I've been quarantined!

Rodrigo · 2021-03-06, 06:49

No, this is not a COVID-19 thread.

Tonight I reported a batch of TF results, and at the bottom of the page was a notice in red boldface that said:

[see attachment below]

Never seen anything like this before, and I'm sure that over the years I've submitted batches with equally "unusual" factor rates. Is this something new? A forum search for "quarantine" yielded only pandemic-related results.

Comments welcome.

Viliam Furik · 2021-03-06, 15:25

Wow, that must be some new feature... What exponent range are you doing TF in? I noticed from the picture the bit depth is 74, for the one factor you found. So I guess you are doing TF for exponents in the 104M range (or somewhere near), is that right?

If so, then I don't think that should be something abnormal unless it happened more than once, but even then, random distribution doesn't always behave as we would like.

I think the suspicion bound should be at most 0.25%.

petrw1 · 2021-03-06, 16:22

Quote:

Originally Posted by Rodrigo

No, this is not a COVID-19 thread.

Tonight I reported a batch of TF results, and at the bottom of the page was a notice in red boldface that said:

[see attachment below]

Never seen anything like this before, and I'm sure that over the years I've submitted batches with equally "unusual" factor rates. Is this something new? A forum search for "quarantine" yielded only pandemic-related results.

Comments welcome.

I've been GPU-TF'ing for a few years now in ranges that have had decent P-1.
I'm averaging 1% success rate ... however ...
36 times I went more than 255 exponents without a factor; as high as 633.

ixfd64 · 2021-03-06, 16:22

Probably related to this thread: https://mersenneforum.org/showthread.php?t=24103

Rodrigo · 2021-03-06, 16:43

Quote:

Originally Posted by Viliam Furik

Wow, that must be some new feature... What exponent range are you doing TF in? I noticed from the picture the bit depth is 74, for the one factor you found. So I guess you are doing TF for exponents in the 104M range (or somewhere near), is that right?

If so, then I don't think that should be something abnormal unless it happened more than once, but even then, random distribution doesn't always behave as we would like.

I think the suspicion bound should be at most 0.25%.

Thanks. FWIW, it was in the 208M range.

Rodrigo · 2021-03-06, 17:09

Quote:

Originally Posted by ixfd64

Probably related to this thread: https://mersenneforum.org/showthread.php?t=24103

That made for some pretty interesting reading.

2021-03-06, 06:49	#1
Rodrigo Jun 2010 Pennsylvania 2×467 Posts	I've been quarantined! No, this is not a COVID-19 thread. Tonight I reported a batch of TF results, and at the bottom of the page was a notice in red boldface that said: [see attachment below] Never seen anything like this before, and I'm sure that over the years I've submitted batches with equally "unusual" factor rates. Is this something new? A forum search for "quarantine" yielded only pandemic-related results. Comments welcome. Attached Thumbnails

2021-03-06, 15:25	#2
Viliam Furik "Viliam Furík" Jul 2018 Martin, Slovakia 2⁶×7 Posts	Wow, that must be some new feature... What exponent range are you doing TF in? I noticed from the picture the bit depth is 74, for the one factor you found. So I guess you are doing TF for exponents in the 104M range (or somewhere near), is that right? If so, then I don't think that should be something abnormal unless it happened more than once, but even then, random distribution doesn't always behave as we would like. I think the suspicion bound should be at most 0.25%.

2021-03-06, 16:22	#4
ixfd64 Bemusing Prompter "Danny" Dec 2002 California 4513₈ Posts	Probably related to this thread: https://mersenneforum.org/showthread.php?t=24103