Folding Forum

Jester wrote:All too late sadly,
I'd guess a lot of those "alternative formula" posts were not looked into in great detail anyway,
Once a tough choice is made I can't see any avenue for modifications that wouldn't cause more anger on one or both sides,

I have no stake in this, since I'm not folding bonus-eligible WUs anyway. Tried my damnedest to come up with a "soft landing" formula for a cap nevertheless, something that could be adjusted on a project basis and wouldn't rock the cradle too much. But like you pointed out... "A day late, a dollar short" doesn't cut it.

GreyWhiskers, if you have dabbled into overclocking, I think you are ready to start building your own machines. When you build your own and launch it at one of those vicious, dangerous, protein molecules, it's just such a joy to watch the battle.

FlipBack wrote:

GreyWhiskers wrote: b. Given all that, my own personal feeling is that the points structure is relative - I haven't gotten emotionally attached to the the absolute numbers. Going forward from here, being able to complete a P6901 in 50 hours will still give me more points that someone who takes 72 hours to complete the WU, and fewer points than someone who can complete it in 24 hours. The bigger the iron brought to the party, the higher the points awarded, still. The "hierarchy" of the numbers is still the same - but the absolute numbers are different.

Agree 100%. I was going to bring up this point earlier but I didn't get it to make sense. I couldn't have said it better.

Again, taking each project in isolation I can find nothing to not agree with either, any changes in value affect everyone equally,
where things start to get "murky" is when some (like me) don't have an infinite supply of power, and due to significant increases don't
have the regular funds to provide it, so to maximise contributions it's a very carefully thought out "juggjing act" between various hardware, projects and
power consumption, within that self imposed power budget until the current changes were made it was leaving 4 good Gpu's idle and running an extra 970
Bigadv folding rig, that maximised my contributions within that limit,
Now it's swung more in favour of shutting down the extra 970 rig and running 4 x Gpu clients, if the rumoured QRB is then applied to the Gpu clients it will swing
further in favour of Gpu folding, and still, within the Gpu project itself any changes there will be equal to all running the Gpu client....
So "where's the problem ?" you may rightly ask,
Any idea how much that now almost redundant brand new high end fully watercooled 970 cost to build less than 2 months ago ?
I can fully understand why these changes don't seem overly important to some, but I hope you can now see it a little more clearly from my side too...

Leonardo wrote:GreyWhiskers, if you have dabbled into overclocking, I think you are ready to start building your own machines. When you build your own and launch it at one of those vicious, dangerous, protein molecules, it's just such a joy to watch the battle.

Indeed,
there's nothing like building a new Folding rig,
carefully choose all the right parts,
assemble with the utmost of care,
install the O/S and enjoy your new creation humming away,
Then overclock it 'till it bleeds,
then back it off to 100% Folding stable,
and finally release it on an unsuspecting Wu.....

Pandora's box once opened... This is a huge project, which significantly complicates all issues. Of course everyone is behind the science in attacking these diseases. Folks crunch for many reasons. A great number are in it in much greater fervor due to the competition aspect for their "team" and also their personal placement. So in that regard the huge point units are both a blessing and a curse. The nature of the beast is that the smaller point units and contributors have a much smaller impact in the team points game and it becomes a game between the "haves" that have the resources (private or corporate) to front the heavy hardware to wrack up the huge points. So in that aspect the fun part of the team game goes out the door. Maybe the project should have been split so there was the plain vanilla F@H, and then F@H BigAdv. Of course then I imagine that ego's would be crushed too. So it's a no-win situation for PG.

The other problem here that isn't out in the open is what the "considering renormalizing" other aspects of the project will impact and entail. So now folks from much more than the bigadv crowd are also left wondering how that will impact them and their hardware and budgets. I had been contemplating hardware purchases for bigadv, which will now go on hold, as well also anything further in the GPU direction. So the result of all this is that the project will probably see some reductions in output from more than just the bigadv group. Reality is that this is not the only very worthwhile project in the DC community. Some folks will re-distribute their hardware to other efforts. It is all very unfortunate as there is going to be re-evaluation on the part of more than just PG due to the manner in which this whole matter has been handled. And maybe all the drama could not be avoided no matter how it was handled, hard to tell. But no matter what, for a number of folks, it isn't going to be as "shiny" or fun anymore.

Hi Brucifer, I haven't seen you around in a long time. How's things going?

Do you still hear from Caveman Basser?

For what it's worth (not much) I do not mind the change. I use the resources I have at hand to fold. I take whatever points happen to be assigned me.

I do wish there was a linux-native GPU client, though.

Do I like this decision, "NO", but you have to remember it all about the science anyway.

angra wrote:For what it's worth (not much) I do not mind the change. I use the resources I have at hand to fold. I take whatever points happen to be assigned me.

I do wish there was a linux-native GPU client, though.

Meaning to reinforce, rather than to contradict:

To put this in to a personal perspective. My wife is suffering from multiple cancers. She has chronic pain that adds to her misery. I fold for her and for all of your loved ones.

Points? Why should you care?

Recently built 2 X 2600K systems running Linux for bigadv folding. Even with the latest points reduction each is still 4 times as efficient as my GTX 460 (PPD/unit cost of electricity) unless there are other changes to the points system that is!!

This is in a nutshell my problem.

I understand the reasoning for the change but I now have no idea what I should consider for future FAH investment (if anything now).

How can one make sensible plans when the goal posts are being moved?

I very much agree with the points phoenicis raised in his posts.

Surely to aid the science (which is why we fold) Stanford needs to set out a clear path for the points system and then STICK TO IT.

Well, I made this suggestion at the end of GPU1.

Here's another go:

How about Predefined mileposts and review schedules

If you release a "beta" program, the expectation is that there is some plan, some method of evaluation that will lead to it either being adopted or rejected.

While PG may have such a plan, donors are left to guess.

If a formal review at 3 and six months were published on the bigadv and QRB, donors would have know if they could have made purchasing decisions based on some expectation of continuity.

Leaving donors to guess for 2 years, and then claiming to

making hard calls when we feel it is necessary.

is just irresponsible.

If there is no plan to change the structure, the documentation, the review and communication process; this unpleasant situation is guaranteed to repeat itself. :twisted:

Ultimately it is not the final decisions that are flawed, but the process by which they are reached.

Anyone can read the thread where this was debated, but even now after the fact, no donor knows the factors, the numbers, the equations, or the statistics that went into PG "re-normalizing"

7im, MtM, and I can continue to debate the time value of science ad nauseam. PG still hasn't given any direction on how they determined the rebalanced.

Based on this: how can anyone hope to anticipate future changes?

The whole process is in fact capricious and arbitrary.

Simply stating that

Finally, I would like to remind all that we do listen to donors and take their input very seriously.

does nothing to rectify this situation.

Without the ability to audit if not take part in the review process, donors are left to simply trust to luck in their planning.

I have been called a forum troll dozens of time on this forum. However, I have predicted and warned about this for years.

If donors don't demand this change, they really can't expect any different results in the future.

.
Hear, hear!
.

The PPD for bigadv were plain and simple out of line and needed to be adjusted, as VJ said they are in a tough spot and had to make a decision one way or the other. I completely agree with the decision thing were out of line and I believe there still should be more cuts to the bigadv folding. If you want to reward the so called big guns out there then the deadlines need to be shortened on the 6903's and 6904 so that those individuals that are running them on hardware that they were not intended for no longer have a reason to hack the system and run them. I believe allot of the big guns are not able to get the new big bigadv units because they are being run on 2600k systems, so what is the incentive going to be to encourage the so called elitist folders to build these so called elitist rigs. ? Put a preferred deadline that is reasonable, why is there a 7 day preferred deadline on them your 48 core machines are folding the 6903's in less than 24 hours I believe the SR2 rigs are also folding them in a day reward them and take the initiative away from the the hackers and you might find a few more satisfied customers out there. I really do not care if I am able to fold the new big bigadv WU's with my hex rigs, or if the hackers are able to. But I do believe you could get rid of some of this unrest if you reward the different classes of hardware appropriately.

Re: Bigadv points change

Postby VijayPande » Sat Jul 02, 2011 4:07 pm
This one was a tough call. After much donor concern on this issue, we did our own research into the matter which showed that bigadv was out of balance. However, we know that any sort of points change will be unpopular. This puts us in a very tough position. If we do nothing, we're ignoring the majority of donors (i.e. "Pande Group doesn't care.") and going against our own data, in favor of the high end FAH donors (and some have called FAH "elitist" because of this). If we make the change that's right based on our own numbers (i.e. the change we've just made), we get the bigadv donors naturally upset, which obviously is very bad too. No way to win here.

The PPD for bigadv were plain and simple out of line and needed to be adjusted, as VJ said they are in a tough spot and had to make a decision one way or the other.

If you recall, I argued for a reduction of the base points instead of a capping of the upper limit.

The final result to date is not far from what I argued for in regard to a possible solution.

My point is that the process is still flawed.

Rather than assuming

Moreover, with points, there will never be any system which makes everyone happy,

perhaps some thought should be given to changing the process.

Process Quality control 101

You don't attack the variable outcomes, you attack the process that leads to them.

So, when the Bulldozer chips come out and people start complaining that 4p Bulldozers are scoring "too many" points, will all of bigadv be re-normalized again, again penalising everyone who runs bigadv?

This reduction in base points has not changed the fact that as tpf decreases on the high end machines, the bonus increases exponentially. Rather than tinker with the points, just fix the whole system once and for all.

H.