WARNING - 5850 Overclockers!

[Zagitta]

As the topic title says this is a warning to all 5850 overclockers and potentially all 58xx overclockers!

Short version: CHECK YOUR VRM TEMPS!

Long version: Today while setting up my old server as a mining rig with a reference ASUS 5850 i bought recently i was playing around with overclocking and i got up on 1000 core, 900 memory (I'm on a headless linux so i can't downclock memory more than the stupid limit), my Vcore is on 1,21 and set with RadeonVolt...

Here's the issue: I don't know if radeonVolt's reading are accurate but with 100% fan and the core @ 71 degrees my VRMs were COOKING! i kid you not they were between 135 and 140 degrees!!!

This in itself isn't a huge problem as VRMs are build to deal with those temperatures however as far as i can see the VRM's on reference 5850 cards are uP6208 (datasheet: http://ceemic.pri.ee/hardware/datasheets/uP6208.pdf)
Take a look at page 19 and you'll see this: Operating Junction Temperature Range (Note 4): -40°C to +125°C and Junction Temperature: 150°C

So as you can see my card is above the supported operating temperature but still below the absolute max temperature of the VRM and my point is that when you do hefty OC's like this you most likely WILL shorten the boards life span, even with 100% fan and good core temperatures

[gnukix]

I don't think this is news to anyone who overclocks, or if it is, they shouldn't be overclocking.  You really need to watch ALL of your available temperatures.  The problems happen when even one exceeds the limits.  The sensor is invariably some distance away from the heat source even so it's reading will be a bit cooler (usually only a couple degC)

Many stock cards have very poor VRM cooling, or poor cooling in general.  I just got a couple 6870 Sapphires (never owned a sapphire, nor a 6870) and the VRM heatsinks are way too small to be doing overclocking.  The primary HSF looks decent though.  Untested since I'm still waiting on the motherboard to show up.

[Zagitta]

Well it certainly was news to me because i never saw any of the OC tools report VRM temps of this card untill RadeonVolt (I had it in my windows machine the first couple of days where i played around with different OC's etc)

Secondly, AFAIK the Asus TOP directCU actually has worse VRM cooling than reference :p

[gnukix]

5870's have 11 different sensors.  I dunno about others.  They're all visible ont he i2c/smbus connection, the AMD GPU Clock tool in windows shows them all, I think GPU-Z does too, I haven't ever actualyl had a GPU based card in a Linux box, I'll know more about that next week (I work with servers.  Most of the machines I use only see a keyboard long enough to get serial console or their IPKVM working)

[dikidera]

Yeah, warned too a tad bit too late. I OCed my 5850 to 950/1000 before while mining and although it crashed a couple of time i ignored it. Then the damage that was caused did not cease with going to stock. So far, i am 720 core and 500 mem and no problems, however my VRMs got a bit damaged or something and my temps are always near 110C and only once did they reach ~120. Also, if i stress the card with some stressing program specifically for GPUs i would see the the card shutdown(not PC) problem. So i keep my card cool and below 99% usage.

[qed]

Yeah, warned too a tad bit too late. I OCed my 5850 to 950/1000 before while mining and although it crashed a couple of time i ignored it. Then the damage that was caused did not cease with going to stock. So far, i am 720 core and 500 mem and no problems, however my VRMs got a bit damaged or something and my temps are always near 110C and only once did they reach ~120. Also, if i stress the card with some stressing program specifically for GPUs i would see the the card shutdown(not PC) problem. So i keep my card cool and below 99% usage.

A damaged VRM may drain too much current from the PCI-E slot damaging the motherboard too.

[Basiley]

keep SOLID/CONSISTENT airflow in you PC case. add extra fans/switch case to more capable/adequate and etc and etc.
stick with reference-based-cooler-build GPU's/PCB's[they usually more fool/overclock-proof, than "improved" mimics].

[gigabytecoin]

I don't think this is news to anyone who overclocks, or if it is, they shouldn't be overclocking.  You really need to watch ALL of your available temperatures.  The problems happen when even one exceeds the limits.  The sensor is invariably some distance away from the heat source even so it's reading will be a bit cooler (usually only a couple degC)

Many stock cards have very poor VRM cooling, or poor cooling in general.  I just got a couple 6870 Sapphires (never owned a sapphire, nor a 6870) and the VRM heatsinks are way too small to be doing overclocking.  The primary HSF looks decent though.  Untested since I'm still waiting on the motherboard to show up.

It was news to me! Lowering my OC on some of my 58xx cards until I can get an infared heat gun.

[AngelusWebDesign]

I do some OC myself, but never push it to the limit. It's not worth it. Better to buy another card and have no problems than to be messing with fried cards/hung machines all the time.

I guess it depends on how much time you have, and how long you plan to keep your cards.