• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Screen Corruption

T

Trit

n00b
Joined
Feb 3, 2005
Messages
20
Since I upgraded my CPU (to an Athlon XP 2600+ Barton) and graphics card (Radeon 9600 XT 128MB) I've been getting this weird screen corruption, which look like random yellow dots around text, icons, and where a drop down menu had been used. This mostly happens when I do a cold boot but some cold boots are fine. Also my rig would crash to desktop or randomly reboot during games and on very rare occasions it would randomly reboot in windows.

The first thing I did was to turn off auto reboot to see if it gave me an error message, but all that happened was that I got a blank blue screen when it crashed. So I looked around the net to see if I could find any info on why this was happening. I found some info about it could be either bad system RAM or a bad CPU, so I put my RAM through a couple of goes in Memtest86 v1.50 which it past on all tests every time.

So I decided to change my 2600+ for my 2200+ to see if that would help, which it did in the case of the crashing to desktop and random rebooting, Max Payne 2 still causes a random reboot always at the exact same point, but everything runs fine.

The screen corruption still happens though from a cold boot and on rare occasions after windows has been running for a while. I've changed the cooler on the 9600 XT but still the screen corruption happens. Which makes me think it's either the 9600 XT or it could be my PSU.

So what do you guys think is it the 9600 XT, the PSU, or something else?

My PSU is an Antec TruePower 480W.
 
I am going with the video card. Is it seated properly?
 
I think it's seated correctly, I took it out to change the cooler and the corruption is exactly the same before and after.
 
Have you checked your Video card fan that you changed, is nice and tight? but not to tight as you dont want to crush your core...which cooler you using?
 
The new cooler is the Zalman ZM80A-HP.
When I put it on I checked it all over and it looked right, the temps even came down by a few degrees, but still the corruption.
 
run memtest in dos see if ur memory is screwing over...
 
I ran Memtest86 (v1.50) a couple of days ago and it passed every test.
 
This was one of my thoughts, but it's a 480W PSU surly that can handle a 9600 XT.
Could the PSU be on it's way out?
 
Sounds like artifacting. Was the card oced? Did you buy this card new, refurbished, or used?
 
GlobalFear said:
Sounds like artifacting. Was the card oced? Did you buy this card new, refurbished, or used?
Click to expand...
The card is running at stock and I brought it brand new.


BigBadBiologist said:
are you using an LCD monitor?
Click to expand...
Nope. I'm using a 17 inch CRT.


JTY said:
What brand is the PSU?
Click to expand...
It's an Antec TruePower 480W and it's just coming up to a year old.
 
I had a problem exactly like that. Red and yellow dots all over the screen like it had a disease. Rebooting didnt work, so I pulled the power cord, waited, and plugged it back in. On powerup, it was normal.

Are you getting these dots even in dos or during POST? mine did it only when there were graphics up on the screen, like the XP loading logo or the Asus mobo logo during POST.
 
POST and the windows loading screen are fine, after that is when it starts.
Rebooting gets rid of it, but it's annoying having to reboot nearly eveytime I start my rig.
It's happened after the rig has been on a while too.
Also when it does happen the ATI 3D preview artifacts like mad.
 
Trit said:
POST and the windows loading screen are fine, after that is when it starts.
Rebooting gets rid of it, but it's annoying having to reboot nearly eveytime I start my rig.
It's happened after the rig has been on a while too.
Also when it does happen the ATI 3D preview artifacts like mad.
Click to expand...

have you blown out the VGA connectors? Maybe something like dust or lint in there causing the problem..maybe a broken pin? Doesnt seem like it would be a video card issue seeing that it doesnt happen all the time. Just unplug the monitor and blow it it out, check for bent pins or broken ones. Maybe check the video card connector, blow it out as well.
 
I've just blown out the connector and port and checked for broken and/or bent pins.
All the pins are there and not bent.
When booting up there was no corruption, but I wasn't expecting there to be any as it's most common time to do it is when the rig has been off for a few hours.
 
Well, I guess the blowing out of the connector and port didn't work cause when I turned my rig on today the corruption was back.
 
If it is the video memory then would that explain the crashing to desktop and random reboots that kept happening when I had my 2600+ Barton in there?
 
Trit said:
If it is the video memory then would that explain the crashing to desktop and random reboots that kept happening when I had my 2600+ Barton in there?
Click to expand...

Yes, it would. I would think that your card is either overheating or having memory problems. Double-check that the car is seated and that nothing foreign is in the slot.

If there was a problem with the video connector, you'd see it all over the screen and not in certain spots around things that have recently been redrawn.

.B ekiM
 
It's very unlikely that your power supply is at fault, the 9600XT doesn't use much power at all.
 
mikeblas said:
Yes, it would. I would think that your card is either overheating or having memory problems. Double-check that the car is seated and that nothing foreign is in the slot.

If there was a problem with the video connector, you'd see it all over the screen and not in certain spots around things that have recently been redrawn.

.B ekiM
Click to expand...
Why would I not get the crash to desktop and random reboots with my 2200+?
Would it have something to do with the 2600+ Barton having a faster FSB than the 2200+?
 
I'm going with the video memory also. Get it RMAed and then see. If not you didn't lose anything.
 
ok, I was only answering your screen corruption problem as changing back to the other CPU showed that the crashing stopped once you swapped CPUs.

So the vid card is probably the culprit for your screen corruption (and maybe video game crashing) and the 2600+ is causing the crashing.

What were your CPU temps when you put the 2600+ in? Also did your run memtest with the 2600+ in?
 
Well, if its his Video card....why not just pop in it another machine? If its the video card, then it will show up on another machine dont you think?
 
BigBadBiologist said:
ok, I was only answering your screen corruption problem as changing back to the other CPU showed that the crashing stopped once you swapped CPUs.

So the vid card is probably the culprit for your screen corruption (and maybe video game crashing) and the 2600+ is causing the crashing.

What were your CPU temps when you put the 2600+ in? Also did your run memtest with the 2600+ in?
Click to expand...
I remember the temps for the 2600+ being around 35C when idle.
I didn't run memtest86 when the 2600+ was in there, but I'll try it.
 
that could be it. Since the FSB of your 2600+ is higher than the 2200+, your RAM may not be capable of running the higher speed. If you go into your BIOS, you can probably set the RAM to run at a speed lower than your FSB.
 
I put the 2600+ back in and ran memtest86 and let it run so it would do each test twice, on the first run through I got 22 errors in test 5 with all other tests getting 0 errors, on the second run through the 22 errors went up to 52 errors on test 5, again with all other tests getting 0 errors.


BigBadBiologist said:
that could be it. Since the FSB of your 2600+ is higher than the 2200+, your RAM may not be capable of running the higher speed. If you go into your BIOS, you can probably set the RAM to run at a speed lower than your FSB.
Click to expand...
So it's bad RAM then which is making my 2600+ crash to desktop and reboot ramdomly.
I have 2 sticks of 256MB Corsair XMS 2700LLPT 2-2-2-5, they're running in dual channel but they're not TwinX could that be the problem?
 
It's possible. You could try upping the voltage a little. Corsair reccomends all of their RAM to run at 2.8V. Or you could change your RAM divider to run at a lower speed.
 
I dont think System ram would cause a artifacts on your screen like you mentioned in the first post. That would have to be cause by the video card....i could be wrong, but in my experiences in overclocking, i have never had system ram cause that type of corruption. I would still try your Video card in another machine, before you go out and spend money on ram. Just rule it out completely. When you start adding in more factors, the anwer gets harder and harder to find. Work 1 peice at a time and just use process of elimanation. Ram could be a factor since you get errors in Memtest.

What ram is it? Make/model/Chips? Is is rated for 200fsb? Ram timings? I would look at these before i start messing with ram voltage. Some ram is very sensitive to voltage increases. My TCCD (437 chip) doesnt like anything over 2.7. It DOES NOT increase FSB to raise my ram voltage. However, chips like BH-5 LOVES VOLTAGE.....like 3.3v to it.....people use Ram boosters to get their ram that high, since no board has a higher setting then 2.9v in the bios.

Like i said, get a friend with a PC like yours, at least one that can use your Ram and video card. Test them seperate.....try to mimic the program or whatever you run that causes the corruption. But only troubleshoot 1 peice at a time. Trust me you will thank yourself in the long run.
 
I have 2 sticks of 256MB Corsair XMS 2700LLPT 2-2-2-5 running in dual channel but they arn't TwinX. I'll try them in Memtest86 with the 2600+ one at a time to see if it's to do with the dual channel or just one of the sticks.

As to the screen corruption, it hasn't happened in the last 2 days, which is odd cause I haven't done anything to my rig so, if it comes back I'll try my graphics card in another rig.
 
Highlife said:
I would still try your Video card in another machine, before you go out and spend money on ram. Just rule it out completely.
Click to expand...

Say the video card works in the other machine. Have you really ruled it out?

It's possible there's a marginal failure in the card that's demonstrated some attribute of the broken machine. If that attribute doesn't exist in the test machine, then what have you learned?

.B ekiM
 
mikeblas said:
Say the video card works in the other machine. Have you really ruled it out?

It's possible there's a marginal failure in the card that's demonstrated some attribute of the broken machine. If that attribute doesn't exist in the test machine, then what have you learned?

.B ekiM
Click to expand...

Its possible, but with those problems he mentioned it has to due with video card in some small way. If he tests it in another machine and the problem is there, then he knows its the card thats part or all of the problem. You learned that its the card having problems, then move on to next problem....Ram....well drop that ram into another machine and test. Run Memtest, Prime95, whatever you want...if he gets errors in stress test, the he should replace it or his ram isnt able to run at the 200mhz FSB. Even with loose timings.

The odds are, i think, its his video card. Again i have never seen bum ram cause screen corruption. Maybe he doesnt have a problem with his ram. At least testing them in another machine will help him track down his problem.
 
Highlife said:
At least testing them in another machine will help him track down his problem.
Click to expand...

Of course, using the card in another machine provides more information. But I don't think it "eliminates" anything.

.B ekiM
 
I've been testing the RAM to see how it copes in different configurations and speeds, I've tried each stick on it's own and also both sticks at the sametime in single channel and at 166MHz DDR they always get errors during Memtest86, if I slow them down to 133MHz DDR then they don't get an errors.

So with this I thought I would try some Crucial 166MHz DDR that I have and at 166MHz DDR that produced errors too. So I took the 2600+ out and put my 2200+ back in and changed the dividers so that the 2200+ was running at it's stock FSB of 133MHz DDR and the RAM also running at it's stock of 166MHz DDR and that produced errors in Memtest86 as well.

So unless I've got 4 sticks of bad RAM and/or 2 bad CPU's it has to be something else?
 
You must log in or register to reply here.
Back
Top