mersenneforum.org Search Results
 Register FAQ Search Today's Posts Mark Forums Read

 Showing results 1 to 25 of 1000 Search took 0.45 seconds. Search: Posts Made By: ewmayer
 Forum: Lounge 2021-09-16, 21:28 Replies: 1,697 Views: 134,349 Posted By ewmayer My own rule for my GPU crunchers is that only one... My own rule for my GPU crunchers is that only one of A/C or Crunchers may run at a given time. I only ethically justify running the GPU farms (which has added ~\$200/month to my PG&E bill) by way of...
 Forum: Programming 2021-09-09, 05:27 Replies: 9 Views: 1,193 Posted By ewmayer It'll be easier for the rest of us to tell if... It'll be easier for the rest of us to tell if it's implemented correctly if you would provide a sample input, say a random number of the desired bitlength p you are working (or pair of such if you're...
 Forum: Programming 2021-09-07, 20:50 Replies: 9 Views: 1,193 Posted By ewmayer BTW, a useful identity for computing inverse... BTW, a useful identity for computing inverse weights for mod-M(p) transform is as follows - for length-N transform, start with the weights as per the original Crandall/Fagin paper (I dislike their p...
 Forum: Programming 2021-09-07, 00:38 Replies: 9 Views: 1,193 Posted By ewmayer Assuming your inputs are properly normalized,... Assuming your inputs are properly normalized, i.e. x,y < (2^p-1), you break each into 32/33-bit pieces according to the per-word bitlengths in your list. (Balanced-digit representation is easy to...
 Forum: Mlucas 2021-09-05, 03:02 Replies: 6 Views: 486 Posted By ewmayer Congrats, but note that due to a small typo, "77... Congrats, but note that due to a small typo, "77 digits" means binary digits, a.k.a. bits. We can dream of finding a 77-decimal-digit monster, tho. So only a 6-7% stage 2 speedup on Arm, vs the...
 Forum: Mlucas 2021-09-02, 20:36 Replies: 9 Views: 721 Posted By ewmayer v20.1 is out. I suggest users upgrade at their... v20.1 is out. I suggest users upgrade at their earliest convenience to gain the benefit of various bugfixes, as well as a significantly faster p-1 stage 2. Please also check your worktodo.ini...
 Forum: Mlucas 2021-09-01, 22:18 Replies: 6 Views: 486 Posted By ewmayer Brief post illustrating how users who hit the v20... Brief post illustrating how users who hit the v20 assignment-borkage-due-to-missing-string-terminator issue mentioned in the above list can manually patch up affected assignments, which is preferable...
 Forum: Mlucas 2021-08-31, 22:14 Replies: 6 Views: 486 Posted By ewmayer Mlucas v20.1 (latest) available This is an Update-release of v20, but with enough changes as to warrant a minor-version number increment. As always, download via the README page (http://www.mersenneforum.org/mayer/README.html). ...
 Forum: Lounge 2021-08-09, 00:52 Replies: 1,109 Views: 96,229 Posted By ewmayer https://www.resilience.org/stories/2021-08-05/bill... https://www.resilience.org/stories/2021-08-05/billionaire-space-race-the-ultimate-symbol-of-capitalisms-flawed-obsession-with-growth/
 Forum: Msieve 2021-08-08, 00:05 Replies: 46 Sticky: Msieve benchmarking Views: 23,521 Posted By ewmayer Quickly found this from early 2018: Cavium's... Quickly found this from early 2018: Cavium's ThunderX2 processor powers first 64-bit ARMv8 workstation...
 Forum: Msieve 2021-08-07, 00:03 Replies: 46 Sticky: Msieve benchmarking Views: 23,521 Posted By ewmayer By VBITS do you mean #threads running on the CPU?... By VBITS do you mean #threads running on the CPU? (The 9980 is like the Intel KNL, in supporting up to 4 threads per core). Just curious, how much might such a beast cost? (Need not be new.) The...
 Forum: Mlucas 2021-08-01, 21:49 Replies: 9 Views: 721 Posted By ewmayer Did some unclosed-element cleanup (yes, I am a... Did some unclosed-element cleanup (yes, I am a dinosaur, I still hand-create and edit my html) and spell-checking on the README.html file; also realized that my 'ln -s' example in the...
 Forum: Mlucas 2021-08-01, 20:14 Replies: 9 Views: 721 Posted By ewmayer That's gonna be the major feature add for v20.1.... That's gonna be the major feature add for v20.1. I am hopeful that that will need less time than the p-1 proof-of-principle-work-and-code-it-up did. No fundamental new algorithms or flavors-of-modmul...
 Forum: Mlucas 2021-07-31, 20:10 Replies: 9 Views: 721 Posted By ewmayer Mlucas v20 available Mlucas v20 has gone live (http://www.mersenneforum.org/mayer/README.html). Use this thread to report bugs, build issues, and for any other related discussion. Major feature-add: p-1 support Major...
 Forum: Hardware 2021-07-29, 20:15 Replies: 40 Views: 4,045 Posted By ewmayer Axn beat me to it - does his 300-some-GB estimate... Axn beat me to it - does his 300-some-GB estimate jibe with what your system load app shows? The 'pair%=95.65' means that out of every 10000 stage 2 primes, on average 9565 end up in bitmap pairs,...
 Forum: Hardware 2021-07-24, 23:04 Replies: 40 Views: 4,045 Posted By ewmayer @drkirby: I believe once you get beyond... @drkirby: I believe once you get beyond 500-1000 memory buffers in stage 2, you get rapidly diminishing returns, as the stage 2 prime-pairing percentage approaches 100%. Do your mprime logs list...
 Forum: Hardware 2021-07-19, 19:53 Replies: 221 Views: 33,410 Posted By ewmayer Ah, you must have a 64-core rig, not 68. What... Ah, you must have a 64-core rig, not 68. What does 'sensors | grep Package' show? Here's output for mine: Package id 0: +68.0°C (high = +80.0°C, crit = +90.0°C) ALARM (CRIT) This system is...
 Forum: Hardware 2021-07-19, 01:35 Replies: 221 Views: 33,410 Posted By ewmayer I'd have to dig out the data I generated when... I'd have to dig out the data I generated when first testing my KNL, but IIRC I got close to 100% || scaling up to 4-threads for current-wavefront FFTs, rapid ||-scaling deterioration above that...
 Forum: Software 2021-07-18, 21:54 Replies: 50 Views: 2,467 Posted By ewmayer One of my college thermodynamics professors used... One of my college thermodynamics professors used the following example to illustrate some of the counterintuitive aspects of statistical thermodynamics. He was standing in front of a blackboard and...
 Forum: Software 2021-07-18, 20:21 Replies: 50 Views: 2,467 Posted By ewmayer I gave you a link to the original Crandall/Fagin... I gave you a link to the original Crandall/Fagin DWT paper 2 full pages worth of posts ago. I suggest you cool the posting byterate and actually read it and work through the short-length examples it...
 Forum: Software 2021-07-18, 04:36 Replies: 50 Views: 2,467 Posted By ewmayer http://www.faginfamily.net/barry/Papers/Discrete%2... http://www.faginfamily.net/barry/Papers/Discrete%20Weighted%20Transforms.pdf IBDWT leverages the mathematical structure of the FFT (or suitably chosen NTT) to do an "implicit mod" modulo M(p)...
 Forum: Mlucas 2021-07-06, 19:40 Replies: 46 Views: 5,752 Posted By ewmayer Thanks, Teal - that's a very useful fiddle, I... Thanks, Teal - that's a very useful fiddle, I will post a link to the so-hacked Loarer-script in a v19 patch alert to the README. It should be moot in a few weeks when v20 is up, but for folks not...
 Forum: Mlucas 2021-07-05, 20:19 Replies: 46 Views: 5,752 Posted By ewmayer Bug Alert: This bug will cause around 6% of... Bug Alert: This bug will cause around 6% of completed results to not be successfully submitted to the server via the v5-APIified primenet.py script. I first discovered the issue in March, around a...
 2021-07-02, 18:57 Replies: 80 Views: 29,996 Posted By ewmayer Fermat-modmul with circularly shifted residues I spent an appreciable amount of time last year wrestling with basic concepts and implementation-in-code of Fermat-mod squaring chains in the presence of residue shift. This proves rather more...
 Forum: GpuOwl 2021-06-26, 19:34 Replies: 29 Views: 1,836 Posted By ewmayer It means you don't need to do a thing - nonzero... It means you don't need to do a thing - nonzero means p-1 will get done automatically before PRP.
 Showing results 1 to 25 of 1000

All times are UTC. The time now is 17:14.

Sun Sep 19 17:14:07 UTC 2021 up 58 days, 11:43, 0 users, load averages: 1.74, 1.47, 1.31