Strategic Double Clicking

[kladner]

35904383 mismatched
DoubleCheck=35904383,71,1

[dragonbud20]

Originally Posted by kladner

35904383 mismatched
DoubleCheck=35904383,71,1

Got it ETA: 4 days

[Dubslow]

Originally Posted by kladner

If the user-of-record isn't on the Forum, then it seems that PrimeNet is the only other source of contact information.

Hence why I brought it up to Madpoo who can do something about it (though as you say, said something might not be the slightest bit effective).

The 35.273M bad test in the middle image was verified by me, which sparked my initial interest in this user.

[kladner]

Originally Posted by dragonbud20

Got it ETA: 4 days

Thanks! I look forward to knowing one way or the other(s).

[kladner]

Originally Posted by Dubslow

Hence why I brought it up to Madpoo who can do something about it (though as you say, said something might not be the slightest bit effective).

The 35.273M bad test in the middle image was verified by me, which sparked my initial interest in this user.

There have been a number of other boxes, of which one turned in some bad results.

i5-4670-main shows up around Feb 2014. It has logged "more than a hundred" verified results. Something happens around Nov 20, 2015. Then it churns out a lot of bad ones.

Looking at the sheer number of machine names, it makes me think this is some university IT project. Amy has actually contributed quite a lot. This could suggest that they might not mind a gentle contact:

Good Evening, Sir or Madam

As you are a valued contributor to the Great Internet Mersenne Prime Search, we have lately taken note that one of your formerly-productive computers, one i5-4670-main, has been spewing a lot of crap in the last few months. We respectfully request that you attend to this ailing machine immediately, and restore it to its former productive state. You are wasting your electricity, and our cycles.
Let's have a bit of responsibility here, eh? Curtis Cooper farts in your general direction! He waves mule private parts at your aunties!
Sincerely,
The GIMPS Folks
Have a Nice Day!
[/SNARK,/SATIRE./FACETIOUS]

That would be up to others, though.

[dragonbud20]

Kladner the exponent M37334837 came back with no match someone's going to have to quad check it

as a note to your comments about contacting ailing machines I would support that addition as I use the email feature to let me know when results mismatch currently and while my heart skips a beat when I see one I feel good knowing what's going on and I expect many others would as well.

[kladner]

Originally Posted by dragonbud20

Kladner the exponent M37334837 came back with no match someone's going to have to quad check it

as a note to your comments about contacting ailing machines I would support that addition as I use the email feature to let me know when results mismatch currently and while my heart skips a beat when I see one I feel good knowing what's going on and I expect many others would as well.

EDIT: Perhaps this is a "Set-and-Forget" situation.

I continue to have doubts about the GTX 580 that is running these. Thanks for the feedback.

At this point, it is running at stock frequencies (for an MSI Lightning: GPU: 833 MHz, RAM: 2100 MHz. Voltage is boosted for GPU and RAM, which got it to pass the long ST.

With other cards and CUDALucas, I have always had to turn down the RAM clock, to get good results. Trouble is, if the program times out and resets, the RAM will revert to the BIOS value. If I can't be sure of the card giving good results at BIOS RAM clock, I'll have to stop using the loop batch file to run CuLu, so I can reset the values when it hangs.

[0PolarBearsHere]

Originally Posted by dragonbud20

Kladner the exponent M37334837 came back with no match someone's going to have to quad check it

I'll bung it at the top of my queue.

[ET_]

DoubleCheck=37237547,72,1

Verified!! :ok:

[kracker]

Originally Posted by ATH

It did not match unfortunately: http://mersenne.org/M38194643
Titan Black 2048K FFT. Max error: 0.25781.

If I had noticed the "high" error level I would have forced the next FFT probably.

We matched!

[ixfd64]

One of the computers at my old workplace returned a bunch of suspect results (with at least two known bad ones) before I pulled it from GIMPS:

• 52123739 (suspect, residue mismatch)
• 52123733 (suspect, residue mismatch)
• 48424561 (suspect, residue mismatch)
• 48343457 (residue mismatch)
• 24670517 (bad)
• 24670501 (bad)
• 50546803 (suspect, residue mismatch)
• 50546777 (suspect, residue mismatch)
• 50546803 (suspect, residue mismatch)
• 50546777 (suspect, residue mismatch)

It's also completed the following P-1 assignments:

• 52123733
• 52123739
• 48424561
• 48343457
• 50546803
• 50546777

Just putting this out there. For the record, the results.txt file contains four ROUND OFF > 0.4 errors.