mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Marin's Mersenne-aries (https://www.mersenneforum.org/forumdisplay.php?f=30)
-   -   Trippple Checks (https://www.mersenneforum.org/showthread.php?t=17108)

LaurV 2015-04-02 08:09

I am with you here (read: I will not blame you for it, unless someone else blames me for it :razz:)

Madpoo 2015-04-04 00:10

1 Attachment(s)
[QUOTE=LaurV;399196]I am with you here (read: I will not blame you for it, unless someone else blames me for it :razz:)[/QUOTE]

Here's an updated list. I'm laying claim to doing triple-checks on the ones between 21M and 22M since I finished the last of the 20M-21M numbers.

cuBerBruce 2015-04-04 02:49

[QUOTE=Madpoo;399310]Here's an updated list. I'm laying claim to doing triple-checks on the ones between 21M and 22M since I finished the last of the 20M-21M numbers.[/QUOTE]

I'll take the next 10, M220xxxxx thru M2216xxxx.

LaurV 2015-04-04 03:48

And why do you quote my message with that list? :blush:

Madpoo 2015-04-04 04:59

[QUOTE=LaurV;399318]And why do you quote my message with that list? :blush:[/QUOTE]

Because I'm your #1 fan!

Honestly, I have no idea. Sometimes I just do things? :smile:

VictordeHolland 2015-04-05 09:10

M24939227 has a factor: 2216644110838629719887
M25901983 has a factor: 1877785088209838147849
M26501791 has a factor: 1887069080169066172433
M26838139 has a factor: 1656114794350730091919
M28603763 has a factor: 1250967656026180632049

This concludes my TF work on the triple check candidates (in other words, you can mark my TF ranges as complete).

Madpoo 2015-04-05 17:58

1 Attachment(s)
[QUOTE=VictordeHolland;399380]M24939227 has a factor: 2216644110838629719887
M25901983 has a factor: 1877785088209838147849
M26501791 has a factor: 1887069080169066172433
M26838139 has a factor: 1656114794350730091919
M28603763 has a factor: 1250967656026180632049

This concludes my TF work on the triple check candidates (in other words, you can mark my TF ranges as complete).[/QUOTE]

Nice work.

"tigreroars" is doing the 35M-36M range up to 72 bits but other than that there's no more TF work anyone has signed up for (and maybe no more really needed).

You can look over this latest list and see if there's anything that anyone would care to do to higher bit levels, but it's fine with me if it were left with what's there now.

I also signed up to do the rest of the 22M range for triple-checking.

manfred4 2015-04-05 22:40

There is some more TF work, that should be done first, i.e. 39M+ is not ready yet. It should be at 71 up to 40M, at 72 from there on and even to 73 from ~50M. But I won't be doing that now, only if it is needed.

tigreroars 2015-04-06 12:33

[QUOTE=Madpoo;399393]Nice work.

"tigreroars" is doing the 35M-36M range up to 72 bits but other than that there's no more TF work anyone has signed up for (and maybe no more really needed).

You can look over this latest list and see if there's anything that anyone would care to do to higher bit levels, but it's fine with me if it were left with what's there now.

I also signed up to do the rest of the 22M range for triple-checking.[/QUOTE]

Well, finished the 35M-36M TF to 72. No Factors. Are we done with TF?

Madpoo 2015-04-06 15:13

[QUOTE=tigreroars;399452]Well, finished the 35M-36M TF to 72. No Factors. Are we done with TF?[/QUOTE]

Well, I guess that depends on what people would consider "finished" as far as TF is concerned. :smile:

manfred4 says:
[QUOTE]There is some more TF work, that should be done first, i.e. 39M+ is not ready yet. It should be at 71 up to 40M, at 72 from there on and even to 73 from ~50M. But I won't be doing that now, only if it is needed.[/QUOTE]

If we go by that, then up to 40M should be okay. Right now there are still 12 exponents in the 39M range that are only at 70 bits, but user Nesowa has been doing their own work on bringing TF up to that level, and they just checked in a batch up to (from our sub-list) M39585463 early this morning. I'd suggest we leave those be since Nesowa is going through them all anyway? I think by the time I, or someone else, gets around to triple-checking those 39M exponents they'll probably be done.

< 20M are all up to 69 bits
20M - 30M are all up to at least 71 bits
30M - 40M are all up to at least 71 bits except those aforementioned 12 that are still at 70
40M - 55M are all up to 74 bits
55M - 60M are all up to at least 72 bits with 3 exceptions (listed below)
60M + seem okay with a couple I question...

Those 55-60M exceptions that are only at 71 bits are:
DoubleCheck=55018393,71,1
DoubleCheck=55837657,71,1
DoubleCheck=58317893,71,1

This one in the 60M range is only at 73 bits but the rest of it's buddies are at 74:
DoubleCheck=61088273,73,1

And I don't know about this one in the 90M range:
DoubleCheck=90000607,72,1

I have a machine with that 90M exponent queued up to start running the triple-check in 21 hours, as soon as it finishes it's current check. If someone wanted to take that one up to a higher TF level before then, that'd be great. If a factor was found it'd save me an estimated 5 days on this 10-core worker.

As for those other 4 exceptions I mentioned, I guess if someone wanted to take those a bit higher they could?

As for the rest, I know VictordeHolland was taking his things to higher bits than manfred4 so any other TF work would depend on whatever you feel like doing. :smile: For instance, you took 35M-36M up to 72 bits, but the work that Nesowa has been doing was only taking the rest of the 30M range up to 71 bits. That's where most of them from 36M-40M live right now.

tigreroars 2015-04-06 22:12

[QUOTE=Madpoo;399473]
< 20M are all up to 69 bits
20M - 30M are all up to at least 71 bits
30M - 40M are all up to at least 71 bits except those aforementioned 12 that are still at 70
40M - 55M are all up to 74 bits
55M - 60M are all up to at least 72 bits with 3 exceptions (listed below)
60M + seem okay with a couple I question...

Those 55-60M exceptions that are only at 71 bits are:
DoubleCheck=55018393,71,1
DoubleCheck=55837657,71,1
DoubleCheck=58317893,71,1

This one in the 60M range is only at 73 bits but the rest of it's buddies are at 74:
DoubleCheck=61088273,73,1

And I don't know about this one in the 90M range:
DoubleCheck=90000607,72,1

I have a machine with that 90M exponent queued up to start running the triple-check in 21 hours, as soon as it finishes it's current check. If someone wanted to take that one up to a higher TF level before then, that'd be great. If a factor was found it'd save me an estimated 5 days on this 10-core worker.

As for those other 4 exceptions I mentioned, I guess if someone wanted to take those a bit higher they could?

As for the rest, I know VictordeHolland was taking his things to higher bits than manfred4 so any other TF work would depend on whatever you feel like doing. :smile: For instance, you took 35M-36M up to 72 bits, but the work that Nesowa has been doing was only taking the rest of the 30M range up to 71 bits. That's where most of them from 36M-40M live right now.[/QUOTE]

Ohh, I thought the 30M range was where the 72bit range was.

I don't have the optimal GPU graph infront of me, so I was thinking the 90M resided in the 75 or 76 bit range?

Otherwise, it takes me 2 weeks for a CPU LL double check in the 33M range, so I'll let other people work on those triple chucks. I'll move my GPU back to closing the TF DC pile.


All times are UTC. The time now is 10:00.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.