mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   GPU to 72 (https://www.mersenneforum.org/forumdisplay.php?f=95)
-   -   GPU to 72 status... (https://www.mersenneforum.org/showthread.php?t=16263)

chalsall 2012-08-11 01:29

[QUOTE=nucleon;307616]So are people taking to 2^73 bit work on?[/QUOTE]

Yes -- most people are doing "What Makes Sense" with a pledge of 73. If the ranges aren't altered, this translates into lowest exponent above 57M to 73.

[QUOTE=nucleon;307616]Given, what I've just written, I grabbed a whole swag of 2^73bit work.[/QUOTE]

Thanks. While I very much appreciate your 71 -> 73 work in 59M and above, we really need to get most of 57M and 58M to 73 first. If things go to plan, the work completed in 59M won't be actually be distributed for LLing for several months.

nucleon 2012-08-11 01:48

[QUOTE=chalsall;307618]Yes -- most people are doing "What Makes Sense" with a pledge of 73. If the ranges aren't altered, this translates into lowest exponent above 57M to 73.[/QUOTE]

Cool.


[QUOTE=chalsall;307618]Thanks. While I very much appreciate your 71 -> 73 work in 59M and above, we really need to get most of 57M and 58M to 73 first. If things go to plan, the work completed in 59M won't be actually be distributed for LLing for several months.[/QUOTE]

Yeah I was just clicking lowest TF level. I'll go back to what makes most sense.

-- Craig

sonjohan 2012-08-13 18:39

[QUOTE=chalsall;307618]Thanks. While I very much appreciate your 71 -> 73 work in 59M and above, we really need to get most of 57M and 58M to 73 first. If things go to plan, the work completed in 59M won't be actually be distributed for LLing for several months.[/QUOTE]
So, I guess my selection is just as wrong, as I picked lowest TF, which got me 62M...

kladner 2012-08-13 19:45

[QUOTE=nucleon;307620]..... I was just clicking lowest TF level. I'll go back to what makes most sense.
-- Craig[/QUOTE]

Same here, then, on both counts. What Makes Sense it is!:grin:

chalsall 2012-08-13 23:17

[QUOTE=sonjohan;307831]So, I guess my selection is just as wrong, as I picked lowest TF, which got me 62M...[/QUOTE]

Not really wrong -- all of the work needs to be done (eventually). But "What makes sense", well, makes the most sense in so far as the work done will be utilized sooner.

lycorn 2012-08-15 15:04

[QUOTE=chalsall;307010]Unfortunately the Internet connection for the machine here in Barbados which collects and processes the data for that is having problems at the moment. [/QUOTE]

Still out?

chalsall 2012-08-16 04:31

[QUOTE=lycorn;308041]Still out?[/QUOTE]

Long and stupid story...

Short version -- Clueless and Worthless (C&W) DBA LIME (Lousy Internet; Massive Expense) didn't think it would be a problem if they changed everyone's LAN subnet. Fine if you're a home or a SME with only a few machines all using DHCP. A bit of a problem if you have servers with static IPs and routes...

The data is current now. Unfortunately there's ten days of data missing, so the delta reports will be a bit wonky for a while.

lycorn 2012-08-16 07:57

[QUOTE=chalsall;308098]
The data is current now. [/QUOTE]

That´s great. Thanks.

Dubslow 2012-08-19 04:55

Complain complain complain: If I don't have assignments, the Production Heuristics disappear :razz:

petrw1 2012-08-26 14:39

Some milestones.....
 
Over 100,000 total P-1 work done.
Craig Meyers: Over 500,000 work done; over 1,000 factors found.

LaurV 2012-08-29 18:26

@chalsall: waaaa we have now sortable columns for individual assignment tables. That is good. I can see my computers which are lazy to report, and I can push the one lazy to work closer to the finish line. Still wonder why plenty of things always happen in my absence... :D

Now if we are here, how about the "factor cost" tables? Are they all correct? for example in [URL="http://www.gpu72.com/reports/factoring_cost/"]"tf cost per factor found in dc range[/URL]" table, the total at 69 bits, shouldn't be divided by a thousand, or so? (it seems too big). The same for "[URL="http://www.gpu72.com/reports/factoring_cost/p-1/"]p-1 cost per f f in ll range[/URL]" table, at 72 bits, there should be a missing division somewhere.

And hey! found out I am still missing those 60GhzDays of DC. One full month since I notified you... If you look [URL="http://www.gpu72.com/reports/workers/dc/"]here[/URL], [URL="http://www.gpu72.com/reports/workers/ll/"]here [/URL]and [URL="http://www.gpu72.com/reports/worker/2423ae6e8f696d5e7d1447de91ca35a6/"]here[/URL], there is a discrepancy of 60GHzDays recorded correctly as DC in my statistics, but recorded wrongly as LL in the top list(s), where I would need it mo(n)st... :razz: They were 60 Gigs of DC, and should appear in the top, otherwise how can I beat that monst guy? :razz: As a protest I have not done any DC in the last days (which would make you and Sid and Andy very happy, grrr, I should better find a different way of protesting)...[SIZE=1] [COLOR=Silver](i had no internet and no work queued and I was few days out of the buttons, but you better don't know that and assume I am protesting). [/COLOR][/SIZE]


All times are UTC. The time now is 23:16.

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