Did you check the LEDs when you went black? I'm assuming you're using a 4870.Quote:
Originally Posted by dnottis
Printable View
Did you check the LEDs when you went black? I'm assuming you're using a 4870.Quote:
Originally Posted by dnottis
no idea but RTHDRIBL pushes it to 90c max. Wish I knew where it was so I could sink it.Quote:
Originally Posted by dennismagic
yea, D601 comes onQuote:
Originally Posted by chawks2
Damn 126C ouch. I know they are rated for hot, but thats really hot. Some reason 110-120C rings a bell at least for most VRM's on GPU's but not sure.
I dont even think this is accurate - I can touch them and the back of the card. Nothing is hitting 250F.
Welp, this was my problem. New PSU solved it. Test disconnecting some components and try stress testing, I would bet you there will be no issue. :up:Quote:
Originally Posted by dnottis
I'm using a TT ToughPower 1000 watt quad rail PSU lol. Each of the 6pin PCI-e are connected to a different 36A rail. I hardly doubt mine is a power issue. Its even certified with AMD on their site with 4870.Quote:
Originally Posted by chawks2
Hey, you can cross one thing off your list. Why else would you get a crit PWR failure tho? Fortunately for me, the LED made me realize my PSU was underweight. Are these the first cards with diagnostic LEDs?Quote:
Originally Posted by dnottis
You would figure if temps were the issue you would get the temp LED light up. Maybe I just dont have the proper definition of the LEDs?
I dont know where to post this but maybe someone can help me. While playing Cryis or any game with my new 4870 the games freezes for 2 seconds anytime I move my mouse in any direction. What's weird is my gpu fan slows down during this freeze then speeds back up when the freeze ends. Crysis runs smoothy when im done moving my mouse in every direction but when I move forward and enter a new area it does it again. Not only does this happen in games but in windows as well. If I move a window around the screen too fast it freezes for a split second then becomes smooth motion. Is it a bottleneck issue where the gpu is waiting. Or is the gpu not getting enough power to load the information, or is some of my ram bad that it cant load the information. I dont know and Im at my wits end. ive tried almost everything accept try my gpu on someone elses computer system or use a different psu.
Ive tried every driver
Os reinstalls
Every Netframework is installed
Tried overclocking everything and underclocking everything
My old 7900 never had this problem
Tried every setting in CCC and in Ati tray tools
One last thing, when i got the card one of those 12v regulator thingys fell off and I reattached it as best I could with superglue because I do not have a solder gun. (10bucks says this is my problem)
Anyone have an idea?
Windows Vista 32
2x1gig Ballistx 6400
Seasonic 650w psu
q6600 go 3.2
Ip35 pro
Thanks all...
http://forums.techpowerup.com/showpo...8&postcount=59
WTF? direct 80mm airflow, bigger VRM sinks - same temps, same crash....
What are the rails on the new PSU - check my post out on TechPowerup - 72amps the card is drawing in Furmark... my rails are 36a each - wtf?Quote:
Originally Posted by chawks2
Maybe you have stumbled across the cause of the problem? The voltage regulators become unstable with a certain type of load and crap out feeding all juice to the card, making all the over-voltage switches built in everywhere turn off the card.
yea with the reg where it fell off its probably not getting enough power now or somethingQuote:
Originally Posted by jpercel
also what PSU are you using?
Also where are the leds :S?
and why doesnt VRM show up for me in everest:(
Here is what I believe and what we need to test. The latest build of Everest shows the GPU VRM amperage draw (last reading) for the HD4870s. I need some others to run Furmark and watch the amperage draw. I dont think it's a temperature issue anymore, nothing is really hitting 125C / 250F as there is nothing on the card that hot. However, I see the GPU VRM amperage draw hit 72amps before I shutdown and the D601 light is a critical power trip on the 4870.
So I need you guys to run Furmark with latest Everest (build 1455) and see if your amperage draw is maxing your PSU rails. I have a TT1000 with quad rails. Rail 3 and 4 are dedicated to the PCI-e connectors. Each rail is 36a. My PC is going to a black screen when the HD4870 hit 72a draw, the maximum amperage I have available.
Links for latest Everest are here -
http://www.9down.com/EVEREST-Ultimat...55-Beta-29037/
http://middown.com/news_3945_Everest...1455-Beta.html
GPU VRM is the bottom reading in Everest -
http://3dxtreme.net/other/VRM%20more/vrmwith80mmfan.jpg
that is some insane draw if that really is the case. I have a single railed 60a unit and not even that would be possible to hold up to this thing. How is anyone to successfully stability test with furmark if this is the case?
I have a TT700W.Quote:
Originally Posted by dnottis
Furmark test 15 minutes
Maximum GPU VRM Temp 120°C, Maximum GPU VRM draw is 54.19A
Btw, Furmark didn't crash nor did I see the red LED
4870 or 4850? What brand? What drivers? What OS?Quote:
Originally Posted by MadMonkey
Gentlemen - here is the fix.
http://3dxtreme.net/other/VRM%20more...640x480%5d.JPG
http://3dxtreme.net/other/VRM%20more/sucess!!.jpg
Sapphire HD4870 750/900Quote:
Originally Posted by dnottis
ATI Catalyst 8.7 Vista 64-bit
OS: Vista 64-bit
run the furmarks with 2 or 4 MSAA enabled, it wont crash in my case! But it crashes with no AA.......Quote:
Originally Posted by dnottis
Is it a driver problem?
I don't understand how this is the fix when in the other thread you were doing the exact same thing with cooling the back side of the card, however with an even larger fan? What else has changed?Quote:
Originally Posted by dnottis
Ok, I ran Furmark with the latest build of Everest and the highest temps on the VRM's was 109c and the amps being pulled were 63a.
One other thing that sprang to mind is, I was always under the impression that 1xpci-e-16 supplies 75 w and each 6 pin pci-e connector can only supply the same, Therefore you have 225w which when divided by 12 gives you your amps 18.9a so how can it be drawing 63a+?:confused:
Glad you identified your issue, but I still think it's strange it would come back with a CRIT PWR failure if the issue was temps.:shrug:Quote:
Originally Posted by dnottis
I think either the sensor is completely wrong, or the reading is completely wrong (or both ?) or it shows the current output from the VRM at a lower voltage like ~1.2V:Quote:
Originally Posted by holwill
1.2 * 63 = 75.6W
On top of that, generally a good PSU can easily go higher than the specifications of PCI-E. But there is just no way in hell that the card is pulling 12 * 63 = 756W