| Burning in new vid card - How to tell if bad ram crashing PC I am burning in a new video card, no over clocking or anything fancy. I am just wondering when a problem that freezes your computer is because of bad video card memory or because you are using high or extreme settings in a game or burn in program.
My example would be with games such as Team Fortress 2 and Starcraft 2 if I use the highest / extreme settings the games look awesome and the FPS range from 40-60 depending on the content on screen, no slow down or anything but they crash once every 30-40 minutes either square boxed artifacts or jibberish on screen, at times freezing and have to restart. They also crash on medium settings but less frequent. Next I use burn in programs for testing (Furmark & BurnInTest) and on normal settings they can go on for hours but then I push them to the max and once in a while I get errors like bad video ram, with thousand of errors every minute, sometimes the computer freezes and I have to restart and the rare times it finishes there are errors in the 100,000+ range.
So how am I supposed to know in these cases if its bad video card ram or just being pushy? The truth is too not getting crashing I would have to have my settings close to low in games.
Here are my specs:
-Sapphire Radeon HD 5570
-Intel Core2 Duo E6750
-4GB Kingston 800 DD2 ram
-Asus Mother board
-Windows 7 x64
-Bios, Drivers and DX and all that junk needed for gaming all up to date as of yesterday.
Also I have a question about windows detecting my video card as having 2+GB of ram by sharing system ram. I am wondering if when I put it too the max if my ram is too slow to match my video card ram, causing problems. How can I disable that and make it only use video card ram? |
|
 ArchivisYour DaddyPremium
join:2001-11-26
Earth
kudos:17
 ·Verizon FiOS
| Sounds like a heat issue.
You need to make sure that your shit is properly ventilated and that everything is properly secured. I've dealt on and off with issues like this since Quake 1 and overclocking. Same symptoms, same problems, same cause: Heat.
You're not overclocking, so it's not like you need better cooling equipment. What you need to do is to make sure that your cooling equipment is actually functioning.
#1. Are the vents to your PC closed off or pushed up against a wall?
#2. How much dust is in the system?
#3. Is the heat-sink/fan combo properly secured to your processor? I don't care how you answer, re-seat it anyways and while you're doing it, inspect the heat sink compound. Make sure it didn't dissolve. Apply new, if necessary and then reseat and make sure that everything is 100% secured and latched on properly. Some of that stuff can be a bitch.
You could have had an issue all along and never knew it because games never stressed your CPU out at a constant rate long enough until recently. I had this issue with Crysis when I got my machine. For months, my machine was fine, playing games fine. Crysis, which ball-busted my computer, caused PC hangs, artifacts, etc. Turns out, my heatsink wasn't secured on entirely.
--
A government big enough to give you everything you want, is strong enough to take everything you have. -MLK |
|
 KrisnatharokCaveat EmptorPremium
join:2009-02-11
Earth Orbit
kudos:3
 ·Comcast
| reply to dhtveso
First, use a temp monitor system like MSI Afterburner and monitor your temps.
1) Make sure your AMD drivers are up-to-date
2) Take a can of compressed air to the inside of your PC, and consider using an air cleaner or vaccuum
3) At the extreme end of DIY, take the heat spreader/shroud off the GPU, apply some TIM cleaner, then apply a fresh dot of TIM and re-seat the heat spreader. Note if temps improve.
--
If we lose this freedom of ours, history will record with the greatest astonishment, those who had the most to lose, did the least to prevent its happening. |
|
| reply to dhtveso
You can never push a card too hard, if it is giving memory errors you have a bad card or heat issue, you posted a new thread with numbers however you did not mention when you got those numbers so they are pretty useless...
We need under load numbers (intel burn test for cpu furkmark for gpu)
If you are getting memory errors with out heat issues the card is shot, plain and simple. |
|
| reply to dhtveso
Being "pushy" should never result in the stuff you are talking about. Is this a new build or did you just replace your old card with a new one?
Check to make sure the fan on your actual video card is spinning free. Next, I'd just take the side off my case and put a room fan there blowing into the case. Quick easy way to see if overall system temp is causing the problem like some others have mentioned. |
|
| reply to anompowa
The numbers I posted were these with the case closed, about a minute before crashing playing starcraft 2 at ultra extreme settings. The other thing I forgot to post is my PSU its a 450mhz ThermalTake, I know its low but even at high stress situations the voltages don't deviate more then 2% up or down. The only thing I am worried about is how much of the 450w is being used when everything is running and I can't find any tools to check that. I might have to buy a new one, is there any way I can lower the power usage of the video card to test?
Motherboard: 45c
CPU Main: 55c
CPU cores: 28c
MCP (Have no idea what this is): 60c
GPU: 62c
HDD: 34c
With case open there about 5c less but problems still occur. |
|
| The psu is not mhz, it is watts.
450 is fine... Size wise...
We need again LOAD temps...
MCP is the north bridge, it runs hot.
TBH 62C on the gpu under NO load seems high. I aint a huge AMD guy so off the top of my head I can not say 100% yes... but 99% yes...
Also whatever you are using is giving wrong temps, core can NOT be lower than "main".
So again, LOAD temps. |
|
| reply to dhtveso
Underclocking memory by 10% makes that card work like a champ, so what does this mean? |
|
ArchivisYour DaddyPremium
join:2001-11-26
Earth
kudos:17 | You have heat issues and whatever you used to gather heat information is incorrect. |
|
Reviews:
 ·Primus Talkbroad..
1 edit | reply to dhtveso
Sapphire graphics cards use the cheapest components they can find and are grossly under-cooled. The GPU temp may not be getting that high but that doesn't mean the components aren't overheating.
Good luck getting them to recognize the deficiencies (or even just answering your emails).
Try getting a good quality graphics card instead. Or keep the Sapphire downclocked. |
|
