[QUOTE=henryzz;310916]Why Q? P is what is most commonly used on the forum for 10^15. It stands for Peta-.[/QUOTE]

I didn't know. I chose Q for quadrillion. I'll change it to P next time.

[QUOTE=Prime95;310919]I didn't know. I chose Q for quadrillion. I'll change it to P next time.[/QUOTE]
It doesn't happen that often:smile: but it would confuse some people. It is also being consistent since you are using G instead of B for Billion.

[QUOTE=henryzz;310920]It is also being consistent since you are using G instead of B for Billion.[/QUOTE]

I probably would have used B for billion except that I remembered Oliver used G in his output line count of candidates.

I hope the Brits don't complain about using T for 10^12 (trillion).

[QUOTE=Prime95;310921]I probably would have used B for billion except that I remembered Oliver used G in his output line count of candidates.

I hope the Brits don't complain about using T for 10^12 (trillion).[/QUOTE]
Again T is standard on the forum. It also stands for Tera-. G is Giga-. We have been using [url]http://en.wikipedia.org/wiki/SI_prefix[/url]
Also I am a Brit so you haven't upset at least 1:smile:

[QUOTE=Prime95;310915]Here is the next version of mmff.

The big change is how Fermat factoring is handled. Instead of specifying the Fermat number you want to factor, you now specify the factors (k*2^N+1) to try. This is compatible with the way fermatsearch.org has always done things. If you want to try to find a factor of F33 you'll need to choose an N >= 35 (and you might find a factor of F32, F31, F30, ... instead).

If you reserved a Fermat range from fermatsearch.org, please retest it with the new executable (but add 2 to the exponent in worktodo.txt). For example, if you tested a k range for F30 (FermatFactor=30,1.5e15,1.6e15) using the previous mmff, Luigi had you reserve a range using N=32. With this executable, retest the range using FermatFactor=32,1.5e15,1.6e15.

The good news? This change will double our Fermat factoring throughput.
[/QUOTE]
Cool beans!!

Let's see if it will find this factor...
m=28 25709319373 N=36 05 Feb 1997 T.Taura..
Indeed,
[COLOR=green]F28 has a factor: 1766730974551267606529[/COLOR]
Test passed.

[COLOR=green]F86 has a factor: 6195449970597928748332522715641578258433
Test passed, too! (88,20e12,21e12)[/COLOR]
______________

[I]Bonus feature[/I]: MMFF = Merck Molecular Force Field (fairly well used in protein modeling area). The almost single author of it is Thomas A. Halgren. [I]J. Comp. Chem,[/I] [URL="http://onlinelibrary.wiley.com/doi/10.1002/(SICI)1096-987X(199604)17:5/6%3C%3E1.0.CO;2-C/issuetoc"]1996 April issue[/URL] is almost entirely written by him.

hi,
an executable for win7 sp1 64bit would be nice

BTW, save files will not be compatible (since mfaktc writes the version number to the save file). However, if you use the -nocheck command line argument you may be able to upgrade in the middle of a range. This bypasses the checksum test when reading a save file. This "feature" is only lightly tested, so make backups first.

Windows binaries
 

mmff .21 beta Windows 32 & 64 bit executables attached.

stupid question again.
my mmff_0.20 reservation was
[code]
FermatFactor=46,300e12,325e12
[/code]

should now be
[code]
FermatFactor=48,300e12,325e12
[/code]
right?

(will run the 300e12-375e12 range when I get back)

[QUOTE=flashjh;310928]mmff .21 beta Windows 32 & 64 bit executables attached.[/QUOTE]

hi,
the win64 executable works fine for fermatnumbers.
all other i did not tried out.

[QUOTE=Prime95;310926]BTW, save files will not be compatible (since mfaktc writes the version number to the save file). However, if you use the -nocheck command line argument you may be able to upgrade in the middle of a range. This bypasses the checksum test when reading a save file. This "feature" is only lightly tested, so make backups first.[/QUOTE]
The -nocheck didn't work for me; I'll switch when my current assignment is done.

My first MM89 reservation is done:
[CODE]no factor for MM89 from 2^128 to 2^130 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^130 to 2^131 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^131 to 2^132 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^132 to 2^133 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^133 to 2^134 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^134 to 2^135 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^135 to 2^136 [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM89 from 2^136 to 2^137 (k range: 0 to 100000000000000) [mmff 0.20mmff mfaktc_barrett140_M89gs]
[/CODE]

Next I'll take
MMFactor=89,2e14,3e14

After new version came out, then on the [URL="http://www.moregimps.it/doublemersennes/download.php"]download[/URL] page, the last paragraph should read: "The second example looks for factors of F[SUB][COLOR=Red]29[/COLOR][/SUB] between 54600..."

[QUOTE=firejuggler;310885]sorry if it's dumb but when i see
[code]
61 132,732,600,000,000 135,795,660,000,000 Andrea (TF) 2012-09-01 2012-09-01
61 3,573,570,000,000,000 5,000,000,000,000,000 Rocke Verser 2012-09-09
[/code]

I wonder, what happened to the k between 135,795,660,000,000 and 3,573,570,000,000,000 ?[/QUOTE]

I am updating the previous results from Tony Forbes' site, and I've not yet completed the update.

I thought that the presence of an actual live list of reserved and completed ranges was more important that a coherence in the already done ranges.

Luigi

Finished for MM89. Partially finished for MM107.
Check in:

[CODE][Mon Sep 10 11:17:23 2012]
no factor for MM89 from 2^136 to 2^137 (k range: 100000000000000 to 200000000000000) [mmff 0.20mmff mfaktc_barrett140_M89gs]
[Mon Sep 10 19:31:57 2012]
no factor for MM89 from 2^137 to 2^138 (k range: 0 to 200000000000000) [mmff 0.20mmff mfaktc_barrett140_M89gs]
no factor for MM107 from 2^154 to 2^155 (k range: 100000000000000 to 200000000000000) [mmff 0.20mmff mfaktc_barrett160_M107gs]
[/CODE]
Still going on for the second bitlevel (155 to 156) of MM107 (up to pledged k of 2e14). ETA ~9h, tomorrow morning Thai time I will report that factor which still managed to stay hidden, but he is now cornered and soon I am going to find him.... :razz:

Still don't get it why it says 2000... as end value on the first row and 0 as start value on the second, both should be the last_k for 136 bits factor and the first_k for 137 bit factor (difference of 1 between them).

Same for MM107, instead of 2000... it should be last_k value for 154 bits, which is 140737488354539, and the next line when it will be written tomorrow morning would have to say 140737488354540. Whatever...

I dld the newer version, I will look immediately for a Fermat assignment, like for 8-12 hours or so, let me play a bit with it before.

Small issue: using win7_64 and Jerry's built, it crashes after the assignment is done and have to be stopped from task manager (fortunately, after it saves the work in result.txt file :smile:)
Also, some cosmetic: when i type strange bitlevels in a Fermat assignment it says that the bitlevel is strange in a "M26 assignment" (instead of "F26 assignment").

[edit: the crash was there also for the old version. In the intermediary bitlevels (when it splits the assignments in more bitlevels) it does not crash, only at the end]

[edit2: another cosmetic: the "no factor for k*2^88+1 from 2^114 to 2^128" is quite confuse. I am running "FermatFactor=88,1e8,21e12" trying to reproduce Batalov's factor, and to measure the speed for different things, with two cards (1e8 was the lowest value for which I did not get some cuss from the program) and certainly I wasn't expecting the "no factor for k*2^88+1" :razz: message]

[QUOTE=flashjh;310935]The -nocheck didn't work for me; I'll switch when my current assignment is done.[/QUOTE]
You may need to rename the checkpoint file too. If you had a F86.ckp before, he might now be looking for an F88.ckp :smile:

[QUOTE=LaurV;310976]You may need to rename the checkpoint file too. If you had a F86.ckp before, he might now be looking for an F88.ckp :smile:[/QUOTE]

Well, you're better off discarding old Fermat save files as you really should be redoing the range from scratch.

Also, M89.ckp, M107.ckp, blabla, it can be resumed with -nocheck, if you delete the "mmff" thing inside (after the version string). The older version has mmff, the newer has not, and that mismatches the reading of the last class. I tried now and resumed ok.

@Prime95: Clearly! I know that. I was talking about resuming (i.e. -nocheck switch works for me, Jerry's built).

and by the way, menatime I found Batalov's factor:

[CODE][Mon Sep 10 21:15:59 2012]
F86 has a factor: 6195449970597928748332522715641578258433 [TF:132:133:mmff 0.21 mfaktc_barrett140_F64_95gs]
[Mon Sep 10 21:23:00 2012]
found 1 factor for k*2^88+1 from 2^132 to 2^133 (k range: 20T to 21T) [mmff 0.21 mfaktc_barrett140_F64_95gs]
[/CODE]

The crash happens to me when I use control+C to stop the program. It seems to only happen when it's running faster classes. I haven't had a range finish yet, so I can't confirm the other crash.

Can anyone running the Linux version tell me if it crashes at the end of a run or when breaking the run?

Still need some time to clarify in my head what's going on with those fermat numbers...

Up to then, reserving MMFactor=61,50e14,75e14

I split it in 25 assignments for each 1e14 range, 13 of them in one card and 12 in another card, after current MM107 will be finished (ETA 7:30h). Each slice of 25 takes 9:30 hours, rounded in addition. So the whole ETA would be about 125 hours (9:30*13 in one card, 7:30+9:30*12 in the second).

My DC testing will have a lot to suffer from it, and this is only George's fault! He gave us new toys so we now play like children...

(edit: otoh, my aliquots will see some faster progress, as now they use the CPU all of it. It is no way to run Prime95 with mmff, for thermal reasons, even water cooling can't suffice it! I reduced the CPU clock to stock 3.4G and switched to aliquots)

MM61 reservation page: can it be split in two? (new/old, or actual/past, whatever). Below 1.3e15 and over 1.3e15. It is too long and inconvenient to scroll, and depending on net speed, it may take ages to load. And periodically move the smallest range from the actual to old, when is done. And keep the idea for the other pages too... :smile: they will also get longer and longer, if that factor of mine still manage to hide itself and I don't find it ...

[QUOTE=LaurV;310995]MM61 reservation page: can it be split in two? (new/old, or actual/past, whatever). Below 1.3e15 and over 1.3e15. It is too long and inconvenient to scroll, and depending on net speed, it may take ages to load. And periodically move the smallest range from the actual to old, when is done. And keep the idea for the other pages too... :smile: they will also get longer and longer, if that factor of mine still manage to hide itself and I don't find it ...[/QUOTE]

I published a page with the actual reservations not yet concluded.

Yes, MM61 is too lengthy, I have some idea to apply, now that I finished the data insertion.
I'm working on Double Mersennes while I'm at work, so I can't be 100% reactive on requests, sorry...

Luigi.

Testing for best GPUSievePrimes for fermat numbers with GTX 460 on a Q9450 (Yorksfield) 2.66 Ghz:

[CODE] Best
GPUSievePrimes Speed
FermatFactor=30,x,x 70k-140k 883.51 M/s
FermatFactor=40,x,x 125k-160k 760.80 M/s
FermatFactor=50,x,x 220k-270k 385.49 M/s
FermatFactor=60,x,x 300k-310k 316.46 M/s
FermatFactor=70,x,x 280k-360k 289.77 M/s
FermatFactor=80,x,x 400k-580k 173.17 M/s
FermatFactor=90,x,x 450k-500k 148.47 M/s
FermatFactor=100,x,x 400k-570k 141.48 M/s
FermatFactor=125,x,x 630k-840k 78.01 M/s
FermatFactor=150,x,x 650k-1075k 61.62 M/s[/CODE]

[QUOTE=LaurV;310983]...and by the way, meantime I found [STRIKE]Batalov[/STRIKE] [B]Michael Dangler[/B]'s factor:

[CODE][Mon Sep 10 21:15:59 2012]
F86 has a factor: 6195449970597928748332522715641578258433 [TF:132:133:mmff 0.21 mfaktc_barrett140_F64_95gs]
[Mon Sep 10 21:23:00 2012]
found 1 factor for k*2^88+1 from 2^132 to 2^133 (k range: 20T to 21T) [mmff 0.21 mfaktc_barrett140_F64_95gs]
[/CODE][/QUOTE]
I only wrote that this test passed - it is of course a known (recent) factor by M. Dangler & Rodenkirch.

[QUOTE=flashjh;310992]The crash happens to me when I use control+C to stop the program. It seems to only happen when it's running faster classes. I haven't had a range finish yet, so I can't confirm the other crash.

Can anyone running the Linux version tell me if it crashes at the end of a run or when breaking the run?[/QUOTE]
Both Linux and Windows vesions "crash" after Ctrl-C. (More specifically, the program correctly waits for the end of current class, then reports SEGV on linux or pops a dialog on Win7.)

My GTX560 took about 2.5 hours to complete the range N=30, k=650e12-700e12 :D

[CODE]no factor for k*2^30+1 from 2^79 to 2^80 (k range: 650T to 700T) [mmff 0.21 mfaktc_barrett89_F0_31gs][/CODE] Benchmark: (mmff version 0.21 for Windows 7 64-bit)

[ATTACH]8574[/ATTACH]

I tested different values for GPUSievePrimes but the default one was the best.

Fabrice

Likewise
[CODE]no factor for MM107 from 2^140 to 2^141 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^141 to 2^144 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^144 to 2^148 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^148 to 2^149 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^149 to 2^150 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^150 to 2^151 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^151 to 2^152 [mmff 0.20mmff mfaktc_barrett152_M107gs]
no factor for MM107 from 2^152 to 2^153 [mmff 0.20mmff mfaktc_barrett160_M107gs]
no factor for MM107 from 2^153 to 2^154 [mmff 0.20mmff mfaktc_barrett160_M107gs]
no factor for MM107 from 2^154 to 2^155 (k range: 0 to 100000000000000) [mmff 0.20mmff mfaktc_barrett160_M107gs][/CODE]
and I'll take
MMFactor=107,2e14,3e14

No one found a factor yet?

Are we there yet? Are we there yet?
/smack!/

:smile:

[QUOTE=Batalov;311084]No one found a factor yet?

Are we there yet? Are we there yet?[/QUOTE]Everyone is too scared to find a factor because of all the myriad of webpages that would need to be updated.

[size=1]Also, no one wants to make [url=http://www.mersenneforum.org/showthread.php?p=147152#post147152]this guy[/url] look bad. It is just not manners, isn't it?[/size]

[QUOTE=retina;311085][SIZE=1]Also, no one wants to make [URL="http://www.mersenneforum.org/showthread.php?p=147152#post147152"]this guy[/URL] look bad. It is just not manners, isn't it?[/SIZE][/QUOTE]
I just voted "yes" to that pool :razz: My guts tell me that MM127 is prime :smile:.

Ok. Now seriously:
- our chunk of MM107 finished to the pledged 200T. The factor still managed to stay hidden. When we reach home tonight we will post the result.txt, which is not accessible now.
- the reservation on MM61 which we took last night (see few posts above) still did not appear in the table. We think that lady Night is traveling to different parts of the world now, and there should be dark and people sleeping there, but we hope they will wake up and do the right modification in that table, so we do not step on someone else' toes.
- from the MM61 reservation in cause (5000T to 7500T), the first 100T (i.e. 5000T to 5100T) already done, no factor found. Again, we will post logs when we reach home.

I'll have to admit - I keep coming back to this thread hoping to see that someone's discovered a factor of MM127. :D

I am more keen to see some new Fermat factors, and I am pretty sure that we could see one by the end of the week.

[QUOTE=rcv;310888]I don't know where 3.573e15 came from. Perhaps others have done the range, but it isn't reported. But since I have the first assignment, like frmky above, I am doing "and below" to the limits of mmff-0.20:
MMFactor=61,96,5e15[/QUOTE]
Now that the reservations page shows some numbers below 3.57357e15 are reserved to other parties, I think it would be poaching to test those reserved ranges. But after my initial assignment is complete, I will endeavor to double-check everything from 96 bits to the start of my first mmff range, EXCEPT the assignments that are still outstanding:
[CODE]MMFactor=61,96,105 [done]
MMFactor=61,105,106
MMFactor=61,106,107
MMFactor=61,107,108
MMFactor=61,108,109
MMFactor=61,109,110
MMFactor=61,110,111
MMFactor=61,111,112
MMFactor=61,112,1303842540000000
MMFactor=61,1327326000000000,1347746400000000
MMFactor=61,1349788440000000,1633632000000000
MMFactor=61,1654052400000000,1678556880000000
MMFactor=61,1684683000000000,2920117200000000
MMFactor=61,2960958000000000,3063060000000000
MMFactor=61,3165162000000000,3308104800000000[/CODE]

[CODE]no factor for MM89 from 2^137 to 2^138 (k range: 200T to 300T) [mmff 0.21 mfaktc_barrett140_M89gs]
no factor for MM89 from 2^138 to 2^139 (k range: 200T to 300T) [mmff 0.21 mfaktc_barrett140_M89gs][/CODE]

I'll continue with
MMFactor=89,3e14,5e14

[QUOTE=ATH;311006]Testing for best GPUSievePrimes for fermat numbers with GTX 460 on a Q9450 (Yorksfield) 2.66 Ghz:

[CODE] Best
GPUSievePrimes Speed
FermatFactor=30,x,x 70k-140k 883.51 M/s
FermatFactor=40,x,x 125k-160k 760.80 M/s
FermatFactor=50,x,x 220k-270k 385.49 M/s
FermatFactor=60,x,x 300k-310k 316.46 M/s
FermatFactor=70,x,x 280k-360k 289.77 M/s
FermatFactor=80,x,x 400k-580k 173.17 M/s
FermatFactor=90,x,x 450k-500k 148.47 M/s
FermatFactor=100,x,x 400k-570k 141.48 M/s
FermatFactor=125,x,x 630k-840k 78.01 M/s
FermatFactor=150,x,x 650k-1075k 61.62 M/s
[/CODE]
[/QUOTE]

Waiting for similar tables on other GPU models, to optimize ranges.

Please, :help:

Luigi

[QUOTE=LaurV;311092]When we reach home tonight we will post the result.txt, which is not accessible now[/QUOTE]

as promised:
(finished for MM107, I saw it is already marked on the list as finished, and partially done for the mm61, split in two semi-ranges, each semi-range split in - now 11 remaining - parts, each part would take about 9 hours and half on a gtx580 a bit overclocked and water cooled :D).

[CODE][Tue Sep 11 07:35:46 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5000T to 5100T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Tue Sep 11 05:44:13 2012]
no factor for MM107 from 2^155 to 2^156 (k range: 0T to 200T) [mmff 0.21 mfaktc_barrett160_M107gs]
[Tue Sep 11 15:25:08 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6300T to 6400T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Tue Sep 11 19:24:53 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5100T to 5200T) [mmff 0.21 mfaktc_barrett120_M61gs]
[/CODE]
With flashjh's settings (816000,128,16) this scores between 611M/s and 618M/s and it seems to be close to the optimum for the card/expo/range combination. The computer can't be used (almost, but nobody need it right now).

[QUOTE=LaurV;311125]The computer can't be used (almost, but nobody need it right now).[/QUOTE]
(forgot to mention that part :razz:)

This project could use a boinc setup. [Not volunteering!]

Is "FermatFactor=127,3700000e9,4000000e9" the correct worktodo entry for the reservation of: 127 3,700,000,000,000,000 4,000,000,000,000,000 ?
I'm a bit confused with your k, N, and things...

;)

Yes or

FermatFactor=127,3700e12,4000e12 ;)

It looks like this is still available so I would like to reserve it.

[CODE]MMFactor=127,2.5e15,2.8e15[/CODE]

And it will take 8 days on my GTX 560.

[QUOTE=MrHappy;311176]Is "FermatFactor=127,3700000e9,4000000e9" the correct worktodo entry for the reservation of: 127 3,700,000,000,000,000 4,000,000,000,000,000 ?
I'm a bit confused with your k, N, and things...

;)[/QUOTE]

[QUOTE=Jatheski;311179]Yes or

FermatFactor=127,3700e12,4000e12 ;)[/QUOTE]

I thought you had to add '2' now?: FermatFactor=[B]129[/B],3700e12,4000e12

[QUOTE=Prime95;310915]Here is the next version of mmff.
If you reserved a Fermat range from fermatsearch.org, please retest it with the new executable ([B]but add 2 to the exponent in worktodo.txt[/B]). For example, if you tested a k range for F30 (FermatFactor=30,1.5e15,1.6e15) using the previous mmff, Luigi had you reserve a range using N=32. With this executable, retest the range using FermatFactor=32,1.5e15,1.6e15.[/QUOTE]

[QUOTE=flashjh;311222]I thought you had to add '2' now?: FermatFactor=[B]129[/B],3700e12,4000e12
[quote=prime95]Here is the next version of mmff.
If you reserved a Fermat range from fermatsearch.org, please retest it with the new executable ([B]but add 2 to the exponent in worktodo.txt[/B]). For example, if you tested a k range for F30 (FermatFactor=30,1.5e15,1.6e15) using the previous mmff, Luigi had you reserve a range using N=32. With this executable, retest the range using FermatFactor=32,1.5e15,1.6e15.[/quote][/QUOTE]
My understanding is as follows:

MMFactor work.run under mmff-0.20 is OK.
FermatFactor work run under mmff-0.20 should be discarded.

The interpretation of the FermatFactor line has changed:

[CODE] mmff-0.20: FermatFactor=<N-2>,low,high[/CODE][CODE] mmff-0.21: FermatFactor=N,low,high[/CODE]If you get new reservations, [B]the N and the k range you receive from Luigi should be plugged directly into the FermatFactor line[/B]. Simple. Straightforward. Ergonomic.

If you got reservations while using the old (0.20) version of mmff, you had to tweak the N value, by subtracting 2. To move that reservation to the new (0.21) version of mmff, you must undo your tweak.

"Reservation" means "N range, k range". The "+2" is there already. The correct one is 127....
This is what the site use, and the new software version. This can find factors of Fm with m<=125. Usually N>=m+2.

The older version of the program would have had to reserve the same range as "125....".

So, please do not use 129!

[CODE]no factor for MM107 from 2^155 to 2^156 (k range: 200T to 300T) [mmff 0.21 mfaktc_barrett160_M107gs]
no factor for MM107 from 2^156 to 2^157 (k range: 0T to 300T) [mmff 0.21 mfaktc_barrett160_M107gs]
[/CODE]

I hope it's not gauche to take all of the currently available ranges listed at [url]http://www.fermatsearch.org/wanted.html[/url], namely

[STRIKE]FermatFactor=31,600e12,700e12[/STRIKE]
FermatFactor=32,600e12,650e12
FermatFactor=34,650e12,700e12
FermatFactor=39,600e12,700e12
FermatFactor=44,400e12,450e12

Timing indicates that these will take less than 12 hours total on a GTX 480.

Edit: I see at [url]http://www.fermatsearch.org/stat/running.php[/url] that the first of those is already taken. I'll do the other four.

FermatFactor=31,600e12,700e12 is mine :D (my precious)

[url]http://www.fermatsearch.org/stat/running.php[/url]

[QUOTE=MrHappy;311176]Is "FermatFactor=127,3700000e9,4000000e9" the correct worktodo entry for the reservation of: 127 3,700,000,000,000,000 4,000,000,000,000,000 ?
I'm a bit confused with your k, N, and things...

;)[/QUOTE]

What are you doing that far? :smile:

The actual limit for 127 is 30e12...

Luigi

[QUOTE=RichD;311203]It looks like this is still available so I would like to reserve it.

[CODE]MMFactor=127,2.5e15,2.8e15[/CODE]

And it will take 8 days on my GTX 560.[/QUOTE]

Assigned to RichD :showoff:

Luigi

[QUOTE=frmky;311243]I hope it's not gauche to take all of the currently available ranges listed at [url]http://www.fermatsearch.org/wanted.html[/url], namely

[STRIKE]FermatFactor=31,600e12,700e12[/STRIKE]
FermatFactor=32,600e12,650e12
FermatFactor=34,650e12,700e12
FermatFactor=39,600e12,700e12
FermatFactor=44,400e12,450e12

Timing indicates that these will take less than 12 hours total on a GTX 480.

Edit: I see at [url]http://www.fermatsearch.org/stat/running.php[/url] that the first of those is already taken. I'll do the other four.[/QUOTE]

Greg, the "most wanted" list was designed for CPUs. I'm reworking it to offer best choices to all.

Meanwhile I reserved those ranges to you. You can extend them at will if you prefer going deep into k (notice that N=31 will start at 10[sup]15[/sup], while N=32 at 2.5*10[sup]15[/sup]. Just let me know about extensions before releasing it, and I will update them online.

Luigi

[QUOTE=Jatheski;311245]FermatFactor=31,600e12,700e12 is mine :D (my precious)

[url]http://www.fermatsearch.org/stat/running.php[/url][/QUOTE]

Fabrice? Is that you? N=31 has been reserved up to k=10^15...

Luigi

Yep it's me ! :D

[QUOTE=ET_;311248]Greg, the "most wanted" list was designed for CPUs. I'm reworking it to offer best choices to all. [/QUOTE]
I understand now. Sorry about that! I'll keep all four but extend the last one to
FermatFactor=44,400e12,700e12

[QUOTE=Jatheski;311250]Yep it's me ! :D[/QUOTE]

You told me N=31, k=600e12-1000e-12, not N=31, k=600e12-700e-12.

Luigi

[QUOTE=ET_;311252]You told me N=31, k=600e12-1000e-12, not N=31, k=600e12-700e-12.

Luigi[/QUOTE]

I wanted to warn frmky that the range N=31, k=600e12-700e-12 was a part of my reservation (N=31, k=600e12-1000e-12).
Sorry for the misunderstanding.

Fabrice

Edit: I will reserve my next ranges on this thread.

Range N=30, k=700e12-1000e12 completed

Luigi, it seems you forgot to update the status of the range N=49, k=125e12-150e12 :whistle:

[CODE]no factor for k*2^49+1 from 2^95 to 2^96 (k range: 125T to 150T) [mmff 0.21 mfaktc_barrett96_F32_63gs]
no factor for k*2^49+1 from 2^96 to 2^97 (k range: 125T to 150T) [mmff 0.21 mfaktc_barrett108_F32_63gs]
no factor for k*2^30+1 from 2^79 to 2^80 (k range: 700T to 1000T) [mmff 0.21 mfaktc_barrett89_F0_31gs][/CODE]Fabrice

checking in:

[CODE][Wed Sep 12 05:16:00 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5200T to 5300T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 03:16:24 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6400T to 6500T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 16:43:58 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5300T to 5400T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 14:28:18 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6500T to 6600T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Thu Sep 13 00:12:55 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6600T to 6700T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Thu Sep 13 02:43:39 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5400T to 5500T) [mmff 0.21 mfaktc_barrett120_M61gs]
[/CODE]

[QUOTE=Prime95;310677]Until Luigi gets his reservation system fully working, these ranges would take about 10 days each on a GTX 460:

...
MMFactor=127,3.7e15,4.0e15

etc.[/QUOTE]

127,3.7e15,4.0e15 is mine. :)

I think we can use the wiki to coordinate these assignments. For example, I've created a simple page here: [url]http://mersennewiki.org/index.php/Mmff/status[/url]

How does it look?

[QUOTE=ixfd64;311321]I think we can use the wiki to coordinate these assignments. For example, I've created a simple page here: [url]http://mersennewiki.org/index.php/Mmff/status[/url]

How does it look?[/QUOTE]

Is Luigi's page inadequate? [url]http://www.moregimps.it/doublemersennes/mm127.php[/url]

[QUOTE=LaurV;311312]checking in:

[CODE][Wed Sep 12 05:16:00 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5200T to 5300T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 03:16:24 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6400T to 6500T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 16:43:58 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5300T to 5400T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Wed Sep 12 14:28:18 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6500T to 6600T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Thu Sep 13 00:12:55 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 6600T to 6700T) [mmff 0.21 mfaktc_barrett120_M61gs]
[Thu Sep 13 02:43:39 2012]
no factor for MM61 from 2^114 to 2^115 (k range: 5400T to 5500T) [mmff 0.21 mfaktc_barrett120_M61gs]
[/CODE][/QUOTE]

Thank you LaurV.

May I ask you to post results only when the range is completed?
Otherwise I should add a "notes" field to the db to hold the actual status... :smile:

Luigi

[QUOTE=Prime95;311325]Is Luigi's page inadequate? [url]http://www.moregimps.it/doublemersennes/mm127.php[/url][/QUOTE]

Luigi's website is great, but I was thinking the wiki could be used as a backup - just in case he is too busy in real life or something.

Also, his website currently doesn't cover Fermat numbers (although Wilfrid Keller's website does).

[QUOTE=ixfd64;311321]I think we can use the wiki to coordinate these assignments. For example, I've created a simple page here: [url]http://mersennewiki.org/index.php/Mmff/status[/url]

How does it look?[/QUOTE]

It looks promising, but at the moment I use [URL="http://www.moregimps.it/doublemersennes/reserved.php"]this one[/URL]...

Luigi

P.S. Sorry, I read your post too late :smile: You are welcome, I just try to avoid too many checkins from the users. Unless you are ready t take charge of updating the whole wiki...

P.S. /2: Some posts ago I asked if a [COLOR="Red"]sub-forum related to Double Mersennes and Fermat factoring[/COLOR] could be useful... Mike told me that it is feasibe. We may create reservation and completion threads on both searches (heck, even on each MMp!), and leave this thread to discussions on the software. Should I post a poll, or we may gather votes right here?

www.doublemersennes.org
 

I just bought the domain and the hosting. I hope we will have a new server by next week.

Luigi :et_:

As one who has been an onlooker to this scene so far, it seems like a very good idea. The same thing has worked well for GPU72.

I'm still trying to learn enough about the whole field (MM & Fermat,) and the software to give it a try. This is also partly because I need to whittle down my worktodo load in TF and DC. I'm thinking that when I get there, I'll try it with the GTX 460, which is running CuLu presently.

One question: are the -st and -st2 switches supposed work to in mmff? I tried that on the GTX 570, but it crashed mmff.

[QUOTE=kladner;311340]One question: are the -st and -st2 switches supposed work to in mmff? I tried that on the GTX 570, but it crashed mmff.[/QUOTE]

I'm not surprised -st and -st2 crash. There is some validation going on when you do normal double-Mersenne and Femat factoring. One remainder from the first 2 GPUsieves in each class are verified with slow C code.

[QUOTE=Prime95;311390]I'm not surprised -st and -st2 crash. There is some validation going on when you do normal double-Mersenne and Femat factoring. One remainder from the first 2 GPUsieves in each class are verified with slow C code.[/QUOTE]

OK. Thanks, George! I have been looking for ways to find the flaw in this card (GTX 570), which keeps it from running CudaLucas. This doesn't seem to be it. Though I confess that I did not understand all of the explanation, I do appreciate the response.

I'm kinda late on this, but good job!
One question though, where and how do I grab ranges to test?
I assume the results can be submitted to mersenne.org manually, right ?

[QUOTE=Karl M Johnson;311453]I'm kinda late on this, but good job!
One question though, where and how do I grab ranges to test?
I assume the results can be submitted to mersenne.org manually, right ?[/QUOTE]

I don't think that such numbers apply for mersenne.org... t least te server would answer "result not needed", as the trial-factoring bitlevel is huge.

ATM you can check the ranges [URL="http://www.moregimps.it/doublemersennes"]here[/URL]. The green values on the Reservation page are free for you to reserve.

You may also visit [URL="http://www.mersenneforum.org/forumdisplay.php?f=99"]this[/URL] subthread and post your reservations (thank you, Mike! :bow:).

Luigi

[CODE]no factor for k*2^31+1 from 2^80 to 2^81 (k range: 600T to 1000T) [mmff 0.21 mfaktc_barrett89_F0_31gs][/CODE]I would like to reserve N=33, k=700e12-1000e12

Fabrice

Oh, right, forgot what we are really looking for:smile:

What does GPUSieveProcessSize parameter influence?
It's not documented in the ini.

Also, a feature request (for close/distant future): ability to select the device to work on.
For example, if monitor is connected to GPU0, which atm mmff uses by default, there WILL be lag.
However, if there's a second GPU1, then, if it's assigned all the work, we shall see no lag!
And be able to squeeze all the juice from it, again, without any undesirable desktop lag.
Sounds reasonable?

Already. The "cosmetic" part of the software (i.e. everything except the math itself) is based on mfakt*. If you would read carefully, some of us (me) are using already multiple cards. Try "-d 0", "-d 1", etc, same as mfaktc.

[edit: P.S.: I use 128 "workers" for the GPU's except for the one which is main video card, for which I switched to 16 and everything is much better. There is no big difference in performance, and the computer can be used ok for daily crap]

Oooooh, nice.
Thanks!

Version 0.22

A minor upgrade.

1) Prettied up the way k ranges are output. In v0.21 both the bit range and k range were output and what was actually tested was the intersection of the two --- pretty confusing. Now the range of k's tested is clearly output and the bit size of the factors tested is output in parentheses.
2) Save files are incompatible. The k range is now output to the save file. This prevents an accidental error where editing the k range in worktodo.txt could use an old incompatible save file.

Obviously, finish your current range before doing any upgrading.

Sources:

A question: should mmff crash when finishing assignment and waiting for another class to write checkpoint(when Control+C is pressed) ?
Or is it just me?

[QUOTE=Karl M Johnson;311786]A question: should mmff crash when finishing assignment and waiting for another class to write checkpoint(when Control+C is pressed) ?
Or is it just me?[/QUOTE]

We [URL="http://www.mersenneforum.org/showthread.php?p=311013#post311013"]know it's an issue[/URL]. I haven't had time to look at the code yet.

[QUOTE=Prime95;311784]Version 0.22

A minor upgrade.

1) Prettied up the way k ranges are output. In v0.21 both the bit range and k range were output and what was actually tested was the intersection of the two --- pretty confusing. Now the range of k's tested is clearly output and the bit size of the factors tested is output in parentheses.
2) Save files are incompatible. The k range is now output to the save file. This prevents an accidental error where editing the k range in worktodo.txt could use an old incompatible save file.

Obviously, finish your current range before doing any upgrading.

Sources:[/QUOTE]
I got it compiled for Windows, but my tests didn't go well:

- It's about 3x slower than .20/.21
- I can't get it to resume from a checkpoint (always starts over)
- v.20 with FermatFactor=31,75,76 found a factor at class 243, v.22 does not.

[QUOTE=flashjh;311806]
- I can't get it to resume from a checkpoint (always starts over)
- v.20 with FermatFactor=31,75,76 found a factor at class 243, v.22 does not.[/QUOTE]

Oops, I did my testing of checkpoint files with the -nocheck command line argument. There is a checksum problem. I'll post a 0.23 tomorrow with the fix.

The meaning of the FermatFactor line changed from 0.20 to 0.21. Try one of these lines:
[CODE]FermatFactor=33,5460000000000,5470000000000
FermatFactor=36,25709e6,25710e6
FermatFactor=54,81900e9,81911e9
FermatFactor=101,3334e9,3335e9
FermatFactor=135,88000000000,88100000000
[/CODE]

What were you testing that had a 3x slowdown?

[QUOTE=Prime95;311814]Oops, I did my testing of checkpoint files with the -nocheck command line argument. There is a checksum problem. I'll post a 0.23 tomorrow with the fix.

The meaning of the FermatFactor line changed from 0.20 to 0.21. Try one of these lines:
[CODE]FermatFactor=33,5460000000000,5470000000000
FermatFactor=36,25709e6,25710e6
FermatFactor=54,81900e9,81911e9
FermatFactor=101,3334e9,3335e9
FermatFactor=135,88000000000,88100000000
[/CODE][/QUOTE]
Yes, FermatFactor=33 fixed the no factor problem and the speed. :blush:

Thanks, I'll get v.23 compiled when I see it.

[QUOTE=flashjh;311815]Thanks, I'll get v.23 compiled when I see it.[/QUOTE]

That's one silly mistake for each of us. It's late, I'll retest and upload tomorrow to reduce the chance of making another silly mistake.

Multiple instances and combinations of programs
 

Well I have a couple of questions:

1. When running mmff 0.21 64-bit on windows 7 it appears as if the program is using GPU resources but I see very little CPU activity. If I compare with mfaktc each instance uses one CPU core. Is that correct or am I doing something wrong?

2. I am running prime95 in parallell to mmff. It seems as if I can use all cores for prime95 without affecting the performance of mmff. Is that correct?

3. I tried to run multiple instances of mmff and also a combination of mmff and mfaktc, but one instance of mmff seems to use almost all the capacity of the GPU, so it seems to be useless to run more then one instance or one program at the same time. Is that a correct observation?

[QUOTE=aketilander;311825]1. When running mmff 0.21 64-bit on windows 7 it appears as if the program is using GPU resources but I see very little CPU activity. If I compare with mfaktc each instance uses one CPU core. Is that correct or am I doing something wrong?[/QUOTE]
mmff does the sieving on the GPU (src/gpusieve.cu).

[QUOTE=Prime95;310671]This program uses GPU sieving so is an excellent choice for computers with decent nvidia GPUs but underpowered CPUs.[/QUOTE]

[QUOTE=aketilander;311825]

2. I am running prime95 in parallell to mmff. It seems as if I can use all cores for prime95 without affecting the performance of mmff. Is that correct?[/quote]

Yes.

[quote]3. I tried to run multiple instances of mmff and also a combination of mmff and mfaktc, but one instance of mmff seems to use almost all the capacity of the GPU, so it seems to be useless to run more then one instance or one program at the same time. Is that a correct observation?[/QUOTE]

Yes.

Version 0.23

A minor upgrade.

1) Prettied up the way k ranges are output. In v0.21 both the bit range and k range were output and what was actually tested was the intersection of the two --- pretty confusing. Now the range of k's tested is clearly output and the bit size of the factors tested is output in parentheses.
2) Save files are incompatible. The k range is now output to the save file. This prevents an accidental error where editing the k range in worktodo.txt could use an old incompatible save file.

Obviously, finish your current range before doing any upgrading.

Sources:

Linux 64-bit executable

see post #193 for newer version

Windows 32-bit and 64-bit executables (CUDA 4.2)

Windows 32-bit and 64-bit executables (CUDA 5.0)

CUDA 5.0 Library files are [URL="http://sourceforge.net/projects/cudalucas/files/CUDA%20Libs/"]here[/URL] (if you need them)

[QUOTE=flashjh;311879]Windows 32-bit and 64-bit executables (CUDA 5.0)[/QUOTE]

Does CUDA 5 now work? When I tried it a few months ago mfaktc would not pass the self-test.

[QUOTE=Prime95;311884]Does CUDA 5 now work? When I tried it a few months ago mfaktc would not pass the self-test.[/QUOTE]
It does work and it is a little bit faster for the same assignment, per iteration. For quick assignments it doesn't make much a difference, but for longer ones, it will speed it up.

If anyone else can test it and make sure it's working, let me know.

Again, if you need the CUDA 5.0 library files, they're [URL="https://sourceforge.net/projects/cudalucas/files/CUDA%20Libs/"]here[/URL].

Attached CUDA 4.2 and CUDA 5 tests for the following:[CODE]
FermatFactor=33,5460000000000,5470000000000
FermatFactor=36,25709e6,25710e6
FermatFactor=54,81900e9,81911e9
FermatFactor=101,3334e9,3335e9
FermatFactor=135,88000000000,88100000000

F31 has a factor: 46931635677864055013377 [TF:75:76:mmff 0.23 mfaktc_barrett89_F32_63gs]
found 1 factor for k*2^33+1 in k range: 5460G to 5470G (76-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
F28 has a factor: 1766730974551267606529 [TF:70:71:mmff 0.23 mfaktc_barrett89_F32_63gs]
found 1 factor for k*2^36+1 in k range: 25709M to 25710M (71-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
F52 has a factor: 1475547810493913550438096961537 [TF:100:101:mmff 0.23 mfaktc_barrett108_F32_63gs]
found 1 factor for k*2^54+1 in k range: 81900G to 81911G (101-bit factors) [mmff 0.23 mfaktc_barrett108_F32_63gs]
F96 has a factor: 8453027931784477309850388309101819121893377 [TF:142:143:mmff 0.23 mfaktc_barrett152_F96_127gs]
found 1 factor for k*2^101+1 in k range: 3334G to 3335G (143-bit factors) [mmff 0.23 mfaktc_barrett152_F96_127gs]
F133 has a factor: 3836232386548105510567872577199319351015739156856833 [TF:171:172:mmff 0.23 mfaktc_barrett172_F128_159gs]
found 1 factor for k*2^135+1 in k range: 88000M to 88100M (172-bit factors) [mmff 0.23 mfaktc_barrett172_F128_159gs]

F31 has a factor: 46931635677864055013377 [TF:75:76:mmff 0.23 mfaktc_barrett89_F32_63gs]
found 1 factor for k*2^33+1 in k range: 5460G to 5470G (76-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
F28 has a factor: 1766730974551267606529 [TF:70:71:mmff 0.23 mfaktc_barrett89_F32_63gs]
found 1 factor for k*2^36+1 in k range: 25709M to 25710M (71-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
F52 has a factor: 1475547810493913550438096961537 [TF:100:101:mmff 0.23 mfaktc_barrett108_F32_63gs]
found 1 factor for k*2^54+1 in k range: 81900G to 81911G (101-bit factors) [mmff 0.23 mfaktc_barrett108_F32_63gs]
F96 has a factor: 8453027931784477309850388309101819121893377 [TF:142:143:mmff 0.23 mfaktc_barrett152_F96_127gs]
found 1 factor for k*2^101+1 in k range: 3334G to 3335G (143-bit factors) [mmff 0.23 mfaktc_barrett152_F96_127gs]
F133 has a factor: 3836232386548105510567872577199319351015739156856833 [TF:171:172:mmff 0.23 mfaktc_barrett172_F128_159gs]
found 1 factor for k*2^135+1 in k range: 88000M to 88100M (172-bit factors) [mmff 0.23 mfaktc_barrett172_F128_159gs]
[/CODE]

Works perfectly fine, but I havent noticed the difference between 5.0 and 4.2.
MMFactor=127,600e12,700e12 (not reserved, just benching) yielded in the same 142s per classes and 152M/s raw rate.
Used one GTX 480, with no monitor attached to it, so nothing could have interfered.
Maybe the speedup is specific for some gpu type?
Like sm_21 or sm_30 ?

V4 works good for me except the ctrl/c issue which is still there (crashing when ctrl/c is pressed or when the last assignment is finished). V5 is asking for a new driver ("you have an older driver then the one used to compile the program, need to update", beside of using the right cudart), which I am not going to install, I had lots of headache to make the current one to work as I want, so if it works, don't fix it. V4 works ok and it is bloody fast!
[code]
===========================
v4.2
===========================
[Tue Sep 18 01:59:45 2012]
F31 has a factor: 46931635677864055013377 [TF:75:76:mmff 0.23 mfaktc_barrett89_F32_63gs]
[Tue Sep 18 01:59:47 2012]
found 1 factor for k*2^33+1 in k range: 5460G to 5470G (76-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
[Tue Sep 18 01:59:49 2012]
F28 has a factor: 1766730974551267606529 [TF:70:71:mmff 0.23 mfaktc_barrett89_F32_63gs]
[Tue Sep 18 01:59:49 2012]
found 1 factor for k*2^36+1 in k range: 25709M to 25710M (71-bit factors) [mmff 0.23 mfaktc_barrett89_F32_63gs]
[Tue Sep 18 01:59:50 2012]
F52 has a factor: 1475547810493913550438096961537 [TF:100:101:mmff 0.23 mfaktc_barrett108_F32_63gs]
[Tue Sep 18 01:59:54 2012]
found 1 factor for k*2^54+1 in k range: 81900G to 81911G (101-bit factors) [mmff 0.23 mfaktc_barrett108_F32_63gs]
[Tue Sep 18 01:59:56 2012]
F96 has a factor: 8453027931784477309850388309101819121893377 [TF:142:143:mmff 0.23 mfaktc_barrett152_F96_127gs]
[Tue Sep 18 01:59:56 2012]
found 1 factor for k*2^101+1 in k range: 3334G to 3335G (143-bit factors) [mmff 0.23 mfaktc_barrett152_F96_127gs]
[Tue Sep 18 01:59:59 2012]
F133 has a factor: 3836232386548105510567872577199319351015739156856833 [TF:171:172:mmff 0.23 mfaktc_barrett172_F128_159gs]
[Tue Sep 18 02:00:00 2012]
found 1 factor for k*2^135+1 in k range: 88000M to 88100M (172-bit factors) [mmff 0.23 mfaktc_barrett172_F128_159gs]
<here it crashes>
===========================
v5.0
===========================
not tested

[/code]

A limit of mfaktc is that it cannot trial factor Mersenne numbers with exponents larger then 2[SUP]32[/SUP]-1 = 4,294,967,295.

A good thing with mmff is that it can do trial factoring above that limit. If I would like to trial factor a Mersenne number too large for mfaktc, like a 10Billion digit Mersenne number is there any way I can use mmff for this (now or in the future)?

[QUOTE=aketilander;312190]A good thing with mmff is that it can do trial factoring above that limit. If I would like to trial factor a Mersenne number too large for mfaktc, like a 10Billion digit Mersenne number is there any way I can use mmff for this (now or in the future)?[/QUOTE]

No. I've disabled standard Mersenne factoring in mmff (in case my many changes broke it). It would be easier to add that feature to mfaktc.

New number types like Mersenne factoring exponents more than 32-bits, generalized Fermats, bigger double-Mersennes, bigger Fermats, etc. requires writing new kernels.

[QUOTE=Prime95;312193]No. I've disabled standard Mersenne factoring in mmff (in case my many changes broke it). It would be easier to add that feature to mfaktc.

New number types like Mersenne factoring exponents more than 32-bits, generalized Fermats, bigger double-Mersennes, bigger Fermats, etc. requires writing new kernels.[/QUOTE]

OK I see, thank you for the answer.

Version 0.24

Another minor upgrade.

1) Fixed bug in calculating which Fermat number is divisible by a found factor.
2) If GPUSievePrimes is not set in mmff.ini, then mmff chooses a default value based on each entry in worktodo.txt. It may not choose the optimal GPUSievePrimes value, but it should be in the ballpark. Report to me any instances where it chooses a wildly non-optimal setting.
3) The -st and -st2 mfaktc command line arguments (self-test) are now ignored.
4) Some uninitialized mfaktc CPU sieving pointers are no longer freed at exit. Maybe this will solve the crash some were reporting at exit.

Because the mmff version number is written to save files, finish your current range before doing any upgrading or try the lightly-tested -nocheck command line argument to force using an old save file.

Sources:

Linux 64-bit executable: