Odd temp spikes with Asus ENGTX480 using GPU-Z 0.4.3

[Tyranny]

I'm getting really weird temperature spikes on both ADT7574 and driver temp sensors while using GPU-Z 0.4.3. The problem occurs both with PCB and GPU temps, and occurs along the lines of this:

I boot Windows 7 Ult 64bit and start GPU-Z. I leave the PC alone with only the desktop showing (so that GPU loading should remain constant). Everything will look nice and stable and then out of the blue a GPU/PCB ADT7574/driver temp will spike (up to like 148C, or down to -2 billion C) for 1 second. The reading then returns back in line with the stable temp readings.

E.g.1:
Date , GPU Core Clock [MHz] , GPU Memory Clock [MHz] , GPU Shader Clock [MHz] , GPU Temperature [°C] , PCB Temperature [°C] , GPU Temperature [°C] , PCB Temperature [°C] , Fan Speed [%] , Fan Speed [RPM] , Memory Used [MB] , GPU Load [%] , Memory Controller Load [%] , Video Engine Load [%] , 12V [V] , VDDC [V] , VDDC Current [A] , VDDC Power [W] ,
2010-06-16 15:53:27 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1663 , 135 , 2 , 9 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:29 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1664 , 135 , 4 , 10 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:30 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1663 , 135 , 0 , 9 , 0 , 12.17 , 0.955 , 12.0 , 10.8 ,
2010-06-16 15:53:31 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1663 , 135 , 3 , 10 , 0 , 12.19 , 0.955 , 12.0 , 10.8 ,
2010-06-16 15:53:32 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1664 , 135 , 6 , 11 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:33 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1666 , 135 , 4 , 10 , 0 , 12.19 , 0.955 , 12.0 , 10.8 ,
2010-06-16 15:53:34 , 50.6 , 67.5 , 101.0 , 4294967285.0 , 4.0 , 33.0 , 33.5 , 44 , 1665 , 135 , 0 , 9 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:35 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1663 , 135 , 1 , 9 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:36 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1666 , 135 , 2 , 9 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,
2010-06-16 15:53:37 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1664 , 135 , 2 , 9 , 0 , 12.19 , 0.955 , 12.0 , 10.8 ,
2010-06-16 15:53:38 , 50.6 , 67.5 , 101.0 , 39.0 , 33.0 , 33.0 , 33.5 , 44 , 1665 , 135 , 2 , 9 , 0 , 12.19 , 0.955 , 12.0 , 11.3 ,

E.g.2
2010-06-16 14:48:54 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1655 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.5 , 26.3 ,
2010-06-16 14:48:55 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1647 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,
2010-06-16 14:48:56 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1657 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.5 , 26.3 ,
2010-06-16 14:48:57 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1648 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,
2010-06-16 14:48:58 , 701.0 , 924.0 , 1401.0 , 70.0 , 127.0 , 64.3 , 53.8 , 44 , 1648 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.5 , 26.3 ,
2010-06-16 14:48:59 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1652 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,
2010-06-16 14:49:00 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1647 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,
2010-06-16 14:49:01 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1646 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,
2010-06-16 14:49:02 , 701.0 , 924.0 , 1401.0 , 70.0 , 53.0 , 64.3 , 53.8 , 44 , 1647 , 344 , 0 , 0 , 0 , 12.17 , 1.035 , 25.0 , 25.8 ,

This does not occur when using HWMonitor, or Nvidia's System Monitor *unless* they run concurrently with GPU-Z. GPU-Z does this whether or not any other monitoring software is running concurrently.

I'm assuming this is a bug with GPU-Z and not my hardware. Can anyone confirm this or should I RMA the gfxcard?

mobo: Asus Rampage III Extreme
CPU: Core i7 930 D0
gfxcard: Asus ENGTX480/2DI/1536MD5
gfx driver: nvidia 257.21 drivers

 

[isolaligree]

Odd temp spikes with Asus ENGTX480 using GPU Z 0 4 3

Just a simple question -- Since the Nvidia 480 is supposed to be able to run the GPU3 client with FAHCore_15, what PPDs are they getting?

 

[W1zzard]

this seems to be some kind of misreading problem where the transfer fails. nvidia's driver interface for those transfers is less than robust but i am working with them to improve it.

other apps dont show the problem because they dont read the sensor directly and only ask the driver (that's what the second set of sensors is in gpu-z)

this is not a problem with your graphics card - don't rma it

 

[Tyranny]

Phew. That's taken a weight off. Thanks for the response

 

[Tyranny]

Just a simple question -- Since the Nvidia 480 is supposed to be able to run the GPU3 client with FAHCore_15, what PPDs are they getting?

No idea. I've not thought about Folding@Home.