My 8800GT started getting this Error > MDIO ERROR: cannot open file "restart.coor" on almost every WU it starts since a few days ago. It starts the WU's but after just 3 Seconds it hesitates & then a few Seconds later gives the Computation Error Message.

I've tried the Card in several Box's but it gets the same Error in all of them. I even turned down the Clock Speeds on it but it still gets the Computation Errors ... ???

This mdio error previously occured in every WU which was not stopped and restarted, so the error told you there was no checkpoint file to resume from. Apparently the message has been removed since then, but it may still be active in the error handling code.

The actual error which you get is "Incorrect function. (0x1) - exit code 1 (0x1)", which is a very general one. Not sure what you could do.. it's not a software error and you tried downclocking and tested different machines. What about the temperature, did the fan fail? Otherwise.. looks like a broken card? :/

MrS

---

Scanning for our furry friends since Jan 2002

The Fan is working on it & it's not overheating either so it might just be it won't run the CUDA Wu's anymore. It Displays okay though so it's still usable but just not for the GPU WU's I guess.

Not to worry though, I already ordered a new GTX 260 to take it's place last Friday & should have it in an hour or so today when UPS makes their Delivery ... :)

The old 8800GT OC may not be as Dead as it appears to be, right now it is Merrily crunching away on the Seti@home/AstroPulse Beta GPU Wu's without Error, whether they Validate or not is another story but at least it's running them.

I'm still not completely convinced it was the Cards Fault that it was Erring out all those GPUGrid Wu's either. I put a Brand New GTX 260 in it's place last night & it had a rough start too. Erred a couple WU's just like the 8800GT before it finally settled in & started running them like it should.

So it may be right now something very minor has changed in the Wu's that it just don't like & started erring them out. I'll try every so often & see if it will start running them again or not. But for now it can sit over @ the Seti Beta Site & do their WU's until I see if they Validate or not at least ... :)

The following Seti/Beta GPU WU took 32 Minutes to do with the 8800GT OC Card & is claiming 95+ Credits, will have to see what it actually gets ...

<core_client_version>6.5.0</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_CUDA: Found 1 CUDA device(s):
Device 1 : GeForce 8800 GT
totalGlobalMem = 536543232
sharedMemPerBlock = 16384
regsPerBlock = 8192
warpSize = 32
memPitch = 262144
maxThreadsPerBlock = 512
clockRate = 1566000
totalConstMem = 65536
major = 1
minor = 1
textureAlignment = 256
deviceOverlap = 1
multiProcessorCount = 14
setiathome_CUDA: CUDA Device 1 specified, checking...
Device 1: GeForce 8800 GT is okay
SETI@home using CUDA accelerated device GeForce 8800 GT
setiathome_enhanced 6.03 Visual Studio/Microsoft C++
libboinc: 6.3.22

Work Unit Info:
...............
WU true angle range is : 0.241703
Optimal function choices:
-----------------------------------------------------
name
-----------------------------------------------------
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.00018 0.00000
v_ChirpData 0.01462 0.00000
v_Transpose4 0.00525 0.00000
FPU opt folding 0.00325 0.00000

Flopcounter: 38684852338174.469000

Spike count: 0
Pulse count: 0
Triplet count: 0
Gaussian count: 9
called boinc_finish

</stderr_txt>
]]>

Validate state Initial
Claimed credit 95.1409985351768
Granted credit 0
application version 6.08

I think GPU-Grid is currently experimenting with their WUs, so it may be possible they got something wrong and your 8800GT was *lucky* enough to only get such task. But that's just guessing, nothing more.

MrS

---

Scanning for our furry friends since Jan 2002

That Error is not Exclusive to the 8800GT OC of mine, this Host of mine which is a GTX 260 had the same Error yesterday for about a 10 Hour Period. Also back on the 10'th & 11'th of Feb it had the same Error for awhile too.

The 8800GT OC has been running flawlessly over @ the SETI Alpha Project running the SETI CUDA WU's without error so far since yesterday. I Downloaded 2 more GPUGrid Wu's this morning, the 1'st one it Erred out after just 3 seconds running, the 2'nd one is running so far @ the Cards OC'ed setting it came with (Only about 20 Min's so far really) I'm going to let run & see if it will finish or not & report back on the success or failure of the WU.

I've looked @ all the WU's that erred on both Box's & don't see any comparison to the WU's other than they were all Created about the same Time Period on both Box's. Different Time Periods though for the 2 Box's ... Weird

I'm not 110% convinced the message is related to the error. A lot of your tasks that worked also have the same message but they worked well enough to get credit.

I am seeing a similar picture with both my 8600GTS and GTX260 machines.

Phoneman1