enough power.
I'd say test the card though first since craigslist cards are known to be overclock failure ridden and likely Artifacting or just broken in general.
I'd also say the psu could use an upgrade regardless if it's the problem or not. Rosewill like people have said before is less than reputable in my opinion, especially for a PSU for god's sake, essentially what your entire system hinges on. I'd say that 550 watts isn't over 1 rail but multiple rails. Meaning it actually has less peak power than advertised.
You can look at the side of your psu and it has a table of numbers. Looking at a random rosewill 550 watt psu it says there is only 525 watts on the the 12v rail which is decent but maybe not enough for your new card.
Ok here is the thing. First thanks for all your input very much. Second I replaced a Nvidia GTX260 which was doing the same thing, GPU failed and has recovered from an error. Also the mobo is old and only using and supports ddr3 RAM im pretty sure the card is fine as i had my pick of the cards and saw his upgrade card. so i think its probably the PSU and MOBO. Im working on replaceing it and just gonna put my old GTX 260 back in and use the 6970 for a new build.
thesnappyfingers said:I thought you had a 6870?
XFX has a nice 650W PSU on Newegg that's $59.99 after rebate.
Sorry it is a 6870 and yes my 12v is only going at 525 and since i was having a very similar problem with my old GPU i have to believe that the PSU is failing as it only happens when im playing games or stressing the GPU. I will try and replace that and see if it helps. Also by testing the card you just mean a stress test right? ive never done anything like that though i would love to.
Stress tests: Earlier I mentioned MSI Afterburner and Kombustor. Download those and run Afterburner. You can run Kombustor stress test from the Afterburner interface. See if you get errors and stock clocks. See if you get errors at underclocked speeds (& reduced voltage if unlocked). It should use less power when underclocked, so if this gets rid of errors, you know it's the PSU.
So your PSU says 44A on the +12V rail? I was hoping you could tell us the exact model so that I could look it up myself.
So I have the same graphics card and I too am having a similar problem. I'll play games on it and all of a sudden I'll black screen and ill have to reset the PC. I don't know if its a wattage problem or simply a compatibility issue seeing as i built my PC with my selected parts. Any input would be great. I've provided the links of the parts i've bought. Ty.
and some DDR3 4GB RAM chips as well.
If anyone can tell me how to fix this i would greatly appreciate it, having to play SW or SC II and have it black out is becoming a real pain =/
First of all, have you tried uninstalling the drivers, booting into safemode running Driver Sweeper, then get rid of all AMD - Display drivers, as well as any NVIDIA PHYSX or the like, from there (just not the chipset drivers) (also have you tried updating the chipset drivers?)
then reboot, log into windows like normal, install newest drivers from amds website
mouse24 said:First of all, have you tried uninstalling the drivers, booting into safemode running Driver Sweeper, then get rid of all AMD - Display drivers, as well as any NVIDIA PHYSX or the like, from there (just not the chipset drivers) (also have you tried updating the chipset drivers?)
then reboot, log into windows like normal, install newest drivers from amds website
Just did the sweep and installed the drivers from AMD using the AMD Catalyst software the drivers came with. hopefully it'll work and it won't crash again. However, ill post if it does. As far as chipset drivers, were you referring to the graphics card, or the mobo chipset drivers? if indeed u meant the motherboard then, yes i have installed the latest chipset drivers for it.
Ty again for your time =)
Hmmm interesting... what are the temperatures? (after burner should tell you)
open up firefox and note the temperatures, also try setting the fan speed on 100% and see if that helps any also what is the specific model of ram you are using and is it set correctly in the bios (remember ram is double pumped so whatever number you have it set to in the bios has to be 1/2 of what it states on the actual ram) make sure the timings are right and all that.
i narrowed down the problem. it only blacks out when im about to load a match for starcraft ii. the online match will fully load for all players then when its about to jump into the game the screen will go black and wont come back again, forcing me to reset. ive ran Memtest for more than several hours.
Currently using Corsair DDR3 4GB RAM x2 on my set up. as for bios i havnt messed with that and if that could be avoided itd be best.
ohhh, are you using Comodo or other 3rd party Firewalls? I am and sometimes it will actually block your DXinput.dll which causes the same thing.
note the exact time it blacks out, go into event viewer (you can just type it in on your search bar in windows 7/vista) then go into windows logs > application > then look for things that might help explain (normally with a red exclamation mark) if you can't find anything there see if its in security or system
checked all the log files and the only inconsistency i found was that the file DRVSTORE is in blue font while the rest in black. so i donno what that means...
inside it contained a folder named "usbfilter" with a bunch of numbers, and inside that was another folder named amd64, as well as 2 files named usbfilter as well, also in blue font.
Value = 0.91V
Max = 1.42V
-TMPIN0 Value = ~27 C
Max = 29 C
-TMPIN1 Value = ~30 C
Max = 31 C
-TMPIN2 Value = ~22 C
Max = 38 C
Value = 10 C
Max = 27 C
*This was for all 6 Cores*
Radeon HD 6870:
-Value = 0.95 V
-Max = 1.18 V
-TMPIN0 Value = ~41 C
Max = 71 C
*C = Degrees Celsius
I really dont have the slightest clue what could be causing the light blue screen, but I suggest running memtest (or windows memory diagnostic) and uninstalling your drivers, going into safemode running driver sweeper (clean anything that says amd - display, then reinstalling them after rebooting into non safemode