Sapphire x1950pro Freezes, Reboots when in Game

[TMoney468]

I've been having a problem with this video card ever since I got it, and about a month ago I noticed that I can't play games with the card anymore, nor can I run 3DMark. My computer will just turn black, the fans speed up like a jet engine, and sometimes it reboots. When I first experienced this problem, I checked the temp of the card and it was idling at 70C which is way too hot. I removed the heatsink, applied some AS5 and now the idle temps are about 50C. The main problem with the card is that it used to freeze and reboot every once in a while when I first got it (some games would be OK), but now every game/3DMark kills my computer.

The RMA experience at Sapphire has been horrible, they hire a 3rd company to do RMAs and they charge $15 to do it. I posted on their forums and they basically said it's most likely my PS that is causing problems, not the fact it was idling at 70C before I fixed it. My PS is an Antec NeoHE 550 watt. I don't experience any other problems with my system, even under load, it's only when I game.

Is there anything I could try, or should I just RMA? I'll probably end up buying a new card first and then sell the RMA'd card on eBay. I've tried the latest drivers and that doesn't help either. Thanks

 

[Summitdrinker]

well, I started having problems with my x1950 pro shortly after getting it, I thought the drivers were a problem, they weren't, I never really figured the card itself hardware wise was bad and it's just fine. well slowly but surely things got worse gaming wise as time went on, never a problem when none gaming.

the problem is? tired main system ram, the extra heat from x1950 pushed it over the edge
I have improved things by changing my air flow in the system

UT 1999 and Ut 2004 are the worst games, esp in D3D, it seems to be worse than opengl

 

[Keysplayr]

Your power supply may be on it's way out. 70C is normal for your card under load. 50C is good for idle.

Things to try:

1. Remove and reinstall ATI driver using driver cleaner (latest version). Follow the instructions carefully.
2. If #1 doesn't fix the problem, try another powersupply. One that meets minimum recommended specs for your card.
Your Antec 550 should have been enough, but maybe it's on it's way out. Try a power supply that has MORE than enough power so you can rule that out if the problem persists.

 

[TMoney468]

I have tried uninstalling and re-installing the ATI drivers to no avail. I used the Ati-Uninstall All utility and that didn't seem to make a difference. I have another power supply, it's an Antec TrueBlue 480 Watt PS, which was my first PS in the system before getting the NeoHE. I could try putting that back in to see if it makes a difference (I can't really afford to go out and buy another PS right now). Just as a clarification, my temps were 70C idle before I re-seated the heatsink..god knows how high they got when I was trying to run games (probably 90-100C). Now it idles at 50C and I can never get far enough into a game to see how high it goes.

 

[Stumps]

what are the specs for the rest of your system?

 

[bryanW1995]

My x1950xt idles at 54 c using the 2d/3d auto switching in cat 7.7. It idled at 59 when I used ati tool and kept it at 621/900 all the time btw. The auto fan settings on mine are : default 27c, >80c = 36c, >88c = 50%, >91c = 55%, >94 = 68%, >100c=100%. 70c is not very warm AT ALL.

 

[Dainas]

If your card was idling at 70c, then the heatsink was not properly put on and I would worry that something got burned. My sisters Sapphire X1950pro AGP idled at 48c out of the box and can not be coaxed past 70c in oblivion. I've also had an Antec 550W, it handed 7800 GTX SLI fine till a fan died an a capacitor blew from heat. I would try your 480W as even that should be plenty for an X1950pro in a system with its 12v not loaded up with a mess of drives

A 14amp 420w generic with 3 drives has worked flawlessly for my sister's X1950pro for the past months. Won't be surprised if that goes but it certainly is because the PSU is made from crap and isn't because of the specs being too low for a X1950pro.

If the 480W doesn't fix it, hate to say it but its RMA time >_<

 

[sam0t]

When ever a computer reboots itself suddently and especially under 3D load the PSU is the first one to suspect. I hate to say, but I have personally lost every bit of confidence towards Antec PSU, having 3 of them die on me and hearing lots of crap about Antec in the Internet has not helped any either.

First thing is to eliminate heat out of the pack, usually this can be archieved by taking off the computer side/case. Next thing I would do is test memory with memtest and after that give hell to the processor with prime95. If the system survives these, then you can be pretty sure the villain is a bad PSU or bad 3D card.

ps. When ever the computer reboots itself, does it make any noise? I had a bad PSU once which emitted a few high pitched clicks when the system rebooted under 3D load. And as one could guess, it was a bad PSU that time. Also I have seen 1950pro run on quality 350W PSU without a hitch and then again I have heard bad quality > 400W PSU cause problems with the very same card.

 

[TMoney468]

Hi everyone, thanks for the responses. First off, here's my computer scecs:

Intel C2D E6600
Asus P5B-E Mobo
2x1GB Corsair XMS DDR2-800
2 x 400 GB Seagate HDs
Samsung DVD Burner
Antec P180 Case, 1 fan in front and 2 exhaust in back
NeoHE 550 Watt

I am certain that every component in my PC other than the video card and PS (either could be the cause of the freezing and rebooting) are functioning OK. To make a long story short, I had another problem before which turned out to be bad ram and motherboard which I got fixed after getting new ones. I do a lot of video/dvd encoding and backup which uses a lot of CPU and Ram, so my question is would these tax the PS the same as when playing a game? Because I've had no problem whatsoever doing my DVD encodings. The NeoHE has 3 12-volt rails with 17A on each, which seems plently for my system as it isn't overloaded with components. I will try the other Antec TrueBlue PS sometime this week and see if it's the NeoHE.

 

[Summitdrinker]

how about throwing that x1950pro in a different system and give it a workout................that will tell ya the story, good or bad

 

[TMoney468]

I don't have another PCI-express system to test in, so the easiest bet would be for me to swap out the PS's and see if that's the problem.

 

[TMoney468]

Just as an update to my problem and this thread, I tried my other PS the other day (Antec TruBlue 480watt) and I still experienced the same freezing/rebooting problem as I did before. It's most likely the video card at this point, I'll have to wait and see if I can find a deal on a NVidia card somewhere. Also one thing I noticed was that the instant the computer freezes when playing a game, the fans speed up to 100% (both the CPU and GPU) and it sounds like a jet engine in my computer case.

 

[*kjm]

It sounds like you have something else going on here. Do you have a spare hard drive to do a new install on? I'll almost bet that fixes it. Then you can backup your system and do a reinstall.

 

[TMoney468]

I have Vista installed on another drive which I've only booted into about 4 times so that's close enough for a fresh install. Same problem, computer freezes and fans spin up like jets when trying to play a game or run 3DMark

 

[evolucion8]

Strange that both fans, the CPU and GPU fan stays at 100% when the PC freezes, it's very strange, try to test your RAM and switch it from place, clear your CMOS and monitor the 12V line to see if there's any weird spikes, it could be normal if the GPU is overheating that it would lock up and the fan stayed at 100%, but why the CPU fan?

 

[Makaveli]

hi

I also own this card and will tell you the main issue with it. The Stock HSF is crap, and the Auto fan settings in the bios are also crap. They messed up the fan to temperature ratios on this model. That is why i've seen so many post of people getting freezing and locking up which is the card overheating. I'm soon to replace the stock cooler, but I found the fix for this was ATI try tools. The bios has the fan spinning at 40% until about 70c, which is absurd. I installed tray tools and now have proper fan settings. And don't have any issues with it, but had I not done alittle investigating I would have been thinking the same as everyone else, that something was wrong with the card. With everything Stock and no tray tools, I would get major artifacts after an hour of C&C3.

Pro's

Great card,great performance

Con's

horrible stock cooler!

my Suggestion to anyone that owns this card is to replace the Stock HSF ASAP!

 

[TMoney468]

Originally posted by: evolucion8
Strange that both fans, the CPU and GPU fan stays at 100% when the PC freezes, it's very strange, try to test your RAM and switch it from place, clear your CMOS and monitor the 12V line to see if there's any weird spikes, it could be normal if the GPU is overheating that it would lock up and the fan stayed at 100%, but why the CPU fan?

What's the best program to monitor the 12V line of the PS to make sure there aren't any abnormal spikes? My guess as to why the CPU and video card fans spin up is that the CPU fan is set to adjust the speed in the BIOS...and when the computer freezes everything goes out of wack and the fans go to 100%. The RAM is fine because I RMA'd it 2 times with Corsair and finally got 2 sticks that didn't have any errors in Memtest. We have a tax-free holiday coming up in a couple weeks so I plan on trying to find a deal in-store for a new video card. That should be the definitive say in whether it's the card or my PS.

 

[evolucion8]

I do not know of any monitoring program that can be seen in full screen games, but what I do is the following, I use Everest Ultimate Edition sensor utility and then I run the 3D Artifact utility in ATi Tool, then I use a CPU stress Test like Prime95 and then I monitor the readings of the Everest program. Since the CPU and GPU are under heavy stress, if the 12V goes below 10.5V it should be a problem. Also you can switch ATi Tool with the rthdribl_1_2 whatever it's name it is, but ATi Tool really heats up the card.

 

[TMoney468]

That sounds good, I'll test it out. If I remember correctly, I tried running the 3D artifact in ATI tool and my computer froze and rebooted. I'll download Everest and see if it still freezes when running it. Thanks

 

[chevalierdulys]

TMoney468
I have the same problem. I even bought an artic cooller accelero x2. The fan is now at any time at 40/45degrees but still it freezes and reboots. did you work it out?

 

[onlyCOpunk]

I was having this problem with my x1950gt AGP as well. At random times in games (I'm assuming when it was working really hard) the screen would just go black and the fans would spin up, sometimes it would reboot itself sometimes it would just go into this "state" and I would have to hold down the power button to turn the machine off. I bought a new PSU because that's what everyone was saying, everything was fine until yesterday when it did it again. BUT this time when it rebooted, I wasn't getting my wireless signal and when I clicked on the wireless icon in the toolbar and "refreshed wireless networks" it immediately rebooted. I then figured I'd narrowed it down to my crummy 5 year old wireless PC card. And remember reading a while back that it could cause conflicts with the graphics bus I believe, so I went out and bought a 25 dollar new wireless card, and am waiting to see if the same thing happens again. I'm sick of this money pit.

 

[chevalierdulys]

Only on a few games does this happen. After playing Warcraft 3 for about 1 hour I had no problem. Fan at 50º or something. If I try Heroes of Might Magic V, Settlers VI, Bioshock or whatever it just freezes and reboots. So.. There is a problem with only a few games... what does this games have that the olders don't?

 

[TMoney468]

Wow, my thread is revived! I actually finally got an RMA from Newegg and due to the fact that the 1950pro was out of stock, they credited me the original purchase price ($215) and I'm so happy! The card was definitely defective/damaged and I read that it's because the 1st revision of the heatsink wasn't designed correctly, but the latest rev is better. Anyways the end result is that I got a 8800GTS for about $30 after rebate and after the credit from Newegg! I'll never buy a Sapphire product ever again.

 

[PingSpike]

I thought I remember there being some kind of problem with these...where the bridge chip wasn't cooled properly maybe?