Pascal's OPN roadblock files

[henryzz]

Originally Posted by ryanp

I'm down to 1124 composites remaining in my mwrb2100.

Remaining: https://cs.stanford.edu/~rpropper/mwrb2100.txt
Factors found: https://cs.stanford.edu/~rpropper/opn.txt

Originally Posted by RichD

It looks like an SNFS-253 quartic. A few SNFS-251 quartic have made it through so I believe it would fit.

I can send it to yoyo for further ECM work to see how that goes. What is a good t-level to start at?

It appears that ryanp ecmed this in the quoted post. The factors file contains some factors in the low 60 digits. My best guess is that t60 has been done. Maybe a bit more.

[VBCurtis]

It has been a few years since I've test-sieved a quartic; I'm due to refresh myself, so I volunteer to generate params for f-small if desired. I vaguely recall that quartics benefit from a large sieve area but are pretty fast in sec/rel.

Is this number interesting enough to run as a CADO team effort? I'm willing to host the job and donate a machine-month (20 old ivy bridge xeon cores, about as fast as 8 ryzen-7900 cores).

[kruoli]

Please add at least eight Zen 3 core-months from me, please.

[henryzz]

Originally Posted by VBCurtis

It has been a few years since I've test-sieved a quartic; I'm due to refresh myself, so I volunteer to generate params for f-small if desired. I vaguely recall that quartics benefit from a large sieve area but are pretty fast in sec/rel.

Is this number interesting enough to run as a CADO team effort? I'm willing to host the job and donate a machine-month (20 old ivy bridge xeon cores, about as fast as 8 ryzen-7900 cores).

I would be quite interested in a test sieve comparison between CADO and f-small parameters. It wouldn't surprise me if CADO would allow some nice variation compared to lasieve. Could an unusual shape sieve region be more important than the size? The correct sieve-adjust choice could be key.

Would be glad to take part, although my resources are fairly limited.

[RichD]

Originally Posted by henryzz

My best guess is that t60 has been done. Maybe a bit more.

I'll put 10K @ 850e6 at yoyo tonight for starters. I could help with a team sieve but I only contributed 1% the last time around with my Sandy/Ivy Bridge boxes.

@Oliver: Do you know when yoyo might implement the stage 2 only feature on his site?

[VBCurtis]

Originally Posted by henryzz

I would be quite interested in a test sieve comparison between CADO and f-small parameters. It wouldn't surprise me if CADO would allow some nice variation compared to lasieve. Could an unusual shape sieve region be more important than the size? The correct sieve-adjust choice could be key.

Would be glad to take part, although my resources are fairly limited.

Can you post the poly to use for this job? I'll work on f-small params first, then I can test CADO with the various sieve-adjust flags on those lp/mfb params.

[RichD]

Originally Posted by VBCurtis

Can you post the poly to use for this job?

n: 11762658718349861390232119244704668953655520173988132853958736559505227543720741129707951916855819491584483404706118568545301944488029224273208806055167884044998017282642609466917674130929252487860835945397664760737343625974929824544969881973722726911
# 1310825268269643509279336731098526398390609803239319801398048897^5-1, difficulty: 252.47, skewness: 1.00, alpha: 1.45
# cost: 1.00224e+19, est. time: 4772.57 GHz days (not accurate yet!)
skew: 1.000
c4: 1
c3: 1
c2: 1
c1: 1
c0: 1
Y1: -1
Y0: 1310825268269643509279336731098526398390609803239319801398048897

Edit: cownoise suggest using skew: 2.026

[charybdis]

If you're using CADO, I suspect this may be in the range where A=30 with adjust-strategy=2 is fastest. Large prime bounds skewed towards the rational side, 33/31 or 34/32.

My cores are busy with Cunninghams so will not contribute to this - @RichD that'll help you get a higher percentage

[mataje]

c80-89 done.

[kruoli]

Originally Posted by RichD

@Oliver: Do you know when yoyo might implement the stage 2 only feature on his site?

He had been on it before the Pentathlon, until that needed his time more urgently, so maybe he will continue after it? He did not make any promises.

[mataje]

c90-99 done.