[QUOTE=dragonbud20;410720]I was thinking about the predicament we had the past few days with a lack of TF power and it though of something that might be useful for figuring out what type of work to do. Would it be possible to have some sort of indicator on the assignment page or anywhere really on the gpu72 site that indicated when TF or DC needed more work done or we don't have enough buffer for one or the other? I personally do all my assignments manually so I know it could help people who do the same although I'm not sure what could be done for people who get all of their work using misfit or spiders or whatever(I'm not familiar with automating mfactc at all I don't know if those terms are right). I don't know if this has been considered but it would be a nice little feature.[/QUOTE]

DCTF isn't needed for another year or so. But some of us want to finish it all right now, and at our current rate that will be done in about eight months.

LLTF is needed. There are two main "waves" in the LLTF exponents. One for being released for a full LL test, the other for P-1 factoring. Ideally, for current exponents, everything should be factored to 75 bits. This is largely happening before exponents are being released for full LL, but not for P-1. Ideally it would be done before both, but it's more important that 75 bits be reached before LL, so P-1 gets released early. If you get exponents with Let GPU72 Decide it will take care of handing out the most important work first. If you pick What Make Sense, a bit level of 74 or 75 is the most helpful.

[QUOTE=dragonbud20;410720]Would it be possible to have some sort of indicator on the assignment page or anywhere really on the gpu72 site that indicated when TF or DC needed more work done or we don't have enough buffer for one or the other?[/QUOTE]
This "feature" exists already, for automatic assignments, it is called "what makes sense" or "let gpu72 decide".
Misfit is a bliss, unless you use linux, and even on linux there is a good script to fetch exponents from gpu72, so why do "manual" assignments?

:grin:I really should figure out how to get misfit working. From what everyone says I'm guessing it would make keeping my cards fed a lot easier. honestly I've downloaded it but I;m not really sure how to get it working.

EDIT: spent a few minutes cracking at it and I think I have all my gpu work getting and result sending automated now. thanks guys for pushing me up off my ass for a bit.

Chalsall,

You have asked people to move to LLTF over DCTF to help keep ahead of the P-1 wave, but the Chris Halsall account is doing DCTF. Are you doing some of the triple checking that has been discussed and that's why you aren't taking your own advice? :smile:

-Walt

Haha, that is a good one, and I would like to sustain you and hit this guy in the nose :razz: but actually he took the advice, you look to his assignment queue (first table) and he has 3100++ LLTF expos, and only 14 DCTF. What you see now is liquidating the DCTF queue. I mean, if he is like me, we queue work for at least 3-4-5 days or a week.

[QUOTE=Walt;410817]You have asked people to move to LLTF over DCTF to help keep ahead of the P-1 wave, but the Chris Halsall account is doing DCTF. Are you doing some of the triple checking that has been discussed and that's why you aren't taking your own advice? :smile:[/QUOTE]

LOL... I was wondering when someone was going to call me out on that... :smile:

The reason is (ironically) I don't have have a running GPU myself (well, actually that's not entirely true, I have a 580 kindly donated by Jerry and a 480 I purchased here in Bim, but I use those for private, sensitive computer vision compute; electricity is simply too expensive here to run such hardware locally unless there's a privacy reason to do so).

The little (~500 GHd/d) I contribute is via a rented EC2 "cg1.4xlarge" spot instance which I constrain to be at no more than $0.145 an hour -- rather unpredictable as to its availability. For the most part I tend to use this to "clean up" certain ranges; make things "round", finish work abandoned by others, etc.

Ever heard the saying "Do what I say, not what I do."? I hope this doesn't come across as disingenuous.

Thanks. I just wondered. It seemed strange. I assumed since you have a good view of the big picture, you were doing specific tasks that needed to be done. Sounds like that's the case.

-Walt

[QUOTE=chalsall;410855]LOL... I was wondering when someone was going to call me out on that...

Ever heard the saying "Do what I say, not what I do."? I hope this doesn't come across as disingenuous.[/QUOTE]

Your labor and brainpower are vastly more valuable than your silicon or electronpower. Thank you!

[QUOTE=VBCurtis;410864]Your labor and brainpower are vastly more valuable than your silicon or electronpower. Thank you![/QUOTE]

You are very kind. Thank you.

I have betrayed AMD and added an NVIDIA Rig to the mix with an array of cards (690, two 980s and a Titan X). Eventually these will be doing CudaLucas LL work but I have them also contributed to LLTF for now. Should add some extra boost.

[QUOTE=airsquirrels;410908]I have betrayed AMD and added an NVIDIA Rig to the mix with an array of cards (690, two 980s and a Titan X). Eventually these will be doing CudaLucas LL work but I have them also contributed to LLTF for now. Should add some extra boost.[/QUOTE]

What system can host that many cards?