• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

OcUK X1950Pro 512mb - Artifacts / Errors - Help

Puzzled

Puzzled

Puzzled

Soldato
Joined
9 Jul 2003
Posts
9,623
Edit: Better pics / video further down.

Hi

I bought one of the OcUK ATI Radeon X1950 Pro 512MB PCI-E cards that were on special offer last week, and I'm having a few issues with it. I've tried several drivers - 7.9, 7.10 and 7.7 in XP and 7.9 in Vista 64bit.

At the moment I'm not sure if it's a hardware or software fault as it only shows in certain programs. For example I can 3dMark06 perfectly with no display errors, but in HL2 Lost Coast I keep getting lots of little squares popping up all over the place. They can be black, white or grey and tend to disappear when you get close to the object. I've attached some pics below but it only captured a couple of them. I also get a couple of squares on the Ati Toy Shop demo but couldn't capture it.

LostCoastZoom.jpg


I know they look like bullet holes but they shouldn't be there. I also made a small video on my phone but it's quite hard to see them, for those who wish to try it can be found here :D

I've also tried ATI Tool "Scan For Artifacts" after reading it's good at pushing cards to the limit and get this:
AtiTool.jpg

The card is at stock speeds but I've heard it's not really compatible with 1950Pro cards though, so not sure if I can trust it or not.

The cards has no temperature sensor so I can't tell if it's a heat issue, but my old 6800GT worked fine.

PC Spec:
Core 2 Duo 6300
2GB OCZ PC2-6400 Platinum Revision 2
ASROCK 775 Dual-VSTA Mobo
1950pro
Creative X-fi
Seasonic 430w S12 PSU (Using the PCI Express power connector)
Win XP SP2 & Vista 64bit. (Same result in both)

The fact that 3D Mark 06 shows no errors is what confuses me as doesn't it stress the card just as much as Lost Coast?

I'll try to get some better pics, but any suggestions?

Cheers
Steve
 
Last edited:
I take it the recommended 12v is 30a on these cards but at a guess i would say the vru area is overheating like it did on my sapphire model.
 
Hi Monkey man, I’ve installed a couple of these OcUk 512mb X1950's in PC's built lately for friends, friends of friends etc. Had concerns about lack of direct cooling on memory + the other 256mb which appears to be mounted at the rear of PCB. So I attached brackets and mounted one 120mm 1200rpm fan blowing across gfx card. Without the 120mm fan I did notice things were getting a tad toasty. I'm by no means saying my method will cure your problem, but if the cards only suffering issue's once warmed up then I suggest 120mm fan DIY fix.
 
Yes i did i kept getting worse than that too i got random black screens restarts and what your are experiencing and the mods on the sapphire forums said my corsair 520hx wasn't up to the job which made me laugh.Upon further investigation the problem that i finally decided upon was that theres a certain area on some models of these cards which is not adequately cooled i stated this in my rma request and didn't have to pay for return as it was deemed an inherent fault which it is.
It's either called the vmu or vru cant remember which also you need to check your power supply as these cards are 30a on the 12v rail recommended 26a minimum.
 
Last edited:
Sean_UK said:
Hi Monkey man, I’ve installed a couple of these OcUk 512mb X1950's in PC's built lately for friends, friends of friends etc. Had concerns about lack of direct cooling on memory + the other 256mb which appears to be mounted at the rear of PCB. So I attached brackets and mounted one 120mm 1200rpm fan blowing across gfx card. Without the 120mm fan I did notice things were getting a tad toasty. I'm by no means saying my method will cure your problem, but if the cards only suffering issue's once warmed up then I suggest 120mm fan DIY fix.
Click to expand...

TBH, they should run at stock perfectly fine without any extra cooling.

If not somethings up :)
 
Hmm looks like I'll be sending it back then. First Item I've had go wrong from OcUK so I suppose it had to happen at some point.

The cooling really is poor, it just has a cheap zalman imitation fan on the core nothing on the memory chips.

I've also had a look at my PSU specs:
3.3v = 30A
5v = 30A
12v1 = 14A
12v2 = 15A

I'll try to get some better pics and try some different games.

Cheers
Steve
 
EDIT: You really need a single rail psu, but it might do. Any chance you can borrow a 'better' psu from a friend to try out?

The mods on sapphire are a waste of time. They were saying the 520hx wasn't upto the job of support a pro. I had my c2d o/c'd, and a 1950xt o/c'd to 695mhz/2050 mhz without any issues for prolonged used use with gaming :)

Sapphire forum didn't even suggest altering the pci-e bus to 99/101 etc. I had to go on the corsair forum and they suggested it. Sorted the initial problems immediately with my card :)

One thing I did notice with the ati stuff. Once it started going wrong you had to reboot. There was no chance of recovering even if you lowered the clocks right down. I would suggest that maybe you drivers are corrupted or you need a reinstall.

Matthew
 
ooo noticed you have a seasonic.. that should be fine.

ALSO NOTICED: You have a asrock board. Make these changes (in this order).

1) Reset you bios! See if it helps :) (DO NOT OVERCLOCK CPU - IMPORTANT)
2) Manually change the pci-e speed to 101mhz. see if it helps.
3) Use driver cleaner and remove all ati traces after you have uninstalled the ati stuff. Restart and reinstall JUST the driver, not CCC.

Come back and let us know how you got on. I say not to o/c you chip any is because it caused probs above 2.4ghz when using my xt, I have no idea why as before it was stable at 2.6ghz. The other thing is that the bios's seemingly corrupt or something on the asrocks, and only thing that stopped freezes and crashes was resetting the bios. It's wierd, but my pc has been super stable ever since :D

Other than that... go here: http://forums.overclockers.co.uk/showthread.php?t=17301623

Matthew
 
Last edited:
The power supply is not fine look on the ATI site it says clearly recommended 12v rail rating 30A no idea where people get these numbers from.You have 29a on the 12v rail with everything else it's probably not getting enough juice try with the bare minimum connected ram card 1 x hd turn case fans off etc and see if the same happens.
 
The PSU is perfectly fine, 30A is way over estimated, i ran my old x1800 XT (which needs more power than the x1950 pro) on a single 12v rail PSU that was rated at 28A, and that same PSU is powering my 8800 GTS, and it would even power a GTX which both use more power than the x1950pro as well, even a 2900 XT manages on less than 30A, and thats the most power hungry card going.
 
Last edited:
21A on 12v and I have a x1950pro working just fine. I just removed 4 (of 6)HDD recently, it worked fine with those also

The card doesnt draw a lot of power or produce a lot of heat imo. Most gfx will heat up beyond the cpu temp and burn your finger. This card never gets hot like that, the 1800xt mentioned above will at stock reach temps of 80c and carry on working fine.

I think the whole point of these cards is the cheap to make reduced die size, it also has these other benefits


10rsiog.png
 
Last edited:
Seasonic PSU's are a good brand and you have more than enough power unless the PSU is faulty. THB though, if the PSU was at fault I'd be expecting random lockups restarts and instability rather than a few pixel artifacts.

My only other thought is case cooling, Do you have decent airflow through the case as these card dump there full heatload in there you need to make sure it's flushed through with fresh cool air quickly.

If all is well on the case front, RMA the card. I had a Ti4200 which used to overheat so I fitted a fan in the side of the case. I thought getting round the problem was a great idea at the time but in reality if you've paid for a fully working card you should get one.

AD
 
Sorry for the late reply, been a bit busy.

Thanks for all the advice guys.

I've tried changing Bios setting, running driver cleaner and trying various drivers (I did do a clean install anyway, but thought it was worth a try) but it kept on messing up.

I then tried taking the side of the case off and having a desk fan blowing on the card. This delayed the sparklies but they still came up. The card itself is really hot, the amount of heat coming of it is incredible so I think the cooling is a bit messed up.

I've done a few fraps films (reduced file size) so you can get a better idea of whats happening.

Lost Coast
Call of Duty 4
3D Mark 03

COD4.jpg


It's weird how 3d Mark 06 runs fine, but 3D Mark 03 and all games I tried (HL2, COD4, Unreal 3) mess up.

Guess I better start the returns process.

Cheers

Steve
 
Sadly not, but reading around my PSU should be able to handle it fine.

Plus I tried running with just the essentials and it made no difference.
 
Sorry to bump this up, but I have now tried using a 500w Antec Earthwatts PSU and it still does exactly the same thing.

Does anyone have any other suggestions to fixing this issue?

Cheers
 
You must log in or register to reply here.
Back
Top Bottom