Sign in with
Sign up | Sign in
Your question

should a 8800GT perform this way?

Last response: in Graphics & Displays
Share
June 14, 2008 3:54:57 AM

Hi guys I recently built a system with below specification for my cousin he is 14 and likes to play games and because of his old CRT he can only play on low res like 1024X768 or 1280X1024 but this new system acts strange here are his system specs:
CPU:Intel E6550 @ stock with Intel Box Fan
MB:Gigabyte GA-P35 DS3L
Mem:2GB Dual channel DDR2 800 Patriot Memory
VGA:Gigabyte 8800GT 512MB GPU@700 Memory@1940
HDD:Western Digital 250GB Sata2
PSU:Green 580 Watts Active PFC

when i assembled his system i tried to play "Never Winter Nights 2" which was bundled with his VGA but the strange thing is that on 1024X768 and high setting which was default setting when i installed the game after 5 or 10 mins it crashed then i installed COD4 on 1280X1024 and high setting it crashed too but on 1024X768 i was able to play COD4 for 3 hours i am a little worried is something wrong with his VGA or CPU?can a system with this spec run NWN2 or COD4 @1600X1200
and high setting can someone tell me what is wrong.

More about : 8800gt perform

June 14, 2008 4:22:00 AM

Make sure you update all of the drivers, thats all I can think of but I'm sure others will have more advice.

Also, you're a pretty cool cousin :sol: 
June 14, 2008 4:27:29 AM

The graphics card should be capable.

The problem may be overheating, you should download a program to read the temperatures. Consider SpeedFan.

Another issue may be the quality or quantity of power output from the PSU. On the PSU, there should be a listing for the amount of 12V power the PSU can generate. It may have multiple "rails" listed as 12V1. Find out the amperage and list them here.
Related resources
June 15, 2008 3:38:42 AM

But it has dual rails and each of them has 18A it says 12V1 18A and 12V2 18A.so doesn't this mean that this PSU can deliver 36A?
June 15, 2008 3:53:47 AM

You can not add the amps together. The card is not getting enough power because it is feeding off of only 1 12v rail, which is a symptom of a bad PSU I am afraid, a better quality one with the same amperage would have been fine. Get this PSU, it will run it fine and is cheap:

http://www.newegg.com/Product/Product.aspx?Item=N82E168...
June 15, 2008 3:30:47 PM

thanks for your answers i have teo more questions 1.can my problem be overheating cause i had coretemp running in background while playing and it was logging temps after crash happened i looked in coretemps log and isaw strang numbers like 95, 100 and then 0.
2.considering that my problem might be PSU related and the PSU you recommended is not available in my country (Iran) will following PSU be enough and safe "OCZ stealthXstream 600 watt"
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
June 15, 2008 3:42:59 PM

daryoush said:
thanks for your answers i have teo more questions 1.can my problem be overheating cause i had coretemp running in background while playing and it was logging temps after crash happened i looked in coretemps log and isaw strang numbers like 95, 100 and then 0.
2.considering that my problem might be PSU related and the PSU you recommended is not available in my country (Iran) will following PSU be enough and safe "OCZ stealthXstream 600 watt"
http://www.newegg.com/Product/Product.aspx?Item=N82E168...

It's likely temperature. 100 degrees C will cause crush. Officially, many gpus tolerate 100+ temperature, but this doesn't work well in real life. And the information on the rail is wrong. It does not feed only off 1 rail. The "rail" is internal and has nothing to do with the physical outside wire leading to your pcie connector. It can feed off multiple rails. Of course, 18x2 does not equal 36amps, there's some loss on switch. But if not defective, the psu should be enough. Also, if it's power problems, your game should crash as soon as the gpu hits high load, not a while after, as it heated up.
June 15, 2008 3:53:01 PM

@dagger thanks for your help but by temp i meant CPU temp not GPU temp will cpu temp of 100 cause this kind of crash?
June 15, 2008 4:04:51 PM

daryoush said:
@dagger thanks for your help but by temp i meant CPU temp not GPU temp will cpu temp of 100 cause this kind of crash?



Wow, yes, definately. The highest tolerance cpu can handle no more than 75 degrees C. I'm surprised it haven't burned out yet.

Get a new cooler.
June 15, 2008 4:49:48 PM

@dagger why would intel box cooler shipped with CPU perform this bad
is it even possible or might coretemp be wrong?
June 15, 2008 4:54:17 PM

daryoush said:
@dagger why would intel box cooler shipped with CPU perform this bad
is it even possible or might coretemp be wrong?

The e6550 temp sensors are solid, and unlikely to be wrong. And the current stock Intel cooler is basically the worst in existance. The heatsink is one third the size of the stock cooler that comes with an old P4. Get a real cooler. You can also try to reset it. But might as well ditch it completely.
June 15, 2008 7:50:08 PM

@dagger thanks for your comment if its ok i have one more question here in Iran two good HSF are available 1.Tuniq tower 120 2.OCZ vandetta which one do you think is enough for this system?
Thanks
June 16, 2008 3:46:02 PM

@dagger sorry but i faced a strange problem my cousin's VGA which as i mentioned before is a Gigabyte 8800GT factory OCed has a software called HUD i closed that app then played with 1600X1200 res on my monitor i was able to play for more than 2 hours then i got tired and exited the game before i started to play i set coretemp to log temp every 2 seconds after exitting the game i looked at that log i saw strange numbers like 0,100,95 but stranger than temps is that these temps only occur on second core while first core temp is normal like 45 or 47.so my questions are these 1.can HUD cause crashing maybe by OCing VGA 2.why CPU's second core temp is so high and sometimes 0 when first core's temp is normal?
Thanks
June 17, 2008 3:25:37 AM

Hi
I used CoreTemp 0.99 for monitoring the temps for core#1 they are normal in the game they are like 45c,47c,40c,... but temps for core#2 are odd like 0c,85c,100c,... i was wondering how could core#1 be 45 and core#2 be 100c or 85c or 0c at the same time? and if these temps can't exist at the same time why CoreTemp shows these numbers?
June 17, 2008 10:46:34 PM

daryoush said:
Hi
I used CoreTemp 0.99 for monitoring the temps for core#1 they are normal in the game they are like 45c,47c,40c,... but temps for core#2 are odd like 0c,85c,100c,... i was wondering how could core#1 be 45 and core#2 be 100c or 85c or 0c at the same time? and if these temps can't exist at the same time why CoreTemp shows these numbers?

No, such temperatures are not possible. Looks like bad cpu temp sensor. Consider rma it if you can. You can't fix bad sensor.
June 18, 2008 11:19:04 AM

Unfortunately i don't have the option to RMA so i have to stick with it but i wanted to know that if core2 duo CPUs have two temp sensors one for each core and if they have can i rely on the other sensor for temp readings cause in that case two cores shouldn't be more than 2C-5C apart right?
June 18, 2008 11:14:09 PM

daryoush said:
Unfortunately i don't have the option to RMA so i have to stick with it but i wanted to know that if core2 duo CPUs have two temp sensors one for each core and if they have can i rely on the other sensor for temp readings cause in that case two cores shouldn't be more than 2C-5C apart right?

Good point. I was thinking of quads (which I use). If I remember correctly, duo should only have one for 2 cores, while quad has 2 for 4 cores. But Coretemp display 4 slots for quad, with first 2 temps the same, and second 2 temps the same. For duo, it should display 2 slots, but both at the same temperature. Not sure what happened in your case.
June 19, 2008 8:16:17 PM

Thanks man you helped alot.
!