Riesel base 6 - team drive #4 - EIGHT OR BUST!

[em99010pepe]

In the reservation board the letter K should be replaced by k. K is for Kelvin degree, k is a SI prefix than means kilo (1,000) and that what it should be used.

[Mini-Geek]

Originally Posted by Flatlander

Are we anywhere near the point where it would be worth double-checking k=1597?

I mean,
When would the probability of finding a missed prime during a month, say, of double-checking all previous k=1597 be greater than finding a new prime during a month of new work?

Well I know what I mean anyway. lol

Funny you should mention that: I figured a little double checking would be well worth it, so yesterday I reran k=1597 for 1<=n<=31377. I also just ran k=36772 for 1<=n<=31927 and sieved both for n<=150K to 10G. The sieve files and what results I have are attached, though I don't know whether it's really worth it to continue, and if so how far.

Originally Posted by gd_barnes

Tim, my preference as this point is to test all of the pairs in the sieve file. Since there were some bugs in PFGW (and I think with LLR) back in 2009 before PFGW version 3.4, I would kindly request a doublecheck of the few pairs that have already been tested. Thanks!

Ok, I've added them back in.

[Mini-Geek]

Originally Posted by Mini-Geek

... it should be about 246 minutes per candidate. ...
If I get drastically different times once I start actually running it, I'll repost.

The first four results are in, and it looks to be closer to 314 minutes per test, or 18.5 per day, or 9.2 days for the whole range. Not sure why my earlier estimate was so far off, except that that was on k=1597, and these four tests were all on k=36772. If that's really the only difference, the average should be a bit lower, but closer to 314 than 246 because there are about twice as many with the larger k.

[Mini-Geek]

Originally Posted by Mini-Geek

The first four results are in, and it looks to be closer to 314 minutes per test, or 18.5 per day, or 9.2 days for the whole range. Not sure why my earlier estimate was so far off, except that that was on k=1597, and these four tests were all on k=36772. If that's really the only difference, the average should be a bit lower, but closer to 314 than 246 because there are about twice as many with the larger k.

After some more results, it seems very likely that the difference is (mostly, anyway) explained by the k's. The k=1597 tests go significantly faster than the k=36772 tests. Based on this, I estimate that my average is about 293 minutes per test, making the whole range take ~8.6 days. I've done 27 tests, and 142 remain (~7.2 days).

[Flatlander]

Originally Posted by Mini-Geek

After some more results, it seems very likely that the difference is (mostly, anyway) explained by the k's. The k=1597 tests go significantly faster than the k=36772 tests. Based on this, I estimate that my average is about 293 minutes per test, making the whole range take ~8.6 days. I've done 27 tests, and 142 remain (~7.2 days).

Okay, thanks.
I'll probably take a bite soon.

[Mini-Geek]

1000K-1010K complete, no primes, results attached.

[mdettweiler]

Reserving 1010K-1030K for a new private PRPnet server I'll be setting up for Flatlander.

[Flatlander]

1030-1050 please Max.
Est. 2-4 weeks. (I can throw more cores at it if anyone gets fidgety.)

[mdettweiler]

Originally Posted by Flatlander

1030-1050 please Max.
Est. 2-4 weeks. (I can throw more cores at it if anyone gets fidgety.)

Loaded, thanks.

[mdettweiler]

Reserving 1050K-1051K.

[unconnected]

950K-1M done.