Folding Forum

@DocJonz: Work units try to balance a few things out, but equal pay for equal work was never mentioned. The two "equals" I've heard are "equal points for equal cost or value (of the resources needed to run the wu, and "equal points for equal science", which sounds similar, but could have a quite different meaning, depending on how you mean it.

@patonb: Zigzagging the point values around, would be a misstep. Really, more like a nightmare. We must not have that. Slow gradual changes in existing projects, are the way to go, and only if they're really needed. New projects of course, are different - but you have to be very careful what you value them at, for the folders.

@Napoleon: A SLIGHT trimming of the BA points would be OK. Perhaps 5% or so. Certainly not 25%!! It's SMP that needs to be adjusted the most, and obviously, upward - maybe 10% or so now, and another 10% in April. Something along those lines.

@Orion: Just one word - Yes.

The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

orion wrote:If more SMP's need to be run then make it worth everybody’s time to run them. 2p/4p, i7's, c2d's etc...etc... Increase their points...that way everyone who runs them wins.

And everybody becomes really, really fat in the process, with no end in sight, except maybe limitations of 64bit integers. That sounds healthy and sustainable to you?

We can talk all day and forever about what "we think" should happen. But, until there is solid dialog with PG about the actual Bigadv changes, we're basically arguing point/counterpoint amongst ourselves. That may be therapeutic in a way, but it would be much more useful to actually have these discussions with the people that are in control of the changes (PG). I'd like to see PG actively engage us as Bigadv Folders, right here and now. If we can be here just about every day, I think at least one of them could "look-in" and offer some sort of feedback/direction to the conversation. That is what is needed most.

HaloJones wrote:The hockey stick aspect of BA is what needs to change. It is ludicrous to me that returning work faster can have such an exponential effect. The way things are, one big rig (and for $7K I could pick up a 40-core Intel box) could be outproducing entire teams of dedicated folders.

It's not about increasing the others (although I do agree SMP has suffered over the last two years) but decreasing the BA QRB.

Here's what you aren't considering, HJ.

1) FAH gets grants or other types of funding, to conduct some of their (usually large), studies of "THIS" or "THAT" protein action or structure. In order to be accepted, FAH needs to show it can complete said study in a short amount of time. If FAH can't do that, then the grant will go to say, a supercomputer center, or to another protein study and analysis group, instead of FAH. There are several other such Distributed Computer groups that study proteins, around the world

FAH wants those studies/grants! Naturally.

2) Pande Group is run by Prof's and graduate students, at well known Universities (typically). They want (NEED), to publish significant findings, in peer reviewed scientific journals (Cell, etc.). The graduate students need it for their career advancement, and the established Professors want it for their continued contribution and advancement, in their field.

In the University world, getting well published in these scientific journals, is a VERY big deal, and sometimes critical to job security. Always important for advancement.

Now look at the BA project, anew. The BA project was a brilliant idea, but now that the initial period is over, we do need to diminish the disparity between BA and SMP, for those who no longer will qualify. Yes, it will (and should), be a step down in their ppd, but it shouldn't be a push off a 50 ft. cliff. Maybe just a 10 ft. "cliff". Because alienating them is not what we want to do. We want to thank them, and make their continued contribution as an SMP folder, still meaningful.

Napoleon wrote:

orion wrote:If more SMP's need to be run then make it worth everybody’s time to run them. 2p/4p, i7's, c2d's etc...etc... Increase their points...that way everyone who runs them wins.

And everybody becomes really, really fat in the process, with no end in sight, except maybe limitations of 64bit integers. That sounds healthy and sustainable to you?

And the problem with F@H winning by getting the work it wants done is?

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Just one word: No.

OK, more words:
An SMP work unit can't be given the same points per day, as a BA wu. There will always be a "jump" from SMP to BA, which is how it must be. BA rigs do significantly more work, and in a far shorter period of time. (More reliable as well), requiring greater cost and resources.

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Conveniently ignoring GPUs and Core_17, aren't you? Sticking to "equal pay for equal work"... if SMP points are raised, then GPU points need to be raised, too. GPUs would blow BA out of the water, and you'd be suggesting more points for BA... it'd be a vicious circle. On a side note, reminder from http://folding.stanford.edu/home/welcom ... core-17-2/:

VijayPande wrote:We have also successfully tested FahCore 17 with extremely large proteins (500,000+ atoms), which are on par with the ones used by "bigadv" CPU projects.

As flawed as the current points system is, what you're suggesting would reduce it to a mere social engineering tool. Is that what you actually want?

EDIT:

orion wrote:And the problem with F@H winning by getting the work it wants done is?

The points system becoming utterly meaningless. Not that it matters to me, I'm die-hard enough to fold with what little I have anyway. I'd like to think I've been around, so I just figured I'll try to point out what you're really asking for. Points be damned...

Changing the hockey stick would help BA line up better with SMP. And any points bonus that reaches infinity points is unsustainable, hence the need for a revision, or regular adjustments, or both.

Adak wrote:

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Just one word: No.

OK, more words:
An SMP work unit can't be given the same points per day, as a BA wu. There will always be a "jump" from SMP to BA, which is how it must be. BA rigs do significantly more work, and in a far shorter period of time. (More reliable as well), requiring greater cost and resources.

No. Cost has NEVER been a consideration to PG. Only the faster science being done.

texinga wrote:We can talk all day and forever about what "we think" should happen. But, until there is solid dialog with PG about the actual Bigadv changes, we're basically arguing point/counterpoint amongst ourselves. That may be therapeutic in a way, but it would be much more useful to actually have these discussions with the people that are in control of the changes (PG). I'd like to see PG actively engage us as Bigadv Folders, right here and now. If we can be here just about every day, I think at least one of them could "look-in" and offer some sort of feedback/direction to the conversation. That is what is needed most.

I agree with what you're saying to a point.

PG does need to be talking to the donors that run BA's...but not all of us and NOT with DAB.

I would like to see PG taking with tear and Grandpa_01. Both are level headed, run BA's and have contributed allot to F@H.

texinga wrote:We can talk all day and forever about what "we think" should happen. But, until there is solid dialog with PG about the actual Bigadv changes, we're basically arguing point/counterpoint amongst ourselves. That may be therapeutic in a way, but it would be much more useful to actually have these discussions with the people that are in control of the changes (PG). I'd like to see PG actively engage us as Bigadv Folders, right here and now. If we can be here just about every day, I think at least one of them could "look-in" and offer some sort of feedback/direction to the conversation. That is what is needed most.

Consider it a conversation, with someone who prefers to just listen at the moment. There are lots of ideas being brought up, and concerns as well. Until they have a clear vision of what they want to say and do, it's best for them to listen and wait for awhile. They can invite conversation with us, when things have calmed down further, and they've had more time to reflect on it.

Adak wrote:

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Just one word: No.

You took it too literally

Napoleon wrote:

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Conveniently ignoring GPUs and Core_17, aren't you?

No. This thread is about BA's with SMP backlog thrown in. If you want to talk about GPU's then please start a new thread.

and you'd be demanding more points for BA

I'm not demanding anything.

7im wrote:

Adak wrote:

orion wrote:The object shouldn't be bringing BA down to the level of others but raising the level of others up to BA.

Just one word: No.

OK, more words:
An SMP work unit can't be given the same points per day, as a BA wu. There will always be a "jump" from SMP to BA, which is how it must be. BA rigs do significantly more work, and in a far shorter period of time. (More reliable as well), requiring greater cost and resources.

No. Cost has NEVER been a consideration to PG. Only the faster science being done.

Vijay has indeed mentioned the amount of resources needed to fold a wu, as one important factor in deciding the point value. All resources have a cost, and the largest amount of resources required right now, are the BA servers.

@Orion: OK. Now we're headed in the same direction.

Orion wrote: I would like to see PG taking with tear and Grandpa_01. Both are level headed, run BA's and have contributed allot to F@H.

I agree 101% about their contribution - both are magnificent donors to FAH. Right now, both are also clearly angered by this latest BA announcement, however. Until things calm down, I don't see any advantage for Pande Group or the moderators, if they were to join into this thread. I would guess that they would rather go swimming with Great White Sharks, than post here.