@diep
I've created the [url='https://www.rieselprime.de/ziki/Riesel_prime_32767']Riesel k=32767[/url] in the Wiki and reserved it for you (including tested to n=50k).
I've also updated the k=89 reservation. If you got any further results for k=32767 please fill them in.

Kar, c'est bon!

You can be sure i will report it if there is a new one above n=1569 :)

Please note for the few who might not know: testspeed of LLR with larger k's is a lot slower than small k's. k = 89 and k=69 here tests roughly factor 2 faster than k=32767, under the same conditions.

At the CPU's with many cores here what matters most is that LLR FFT-size stays within combined L3 cachesize : #cores * L3_cache_a_core

I do not know details yet the FFT-size seems dependant upon which k-factor you use.

(so if you know the FFT size for that specific k-factor and n-size in units multiply it with 16 bytes - because the size is in doubles and you need 1 double which has t he information and 1 to write new information to,
and a double is 64 bits so that's 8 bytes and you need 2 of that makes 16 bytes. The blabla from manufacturers that you can add up the L2 and L1 cachesize to it seems utter blabla from the manufacturers to me by the way yet that's not relevant here - just L3 cachesize matters there in that comparision - except if you do not have a L3 cache then L2 cache size matters. So to say it technical the algorithm doesn't seem to work in-situ regrettably)

[QUOTE=diep;564551]I do not know details yet the FFT-size seems dependant upon which k-factor you use.[/QUOTE]

According to that LLRTools tool I linked, to go from the max [I]n[/I] for [I]k[/I]=1 of a given FFT length (which I'll call [$]n_m[/$]) to the max [I]n[/I] for any [I]k[/I] for that same FFT length (which I'll call [$]n_k[/$]), the formula is [TEX]n_k = n_m - (\log_2 k + \frac{1}{2}(\log_2 k)(fftlen))[/TEX].

Here is the promised script. It's a little slow, and I haven't tested it on Windows (I don't currently have access to a Windows box), but it's completely automated once started. It uses a binary search (which it prints out) to find the minimum [I]n[/I] for every FFT length below a given limit (provided on the command line; you can give the exact number or a "K" or "M" suffixed one like LLR prints out; the actual length doesn't have to be present on that platform). It dynamically finds the relevant FFT lengths for that platform through the search. Create a "temp" subfolder as a working directory for it to dump all of the temp files (it creates a ton of them). It uses the CLI version of LLR (assuming its name is "llr" in the current directory). Let me know if anyone finds this useful.

[QUOTE=Happy5214;564652]Here is the promised script. It's a little slow, and I haven't tested it on Windows (I don't currently have access to a Windows box), but it's completely automated once started. It uses a binary search (which it prints out) to find the minimum [I]n[/I] for every FFT length below a given limit (provided on the command line; you can give the exact number or a "K" or "M" suffixed one like LLR prints out; the actual length doesn't have to be present on that platform). It dynamically finds the relevant FFT lengths for that platform through the search. Create a "temp" subfolder as a working directory for it to dump all of the temp files (it creates a ton of them). It uses the CLI version of LLR (assuming its name is "llr" in the current directory). Let me know if anyone finds this useful.[/QUOTE]

How to use the script?
What do i type on the command line?

something like this?

python happy.py 32767 5800000

k=2805
 

Releasing k=8847 and reserving k=2805.

[QUOTE=diep;564875]How to use the script?
What do i type on the command line?

something like this?

python happy.py 32767 5800000[/QUOTE]
To be honest, I use Linux, so I'd just set the executable flag and the shebang (the first comment line with the Unix-style interpreter path) would do the rest. You appear to be using Windows, though, given that invocation. It uses Python 3, so I think you'd call "python3" instead of "python" (which calls Python 2). At least that's how it works on Ubuntu.

You have the arguments flipped. The [I]max FFT length[/I] ([I]not[/I] the max [I]n[/I]) is the first one, since it's the only required one. The [I]k[/I] you choose for the optional second argument shouldn't change the result significantly unless it starts zero-padding the FFTs (high 6-digits), and I chose [I]k[/I]=100005 to avoid zero-padding while still shortening the timing tests (since the [I]n[/I]'s will be lower). [I]n[/I]=5.8M is at the 480k FFT length for your [I]k[/I] on my Core 2 Quad, and the Xeon won't be too much different, if at all, so run it to that or 512k. Something like "python3 happy.py 480K".

I've attached another script that reads in the generated maxlen.txt and converts the max [I]n[/I]'s (which are adjusted in the file to match Mersenne values) for any chosen [I]k[/I]. They should closely match the test results from the verbose printouts of the first script. It's called using "python3 fftlen.py [I]k[/I]".

I posted those two scripts (renamed to generate_maxlen.py and adjusted_fftlen.py) and a few others to a new GitHub repo: [url]https://github.com/happy5214/rps-scripts[/url]. I'll post updates there. You'll need the fftlen.py library module for both scripts. I plan on adding a new argument interface to generate_maxlen.py to allow you to specify a max [I]n[/I] instead of a max FFT length, and also to specify verbosity via a flag instead of by editing the code.

I posted an update to the generate_maxlen.py script which revamps the CLI interface. It now uses the Python argparse standard library module, which makes it a lot nicer. I've also added support for specifying a max [I]n[/I] in lieu of a max FFT length and moving the verbose flag to the CLI. It now has a built-in help message generated by the module, so check that for the details.

Perhaps an admin should make a new thread for such script-related posts, better finding later.

The (near-)Woodall [I]k[/I]'s listed in [url]https://www.mersenneforum.org/showpost.php?p=550539&postcount=363[/url] have been completed to [I]n[/I]=425k. The following primes were found:[list][*]1993191*2^405251-1[*]582833*2^417560-1[*]197673*2^421871-1[/list]

I've completed the remaining RPS 9th and 10th Drive [I]k[/I]'s with missing ranges from [I]n[/I]=325k to 350k. 14 primes were found, which are attached. I brought an old laptop into the fold, and I have a new Kubuntu Focus on the way, so this project (to [I]n[/I]=400k) may be done by this summer.

[QUOTE=Happy5214;568280]The (near-)Woodall [I]k[/I]'s listed in [url]https://www.mersenneforum.org/showpost.php?p=550539&postcount=363[/url] have been completed to [I]n[/I]=425k. The following primes were found:[/QUOTE]What are (near-)Woodall k's? I know about near-Woodall primes where k is n-1, but that doesn't seem to be the case here.

(General) Near Woodalls are primes of type (n-1)*b^n-1 or (n+1)*b^n-1, examples [url='https://www.rieselprime.de/ziki/Gen_Woodall_prime_M_2_table']here[/url] or [url='https://www.rieselprime.de/ziki/Gen_Woodall_prime_P_2_table']here[/url].
Most of the are not spotted immediately, because the normal form in for example the Top5000 page is displayed different:
Example: 1060*2^1059-1 is normalized written as 265*2^1061-1

To be clear, these are [I]k[/I]'s where near-Woodall and Woodall primes were found for larger [I]n[/I] values, and I'm filling in the untested missing ranges. I'm not actually finding new primes of those forms, which are searched for largely by Steven Harvey and PrimeGrid, respectively.

Reserving k=3317

As of March 27, the (near-)Woodall [I]k[/I]'s listed in [url]https://www.mersenneforum.org/showpost.php?p=550539&postcount=363[/url] have been completed to [I]n[/I]=450k. The following primes were found:[list][*]1268979*2^426051-1[*]1195203*2^444882-1[*]197673*2^447700-1[/list]

[QUOTE=Happy5214;540140]I'll also tack on k=14549535 for 125k < n < 300k as a double-check.[/QUOTE]


Happy,

From a year ago...was this done? I'd like to go ahead and double-check k=14549535 for all n<=300K and then test it up to n=500K. Is that OK?

Gary

[QUOTE=gd_barnes;575291]Happy,

From a year ago...was this done? I'd like to go ahead and double-check k=14549535 for all n<=300K and then test it up to n=500K. Is that OK?

Gary[/QUOTE]
Funny you mention that. I had started work on that [I]k[/I] in late February, but I had suspended work on my local PRPNet servers for a bit in March to work on the GIMPS PRP-CF-DC backlog. I have since resumed work on both clients. That [I]k[/I] is currently at [I]n[/I]=288k, and I should get to [I]n[/I]=300k by the end of this week. You can take over from there.

[QUOTE=Happy5214;575349]Funny you mention that. I had started work on that [I]k[/I] in late February, but I had suspended work on my local PRPNet servers for a bit in March to work on the GIMPS PRP-CF-DC backlog. I have since resumed work on both clients. That [I]k[/I] is currently at [I]n[/I]=288k, and I should get to [I]n[/I]=300k by the end of this week. You can take over from there.[/QUOTE]


OK I'll start testing within a couple of days for n=300k-500k. Thanks.

k=14549535 is complete for n=300K-500K, 5 primes were found shown below, the k is released. Happy is still working on a double-check for n=125K-300K.

Primes:
14549535*2^376607-1
14549535*2^390393-1
14549535*2^430198-1
14549535*2^446171-1
14549535*2^453217-1

The aforementioned double-check of [I]k[/I]=14549535 for [I]n[/I]=125k-300k is complete. In addition to the known primes, the following new primes were found:[list][*]14549535*2^149070-1[*]14549535*2^160953-1[*]14549535*2^161852-1[*]14549535*2^180548-1[*]14549535*2^199042-1[*]14549535*2^204177-1[/list]

k=8847:

Double-check complete to n=500K. No problems found.

Hole filled for n=500K-700K. In addition to confirming the 3 previously found primes, one additional prime was found:
8847*2^514298-1

Results attached.

k=32767:

This one is currently reserved by Diep for n=5M-30M. Since it is so low weight I thought it would be interesting to see if there were any more primes up to n=1M.

Double-check complete to n=50K. 4 previously found primes confirmed. No problems found.

First-time check complete for n=50K-1M. No primes found.

This is a very low-weight k! :-)

Results attached.

As of May 9, the (near-)Woodall [I]k[/I]'s listed in [url]https://www.mersenneforum.org/showpost.php?p=550539&postcount=363[/url] have been completed to [I]n[/I]=475k. The following primes were found:[list][*]667071*2^450745-1[*]938237*2^455888-1[*]667071*2^465478-1[*]582833*2^467596-1[/list]

k=10005 has a large prime gap so I double-checked it for n=0-1.25M.

34 primes confirmed; no problems found.

Results attached.

I've completed the remaining RPS 9th and 10th Drive [I]k[/I]'s with missing ranges from [I]n[/I]=350k to 375k. 21 primes were found, which are attached. Of the laptops mentioned in my last post on this set of [I]k[/I]'s, the Kubuntu Focus has been dedicated to GIMPS and is not working on this project, but the older laptop is actively working on this. The expected finish date is currently October, though I may dedicate some additional cycles to it once it gets closer to the finish line.

Back about 13-14 years ago Beyond (SB2) did a lot of searching on many low-weight k's. Although he gave details about what he was working on his details were frequently incomplete and he did not complete a lot of what he was sieving. One of the main issues is that we never knew where he started his searches. Sometimes they started at n=1, other times at n=300K, many times we never knew. I found these 4 main status posts of his:

post link / date
[URL]https://www.mersenneforum.org/showpost.php?p=100670[/URL] 2007-03-13
[URL]https://www.mersenneforum.org/showpost.php?p=101240[/URL] 2007-03-18
[URL]https://www.mersenneforum.org/showpost.php?p=110378[/URL] 2007-07-14
[URL]https://www.mersenneforum.org/showpost.php?p=153292[/URL] 2008-12-14

I took all of the k's found in those posts and removed some k's that were already checked by others since then. This left 98 k's all of which are shown below.

All 98 k's have been double-checked by me for n=1 to 1M. 5 missing primes were found for ranges that he said were done and 13 new primes were found for ranges that he sieved but never searched or failed to report.

Missing primes:
9613*2^734907-1
124679*2^12-1
239857*2^161901-1
334147*2^103669-1
36231101*2^34482-1

New primes:
273613*2^397407-1
313979*2^243184-1
412843*2^107943-1
412843*2^142983-1
412843*2^377223-1
434729*2^941408-1
451049*2^281984-1
464909*2^136620-1
469933*2^330667-1
473279*2^103840-1
700477*2^492061-1
783073*2^402635-1
1034503*2^174283-1

All k's searched with all primes listed below.
[B][COLOR=Red]Red = missing[/COLOR][/B]
[COLOR=Red][COLOR=Black][B]Bold = new[/B][/COLOR][/COLOR]
[code]
k / n-primes
2039: 24, 64, 319920, 841312
2171: 14, 202, 1594, 6754, 101974
2279: 900, 6572, 23772, 35252
2591: 134, 326, 830, 2174, 17270
2741: 50, 138, 1610, 1850, 3954, 10202, 23178, 84858, 468594, 737586
2759: 388, 405820, 569812
2797: 21, 6093, 7389, 9861
2983: 11, 671, 767, 2891, 771455
3343: 483, 1299
3389: 92, 848, 2684, 23864, 209996, 504584, 749828
3539: 12, 192, 4236, 869868
3817: 40381, 161197, 218917
3821: 70, 15814, 439006
3823: 7, 287, 335, 599, 1015, 2407, 384223
4327: 13, 37, 133, 1189
4597: 17, 37, 161, 221, 721, 112009, 799781
4813: 515, 11379, 13155
5077: 117, 333, 490629, 881829
5099: 184, 1216, 9936, 19072, 27648
5323: 27, 83, 8651, 107643
6119: 4544, 678080
7331: 34, 87154, 704842
9179: 4120, 10408, 41680
9613: 1479, 34239, 464439, 636159, [B][COLOR=Red]734907[/COLOR][/B]
9913: 1187, 8387
10949: 1492, 1516
11519: 164444
12791: 2818, 29398, 308026
17077: 677
19919: 888, 2688, 8460, 170820
21547: 1, 121, 2233, 16001, 310625, 936473
21673: 7, 775, 129175
22183: 727, 2471, 773447
24067: 1753, 2069, 2881, 700073
24671: 1290, 2250, 11698, 108906, 382066
25229: 238652
30727: 1121, 3353, 15929
43541: 507098, 850394
48973: 43, 115, 259, 32119
65279: 1640, 4952, 66632, 625704
70079: 1428
106303: 1939
114613: 211, 1687, 461551
117037: 1029
124679: [B][COLOR=red]12[/COLOR][/B], 948, 2460, 10308, 497244, 769908
132217: 15281
171607: 21117, 73245
179743: 99731
192509: 52
236377: 19693, 111973, 178285, 824773
239857: 33, 7233, [B][COLOR=red]161901[/COLOR][/B]
248047: 11613
249127: 5189, 31301
273613: 39, 87, 159, [B]397407[/B]
278713: 17159, 57719
279731: 1782
284579: 56, 4736
294907: 545, 203921, 221409
306251: 22, 29038, 37462
313979: 4, 184, [B]243184[/B]
320107: 41, 257, 8249
323563: 243, 9099
332159: 4416
334147: 37, 7981, [B][COLOR=red]103669[/COLOR][/B]
334331: 78, 294, 7998
362737: 153
370421: 201442
384029: 24
403993: 379
412843: 15, 16047, [B]107943, 142983, 377223[/B]
434729: 8, 20, 80, [B]941408[/B]
441907: 837, 42861
451049: 7748, 22784, [B]281984[/B]
464353: 191, 527, 1919, 371279
464909: 73740, [B]136620[/B]
465187: 3949, 16441
469933: 51, [B]330667[/B]
473279: 160, 92032, [B]103840[/B]
492787: 553, 1865, 6961, 58849
516569: 20
648433: 75
700477: 973, [B]492061[/B]
780427:
783073: 11, [B]402635[/B]
808477: 11673
842711: 226, 3034, 3754
844559:
893887: 257
1005449: 204, 41100
1034503: 43, 5971, [B]174283[/B]
1210421: 28674, 655938
10013593: 3847
10108837: 33657
10247561: 730
10284899: 868, 2596
10346561: 32714
36231101: [B][COLOR=red]34482[/COLOR][/B], 469002
37616317:
[/code] Karsten, this cleans up a former mess of k's, many of which need to be entered into the Wiki but couldn't because their accuracy was in question. I will slowly work on getting them all into the Wiki going in descending order from the largest k. Larger k's have less history and so are easier to enter. I will be referencing Beyond's work, my own current and former work, the old low-weight stats page done by Joss, and other misc. work that has popped up by other people or projects that I find from searches on the k's in this forum including work done by NPLB for k=2000-3000.

Attached are primes and residues from the effort.

Gary

The (near-)Woodall [I]k[/I]'s listed in [url]https://www.mersenneforum.org/showpost.php?p=550539&postcount=363[/url] have been completed to [I]n[/I]=500k. The following primes were found:[list][*]197673*2^475399-1[*]1467763*2^480087-1[*]665127*2^483394-1[*]1268979*2^485869-1[*]1268979*2^487073-1[*]665127*2^488224-1[*]938237*2^491576-1[*]1183953*2^496687-1[/list]
I want to reiterate my reservation of [I]k[/I]=454483, which will resume work from this point.