I meant this
 

[QUOTE=davieddy;302170]I rest my case.[/QUOTE]
ANSWER Chalsall.

It is central to the TF effort.

[QUOTE=davieddy;303190]ANSWER Chalsall.

It is central to the TF effort.[/QUOTE]

What, [I][U]exactly[/U][/I], is the question?

[QUOTE=chalsall;303193]What, [I][U]exactly[/U][/I], is the question?[/QUOTE]
If you bothered to follow the quote back, you would find:

"Why hasn't every exponent > 53M TFed to at least 72 bits?"

(Not to mention a suggested answer [B]which you brazenly confirmed}[/B]

Bit more conversation, a little less action please.
(Elvis)

[QUOTE=davieddy;303194]If you bothered to follow the quote back, you would find:

"Why hasn't every exponent > 53M TFed to at least 72 bits?"

(Not to mention a suggested answer [B]which you brazenly confirmed}[/B]

Bit more conversation, a little less action please.
(Elvis)[/QUOTE]

I thought I had already answered that with a quote from Einstein: "The only reason for time is so that everything doesn't happen at once." Perhaps you missed the subtly...

Now, being pedantic, since there are an infinite number of candidates above 53M, they will [I]never[/I] all be TFed to at least 72 bits.

However, I consider the [URL="http://www.gpu72.com/reports/released_level/"]99,655 LL candidates we [B][I][U]have[/U][/I][/B] taken to 72 or above[/URL] a reasonably good start....

Think a semi-apology is due Chris.
I'm sure George is grateful for your efforts
in managing the project, and it is really primenet's fault
more than yours.
Just look at the TF assignments assigned for one extra bit
that hang around forever.
This breadth-first [B]ahead [/B]of the wavefront is almost
counter-productive.
Folk would much prefer to TF from 69 to 73 in one go
than 72 to 73.
I said this a couple of years ago.

Re your Einstein quote, there was no excuse for not taking
every expo >53M to 72 at the rate of 200/day ~ 2 years ago.

David

[QUOTE=chalsall;303195]Now, being pedantic, since there are an infinite number of candidates above 53M, they will [I]never[/I] all be TFed to at least 72 bits[/QUOTE]Not true. In fact, it is demonstrably false.

All primitive factors of M(p) are of the form 2kp+1. The proof of that statement is relatively straightforward and is left as an exercise.

Consequently, all prime factors of Mersenne numbers greater than M(M(71)) are greater than 2^72. Hence all candidates greater than M(M(71)) have been TFed to at least 72 bits.


QED.

RE Breadth, the "pledged" TF level is up to each worker. I think most choose to do the whole schbang all at once. (There are some cases where mfakto is more efficient at lower bit levels.) If a worker does not TF a candidate all the way, the expo is kept and redished until a worker takes it all the way. Expos >58M are TFd to 73.

[QUOTE=xilman;303203]Consequently, all prime factors of Mersenne numbers greater than M(M(71)) are greater than 2^72. Hence all candidates greater than M(M(71)) have been TFed to at least 72 bits.[/QUOTE]

I stand corrected. :smile:

[QUOTE=xilman;303203]Not true. In fact, it is demonstrably false.

All primitive factors of M(p) are of the form 2kp+1. The proof of that statement is relatively straightforward and is left as an exercise.

Consequently, all prime factors of Mersenne numbers greater than M(M(71)) are greater than 2^72. Hence all candidates greater than M(M(71)) have been TFed to at least 72 bits.


QED.[/QUOTE]

I think this is my fault I pm'ed a comment like that just to attempt to stop the arguments.

[QUOTE=chalsall;303206]I stand corrected. :smile:[/QUOTE]
Now bend over my knee and take your knickers down

[QUOTE=chalsall;303162]I have, for several months, been deleting posts by davieddy in the "GPU to 72" sub-forum[/QUOTE]

MUCH more to the point, you may not be interested,
but they are typically replies to someone who might be.
Not to mention the "effort" I put in to write the buggers.


STOP F***ING ABOUT YOU JUMPED UP FART