GPU to 72 status...

[petrw1]

Originally Posted by chalsall

I should have been clearer. When I said "about 40", I meant about 40 more than we were doing.

Covered....as I mentioned a couple days ago....my 42/day will start later today once current LLTF completed

[chalsall]

Originally Posted by petrw1

Covered....as I mentioned a couple days ago....my 42/day will start later today once current LLTF completed

Thanks! To you, and everyone who answered the "surge" request, and/or gave permission for a LLTF to DCTF change for LGPU72D.

Based on what is queued we're looking good for staying ahead of the Cat 4 DC'ers (mostly AKA "the churners"), but probably next week I'll update the assignment code such that LLTF LGPU72D requests may occasionally assign some DCTF work.

This will probably not be needed to be used (unless a new MP is found, and we get a new surge of Newbies like last time). Also, please know that this will only be used to stay ahead of the LL'ing Cat 4 wave front. Or, in other words, only working from 71 bits to 72 at the DCTF'ing wave back.

[casmith789]

I've just switched over to DCTF, and I was assigned this exponent from GPU72 (along with some other similar ones):

78986359 DC TF 71 74

As far as I can tell that exponent has never had an LL done, and it is much too high to be the current wavefront for double checks, so I am a bit confused. What am I missing?

[James Heinrich]

Originally Posted by Mark Rose

I like the idea of a "I Don't Care" option, where GPU72 has complete control.

I would also like to see that option.

[petrw1]

Originally Posted by chalsall

Also, please know that this will only be used to stay ahead of the LL'ing Cat 4 wave front. Or, in other words, only working from 71 bits to 72 at the DCTF'ing wave back.

A little confused....

If I understand correctly the "LL'ing Cat 4 Wave Front" is about 72.6M
Isn't the "DCTF'ing wave back" about 58M?

I am currently taking the lowest DCTF to 72 Bits. About 40.8M.
Is this not what you were needing for the DC Churners?

[chalsall]

Originally Posted by casmith789

What am I missing?

You Let GPU72 Decide, and pledged to go to 75 bits.

This doesn't make sense for DCTF'ing.

| Chris Smith     | 2015-06-02 18:57:52 | DF(9) -- N:    1 G:  150 P: 75 L:         0 H: 100000000 -- n:  100 p: 74 l:  66000000 h:  80000000 -- A: 4

[chalsall]

Originally Posted by petrw1

A little confused....

If I understand correctly the "LL'ing Cat 4 Wave Front" is about 72.6M
Isn't the "DCTF'ing wave back" about 58M?

I am currently taking the lowest DCTF to 72 Bits. About 40.8M.
Is this not what you were needing for the DC Churners?

I misspoke. s/stay ahead of the LL'ing Cat 4 wave front/stay ahead of the DC[LL]'ing Cat 4 wave front/ should be more correct.

[James Heinrich]

Originally Posted by chalsall

You Let GPU72 Decide, and pledged to go to 75 bits

If that was me, I'd assume if I selected "Let GPU72 decide" and "75" that it would interpret it as "Give me something to do, whatever is most useful right now, and I'm willing to TF as high as 2⁷⁵ if that's required".

[chalsall]

Originally Posted by chalsall

You Let GPU72 Decide, and pledged to go to 75 bits.

This doesn't make sense for DCTF'ing.

Fsck me...

Thanks, you've pointed out a bug. LGPU72D for DCTF hasn't had an audit recently. Sorry about that.

[chalsall]

Originally Posted by James Heinrich

If that was me, I'd assume if I selected "Let GPU72 decide" and "75" that it would interpret it as "Give me something to do, whatever is most useful right now, and I'm willing to TF as high as 2⁷⁵ if that's required".

LOL...

I'm doing the best I can with the time allowed.

[casmith789]

Ah okay. MISFIT greys out that option so I just assumed it wasn't important for LGPU72 decide.