The "(W)TF Depth Debate" state of the bunion

[chalsall]

Originally Posted by TheMawn

Well I personally only use the one monitor, but it has 27 juicy inches of 1080p goodness. Still I find myself badly wanting a second.

My secondary monitor is 1920 by 1200 pixels.

As any girl will tell you, it's not the inches that count, but how you use them....

(And, yes. Try working with a second (or more) monitor. You'll find yourself much more productive.)

[davieddy]

Originally Posted by TheMawn

http://mersenneforum.org/attachment....2&d=1387089113

I never know if I'm reading these things correctly and if I am reading enough of them to get the full picture. Let's take a crack at it.

I read these tables daily AFTER NOON and even I find it tricky.
You'll see why. The pages you cite were before the noon update

During the period of December 8 to December 15,

And so the period was actually Dec 7th to Dec 14th

in the 60M range, 2359 exponents were taken from something to 74.

In particular, 322 of which were <63M

1017 LL's were completed in the 60M range.

between Dec 8th and Dec 14th: 6 days.
Now throw in the 50M range and any DCs (which are subtracted from first time LLs total) and you get ~1800 LL completions in 6 days, as we always do.

From this it would seem factoring is massively ahead of LL.

Well I reckon each day we complete 300 LLs.
We want to at least replace these with virgin assignments.
(2359 - 322)/7 = 291

I told you it was tricky.

____________

@ Mods: I am grateful to be allowed back, but if you do not allow me to edit my posts, I won't post a painstaking reply like this again.

I think you have learned what to expect instead:

^MT

[chalsall]

Originally Posted by davieddy

I read these tables daily AFTER NOON and even I find it tricky.

David...

If you are so knowledgeable, why don't you build your own tables?

[davieddy]

Originally Posted by chalsall

David...

If you are so knowledgeable, why don't you build your own tables?

Just fix yours.
I've been teling you incessantly that all you need to do is cut out your midnight fiddling.

[chalsall]

Originally Posted by davieddy

Just fix yours.

I've been accused of not taking direction well....

[davieddy]

Originally Posted by chalsall

I've been accused of not taking direction well....

I can debug, but I prefer to get the code right first time.

If a problem takes two pages of math to solve, you soon learn that it's better to avoid typos than search for it when the answer comes out wrong.

[davieddy]

This page
http://www.mersenne.info/exponent_st...lar_delta/1/0/
always makes most instructive reading in the morning.

[TheMawn]

Thanks, David. That's actually entirely the answer I was looking for. Very little TF going on in the 50M, as we *should* be finished with that, although there is talk about raising the upper 50's to 73, if I understood that correctly? Not sure how much I like that...

Attempt number 2: See attached. This is for a one-month period.

• Negligible TF in 50M
• 10,000 exponents (~> 300 per day) brought to 74 in 60M.
• Of these, 9,000 were from 71 and 3,000 were from 73. Where are the remaining 2,000 exponents?
• 3,500 exponents have had LL done in the 50M. Negligible DC.
• 5,200 exponents have had LL done in the 60M. Negligible DC.

• 9,000 TF from 71 to 74.
• 8,200 exponents with LL.

It seems to me that in the last month, the 71-to-74 effort alone has been enough to keep up with LL. There's an extra few thousand 71-to-73. By the looks of things, we can afford to spend a bit more time in the 50M, but maybe it's in our best interest to increase the TF lead at the wavefront.

[LaurV]

Originally Posted by TheMawn

• Of these, 9,000 were from 71 and 3,000 were from 73. Where are the remaining 2,000 exponents?

Obviously, about a thousand went only to 72 (from 71) and the other thousand were factored, therefore disappeared from the table, see last column. This is written following your rounding system, where 600 is rounded to 1000, but if you sum the columns, there is no exponent disappearing. Everything is fine.

And yes, we keep up with the wave...

[davieddy]

Originally Posted by TheMawn

Attempt number 2: See attached. This is for a one-month period.

It is from Nov 17th to Dec 17th for TF (30 days)
Nov 18th to Dec 17th for LL (29 days)

10,000 exponents (~> 300 per day) brought to 74 in 60M
Of these, 9,000 were from 71 and 3,000 were from 73. Where are the remaining 2,000 exponents?

!000 stopped off at 72 bits and 1000 were factored

3,500 exponents have had LL done in the 50M. Negligible DC.
5,200 exponents have had LL done in the 60M. Negligible DC.

Acurately 8746 in 29 days

9,000 TF from 71 to 74.

7000 from 71. 3000 from 73 of which 1000 were <63M (recycled i.e. not virgins)

8,200 exponents with LL.

87000 in 29 days (~> 300.day)

It seems to me that in the last month, the 71-to-74 effort alone has been enough to keep up with LL. There's an extra few thousand 71-to-73. By the looks of things, we can afford to spend a bit more time in the 50M, but maybe it's in our best interest to increase the TF lead at the wavefront.

I disagree

[davieddy]

Originally Posted by TheMawn

maybe it's in our best interest to increase the TF lead at the wavefront.

What lead?
What wavefront?

But yes, it is certainly in our best interests.
Especially for those flying pigs (74 bits for everything >63M).

Won't life get interesting when we are discussing 75 bits at 75M?

D