20210904, 11:31  #738 
Apr 2010
Over the rainbow
3^{2}·5·59 Posts 
sure i'll take 6.2M for a spin

20210904, 13:11  #739 
Apr 2010
Over the rainbow
3^{2}·5·59 Posts 
and thank you for your detailled explanation.

20210905, 04:44  #740 
1976 Toyota Corona years forever!
"Wayne"
Nov 2006
Saskatchewan, Canada
2^{2}×1,193 Posts 
I see a few results coming in with two very different sets of B1/B2; one taking twice as many GhzDays with probably only an extra 1 percent success rate.
Code:
firejuggler Maison 6263651 NFPM1 20210904 16:27 0.0 4.1311 B1=3000000, B2=243000000 firejuggler Maison 6264023 NFPM1 20210904 19:27 0.0 4.1311 B1=3000000, B2=243000000 firejuggler Maison 6264127 NFPM1 20210904 22:27 0.0 4.1311 B1=3000000, B2=243000000 firejuggler Maison 6264367 NFPM1 20210905 00:11 0.0 1.9890 B1=3000000, B2=90000000 firejuggler Maison 6264413 NFPM1 20210905 01:49 0.0 1.9890 B1=3000000, B2=90000000 firejuggler Maison 6264617 NFPM1 20210905 03:19 0.0 1.9890 B1=3000000, B2=90000000 and some are using PFACTOR those with 3M/243M It is no problem using PFACTOR ... however if you want the same expected success rate you will need a lower B1. That is because the newest version of PFACTOR that determines its own B2 tends to pick it many more multiples of B1. If so and if you prefer PFACTOR experiment a bit with B1 so that the B2 chosen gives you about the same 6.64% success using: https://www.mersenne.ca/prob.php. I might guess something closer to 2M instead of 3M. Of course with your higher B2 you will certainly find the required factors with less exponents but the total GhzDays expended will be quite a bit more. Thanks 
20210905, 05:30  #741 
Romulan Interpreter
"name field"
Jun 2011
Thailand
2645_{16} Posts 
Man, you almost beat me for the length of that post, hehe, did you meet kriesel for a beer over the weekend and he schooled you, or what?
Today and yesterday we only got T4 from Gugu (Pro_non_plus, and free instances), which suck at P1, so we let them doing TF with Chris. We only found a single factor in 9.8M, which isn't much . We will continue to do that for a while, but mainly we will retreat from this project, as now all our local hardware is mining, and Gugu is taking our money for free, as currently, the only difference between the free instances and Pro instances is the running time. Seems they want to force us into paying for Proplus, but such budget we don't have yet. So, 11M is free for now, which will make Chris happy, so he has no reason to wonder anymore "why the 11M fixation" For some reason I don't have access to my computer at work, possible server powered off due to long weekend, as here we'll have holiday on Monday. I wanted to post a summary for my 90days activity  I mean, I was doing occasional work here before, but I joined this project more seriously on June 7, and there should be 90 days by today, good opportunity to celebrate, but also, in the light of my former post, to "retreat in glory". But that summary file is lost somewhere inside of the office computer, I will try to retrieve it if the server comes back, but if not, then see you on Tuesday, hehe. Last fiddled with by LaurV on 20210905 at 05:33 
20210905, 15:49  #742 
1976 Toyota Corona years forever!
"Wayne"
Nov 2006
Saskatchewan, Canada
11244_{8} Posts 

20210905, 22:50  #743 
"GIMFS"
Sep 2002
Oeiras, Portugal
2·3^{2}·83 Posts 

20210906, 21:06  #744 
If I May
"Chris Halsall"
Sep 2002
Barbados
23352_{8} Posts 

20210907, 04:30  #745  
1976 Toyota Corona years forever!
"Wayne"
Nov 2006
Saskatchewan, Canada
11244_{8} Posts 
Quote:
Just sayin' ... but ... you probably don't have to curate 22.8 or 23.3 to TF74 I think P1 can get these. 

20210916, 00:18  #746 
"Seth"
Apr 2019
3^{2}×41 Posts 
Anyone want to guess which will be the final range?
My guess is one of 10.7M has 159 remaining, with average TF=71 (already more TF than 17.0M, but P1 is faster) 15.5M has 153 remaining, with average TF=71.358 (with few low P1) 17.0M has 173 remaining, with average TF=71.279 (expect have ~130 remaining at TF=73) 29.8M has 113 remaining, with average TF=74 Maybe 9.9M has 123 remaining, with average TF=71.02 (but 970 Low P1) 20.8M has 143 remaining, with average TF=70.95 (expect more like 60 remaining by the time TF=73/74) 25.3M has 120 remaining, but slightly more low P1 than 29.8 (and is faster) 
20210916, 04:13  #747  
1976 Toyota Corona years forever!
"Wayne"
Nov 2006
Saskatchewan, Canada
2^{2}·1,193 Posts 
Quote:
Once I finish 3xM in a few months, I expect to spend at least 2 more years doing P1 in 2xM; ending with 29.8. A couple entities are working low to high. In that time period I am hopeful the "lowenders" can be getting close to 10M. Then in the 1xM range I will probably end with 17.0. Now I am secretly hoping (don't tell anyone) that someone takes it as a challenge to prove us all wrong and ambitiously take on all these hardest ranges. My poster child is masser who took on 14.0 a few years ago when it was by far the worse with 297 factors remaining. He's getting close!!! 

20210916, 16:26  #748  
Jul 2003
Behind BB
3×577 Posts 
Quote:
My hunch is that 17.0 is the most difficult of the remaining dense ranges. The project certainly has the firepower to finish off this task: push the TF bit level to an extreme (like 74,75 bits) then P1 with B1 = 10M ought to do it. 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Thinking of Joining GPU to 72  jschwar313  GPU to 72  3  20160131 00:50 
Thinking about lasieve5  Batalov  Factoring  6  20111227 22:40 
Thinking about buying a panda  jasong  jasong  1  20081111 09:43 
Loud thinking on irregular primes  devarajkandadai  Math  4  20070725 03:01 
Question on unfactored numbers...  WraithX  GMPECM  1  20060319 22:16 