Hope to get some advice here. Recently a friend swopped my Geforce 7600GT for his Geforce 7900GTX. Sounds like a good deal for me, only problems once i installed his card, i noticed small allmost dead like pixels on my crt monitor.
Its increase over the last week or so. Ive read on another forum that it might be overheating, 32bit problem or reinserting the graphics card.
Anyone here had similar problem before and what solution can there be.
I have not check if its overheating yet. Tried different drivers. Will try to clean the gold tracks on the card and reinsert card.
Using Win XP.
---------------
......................
We are but a grain of Dust in a Universe of Sand!
I had the same problem on my daughters comp. The GPU was overheating. Silly me forgot to plug in the 4 pin molex connector. I didn't realize it at first cause I tried a card that i knew had a bad GPU and it had the weird pixels and lines on the display. When I went to put the first card back in, I remembered not having to unplug the molex (cause it hadn't been plugged in) when I pulled it out. Luckily, it didn't get fried and had no issues after reinstalling it and having the 4 pin molex plugged into it.
IT is likely overheating or it isn't recieving enough power from the PSU. What kind of PSU is in the PC? I don't remember exactly, but I think the 7900 series requires something like 450 Watts with atleast 26a on the 12V rail/s.
Otherwise, yes, check the temps and make sure it isn't overheating.
I checked if its overheating but it is idling at 51degrees, cant see what it is under load cause it restarts the pc. There is alot of tearing going on when i run 3dMark 05. The graphics are all over the place. I have a 500W power supply but its a no name brand.
oh and here are some images of the dots in dots.jpg and when i run the control panel for nvidia the logo seems to stop while turning and carry on. The whole control panel freezes and carries on moving like its battling to function
I checked if its overheating but it is idling at 51degrees, cant see what it is under load cause it restarts the pc. There is alot of tearing going on when i run 3dMark 05. The graphics are all over the place. I have a 500W power supply but its a no name brand.
51 doesn't seem to bad idle, but it is the temp when under load that we need to know. Like I said, I am pretty sure you are overheating or have a bad PSU.
Have you tried using something like RivaTuner to make your GPU fan spin at 100%?
Try opening the side panel on your PC, aim a house fan into it, and see if you still have this issue. This isn't a fix, but it may help determine if it is overheating. If it takes longer to go crazy (or doesn't at all) with the house fan there, we can assume it is overheating.
will check tonight when i am at home, and put a house fan on it. The cards fan is spinning but will install rivatuner to find out how fast or as you said make it spin at 100%.
I personally think its the psu and not overheating. I still have 1yr on the warranty for the card so maybe i should sent it to Asus to check if i dont come right.
You could be right about the PSU. The problem is, it is hard to figure out, unless you can test it with a different PSU. But, I would still look into the overheating, as the symtoms you are display could also be related to that.
Message edited by basketcase on 03-11-2008 at 05:06:16 PM
OK, seeing as it was a much cooler day and i did what you said and upped the fan speed using RivaTuner the dots dissapeared. Also i could run 3dMark05 and it was fine.
I also put a fan on the side of the graphics card as you suggested.
So basketcase i think you were correct in saying it might be overheating. The reason i am on about hte power supply is because its a no name brand and i dont trust them anymore seeing as i have lost a few in past.
MY next question now is why would it overheat and how to i correct it???
Artefacts are basically caused by the GPU running close to its limits....
This could be heat (transistors switch faster when cold, this is why liquid nitrogen cooled CPUs overclock well)
It could be power (transistors switch faster with more voltage)
If the GPU is not receiving sufficient power it will start to produce artefacts. Cooling the GPU could cause it to be able to cope with the lower voltage.
In short, I'd still replace the PSU.
You say it is a 500W model, but you don't list the specs on the side.
Alot of older PSUs have a strong 40A+ 5V rail from the days when CPUs were powered from the 5V rail, and a weak 12V rail.
These days with all the most power hungry components being fed from the 12V rail, these PSUs are insufficient even though they may have high wattage ratings....
