Let GPU72 Decide and other questions
 

I have some more questions.

How do I set the option to let GPU72 decide?
I only have these options in my GPU72Config.txt
#OPTION
#WhatMakesSense = 0
#LowestTFLevel = 1
#HighestTFLevel = 2
#LowestExponent = 3
#OldestExponent = 4
Option:0

Also, what does pledge:73 mean and should I use 73?

[QUOTE=jschwar313;424770]I have some more questions.

Also, what does pledge:73 mean and should I use 73?[/QUOTE]

"Let GPU72 Decide" (LGPU72D) is option number 9. This was added after the other options, which is why I assume it's not in your spider's config documentation.

So you know, LGPU72D sets the pledge level to what is needed most. At the moment, that's 75. Edit: The pledge is the factoring depth your GPU will trial factor up to.

All other Options honour the pledge level. And, while going to 75 is the most needed at the moment, if you only want to go to 73 that's fine as well; all work is needed and helpful.

Great. thanks a lot Chris. I'll use let GPU72 decide in the future. I was using 73 since that was what was in the config file when I downloaded the software from the site. Also, the LGPU72D was not in the the config file either. Maybe this post will help other newbies like me. Thanks again.

I think we need a glossary of acronyms and other terms for newbies somewhere. I would think putting it in a FAQ somewhere would be nice.

Is it possible to get assignments for the range above 332 million at present, or is the emphasis on 75 million range?
This is for TF and P-1
Thanks

[QUOTE=bayanne;431307]Is it possible to get assignments for the range above 332 million at present, or is the emphasis on 75 million range? This is for TF and P-1[/QUOTE]

GPU72 is currently focusing on feeding the LL'ers at the four LL "wavefronts", and the P-1'ers in 75M and 76M (just about everything below 75M has already been P-1'ed, and those that aren't are already assigned by Primenet to P-1'ers).

In addition, we're (slowly) working the 332.2M range up to at least 78 bits.

333M and above is available via Primenet.

[QUOTE=chalsall;431315]In addition, we're (slowly) working the 332.2M range up to at least 78 bits.[/QUOTE]There is an Anonymous that chewed through a bunch of 74->75 in the 332,38x,xxx range. Did they pull them from GPU72?

[QUOTE=Uncwilly;431329]There is an Anonymous that chewed through a bunch of 74->75 in the 332,38x,xxx range. Did they pull them from GPU72?[/QUOTE]

Probably. If you can give me an explicit example or two I'll do a quick query, and be able to tell you for sure.

332381243
332382007
332382571

among others. Those had factors. I had been toiling away slowly in that area 'off the grid' as it were.

[QUOTE=Uncwilly;431331]...among others. Those had factors. I had been toiling away slowly in that area 'off the grid' as it were.[/QUOTE]

No, these were not through GPU72. I have no idea who did these.

So, in light of the discussions immediately above, what is the best use for a GTX460, which is currently 17:19:45 79.00% away from finishing a DC assignment? With the frequencies I run it at, it can do a little over 200 G-D/D. I would be happy to move it back to TF in a little over 17 hours. :smile:

[QUOTE=kladner;445008]So, in light of the discussions immediately above, what is the best use for a GTX460, which is currently 17:19:45 79.00% away from finishing a DC assignment? With the frequencies I run it at, it can do a little over 200 G-D/D. I would be happy to move it back to TF in a little over 17 hours. :smile:[/QUOTE]

I would suggest you stick with DC'ing.

While there will still always be a need for TF'ing, we are _well_ ahead of both the LL'ers and the P-1'ers.

The choice is yours, of course.