What is a "suspect" result

[Mark Rose]

Originally Posted by sdbardwick

We'll need to wait for Mark for it to be official, but it looks like your result is good! My boring Haswell-E produced a zero-error result that matches your residue (at least as far as I can tell).

I should have an answer tomorrow morning. ETA is 15 hours, but I'll likely be sleeping. As it's already assigned, and I don't want to poach, I'll only submit the result if it differs from all the others.

[CuriousKit]

Originally Posted by sdbardwick

We'll need to wait for Mark for it to be official, but it looks like your result is good! My boring Haswell-E produced a zero-error result that matches your residue (at least as far as I can tell).

That is a huge relief to hear and a nice reduction in my worry - thanks for the effort!

Originally Posted by Mark Rose

I should have an answer tomorrow morning. ETA is 15 hours, but I'll likely be sleeping. As it's already assigned, and I don't want to poach, I'll only submit the result if it differs from all the others.

No worries, I can hopefully wait until the 3rd of September!

[Madpoo]

Originally Posted by CuriousKit

Oh man... three results that don't agree?! Now I have no clue what to think!

If there are exponents with 3+ unmatched results, I usually pick those up pretty quick once they're unassigned.

I don't touch assigned exponents though... there's enough work with triple+ checking that I don't have to worry about picking up any that someone might be working on.

I just added a batch of, I think, 6 exponents that need quad checks and weren't assigned. Some had a new mismatch, some recently became available.

Overall there aren't many with 3+ mismatches left... I've cleared out all of them I could and now just picking off the few that become available.

In rare cases my own result doesn't match any of the others, so I've been mentioning my mismatched *fourth* result to the Marin Mersennearies thread so someone can do a quint check. I have a handful where my check was the third mismatch and it now needs a quad check, but I haven't been pointing those out... probably a dozen or so of them. And quite a few of mine now need triple checks thanks to the "strategic double checking" we've been doing... picking up lots of (probably) bad first time tests.

[pyajve]

Hello,

Sorry if posting this to a wrong thread.
I've recently got a "C-Unverified" result which is still unassigned (for a month now).
Could someone please triple-check it?
http://www.mersenne.org/report_expon...5728993&full=1

Thanks a lot.
Andrey.

[CuriousKit]

Just tried picking it up as a manual test, but it won't let me. Says it's unavailable. How is this normally done?

[Mark Rose]

Originally Posted by sdbardwick

Finished with zero errors, but without matching other results. It would be great if someone else picked it up for a quick re-test, but I'll schedule it for a rerun it on a known-good Sandy Bridge box after endless mike's finishes. So in a week we'll (probably) know either way.

M39070231 is not prime. Res64: D5FE1583820D4C__.

I matched CuriousKit's result.

[kracker]

M36666709 Hmm.. I'm really curious about this one as it was done on a GPU.

[Mark Rose]

Originally Posted by kracker

M36666709 Hmm.. I'm really curious about this one as it was done on a GPU.

I tried to grab it, but someone did in the literal last 30 seconds.

[sdbardwick]

Originally Posted by kracker

M36666709 Hmm.. I'm really curious about this one as it was done on a GPU.

I grabbed this one so we'll know in about a day.
Edit: Sorry Mark :)

[kracker]

Originally Posted by sdbardwick

I grabbed this one so we'll know in about a day.
Edit: Sorry Mark :)

Thanks!

[sdbardwick]

Originally Posted by kracker

M36666709 Hmm.. I'm really curious about this one as it was done on a GPU.

We match!