Message boards : 321 Prime Search : Completed, marked as invalid

 Subscribe SortOldest firstNewest firstHighest rated posts first
Author Message
David Steel

Joined: 14 Feb 11
Posts: 78
ID: 86280
Credit: 213,794,922
RAC: 4,922

Message 95301 - Posted: 25 May 2016 | 7:48:18 UTC

Then the next 6 tasks were fine, completed and validated.

I've done nothing different and never had problems with this PC, any thoughts?

Thanks

http://www.primegrid.com/results.php?hostid=485043&offset=0&show_names=0&state=0&appid=7

JimB
Honorary cruncher

Joined: 4 Aug 11
Posts: 916
ID: 107307
Credit: 974,118,817
RAC: 0

Message 95303 - Posted: 25 May 2016 | 10:35:48 UTC - in response to Message 95301.

Let's look at http://www.primegrid.com/workunit.php?wuid=481654628

The successful computers sent a result that looks like this:

3*2^12412648-1 is not prime. LLR Res64: 889133BBF6262767 Time : 70295.002 sec.
and
3*2^12412648-1 is not prime. LLR Res64: 889133BBF6262767 Time : 73509.455 sec.

And here's what your computer sent:

Iter: 7224320/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7200385 [58.00%]
Iter: 7222144/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7200385 [58.00%]
Iter: 7254912/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7222145 [58.18%]
Iter: 7223040/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7222145 [58.18%]
Iter: 7295360/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7259265 [58.48%]
Iter: 7289856/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7259265 [58.48%]
Iter: 7276032/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7259265 [58.48%]
Iter: 7330048/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7312001 [58.90%]
Iter: 7318400/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7312001 [58.90%]
Iter: 7425280/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7391233 [59.54%]
Iter: 7410048/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7391233 [59.54%]
Iter: 7391360/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7391233 [59.54%]
Iter: 7482496/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7463809 [60.13%]
Iter: 7465472/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7463809 [60.13%]
Iter: 7506048/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7501569 [60.43%]
Iter: 7654144/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7650177 [61.63%]
Iter: 7694720/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7690113 [61.95%]
Iter: 7696128/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7694721 [61.99%]
Iter: 7716992/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7696129 [62.00%]
Iter: 7697152/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7696129 [62.00%]
Iter: 7697408/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7697153 [62.01%]
Iter: 7758208/12412647, ERROR: ROUND OFF (0.498046875) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7733505 [62.30%]
Iter: 7875968/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7865729 [63.36%]
Iter: 7915648/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7912321 [63.74%]
Iter: 7955328/12412647, ERROR: ROUND OFF (0.4985351563) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7951873 [64.06%]
Iter: 7971840/12412647, ERROR: ROUND OFF (0.5) > 0.4
Continuing from last save file.
Resuming LLR test of 3*2^12412648-1 at iteration 7955329 [64.09%]
3*2^12412648-1 is not prime. LLR Res64: 805987DF4872ABAF Time : 207170.960 sec.

The errors wouldn't have mattered if the residue came out the same, but it did not. You're either overclocked, have bad RAM or something else is going on. When only some WU's fail, I'm always thinking of RAM as different jobs could execute in different memory ranges. Alternatively, if all those WU's were running at the same time and they all died, something environmental may have happened (nearby lightning strike?) or your power supply could be having problems. Others may have better explanations.

Paul Underwood

Joined: 14 Jul 08
Posts: 67
ID: 25545
Credit: 3,482,191
RAC: 0

Message 95304 - Posted: 25 May 2016 | 10:39:45 UTC - in response to Message 95301.

I hope you are not overclocking! Your error might be due to a build up of dust on the CPU fan and heatsink causing heat problems in warmer weather. That covers heat, assuming your case is well cooled. Next the PSU might be on its way out -- check the voltages are acceptable in BIOS. Thirdly, make sure all your memory timings and BIOS settings are within spec. Lastly run memtest86+ to check that your RAM is fine. HTH

David Steel

Joined: 14 Feb 11
Posts: 78
ID: 86280
Credit: 213,794,922
RAC: 4,922

Message 95309 - Posted: 25 May 2016 | 16:32:37 UTC - in response to Message 95304.

Thanks for the replies,
The PC is definitely not overclocked, and the PSU is only 12 months old.
Heat is not a problem, it's the start of winter here in Australia and when running Boinc the CPU sits at about 58 degrees celsius.
Nothing on Memtest.

I remember last week my house lights flickered a few times during a storm.
The only other thing I could have done was to start playing Battlefront with Boinc still running, (got a GTX690 in this PC) I sometimes forget and have to ALT TAB out of the game to close Boinc and the RAM is usually maxed out...
Anyway the next 6 tasks all worked so hopefully it's nothing to worry about.

Speaking of RAM, can you recommend what I should update to?
I've been reading how important RAM is for LLR.
Currently running:
CORSAIR DOMINATOR 6GB (3x2GB) DDR3 12800 1600MHz 7-7-7-20 (TR3X6G1600C7D)

CPU: i7-920

http://ark.intel.com/products/37147/Intel-Core-i7-920-Processor-8M-Cache-2_66-GHz-4_80-GTs-Intel-QPI

Motherboard: Gigabyte GA-EX58-EXTREME

http://www.gigabyte.com.au/products/product-page.aspx?pid=2957#ov

Thanks

Michael Goetz
Volunteer moderator

Joined: 21 Jan 10
Posts: 13310
ID: 53948
Credit: 224,137,945
RAC: 34,182

Message 95310 - Posted: 25 May 2016 | 17:55:36 UTC - in response to Message 95309.

Speaking of RAM, can you recommend what I should update to?
I've been reading how important RAM is for LLR.
Currently running:
CORSAIR DOMINATOR 6GB (3x2GB) DDR3 12800 1600MHz 7-7-7-20 (TR3X6G1600C7D)

CPU: i7-920

(Take the following with a grain of salt. If anyone says otherwise, believe them, not me.)

On a first gen i7 (sans AVX) memory's not nearly as important. Your CPU uses the same type memory (DDR3) as Sandy Bridge, Ivy Bridge, and Haswell, but without AVX its internal computation speed is a lot slower. Comparatively speaking, your memory is faster, compared to the CPU, than later generation i7s.

That being said, I suspect that having 3 sticks of ram (rather than 2 or 4) might not be ideal, but it might depend on the memory and chipset. You may get some performance increase by having 2 or 4 sticks and running in dual channel mode. With 3, I think you're probably running in a single channel.
____________
My lucky number is 75898524288+1

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2397
ID: 29980
Credit: 411,470,142
RAC: 23,105

Message 95312 - Posted: 25 May 2016 | 18:02:12 UTC - in response to Message 95310.

i7-920 runs triple channel ram. I had forgotten about those days... anyway, combined with the clock speed and lack of grunt, I think it fairly safe to say this system wont be ram speed limited.

Keith
Volunteer tester

Joined: 8 Dec 13
Posts: 425
ID: 284516
Credit: 397,621,690
RAC: 84,438

Message 95313 - Posted: 25 May 2016 | 18:05:38 UTC - in response to Message 95310.

That being said, I suspect that having 3 sticks of ram (rather than 2 or 4) might not be ideal, but it might depend on the memory and chipset. You may get some performance increase by having 2 or 4 sticks and running in dual channel mode. With 3, I think you're probably running in a single channel.

If memory serves, motherboards with that particular CPU socket (LGA1366) used triple channel memory so it was optimal to use either 3 or 6 sticks of memory.

edit: Mackerel beat me to the punch while I was double checking my memory on memory.
____________
My Primes
Badge Score: 2*1 + 4*2 + 6*6 + 7*7 + 9*2 + 10*1 = 123

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2397
ID: 29980
Credit: 411,470,142
RAC: 23,105

Message 95314 - Posted: 25 May 2016 | 18:07:03 UTC - in response to Message 95312.

Looking more... the CPU only officially supports up to 1066. If the ram is running at 1600, that's an overclock. Personally I'd suggest keep an eye out for any other bad units. If this turns out to be a one-off, just ignore it. If there are more bad units, check the actual running ram speed and if above 1066, try lowering it. I don't believe it would have a significant impact of prime finding speed.

David Steel

Joined: 14 Feb 11
Posts: 78
ID: 86280
Credit: 213,794,922
RAC: 4,922

Message 95319 - Posted: 26 May 2016 | 7:13:00 UTC - in response to Message 95314.

In the BIOS / CMOS or whatever its called, the System Memory Multiplier (SPD) is set to 8.0, and the Memory Frequency (Mhz) is set to 1066.
The memory multiplier can go up to 18.0 = frequency at 2400.

So even though my memory is 1600, because my CPU only supports 1066 that's what i'm stuck at?
Or should I change the memory multiplier to 1600?

I was going to change my RAM to 24GB 2133MHz (6 x 4GB) thinking it would help Primegrid but it wont make any difference?
It should at least help my gaming? - some of the big maps I play on Battlfront get a bit laggy, I'm guessing it's because I only have 6GB of RAM.

https://www.pccasegear.com/products/23781/corsair-vengeance-pro-cmy8gx3m2a2133c11r-8gb-2x4gb-ddr3

Thanks

mackerel
Volunteer tester

Joined: 2 Oct 08
Posts: 2397
ID: 29980
Credit: 411,470,142
RAC: 23,105

Message 95320 - Posted: 26 May 2016 | 8:12:55 UTC - in response to Message 95319.