Failure on blend test after 1 minute!

Destria · 2004-10-17, 08:50

I've run the blend test a couple of times now, and it always trips over after about one minute with this error:

"FATAL ERROR: Rounding was 0.5. Expected less than 0.4"

I'm not overclocking or anything, but needless to say I'm a little concerned about this.

I'm currently running a AMD64 3500+ on an MSI Neo2 Platinum, with 1024mb of 3200 Crucial memory

Any ideas what might be worth looking at?

2004-10-17, 09:39

in get exactly the same error on blend aswell on the first test.

i am running a p4e@2.8 with no overclocking, this is kind of annoying as i am trying to test my ram as i am adjusting the timings and i need to be able to test if its stable.

zombie1000 · 2004-10-17, 09:51

by the way i got the erros before i adjusting the timings aswel. Hopefully this should be a software bug because i have noticed lots of other people with the same problem and in stress.txt it says if its the same error in the same place all the time it is most likely to be a bug

Destria · 2004-10-17, 14:40

I'm inclined to think it might be a software thingy, rather than a hardware thingy (to use technical terminology). I've been able to run 'CPU stability tester', and Memtest86 (or whatever they're called) with no problems. It's still a bit of a concern though. I've just built this 'dream machine' of mine, and I want it to be perfect :)

Oh, and this is great:

Destria · 2004-10-17, 14:45

By the way, my CPU (and my graphic card) seem to perenially be at 50-60 C when idling. Is that a little hot? I'm using the stock heatsink and fan, but the graphics card is a Galaxy Glacier with a hefty heatsink and fan ensemble on it.

Maybe my case is hot...

Prime95 · 2004-10-17, 17:24

It is unlikely that heat is the cause of the problem. My best guess is memory. Try changing your memory timings in the BIOS to the safest (slowest) values possible.

Destria · 2004-10-17, 19:46

To test this, I took the memory out and put in some older PC2100 RAM that I've got sitting about from an old PC. Unfortunately, I couldn't get it to work with the two 256 mb stick and one 512 mb stick together (probably to do with memory configurations... but it only liked either the pair, or the single 512), so I couldn't try it with 1 gig of older memory, but I was able to pass this test (and a fair few more) with both the single 512 stick and the pair of 256 sticks.

The strange part is, I put the newer memory, that was present when I encountered the problem before, back in, and ran the test again. It passed it

.

The only difference is, I put it back into the other pair of slots (slots 1 and 2, not 3 and 4 which they were in originally). It's in the same order (I swapped them over and retested to check if that was it), just a different slot. So it might have been badly seated the first time, or had a fleck of dust or something, or maybe something's up with my slots 3 and 4...

Strange...

zombie1000 · 2004-10-27, 12:52

Quote:

Originally Posted by zombie1000

by the way i got the erros before i adjusting the timings aswel. Hopefully this should be a software bug because i have noticed lots of other people with the same problem and in stress.txt it says if its the same error in the same place all the time it is most likely to be a bug

It must be a software bug as i have now swapped my ram with some other pc3200 dual channel 512mb and i get exactly the same error and trust me i ahve tried everything to get it to work: upping the voltage, lowering the timings, swapping the slots there in, running single channel, nothing works!

Prime95 · 2004-10-27, 13:46

Sorry, I still think its a hardware problem. Unfortunately diagnosing the exact cause is terribly difficult.

Does it pass the in-place small FFT torture test?
Have you tried under-clocking the memory in the blend test?
Have you tried underclocking the CPU?

zombie1000 · 2004-10-28, 07:58

ok i will try these things then, maybe i didnt try everything

zombie1000 · 2004-10-28, 15:13

It fails both of them tests, the in-place large on the first test and the small FFT's won the first 10k length test(it passes the 8k tests). also i tried under clocking by lowering my FSB all the way down to 160mhz! but it still fails.

2004-10-17, 08:50	#1
Destria Oct 2004 2² Posts	Failure on blend test after 1 minute! I've run the blend test a couple of times now, and it always trips over after about one minute with this error: "FATAL ERROR: Rounding was 0.5. Expected less than 0.4" I'm not overclocking or anything, but needless to say I'm a little concerned about this. I'm currently running a AMD64 3500+ on an MSI Neo2 Platinum, with 1024mb of 3200 Crucial memory Any ideas what might be worth looking at?

2004-10-28, 15:13	#11
zombie1000 Oct 2004 2×5 Posts	It fails both of them tests, the in-place large on the first test and the small FFT's won the first 10k length test(it passes the 8k tests). also i tried under clocking by lowering my FSB all the way down to 160mhz! but it still fails. Last fiddled with by zombie1000 on 2004-10-28 at 15:13

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Blend torture test fails.	Unregistered	Information & Answers	2	2008-12-06 07:20
Blend test causing pc to become non-responsive...	enyceexdanny	Information & Answers	4	2008-01-17 18:02
Prime95 crashing while running blend test	day61	Hardware	1	2007-01-30 12:03
blend / small FFT torture test on linux ?	lukasz	Linux	1	2007-01-09 16:16
torture test failing in 1 minute	ghatothkach	Hardware	5	2005-01-14 09:50

2004-10-17, 09:39	#2
Unregistered 2·3·5·7·29 Posts	in get exactly the same error on blend aswell on the first test. i am running a p4e@2.8 with no overclocking, this is kind of annoying as i am trying to test my ram as i am adjusting the timings and i need to be able to test if its stable.

2004-10-17, 09:51	#3
zombie1000 Oct 2004 2×5 Posts	by the way i got the erros before i adjusting the timings aswel. Hopefully this should be a software bug because i have noticed lots of other people with the same problem and in stress.txt it says if its the same error in the same place all the time it is most likely to be a bug

2004-10-17, 14:40	#4
Destria Oct 2004 2² Posts	I'm inclined to think it might be a software thingy, rather than a hardware thingy (to use technical terminology). I've been able to run 'CPU stability tester', and Memtest86 (or whatever they're called) with no problems. It's still a bit of a concern though. I've just built this 'dream machine' of mine, and I want it to be perfect :) Oh, and this is great:

2004-10-17, 14:45	#5
Destria Oct 2004 2² Posts	By the way, my CPU (and my graphic card) seem to perenially be at 50-60 C when idling. Is that a little hot? I'm using the stock heatsink and fan, but the graphics card is a Galaxy Glacier with a hefty heatsink and fan ensemble on it. Maybe my case is hot...

2004-10-17, 17:24	#6
Prime95 P90 years forever! Aug 2002 Yeehaw, FL 3⁵×31 Posts	It is unlikely that heat is the cause of the problem. My best guess is memory. Try changing your memory timings in the BIOS to the safest (slowest) values possible.

2004-10-17, 19:46	#7
Destria Oct 2004 4₁₆ Posts	To test this, I took the memory out and put in some older PC2100 RAM that I've got sitting about from an old PC. Unfortunately, I couldn't get it to work with the two 256 mb stick and one 512 mb stick together (probably to do with memory configurations... but it only liked either the pair, or the single 512), so I couldn't try it with 1 gig of older memory, but I was able to pass this test (and a fair few more) with both the single 512 stick and the pair of 256 sticks. The strange part is, I put the newer memory, that was present when I encountered the problem before, back in, and ran the test again. It passed it . The only difference is, I put it back into the other pair of slots (slots 1 and 2, not 3 and 4 which they were in originally). It's in the same order (I swapped them over and retested to check if that was it), just a different slot. So it might have been badly seated the first time, or had a fleck of dust or something, or maybe something's up with my slots 3 and 4... Strange...

2004-10-27, 13:46	#9
Prime95 P90 years forever! Aug 2002 Yeehaw, FL 3⁵×31 Posts	Sorry, I still think its a hardware problem. Unfortunately diagnosing the exact cause is terribly difficult. Does it pass the in-place small FFT torture test? Have you tried under-clocking the memory in the blend test? Have you tried underclocking the CPU?

2004-10-28, 07:58	#10
zombie1000 Oct 2004 2×5 Posts	ok i will try these things then, maybe i didnt try everything