Change in BA requirements

Moderators: Site Moderators, FAHC Science Team

Locked
Viper97
Posts: 38
Joined: Thu Apr 18, 2013 8:17 pm

Re: Change in BA requirements

Post by Viper97 »

I'm folding SMP now for a member of our forum on my BA machine. He wants some measurements of SMP VS BA points as a way of perhaps presenting a solution that might benefit everyone folding.

The problem is I see it as a wasted exercise. I believe the course has already been set and the rest is all lip service.

Oh well... it was fun and costly and a great way to feel good about helping out humanity. Then when you pull the curtain back all I see is the ruins of a once great idea. I firmly believe it is time for FAH to become FAC (Folding @ Corporations).

I suspect the donors are more upset than is being realized here and that reflection has resulted in more machines being taken off line since yesterday. We are now down to 222,988 from 224,908 from yesterday... 1920 lost according to todays official stats.
Macaholic
Site Moderator
Posts: 811
Joined: Thu Nov 29, 2007 11:57 pm
Location: 1 Infinite Loop

Re: Change in BA requirements

Post by Macaholic »

sbinh wrote:P.S: I might be BANNED from this forum... :D :D
It is unfortunate, that seems to be the goal of several in this thread. I am quite certain that personal respect, professionalism, sound arguments, and sensible discussion will gain an appropriate reply in due time. The rest is just distracting noise.
Fold! It does a body good!™
Grandpa_01
Posts: 1122
Joined: Wed Mar 04, 2009 7:36 am
Hardware configuration: 3 - Supermicro H8QGi-F AMD MC 6174=144 cores 2.5Ghz, 96GB G.Skill DDR3 1333Mhz Ubuntu 10.10
2 - Asus P6X58D-E i7 980X 4.4Ghz 6GB DDR3 2000 A-Data 64GB SSD Ubuntu 10.10
1 - Asus Rampage Gene III 17 970 4.3Ghz DDR3 2000 2-500GB Segate 7200.11 0-Raid Ubuntu 10.10
1 - Asus G73JH Laptop i7 740QM 1.86Ghz ATI 5870M

Re: Change in BA requirements

Post by Grandpa_01 »

Adak

You missed it I was not comparing the two I was stating that the points incentive works for bigadv, How is it possible that with my boxen I can be the #1 folder / cruncher on 2 different boinc projects in the same day. Boinc clearly does not have the attraction of fah for this class of machine, they are not even set up to utilize the potential of this class of machine, the point is if you keep cutting bigadv then fah will have no attraction for this type of machine either.
Bigadv was designed for the top end of the market as the market moves so does the requirements. If there is no incentive to move with it then it will just go away. Right now there are many machines running bigadv that have been out dated, that is part of the risk of running them the rigs become out dated very rapidly comparatively to other machines.

Everybody knows this they just do not want it to happen to their investment but it was stated when bigadv was started. The thing that was not given though was a road map which was / is wrong a person should have some idea of how long there investment is going to be good for before they make that investment, with no roadmap then people tend to believe there investment is going to be good forever.

The only reason I believe SMP is undervalued is because of the supposed backlog (Stanford has never confirmed this) I am pretty sure that technically there is a backlog of both bigadv and smp work to be done. As far as scientific value goes the points are correct as far as incentive value goes they are wrong. Stanford has made a mistake in using an incentive program as a scientific value program which is fine until you get an imbalance in the type of work that needs to be done and the one that has the most work also has the least value (what do you expect)

Any way I believe bigadv and GPU should be left alone and SMP needs a bump whether temporary or permanent is irrelevant, a points system is supposed to guide the flow of work (the proverbial carrot) if you put the carrot out in front of the mule he will follow the carrot. The lowering of bigadv points values will only lower the class of machine running them and defeat there original purpose, if the return is not there why would a person invest in it.
If you go back to the inception of bigadv it was only intended for a very small % of donors to run them and the amount of points granted was set to entice that % to run them.
Image
2 - SM H8QGi-F AMD 6xxx=112 cores @ 3.2 & 3.9Ghz
5 - SM X9QRI-f+ Intel 4650 = 320 cores @ 3.15Ghz
2 - I7 980X 4.4Ghz 2-GTX680
1 - 2700k 4.4Ghz GTX680
Total = 464 cores folding
orion
Posts: 135
Joined: Sun Dec 02, 2007 12:45 pm
Hardware configuration: 4p/4 MC ES @ 3.0GHz/32GB
4p/4x6128 @ 2.47GHz/32GB
2p/2 IL ES @ 2.7GHz/16GB
1p/8150/8GB
1p/1090T/4GB
Location: neither here nor there

Re: Change in BA requirements

Post by orion »

Macaholic wrote:
sbinh wrote:P.S: I might be BANNED from this forum... :D :D
It is unfortunate, that seems to be the goal of several in this thread. I am quite certain that personal respect, professionalism, sound arguments, and sensible discussion will gain an appropriate reply in due time. The rest is just distracting noise.
You call it the way you see it.
iustus quia...
mdk777
Posts: 480
Joined: Fri Dec 21, 2007 4:12 am

Re: Change in BA requirements

Post by mdk777 »

will gain an appropriate reply in due time.
Well lets see...I've posted several direct quotes from 3 + years ago.

I guess we are all entitled to our opinions on what is "appropriate" and "in due time".
Transparency and Accountability, the necessary foundation of any great endeavor!
tear
Posts: 254
Joined: Sun Dec 02, 2007 4:08 am
Hardware configuration: None
Location: Rocky Mountains

Re: Change in BA requirements

Post by tear »

Macaholic wrote:
sbinh wrote:P.S: I might be BANNED from this forum... :D :D
It is unfortunate, that seems to be the goal of several in this thread. I am quite certain that personal respect, professionalism, sound arguments, and sensible discussion will gain an appropriate reply in due time. The rest is just distracting noise.
The only thing that's unfortunate is that people posting here are aware of questionable standards of FF administration (which cause their concern and reservation).

P.S. This post has been preemptively recorded so evidence of tampering, if any, can be provided.
One man's ceiling is another man's floor.
Image
mdk777
Posts: 480
Joined: Fri Dec 21, 2007 4:12 am

Re: Change in BA requirements

Post by mdk777 »

PS, not afraid of my record.

Last post in the thread mentioned.

http://www.amdzone.com/phpbb3/viewtopic ... 30#p226689
I feel your pain...like money down a rat-hole. :wink:
Over the last few years, I've been tempted to join in again...but finances and a healthy skepticism have always won out. :lol:

Oh well, some day they will help find a real cure and we will say we were there. :mrgreen:

Took my son with my dad to Williamsburgh recently. They are making a replication of Washington's campaign Tent. It really is nothing to look at really. Your average high school graduation party lawn party has bigger and nicer examples.

However, if you managed to survive the war, if you managed to endure the years of deprivation and trails...seeing that tent on tour after the war had to be a pretty emotional experience. :wink:

Anyway, hopefully we can look back at our donation some day with similar pride. :mrgreen:
So, there you have it.
:!:

Amazement that donors contribution is so quickly and with so little thought and concern debased...followed with hope.

I'll wait to see which wins out.
Transparency and Accountability, the necessary foundation of any great endeavor!
Punchy
Posts: 125
Joined: Fri Feb 19, 2010 1:49 am

Re: Change in BA requirements

Post by Punchy »

I just thought of another way an economist can help the project (besides my previous suggestion of helping balance points between projects using economic concepts). An economist might actually be able to help PG put a real time value on research. As far as I know, the bonus calculations were basically a swag at rewarding faster returns and have no actual correlation to the scientific value of those faster returns.

Say, for example, the typical research project takes 1 year of simulations followed by 1 year of data analysis and paper-writing. Say you cut the research time to 6 months; is the research worth 4 times as much per unit, when you are reducing the overall project time by only 25%? If an infinitely fast machine can complete the research in 0 time, you have only shortened the overall research cycle by 50%. Does a 50% reduction in time call for an infinitely higher value for the research?

His/her conclusions from that research should lead to a points system with an actual foundation, in turn leading to far less argument among donors. It might skew the values in entirely new ways that are likely to antagonize some subset of folders, but if it has a valid foundation, then PG could make the change with the knowledge that it was providing the right incentives to get the most research done.

The economist could also take a look at some simple sanity checks, such as whether a machine making 1 million PPD (I hear with the right hardware and WU this is possible) is generating the same amount of research value as 40 2600K systems (using a 25,000ppd number from a few pages back).
Spazturtle
Posts: 17
Joined: Sun Nov 04, 2012 2:42 am

Re: Change in BA requirements

Post by Spazturtle »

A good point system would be very simply.

10 point for each ns/day of folding done.

A 7970 gets around 40 ns/day of folding done, so it would have a ppd of 400

You could easily apply this retrospectively by looking at how many ns each WU covers.

That would give us a point system that perfectly matches how much science gets done.

Then we could simply buy the best hardware for the job.
Nathan_P
Posts: 1164
Joined: Wed Apr 01, 2009 9:22 pm
Hardware configuration: Asus Z8NA D6C, 2 x5670@3.2 Ghz, , 12gb Ram, GTX 980ti, AX650 PSU, win 10 (daily use)

Asus Z87 WS, Xeon E3-1230L v3, 8gb ram, KFA GTX 1080, EVGA 750ti , AX760 PSU, Mint 18.2 OS

Not currently folding
Asus Z9PE- D8 WS, 2 E5-2665@2.3 Ghz, 16Gb 1.35v Ram, Ubuntu (Fold only)
Asus Z9PA, 2 Ivy 12 core, 16gb Ram, H folding appliance (fold only)
Location: Jersey, Channel islands

Re: Change in BA requirements

Post by Nathan_P »

Macaholic wrote:
sbinh wrote:P.S: I might be BANNED from this forum... :D :D
It is unfortunate, that seems to be the goal of several in this thread. I am quite certain that personal respect, professionalism, sound arguments, and sensible discussion will gain an appropriate reply in due time. The rest is just distracting noise.
Bruce asked that discussion on here be kept civil and within the TOS otherwise the thread would be closed, on the whole we have done so. These conditions do not apply to other forums with different TOS. If venting is allowed we Vent :evil:
Image
Rattledagger
Posts: 128
Joined: Thu Dec 06, 2007 9:48 pm
Location: Norway

Re: Change in BA requirements

Post by Rattledagger »

Spazturtle wrote:A good point system would be very simply.

10 point for each ns/day of folding done.
This point-system doesn't work, for a very obvious reason you've overlooked, and this is ns/day among many things depends on #atoms in the molecule being similated.

Let's make a very simple example, let's use a model there only one type of atom-atom is counted, and this is pair-vise.
If you've got a 10-atom molecule, this would mean, for each time-step you'll need to calculate:
9 interactions for 1st. atom, 8 for 2nd. atom, 7 for 3rd. atom, 6 for 4th. atom, ... for a total of 9 + 8 + 7 + 6 + 5 + 4 + 3 + 2 + 1 = 9 * 10 / 2 = 45

If eact time-step simulates one micro-second, and each pair-vise interaction takes 1 milli-second, for this 10-atom molecule one nano-second would take 45 seconds ==> 1920 ns/day simulated.

If you've got a 100-atom molecule on the other hand, you'll need to do... 99 * 100 / 2 = 4950 calculations per time-step ==> 4950 seconds per nano-second ==> 17.45 ns/day simulated.
If you've got a 1000-atom molecule you'll need 999 * 1000 / 2 = 499500 calculations per time-step ==> 5.78 days/ns simulated ==> 0.17 ns/day simulated.

As you can see, your suggested point-system would severely discourage users from running any wu the more atoms a molecule contains, while in reality FAH often would want to run large molecules.

And no, it's not just #atoms being the deciding factor, so just adjusting for #atoms wouldn't work either.
Punchy
Posts: 125
Joined: Fri Feb 19, 2010 1:49 am

Re: Change in BA requirements

Post by Punchy »

Let's not go down the rathole of points as they relate to atoms, ns, etc, as it has been beaten to death in other threads and other forums. My suggestion had to do with asking an economist to evaluate the actual time value of research.
Grandpa_01
Posts: 1122
Joined: Wed Mar 04, 2009 7:36 am
Hardware configuration: 3 - Supermicro H8QGi-F AMD MC 6174=144 cores 2.5Ghz, 96GB G.Skill DDR3 1333Mhz Ubuntu 10.10
2 - Asus P6X58D-E i7 980X 4.4Ghz 6GB DDR3 2000 A-Data 64GB SSD Ubuntu 10.10
1 - Asus Rampage Gene III 17 970 4.3Ghz DDR3 2000 2-500GB Segate 7200.11 0-Raid Ubuntu 10.10
1 - Asus G73JH Laptop i7 740QM 1.86Ghz ATI 5870M

Re: Change in BA requirements

Post by Grandpa_01 »

Macaholic wrote:
sbinh wrote:P.S: I might be BANNED from this forum... :D :D
It is unfortunate, that seems to be the goal of several in this thread. I am quite certain that personal respect, professionalism, sound arguments, and sensible discussion will gain an appropriate reply in due time. The rest is just distracting noise.
It is interesting that you did not include a link to the largest threads out there about the bigadv smp problem, why is that? Could it be that you are affiliated or ........ Just one more example of mod abuse. If you are going to share 1 share them all. There are a lot of unhappy people out there right now and many of them are just as unhappy with this forum as they are with the points system. :ewink:

I wonder if I need a disclaimer here:
Image
2 - SM H8QGi-F AMD 6xxx=112 cores @ 3.2 & 3.9Ghz
5 - SM X9QRI-f+ Intel 4650 = 320 cores @ 3.15Ghz
2 - I7 980X 4.4Ghz 2-GTX680
1 - 2700k 4.4Ghz GTX680
Total = 464 cores folding
Bill1024
Posts: 75
Joined: Mon Jun 30, 2008 2:45 am

Re: Change in BA requirements

Post by Bill1024 »

Adak wrote:
Bill1024 wrote:I only have a 9th grade education, I am not the sharpest knife in the chandelier. After 30 pages, I still can not figure out what the problem is in the first place.
Why are they changing anything at all? 8/16 core Intel desktops will not be out until Q3 or Q4 I have learned.
PG knew that the requirements for BA work units would need to change, before long. They said that when they brought BA out, and they repeated that alert when they increased the threshold for BA, the last time. (over a year ago)

I don't know the ins and outs of the protein folding field, but I'm sure that PG knows it, and these changes are what they want or need. Some faith is required it seems, even dealing with a science project.
Bill1024 wrote: But why, is there a backlog of SMP or not?
I'm not sure. Whether there is a backlog or not, we need to adjust the point system for the BA PC's that will be cut out of BA, in a few months. We don't want them feeling like they're not a vital part of FAH anymore, because they are. Their ppd should reflect that.
Bill1024 wrote: What exactly does move 16 and 24 core server to "enhance SMP" mean? Enhance as to make better? To fold more of?
Looks like -smp 24 or -smp 48 is not even folding the WUs that there is 180,000 of. Breaking it down to 4+8+12+24 made it so too many cores were left idle and the PPD or TPF was really out of line.
Enhanced smp would be a new smp category, that would fit between regular SMP and BA. It's just an idea at this time. Doubt if it will be adopted.
Bill1024 wrote: It's their project and they're going to do what they need to do, I understand that. But the posts are somewhat vague and not to the point.
Too much reading between the lines, guessing what PG is saying. Going to do exactly what "sooner rather than later".
I am not all that bright, I need clarity, say exactly what you mean in simple terms a simple man can understand.
With all due respect, say what you mean, mean what you say, say it so all of us understand exactly whats going on.
Please.
Dr. Kasson was just letting us know that they had heard the complaints and suggestions in this thread, and will work on it soon. Nothing is being proposed as an answer by anyone at PG, yet. It is common for academics to take some vacation days in late December to early January, since the University is closed for classes, at that time.

These things take time. Be patient. Clarity and simplification of all the above, will arrive, but not right away.
Adak you are the new relations person PG hired?

1: I know they said "things will change from time to time" ..................I asked why this time?....If you do not know don't answer.
2:I don't know the ins and outs of the protein folding field, but I'm sure that PG knows it..............If you do not know don't answer.
3: I'm not sure. Whether there is a backlog or not,..........................................................If you don't know don't answer.
4:Enhanced smp would be a new smp category, that would fit between regular SMP and BA............Where did you get that from? Reading between the lines? I do not see that in 30 pages., Not clearly anyway.
5:Dr. Kasson was just letting us know that they had heard the complaints and suggestions in this thread, and will work on it soon........ Not everyone on a few forums see it that way. Please let PG answer.
6: These things take time. Be patient. Clarity and simplification of all the above, will arrive, but not right away...... When then, Christmas is coming you know.
Macaholic
Site Moderator
Posts: 811
Joined: Thu Nov 29, 2007 11:57 pm
Location: 1 Infinite Loop

Re: Change in BA requirements

Post by Macaholic »

Grandpa_01 wrote:It is interesting that you did not include a link to the largest threads out there about the bigadv smp problem, why is that?
Did not need to, really, since all of the looking for a "shiny badge of honor" members (i.e. PS - I might get banned) were in those two threads on those two particular forums. How one acts on their home forum is usually indicative of how one acts on a neutral help forum.
Grandpa_01 wrote:Could it be that you are affiliated or ........ Just one more example of mod abuse. If you are going to share 1 share them all.
Since you asked, please visit here, here, here, and here, if it will help clear the air. Nothing found on MaxPC, OCAU, HWC, Team MacOS X, etc.
Grandpa_01 wrote:There are a lot of unhappy people out there right now and many of them are just as unhappy with this forum as they are with the points system.
Yes, they are all linked in the two threads in my previous post and 95% happen to be from Teams #1 and #2 on the points list. I get it, understandable, and I think you understand my point as well. Let's just end it there as far as OUR dialogue. Now continue with ideas on how to improve the points system and communication coming from the Pande Group with regards to future changes in a positive and professional manner. Thank you.
Fold! It does a body good!™
Locked