mersenneforum.org > Data Thinking out loud about getting under 20M unfactored exponents
 Register FAQ Search Today's Posts Mark Forums Read

 2021-10-21, 08:17 #815 LaurV Romulan Interpreter     "name field" Jun 2011 Thailand 2×3×5×7×47 Posts Yay! Found a gem! Code: processing: P-1 factor 67384156745900300766317103355688421517561 for M11321579 (B1=6,189,107, B2=185,673,210) (135.630 bits) CPU credit is 8.9141 GHz-days.
2021-10-21, 16:56   #816
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

499710 Posts

Quote:
 Originally Posted by LaurV Yay! Found a gem! Code: processing: P-1 factor 67384156745900300766317103355688421517561 for M11321579 (B1=6,189,107, B2=185,673,210) (135.630 bits) CPU credit is 8.9141 GHz-days.

2021-10-22, 01:49   #817
LaurV
Romulan Interpreter

"name field"
Jun 2011
Thailand

232168 Posts

Quote:
 Originally Posted by petrw1 When I started this project just over 4 years ago there were 498 ranges to go. Now with 166 ranges to go we are officially 2/3 complete.
We just passed another milestone:
Less than 333333 candidates left to test (that is, the total number of exponents in the remaining ranges, yesterday evening Thai time, was exact: 333000).

Something to do with being half evil...

Last fiddled with by LaurV on 2021-10-22 at 02:12

2021-10-22, 02:26   #818
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

116058 Posts

Quote:
 Originally Posted by LaurV We just passed another milestone: Less than 333333 candidates left to test (that is, the total number of exponents in the remaining ranges, yesterday evening Thai time, was exact: 333000). Something to do with being half evil...
When I started (July 2017) there were just under 1,050,000 to test.
This makes us over 67% complete by this measure.

Also when I started I needed 55,228 factors.
As of now there are 9,471 remaining.
82.5% complete by this measure.

There were 497 incomplete ranges.
Now there are 158.
69.2% by this measure.

However, all things considered the number 333,333 is the most interesting.

 2021-10-23, 00:32 #819 LaurV Romulan Interpreter     "name field" Jun 2011 Thailand 100110100011102 Posts And one more... How does 179 bits sound? Unfortunately, not prime... Code: processing: P-1 no-factor for M11318603 (B1=6,189,107, B2=185,673,210) CPU credit is 8.9141 GHz-days. Splitting composite factor 735438206294405813130223741791584805574508614902245969 into: * 179701183696182915867337 * 4092561836085598728177371678537 processing: P-1 factor 179701183696182915867337 for M11318563 (B1=6,189,107, B2=185,673,210) (77.250 bits) CPU credit is 8.9141 GHz-days. processing: P-1 factor 4092561836085598728177371678537 for M11318563 (B1=6,189,107, B2=185,673,210) (101.691 bits) CPU credit is 8.9141 GHz-days.
2021-10-23, 05:42   #820
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

19·263 Posts

Quote:
 Originally Posted by LaurV M113....
It's your compute but with the 1523 TF held by GPU72 in this range; expected to be completed next Friday; you do have the option to see if that completes M11.3 and move to another range in the mean time....11.2M ???

 2021-10-26, 05:37 #821 LaurV Romulan Interpreter     "name field" Jun 2011 Thailand 232168 Posts That is actually not a bad idea. The order I went for, in 11.3xM, was x=6,5,4,3,2,1,8,9,7,0, in order, i.e. I started with those sub-sub-ranges that had more-than-200 candidates and ended with those having fewer. The idea was that if I find enough factors in the first sub-sub-ranges, I won't need to bother with the last. And do a little step towards a possible utopian futuristic goal of having "less than 200 unfactored candidates in each sub-sub-range". Now, I already finished 6,5,4,3,2, and I am currently working on 1, all of these had more than 200 (some still have) candidates. Now, 8 still have 201 candidates, and the rest of the sub-sub-ranges have less than 200 candidates. There are only few candidates left in 1, but the progress is very slow, nothing like I was doing a month ago or so. I was thinking to take a break anyhow after finishing 1, or latest, after I found one factor in 8, then wait and see what gpu72 TF will provide. For how much P-1 was done here, I don't believe that the TF work will bring more than 10 factors or so, per bitlevel, and the next bitlevel is A LOT LESS efficient than P-1, at this size of the exponents (per wall clock, regardless of your GPU). So, I still think some P-1 will need to be done at the end to get the last 1-2-3-few factors.
2021-10-26, 05:58   #822
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

10011100001012 Posts

Quote:
 Originally Posted by LaurV ... but the progress is very slow...
I seem to recall you have a few tit-ans that would love to chew through P1 work

2021-10-26, 06:55   #823
LaurV
Romulan Interpreter

"name field"
Jun 2011
Thailand

2×3×5×7×47 Posts

Quote:
 Originally Posted by petrw1 I seem to recall you have a few tit-ans that would love to chew through P1 work
Haha

Still have 6 working Titans (5 classic and 1 black), plus 4x 580s (these are the old NV Fermi, not the much newer AMD 580s), plus 2x 1080Ti, pus 1x R6990 (or 7990, not sure), which don't do anything currently, they are boxed and shelved, because I have no rig to put them in, and running them doesn't really justify the electricity costs. Plus a lot of rubbish, about 20 other cards, including 7 more Titans from which I took components to repair the 6 working cards (totally 13 Titans, including working and rubbish).

Some day... :hope:

Last fiddled with by LaurV on 2021-10-26 at 06:57 Reason: s/to rig/no rig/

2021-10-26, 08:30   #824
Luminescence

Oct 2021
Germany

1010112 Posts

Quote:
 Originally Posted by LaurV Haha Still have 6 working Titans (5 classic and 1 black)…
Woah, 6 Titans. How is the throughput with those six? I have two RX 6900 XTs (only one hooked up) and a run with 1M/30M in the 2xM range takes 10 minutes. I am still trying to fit my second one in the case… they are huge cards

2021-10-26, 15:38   #825
petrw1
1976 Toyota Corona years forever!

"Wayne"
Nov 2006

19×263 Posts

Quote:
 Originally Posted by Luminescence I have two RX 6900 XTs (only one hooked up) and a run with 1M/30M in the 2xM range takes 10 minutes.
That explains how quickly you are clearing the ranges....I need one of those....or two

 Similar Threads Thread Thread Starter Forum Replies Last Post jschwar313 GPU to 72 3 2016-01-31 00:50 Batalov Factoring 6 2011-12-27 22:40 jasong jasong 1 2008-11-11 09:43 devarajkandadai Math 4 2007-07-25 03:01 WraithX GMP-ECM 1 2006-03-19 22:16

All times are UTC. The time now is 04:16.

Tue Jan 25 04:16:29 UTC 2022 up 185 days, 22:45, 0 users, load averages: 2.35, 1.77, 1.54