Radeon HD 4870 Crashing My Games

E

equal6

n00b
Joined
Dec 12, 2006
Messages
60
Hey,

2 days ago I purchased a new video card being a Radeon HD 4870 1GB.

Since then I have tried running WoW and it usually lasts about an hour before my screen just freezes, no bsod no black screen just frozen on my current screen.

I have tried different drivers, updating them, using the stock ones. Nothing seems to work, I've also tried adjusting the WoW in-game video effect settings, I read someone had a similar issue and just changed the multisampler down to 1x, hasn't worked for me though it seems to pro-long the duration before my comp freezes. (not entirely sure though)

I am kinda stuck what the problem could be, if anyone has any solutions it would be more than appreciated.

Thanks
 
Are you running auto fan speeds?
It's possible it's a heat buildup causing the freeze-up.

Is your case in a desk that might restrict air flow?
I'd try running with the side cover off and a desktop fan directed towards the card to see if that cures it. If so, look at upgrading intake/exhaust for your case.
 
Hey,

Thanks a lot for your reply, I'm currently running auto clock speeds, using ATI Tray tools, is there an option to change clock speeds?

I've added another fan directing towards my card, maybe this could be the problem. My card does seem pyschically quite hot, though I can't seem to measure the card with tray tools but when I used CC it was running at about 78 when loaded.
 
78 is fine under load, i think these cards can go up to like 98 or so.

Ive been having the same issue kinda cept i will egt a bsod once in awhile or at least a reboot and always tells me my drivers are bad, even though ive tried about 4 diff types :p I have the Vision Tek 512mb version
 
Run FurMark and see if it's the heat issues you have, make sure to monitor the fan speed and temps with GPU z
 
Run program called FurMark and during the session check your temps.
 
Vlad_13 is suggesting FurMark since it heats a card up and can show if the card is not stable.
The current drivers allow fan adjustment through CCC. Sorry, I'm not currently running a ATi card, so another will have to point to where the adjustment is in CCC.
 
It's hard to read the text, but it looks like the fan speed is at 29%?
I did a search at Rage3D, 4870 fan speed . You might look through their suggestions for creating multi-fan speed profiles.
 
Thats correct after running FurMark my fan speeds where running at 29% according to GPU_Z

I will read through that link you gave me, again, thanks a bunch Vlad and Kowan :p
 
You should check to see that the fan is properly ramping up with temperature increase when you put a load on the card. However, also check that you properly removed the drivers from your old card (especially if it was from another company) and that your PSU can handle the load of the new card.
 
Okay little update: Just ran some checks to make sure the fan is changing correctly when load is applied and can confirm it increased with load.

I have just completely removed all previous drivers using driversweeper program and it picked up ati registry entries so I cleaned those up.

Then installed the stock drivers that came with my box and cd, then I run my game for about an hour then suddenly out of no where a complete freeze lock up.

I am out of ideas, but leaning towards my CPU but I am unsure how to measure this.

Any advice welcome,

Thanks !
 
Let me just add that it is only after a prolonged period, it works flawlessly for a good hour or so then suddenly bam.
 
I test my CPU with Prime95. I don't know if there are other better and more sensitive apps out there though, hopefully someone else will suggest it

And I use Memtest86 for memory test

As a last resort, you could probably send the card back for RMA

ps: did you run any other 3D games besides WOW? Any such problems with other games?

And what is your system spec anyway? Ensure that the PSU has enough power for this new card of yours
 
patrickgerry said:
Do you have anything in the event viewer close to the time your PC froze?
Click to expand...

Sorry to sound like a mong but where can I look at this event viewer?

Ps. Pc has been on for around 4 hours without running anything 3d and fine.
 
Update.

I have just been informed it could be my PSU and I would need a 600watt PSU with a minimum of 34amps on the +12 volt rail

I am currently using:

OCZ ModStream 520w ATX2.2 Power Supply (CA-017-OC)

So I am leading to believe my PSU is failing me.. but I would like some clarification before I go blow more money on a psu :)

Any advice welcome !
 
You could run the install of the drivers again and select to install the folding@home GPU client. I was stress testing my system last night and did so using one instance of Orthos and running the folding@home client. This put all 4 cores of my CPU to 100% load and the video card stayed pegged at 98% utilization.

If you have a program that can monitor your voltages, you could watch them while you do this stress testing and see what happens.
 
Got the temperature down to 50 whilst i was running the game, then it crashed, this leads me to believe it can't be the temperature.

Think I will buy a new PSU, if anyone could suggest a suitable one for my card off overclockers.com i would be greatfull :)
 
Are you overclocking anything on your system at the moment? If so, change it back to stock and try again.
 
Still looking for some advice if I should go ahead and buy a new PSU, if anyone has had a similar issue to what I've had or if they know something I can check to see my psu can't handle my graphics card etc.

Thanks !
 
With that card being the power hog it is I wouldn't put it past it being the PSU as the issue.
 
sounds like a psu problem. i had this with my ocz gamerxstream 700 psu when i had 2x 3870's it would play find and dandy and then out of nowhere once in a while it would reboot the pc. i ended up replacing that psu 3 times, on the 4th time i got them to upgrade it to an ocz elitexstream 800w psu and lived happily ever after.
 
That just demonstrates how crappy some power supplies are. I've had an HD4870 running stable on a 300W PSU with 15A on the 12V rail. Admittedly it only had a Sempron64 1.6GHz, and the case was open, but even so ...

Based on the symptoms you're describing, it is likely that it's a power or heat issue, but it could be any number of things. I'm currently going through a similar thing with a co-worker (hence why I was testing the HD4870 with a 300W PSU). So far we've eliminated the video card (it's got problems with my spare X1800GTO), PSU, operating system and hard drive. But it's still possible that any other component could be causing it - motherboard, sound card driver, optical drive misbehaving (had that look like a mobo problem before), etc.

[edit]Looks like it was his factory-overclocked RAM not working with his motherboard. And I asked if he was overclocking anything ...[/edit].

Can you borrow a PSU from someone else? Does anyone else you know have an HD4870 you could test with. With this type of thing you need to methodically eliminate components, starting with the most likely.
 
Thanks alot for your reply, I could borrow a PSU from a friend as a option also.
 
is there good air flow in your case? what kind of case, mid, full size? any intake and outtake? thats important.
 
UPDATE:

Okay, had some wierd news. I have compeltly removed my new Radeon hd4870 from my PC, removed all of its drivers and started a fresh with my old graphics card with new drivers aswell.

Everything seemed to be running fine today with my old graphics card and suddenly the same thing as what was going wrong with my new graphics card happened again.

So I am really clueless now about what the issue is, I will basically tell you from start to finish what happend.

Last week I ordered a new Samsung 24" I plugged that in is working fine, but this could also lead to the problem? I used to use a 19" and the resolution has increased alot to 1200x1920 now.

I also ordered a new mouse, which I doubt is the issue, drivers are updated etc.

So I then installed my new graphics card and the problems started, it seems the mix of new monitor and graphics card is the issue, but the same thing is happening with my WORKING old Video card.

So I am really confused !
 
I am now considering a complete PC reformat, but don't know if this would solve anything or it is hardware related !?!
 
If you determine it is heat, or just feel that you want to try this (I recommend it anyway), Then just follow these steps:

1) Go into CCC, and select ATI Overdrive from the left.
2) Click the lock to enable Overdrive, and accept the message that warns you.
3) Check "Enable Manual Fan Control" and set the fan to about 35%, or whatever you fancy.

I have mine at 35% and temps never get above 70 under full load. When I play FarCry2 or Burnout Paradise, I will bump it up to 45 to be safe. Do whatever you like.

You can also use RivaTuner to do the same thing if you like. I can also write a tutorial with screenshots if you want. =D

EDIT: I saw your recent posts. What hardware are you running other than the 4870? Go into detail, so that we can better help you =D
 
Before you do anything else, get a newer, better PSU. It could be that your PSU was already weak and the 4870 was the last straw for a PSU that was gonna go bad soon anyway.

I suggest Corsair's HX series. I use the HX620. High quality and takes care of most rigs PSU needs.
 
Okay heres my rig if it helps :)

AMD Athlon 64 x2 4600+, Corsair 3.0 GB of ram, (Video card inside right now - Radeon HD 2600 XT)

Mobo : MSI RD480 Neo2 - Havn't updated these drivers in awhile.

PSU : OCZ ModStream 520w ATX2.2 Power Supply

Hope that helps, thanks for all the replies.
 
You must log in or register to reply here.
Back
Top