Help, I think I did something wrong!

[GamE34]

Hi there.

The problems started when I decided to install an Arctic-Cooling NV silencer 5 on my MSi 6800GT AGP. The screws used for the Voltage Regulator heatsink interfered with the cooler (They were taller than the heatsink itself), so I had to replace them. I accidentally knocked off the Voltage Regulator heatsink in the process, and put it back on without doing anything. I heard that you should never re-use thermal paste / pads, but this heatsink had a very thick thermal pad underneath it, and replacing it with paste wouldn't be very smart since it would probably start leaking everywhere.

Everything was fine after this, and I even managed to OC to 425/1225 without any problems. However, about 2 months later my games started to crash to desktop very often, and I would get occiasional nv4-disp.dll BSODs. I thought it was a driver issue, since my core temps never exceeded 60c. However, installing different drivers didn't help, and one time when I was playing C&C Generals I got a "Serious" error, stating that it was probably caused by malfunctioning / overheating hardware. This gave me an idea. I decided to underclock my graphics card to 320/900, and.. it worked! No more crashes to desktop!

Ofcourse, I was not satisfied yet. Underclocking my graphics card is not my idea of fun, so I decided to try and fix it. I took off the Silencer and the Voltage regulator heatsink. I then removed the thermal pad, and dremeled a bit of copper off the screwholes to decrease the distance between the voltage regulator chips (The heatsink was cooling 3 tiny chips) and the heatsink. Note that I didn't touch the part of the heatsink that makes contact with the chips, only the screwholes were too thick. I then put a very small amount of thermal paste (Cooler Master PTK-blabla) on the chips, put the heatsink on top of them and used screws to tighten it. I cleaned the memory chips / core / Silencer, and put new thermal paste on the core and silencer. I then put the silencer back on, and put the card in my system. So far so good.

However.. When I turned on my system, the monitor didn't get a signal! So I took out the graphics card, removed the heatsinks and made sure there was no thermal paste shorting the card. This was not the case, and I don't even think the Coolermaster stuff conducts electricity, so I put the card back in to try again. This time the monitor did get a signal, but the screen was messed up. All I saw was miscoloured pixels. No BIOS screen or anything resembling it. I turned off the computer and slapped a 80mm fan on the side of the Silencer to blow air on the VRM heatsink using strong double sided tape.

I let it rest for 5 minutes, and turned it back on. Again, the screen was completely messed up. At this point I was afraid that I had wrecked my card, since I don't have the money to buy another one. I removed the graphics card again, and made sure there was no dust on the card or in the AGP slot. I also made sure the Silencer made contact with the GPU core. I put it back in and turned on my computer. Surprisingly, the screen was fine, and it booted into Windows. However, as soon as I accesed the Nvidia Control panel to check my temperatures, the screen got all messed up again. I rebooted again, and decided to use Rivatuner to check my temps. 47c core and 30c ambient. At this point I reinstalled drivers, knowing that it probably wouldn't help, but I had to do something. After reinstalling the drivers (84.21) I checked my temperatures in the Nvidia control panel again, and it showed 47c. The card did not crash. I decided to test the card by playing the Dark Messiah demo. I launched the game, and watched the intro movies.. so far so good.. However, as soon as it reached the main menu, where it was supposed to render a 3d character, the screen got all messed up. I rebooted and tried it again, with the same result. I tried downclocking the card with RivaTuner, but as soon as I pressed the test button, the screen got messed up.

I messed something up, but I don't know what. Was it a bad idea to put thermal paste on the VRM heatsink? I am certain that the heatsink makes contact with the chips, even without paste. Should I get a new VRM heatsink, or does it sound like something else is the problem?

I'm going to bed now, but I will test the card again tomorrow. It appears to improve with every reboot, so who knows..

EDIT: If it's important.. I've had this card for about a year now.

 

[elpresidente2075]

First I must say that you need to make smaller posts.

Second, I wonder how long you had this overclocked. If it wasn't much of a long time, then you probably got it back on wrong and it fried. Otherwise it was a circuit that burnt out from over use.

I overclocked an fx5200 and it worked real great for a time. Then one day I pushed it just a little too far, and the same thing happened. Turns out, it fried the 3d portions of the chip, as I could still use the 2D functions, but when 3d was enabled, it would crash and would take a restart to get it back.

Bottom line: get a new video card, yours is screwed up. Oh, and make your posts shorter. It'll be difficult to get any respect around here with novels like that.

 

[Dahak]

I have to agree with the other guy.you fried your card.nothing to do about it but get a new one.sorry for the bad news.goodluck.

Dahak


EVGA NF4-SLI MB
ACE 520WATT PSU
AMD 4400+ DUAL CORE @2.4
EVGA 7800GT(SECOND ONE HERE SOON FOR SLI)
WD 300GIG HD
2X1GIG SAMSUNG DDR400
19IN EXTREME CRT MONITOR

 

[nanoprobs]

Yep, it seems like your Video card is screwed. Probably the GPU (3d) part or the memory part. If it's still under warranty you can try RMA it, other than that you can't really repair it. :-(

Before you RMA it you should return everything back to factory default, like the OEM heatsin/fan.

 

[GamE34]

I had this card overclocked to 400/1100 from the day I got it. I could go to 420/1200 after I put on the Silencer, but decided to keep it at 400/1100 since I could barely see a performance difference.

I refuse te believe it.. At first, it wouldn't even load the BIOS. Now I can get into windows. Also, I noticed that it crashes while doing nothing sometimes as well. It also crashes while changing the 2d clock speeds..

I am gonna try changing the clock speeds again. If it works.. /dance. If it doesn't, I am gonna be so pissed. I was hoping it was crashing because I didn't clean it properly with isopropyl alcohol and it was causing parts of the card to overheat severely or something..

EDIT: Okay, it didn't work.. I'm gonna take out the card and clean it as much as possible, then I'll let it rest naked while I watch Conan The Barbarian. Then I'll install everything again using different thermal paste.. Cmon! I can't afford a new card. :x

 

[GamE34]

Alright.. This does not look good. When I used a different kind of thermal pste and put te card back into the computer, it booted into Windows just like before. however, after a couple of minutes the screen was all messed up again, I heard a couple of short beeps and the computer shut itself down, for the first time. This leads me to believe the VRM is overheating, badly.. I waited for 1 minute and turned the computer back on. I heard 1 beep, and it shut itself off again.. about 1 second after I turned it on. I am using a friend's computer to write this.. DAMNIT!

 

[177ine177ine]

Your card's screwed. Get a cheap card to replace it till you can get more money.

 

[w00ten]

find your manual and check the beep code(if it has it)... Either way, the card is pooched in my opinion.

edit: taking a dremil(sp?) to any copper area of your card also isn't overly bright. Lets say a small shaving gets into a little crevace around a capacitor or transistor or IC and you cannot get it out because you cannot see it or its jammed. Things start to short out because copper happens to be an amazing conductor of electricity

 

[elpresidente2075]

After more looking through your post, it seems that you have fried your ram. I have consulted a friend and he says that the ram overclock that you achieved was quite a bit higher than what he was able to achieve. In any case, your card is out of commssion. It may still be able to be used, but it is usable in the sense that you can drive on a blown engine: it'll run, but it won't get much performance out of it.

Fortunately it still works for you. You have the ability now to limp along until you can afford a replacement. Fortunately, the prices have come down, but I do know how long it can take to save up that much cash. Maybe you can find some sucker who would just give you one (christmas?).

In any case, you won't be able to use it for gaming anymore. I know, it sucks, but you will have to learn to live with it until you have a replacement. And now you know why you need to make sure you have proper cooling on a component BEFORE you start overclocking.

 

[GamE34]

It's dead now.

It refuses to boot, and there's an awful smell of burnt plastic coming from the VRM. I didn't kill it by overclocking, I was just being stupid. I thought the VRM heatsink was only cooling 3 chips, so I only applied paste to those chips. However, after inspecting the thermal pad that came with it, I noticed there were several small marks left behind by tiny little.. things on the graphics card. I probably could've saved my card when I started this thread, by putting on a new, proper VRM heatsink. However, I couldn't find one, and now my VRM fried.

The worst part is that it's gonna take me at least 2 months before I can get the money for a new card 9I'm a poor college student). It's not going to be a fast card like the 6800GT either. Probably an X1600 or 6800LE or something.

Actually.. what's even worse is that I have no card to replace it with, and don't know anyone that does. I'm currently typing this on a friend's computer. I could buy a a slow card to temporarily replace it, but that's gonna cost me at least 30 euros, and will delay my purchase of a new card.

 

[GamE34]

Alright, I decided to get a replacement card. I can get a powercolor Radeon 9250 for 30 euros, and an EVGA FX5200 for 35 euros. Which one should I get? I really don't want to spend any more money, since it's just a temporary replacement card and I can live without the latest games.

 

[elpresidente2075]

Go for the 9250. The 5200 is really not very good. In that product generation, the ATI cards swept the board, and often were less expensive. Not to mention that the 9250s had more features compared to the 5200s. I have seen both side by side and the 9250 is much better as far as having more options that you can enable while still having a playable framerate.

Good luck in saving for the new card! Which one did you intend to get, if you don't mind me asking?

P.S.
Normally, I wouldn't recommend going from NVidia to ATI or vice versa due to driver issues, but I think that it is warranted in this case due to the vast difference between the two. Just make sure to get rid of all traces of the old driver before you start installing the new one, and that includes a restart. Some issues can happen that are difficult to resolve, and it is always easier to do the right thing first rather than doing the wrong thing twice and then the right thing once.

 

[GamE34]

Go for the 9250. The 5200 is really not very good. In that product generation, the ATI cards swept the board, and often were less expensive. Not to mention that the 9250s had more features compared to the 5200s. I have seen both side by side and the 9250 is much better as far as having more options that you can enable while still having a playable framerate.

Good luck in saving for the new card! Which one did you intend to get, if you don't mind me asking?

P.S.
Normally, I wouldn't recommend going from NVidia to ATI or vice versa due to driver issues, but I think that it is warranted in this case due to the vast difference between the two. Just make sure to get rid of all traces of the old driver before you start installing the new one, and that includes a restart. Some issues can happen that are difficult to resolve, and it is always easier to do the right thing first rather than doing the wrong thing twice and then the right thing once.

Thanks! I'll remove the nvidia drivers with Driver Cleaner in safe mode, reboot in safe mode to install the ATI drivers and reboot again. Is that the proper method?

I am probably going to get a 6600GT or 6800LE (And this time I will NOT touch the VRM heatsink!) when I get the money. I might also consider a X1600 or different ATI card, but I'd have to do some price checks and read a couple of reviews first, since I've never owned an ATI card before. The 9250 will allow me to experiment with their drivers a bit.

 

[elpresidente2075]

Thanks! I'll remove the nvidia drivers with Driver Cleaner in safe mode, reboot in safe mode to install the ATI drivers and reboot again. Is that the proper method?

You got it.

Good luck with the experimentation! Fortunately you will be getting a card that it won't be too horrible if it burns out. And by that I mean the 9250. But make sure your new one is on the way before you start doing it. You don't want another thing happening like what caused this problem now, do you?

Hopefully you will have a good experience with the ati stuff. Get a feel for it, as the ATI cards have been dominating recently, and anything within a college student's budget will rock, comparitavely. Have fun!

Edit:
I just saw which ATI card you were looking at. If it is within your budget, go for the X1600. It's the latest generation and will have better specs than the last generation's (6600) cards.

 

[GamE34]

Edit:
I just saw which ATI card you were looking at. If it is within your budget, go for the X1600. It's the latest generation and will have better specs than the last generation's (6600) cards.

I might just do that. The X1600pro is pretty cheap and has good specs for its price! One thing though.. Is there going to be an AGP version of the X1650XT? If so, then do you know how much it will cost?

 

[GamE34]

is the corruption on the screen a bunch of blocks that are one of the RGB colors?

I can't remember exactly which colors, but the RGB colors were definitely there. Most of the screen was white though.

By the way.. I'm having some trouble with the ATI drivers! o.0 I removed the Nvidia drivers in safe mode, and rebooted ins afe mode to install the drivers.. However, I couldn't install them. Probably because the .net framework doesn't work properly in Safe Mode. So I rebooted and installed the drivers in Windows XP. I rebooted again, and for some reason my screen resolution is 800x600 and the colours are 4 bits per pixel. I can't seem to change those option with the Catalyst Control center or with Windows. Also, CCC told me it did not support my drivers. I installed Catalyst 6.5, since it was the latest version that supported the 9250.

EDIT: Alright, I don't know what went wrong the first time, but I fixed it by uninstalling both the drivers and the CCC, and only installing the drivers. Everything works fine now, though I don't have the CCC.. is it required?

Oh, and both Windows and RivaTuner say that my card is a 256 MB card. I thought I had a 128 MB card, which is what the OEM package said.. Oh well, more memory doesn't hurt.

 

[Slava]

First I must say that you need to make smaller posts.

First, you are wrong. His post is fine. At least he told us every step he took and we do not need to ask him things like "Dude, did you do this? Have you checked that?"

Second, I wonder how long you had this overclocked.

Second, he stated that it was overclocked for two months.

Bottom line: get a new video card, yours is screwed up.

Finally, a useful comment.

Oh, and make your posts shorter. It'll be difficult to get any respect around here with novels like that.

Uh, cut it out already? Would you? The length of and the level of detail in his post saved a lot of people a lot of trouble. If he did not take the time to write all that I would have to ask him approximately 9 questions. He then would have to reply to all of them....

*********************

Answering the OPs original question, no, friend, you were careful and installed aftermarket coolers properly. The only thing you did wrong was OC your card in the first place.

I can tell you with all certainty that PC components (except those that have moving parts such as hard drives, PSUs and fans) do not break just like that. I have been building my own machines for almost 17 years and I have dealt with thousands of computers at work for close to 9 years now. I've never seen anything burn out as a result of power outage. The only times I've seen motherboards, CPUs, video cards and memory go bust is when they were overclocked.

Bottom line is twofold:

1. You did fry your video card by overclocking it.

Idle temps at 47C do not necessarily indicate that your card was OCed properly and "safely". (Not that there is such a thing as safe overclocking)

- I suspect that your video card was not the only thing you overclocked;
- I suspect you did not fix your AGP frequency in the BIOS at 66/33 MHz and when you OCed your CPU you overstressed the video card;
- I suspect you did not adjust AGP voltage in the BIOS.

2. Overclocking always shortens the lifespan of a component and when it dies it is likely to take another one with it.

In my personal overclocking experience, OCing has always caused a lot of trouble while providing little benefit. Yes I've read those articles about how THG overclocked Pentium D to like 5 GHz. That's nice and that's a lot of benefit, but I bet you that their cooling solution cost as much as their test rig and if they let it run like that for 3 months it would eventually crap out... So, there.

 

[Slava]

Actually.. what's even worse is that I have no card to replace it with, and don't know anyone that does. I'm currently typing this on a friend's computer. I could buy a a slow card to temporarily replace it, but that's gonna cost me at least 30 euros, and will delay my purchase of a new card.

Sheeeeeet, man. I feel for you. Were you in the US I'd send you my old video card so you could at least use your PC for email and school...

 

[GamE34]

Answering the OPs original question, no, friend, you were careful and installed aftermarket coolers properly. The only thing you did wrong was OC your card in the first place.

Actually.. I'm fairly certain I fried it because I didn't reinstall the VRM heatsink properly. After all, the card died after I messed with the VRM heatsink.. And it wasn't overclocked at that time. Not to mention the foul stench coming from the VRM o.0.

- I suspect that your video card was not the only thing you overclocked
- I suspect you did not fix your AGP frequency in the BIOS at 66/33 MHz

This is actually partially true.. I used to OC my CPU to 13x180 (XP 3000+), but started experiencing harddisk related problems after about a day, because my board does not have a SATA / PCI lock.

- I suspect you did not adjust AGP voltage in the BIOS.

You're right.. I didn't.. I thought it was unnecessary. :?


EDIT: Alright folks.. I am obviously going to need a new, fast graphics card. I've been looking at the X1600pro. It's pretty cheap.. But for some reason it doesn't appeal to me that much. The 6800GT is often 10-25% faster. This is why I asked whether or not there would be an AGP version of the X1650XT, since that would probably outperform the 6800gt ($300+), for under $200.. Assuming they won't make the AGP version twice as expensive. :x
I've been spoiled by the 6800GT, so the slower X1600pro might bother me. However, if they are not realeasing an AGP version of the X1650 then I am probably going to get the X1600pro anyways.. Or perhaps a 7600GS.

 

[elpresidente2075]

I did a couple of google searches and didn't find anything indicating that they are going to release it in agp. However, there does seem to be a bit of a call for it, as on several of the message boards they seemed to be asking for it. The agp version may still turn up, but I wouldn't bet on it. I did a bit of searching on their website too, and I couldn't find anything neither confirming or denying its existience.

@slava
I am getting some undue hostility from you. I was only trying to help him learn a lesson that I had to learn the hard way when I was brand new to the forums. I also wonder if you even read the whole thread. We were having a civil conversation, regardless of the comments about the length of his posts. Please don't pick apart my statements as some way of showing how much smarter you are than me, it only goes to show how immature you are when it comes to dealing with people. I have seen this sort of discourse before, and it generally leads to a flame war, which is something that this thread really doesn't need.

No worries though, there's just no need for it.

 

[GamE34]

Take a look at this card..: http://www.xfxforce.com/web/product/listConfigurationDetails.jspa?productConfigurationId=268341

I can get it for 150 Euros shipped. It appears to be identical to a 6800GT except for the fact that it has 8 pipelines instead of 16. Any chance it can be unlocked to 12 or 16 pipes using Rivatuner?

 

[elpresidente2075]

Unfortunately that link only redirects you to the /index.asp page. I don't know much about riva tuner or unlocking pipelines, but I think I have heard that it is possible with some cards, not with others. If anything, you can try, but if it doesn't work, you'll have to live with slightly lower performance until you can get around to getting a better card. But if it does work, then you have a better card than what you paid for and you will (I hope) be very pleased with your results.

On the other hand, I believe you run the risk of burning out some components if you don't properly get the card unlocked. This is speculation of course, I'm sure someone has actually done it and knows exactly what could happen.

 

[gomerpile]

I have to disgree with you that ocing shortens the lifespan, currently I have 3 computers
xp 2500 oc,ed the day I got it, asus 6800ge oc'ed 420- 1150. Processor running strong for 4 years or close to 4. vcard 1 year and running fine.
2nd system xp3200 oc'ed the day I got it, watercooled and chiller 5700le oc'ed as well about 3 years that system is running perfect.
3rd system 805d oc,ed the day bought 2 or three monhts ago 3ghz. Vcard 7800gs oc,ed day bought 6 months ago running fine. 473- 1396 Knowing the parts and limits of burn out is the key.
one other computer is the slot one 750 cannot remember the year and it too is oced to 825, vcard v8200 g3 oc'ed, it too is running all day I gave that to my buddy.

 

[elpresidente2075]

Knowing the parts and limits of burn out is the key.

Exactly. Anything over that magical "limit" will significantly reduce lifespan, even with proper, even overkill, cooling.

 

[GamE34]

Unfortunately that link only redirects you to the /index.asp page. I don't know much about riva tuner or unlocking pipelines, but I think I have heard that it is possible with some cards, not with others. If anything, you can try, but if it doesn't work, you'll have to live with slightly lower performance until you can get around to getting a better card. But if it does work, then you have a better card than what you paid for and you will (I hope) be very pleased with your results.

On the other hand, I believe you run the risk of burning out some components if you don't properly get the card unlocked. This is speculation of course, I'm sure someone has actually done it and knows exactly what could happen.

Strange.. It takes me to the right page when I click it. It's an 6800XT with 256MB DDR3 memory with a 256-bit Memory bus. It has clock speeds of 350/1000. The only difference between that card and the rare and ridiculously expensive 6800GT is that it has 8 pipelines instead of 16. I just wonder if those pipelines can be unlocked.. Even if they can't I might still buy it, because I think it performs better than an X1600pro. I jsut hope ATI is gonna release an AGP version of the X1650XT/pro, so I can get more bang for the buck.