texingaAccelerating the date to eliminate lower-end Bigadv rigs...guess they must have figured "well whada we got to lose"! 

How many systems will just shut down when this happens?

cokeman54

texinga
Well, I was hoping for a different response, and at the same time was not very surprised. 
 
Accelerating the date to eliminate lower-end Bigadv rigs...guess they must have figured "well whada we got to lose"! 

Well they turned me into a bucks folder and my two 4P's are crunching away. Too much drama in FAH to be enjoyable anymore.

Sorry to hear you have stopped Folding.
Are you going to Crunch for Crunching@EVGA?

Nothing wrong with their response. Happy they responded and allow folks to make informed decisions instead of assumptive ones.
 
To me this is just part time of life. They decided what offers the biggest value to them. Now we all can do the same.

bcavnaugh

cokeman54

texinga
Well, I was hoping for a different response, and at the same time was not very surprised. 
 
Accelerating the date to eliminate lower-end Bigadv rigs...guess they must have figured "well whada we got to lose"! 

Well they turned me into a bucks folder and my two 4P's are crunching away. Too much drama in FAH to be enjoyable anymore.

Sorry to hear you have stopped Folding.
Are you going to Crunch for Crunching@EVGA?

Have been on the Crunching@EVGA since the first announcement about bigadv. Really like the look and feel of it with the plus, that you can suspend (game) resume and NOT LOSE POINTS. Will fold for bucks with one 24 core 4P until BA is over. Everything else is crunching away for EVGA. I am not going to second guess PG or worry about their motives as that has been said many times, in many ways, by many folders. They refuse to listen to donors and there are other DC projects that make you not only feel useful but welcome. That is why I chose to crunch and my points will go from over 9 million a month to 500K for the bucks. No bad feelings just found a new home ([link=mailto:crunching@EVGA]crunching@EVGA[/link])

Yeah Punchy, the different reading (for me) came from reading Kasson's reply coming on the heels of the last paragraph of Troy's excellent comments.  Troy was suggesting that they possibly delay implementing the change and then Kasson says the whole "sooner than later thing.  So, I read that "sooner than later" reply as possibly speaking to Troy's suggestion and that it had to do with the Bigadv changes.  Maybe I read that wrong too, but it seemed to connect for me.  As you said, clear communications isn't an easy thing to come by with PG.
 
I'm happily Crunching with both my SR-2 and 4P where every Core is being put to work and none of these Bigadv changes matters one bit. 

I tell ya Viper, unless I see something that indicates they actually want (or would use) Troy's analysis, I'd question if it was worthwhile.  He made a very good initial case today right there in the FF.  While Kasson acknowledged all the posts as being helpful, his reply sounded (to me) rather weak and a little dismissive.  The only glimmer was that he acknowledged not having the final say and needing to consult with others.  I'm not expecting a whole lot to change from reading their replies, but wouldn't mind having them prove otherwise either.

BA is Over inflated or SMP is under inflated or both
 
take for instance My 3930k last time was doing 130k PPD on BIGADV(no that's not my best PPD on BA) ...then to be relegated to SMP for 34-45K after the change
work that out it's a 2/3 to 3/4 cut in PPD (that's using all cores no GPU's)
there is no scaling issues because SMP is for Smaller CPU's like the 3930k - so it shouldn't be that
Bigadv shouldn't have worked that well on a 3930k then 
It's the same issue in going from a 4P system to SMP  with the PPD now
 
my belief is it's a QRB formula and benchmarking issue - QRB rewards points too quickly - and we are seeing that under GPU QRB now (huge points)
use the hockey stick analogy - the curve is TOO SHARP - so minor improvement nets huge gains
so it's an issue of speed - if you return quick you get more points (which is why a 3930k did so well)
 
BA is not benchmarked on a server setup as klasson has said- but rather a desktop - wrong way to do it
if your going to do it then benchmark on appropriate machine and try to keep BIGADV over SMP by 30-50% more - (split at 40%)
that would mean re-evaluating all points on a yearly basis - and doing rough justice to points(PPD values on current WU's going down) as newer and bigger WU's come out with more Points (SMP will be able to do bigger jobs with desktop going to 16 cores)
 
I don't know if we are ready to handle that either - but that would be an improvement - because everyone would be hit the same
and that would retain with status quo with hardware differences - equal hit to all parties - and we get new larger WU's as we go along

I might have 6months of fhm data that has both smp and bigadv data. pm me an email and ill send you the data file
 

ArtyD42

texinga

ArtyD42
How many systems will just shut down when this happens?

It's hard to say, but there will be some for sure.  Every time something like this happens there is some fallout of Folders and it is a very individual thing with people.  The main thing is that we don't allow changes like this to affect our team (Folding and Crunching).  I would encourage people to just do whatever DC work makes them happy, and that keeps them engaged.  There is lots of work to do and the  diseases just march on if we do nothing or less.

I thought about this for a while, so I just figured I'd let you know since I've made my decision on this response.
 
Our team's top 10, from 50 to 7 days status.
24/9
11/8
8/5
50/16
6/4
0/0
8/8
3/3
0/0
95/70
 
There's a direct drop in users machines on our list.
 
So I'll look at the top ranked overall.
 
awachs dropped off since 10/17/2013
Grandpa went from 24 to 13
brilong is still cranking 38s
musky has gone down from 5 to 1
Scott_H 39 to 29
-alias- 8 to 5
kennish 88 to 84
Mtnduey 52 to 48
dwdawg 9 to 7
Peter_R holding 31's up
 
There's a direct negative in active machines here.  I can't say why for each one.  But if the goal was to increase users the apparent results at this time is a net decrease from their actions.
 
http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats2
This needs a history page.

221,466 they are losing donors even with the buck folders. Down 1500 in 24 hours. http://folding.stanford.edu/home/
Most folders are computer enthusiasts and there are fewer and fewer of us. Tablets and smartphones are replacing desktops. PC manufactures know this and are moving in other directions. PG in the past has done little in donor recognition and a pattern has been set. Why should THEY change we always let issues like this die before and by their record and lack of response, we can like it or leave it. Looks like many are leaving, and I hope for other DC projects. I got my bucks (thanks EVGA) and am crunching for EVGA 24/7 on two 4P's and my main PC. They lost 25,000 and counting and if they care , they don't show it.

I would be interesting to have a broader (historical) look at the FAH "Client Statistics by OS" data.  That way you could see normal/seasonal trends and spot unusual changes.  I didn't think to take a daily snapshot of this over the past month and chart it.  For me, it is hard to know the total impact of this particular change vs what else may be happening with Folders and the associated equipment used.  It is pretty easy to see the outcry from at least people on the largest teams about this Bigadv change.  But with that being a smaller niche within Folding, how much of the recent decline at FAH is from that and how much is from other things that also occur within the Folding community?  I really don't have a good grasp on that one.  I do see more cross-team Bigadv Folders (this time) either stopping all together or reducing production on their Bigadv rigs.  I hope those that are saying they are stopping all together will find an alternate DC path for their higher-end rigs. 
 
I've been disappointed with the way that PG treats donors for a very long time.  Even after reading people say about "they are scientists", "you just need to understand them", and "be more tolerant", I've reached a threshold.  I'm not even angry, just tired of the aloof way they communicate and handle changes to the donors.  That is largely why I decided to remain Crunching after the recent contests ended.  I have always enjoyed the atmosphere there, the communications, the lack of donor shock, etc, etc.  So, I'm gonna set and just enjoy that for awhile.  I'm not against Folding at all either and wish the project lots of success (if they can hold onto their donor pool). 

Viper97
Actually 220,854 as of this morning.  There has been a steady loss of over 4K machines since Thursday of last week.  Not a good sign.

220463 at the time I checked.  This is the beginning of the freefall.

i can see them loosing a lot of support this year, they are shooting their supporters in the foot and in turn biting the hands that feed them.
 
My opinion is they are making a very bad decision but who am I. Its about as smart as making the manufacturing of light bulbs illegal... that just went into effect, it's just dumb. 
 
http://www.foxnews.com/tech/2013/12/31/end-road-for-incandescent-light-bulb/
 
My opinion is:
we fold to help them
we spend money to help them
We never hear about any of the results, nothing none...
The points mean nothing, take all our points and the money it cost us to generate them to a store... they buy you nothing.
Folding points are worthless so why do people care so much about how many points you can generate, its more like a game and people are crying cause someone dropped 1000's on a dedicated folding machine to generate more points than them... 
 
Honestly, i really have no need to aggravate my self with the crying game and if they go through with the changes. even tho my 4P makes the cut.. i'll stop folding with it possibly sell it off. its not worth the headaches, the lack of known results and the lack of respect that seems to aimed at the very people helping them out.  
 
 
 

Ahhh...data Zagen...thanks!  The data stopped at 12/31/13, but that was a good enough view (for my interests).  So, just putting together a few quick charts from that data here's what came up.  People can draw their own conclusions, but what Viper has been reporting day after day does appear to be a trend (and for longer than just the last month).
 
Looking at the "Total CPUs" from Column AR in that Google sheet (11/28/12 thru 12/31/13), here's an interesting chart.

 
And here's a look at just the "Active Linux CPUs" from Column U of the Google sheet.  I graphed the long look and then also graphed a look just at December 2013.  The Bigadv change announcement was made on 12/17/13 in the post by Dr. Kasson.  Looks like there has been a 364 reduction in "Active Linux CPUs" just since the announcement was made.
 

 

Now it's 218,770 in the last nine hours they have lost 1,693 machines, hosts etc.  Not a good trend, tomorrow should be funny.

Then if I'm understanding correctly and remember I'm old and seniorile'd, we have a lag possibly of 50 days from what looks like the start of a decline to the current state. 
 
That possibly means that the trend is now just showing up and the real fallout is still a few weeks away.