new projects

Hmmmm..... be interesting to see how the adjustment goes. Maybe Dr. Kasson might have a reality check.
 
yep odds are 1 of several diffrent seneraios will play out.

#1 A low % of capable rigs choose to fold them and or not enough people choose to upgrade to machines that are capable of running them the PPD will go up.

#2 Enough people choose to run them and upgrade an all will remain the same.

#3 Enough people choose to run them and upgrade but there is allot of public outcry about too many points. They will get nerfed again.

One thing we all need to remember is the bigadv WU's are a viable and important part of folding from the way I understand it the larger the chunk of a protein that is folded the more accurate it is. (this could be wrong I am no molecular biologist) PG is coming out with new bigadv projects so that in itself says they do not plan on ending the bigadv projects and they are important enough to offer a greater bonus to try and encourage people to build machines to run them. I would not want to be in Stanford's shoes and having to deal with John Q Public all of the time on this issue but I do believe the PPD system is a great motivator and most likely well worth the trouble it creates.

I do believe Stanford is trying to find a happy medium where some people do not run around butt hurt all of the time because there production is not keeping up with the Joneses and feel devalued. I believe most people really do not care one way or the other they just want to feel they are doing something useful. But at the same time there needs to be a large enough incentive to get people to build the rigs needed to do the job.

When it comes to folding there just simply are not enough computers running folding software to do the job as quickly as Stanford would like it done of any class of WU. Some say there was too many computers running bigadv WU's a while back and there was a shortage of bigadv WU's. This just is not true there has never been a shortage of bigadv WU's or enough computers to run them, there was just a bigger over supply of smp WU's at the time, and there was also a server code issue going on at the same time.

Anyway I myself believe Stanford may have done the right thing here by correcting a discrimination between the >12 and the <12 machines where they created a greater value for 1 machine doing the same work as another machine with the recent 8101 move fiasco. If that turns out to be indeed what they have done, (equalised the PPD across the board for the bigadv folders). ;)

Now if they would just give the smp a little loving :D
 
Grandpa_01 said:
...(equalised the PPD across the board for the bigadv folders). ;)....
Click to expand...

You mean please the people who want bigadv points without spending the money.

I also wonder what the ratio of 6905 WUs to 8101/8102 WUs will be.
 
Do not know the answer to that question yet hopefully they are not still penalising one group or the other any more. If they are they will still have some work yo do. ;)
 
Hmmm. k factors of 6905 and 6906 similar to 6903 and 6904.. so is the atom counts. That may translate into a plus for you guys if they are available to anything more than 12 thread rigs.

Noticed the 8102 not listed on the project summary page just yet:
http://fah-web.stanford.edu/psummary.html
 
Nathan_P said:
Just found this over at FF.

New bigadv/bigbeta projects.

Also note the comments at the bottom about reevaluating bigadv again at the end of the year.
Click to expand...

So help a noob out, what does k-factor mean as it relates to PPD? Please explain in detail, I looked at the 3 projects and have trouble relating to what it all means. Thanks,
Best regards, Charlie
 
The 'K factor' is used to calculate bonus points for a WU turned in before the preferred deadline. Time to process the WU is the other factor. Bonus points are then added to the base points.
 
For more detail, read here:

Q:How are you awarding bonuses?
A: Bonuses are awarded according to the following formula:
Total points = base points * bonus factor

The bonus factor is computed based on the time from when our server issues the work unit to when it receives the work unit (WU_time), the time from work issuance to when the deadline would expire (deadline_time), the time from issuance to when the work unit times out and is marked for reissuance (timeout_time), and a constant factor k.
If WU_time > timeout time, bonus factor = 1.
If WU_time <= timeout time, bonus factor = sqrt(deadline_time * k / WU_time)

For an even more complete explaination, read here:
http://foldingforum.org/viewtopic.php?p=105038#p105038
 
Anyone know if these require V7 ...past few units have just been 8101s...no 8102s or other new guys.
 
Guess I will see if I can get v7 to install, try it out and see.

Edit
Can not get F@H v7 to install on Ubuntu 10.10 so unless somebody fixes the installer it would not matter for me anyway. :mad:
 
Last edited:
Got one of the new 8102, looks better than its predecessor, yet only 1% with TPF 14:42 on my Interlagos 6272, but at 2%, we get the truth. I come back with an update sun.

Edit: Update on 8102, very god in compare to 8101, TPF is 11:54, that is 140K better PPD then 8101 on my Interlagos 6272,what do you think boys?;)

After 3% we are down to TPF 11:51, and the PPD is as good as 424K. And 8102 is no visable in the Project Summary http://fah-web.stanford.edu/cgi-bin/fahproject.overusingIPswillbebanned?p=8102

Edit on 4%: Getting even better, TPF is now only 11.47, that means the entire almost 428K PPD. I think Stanford has done the right thing this time!

Uploaded with ImageShack.us

After 6% the TPF is down to 11:44 and 431PPD.

Edit: 11:13 Sunday 29.07.2012 TPF is now 11:39 and 436K PPD after 49% so this is looking good!
 
Last edited:
One of my 4P boxes picked up an 8102 be interesting to see how it does.
 
this 8102 is a nice change, only 3% complete so far but estimated PPD is 556K up from the 360K this same box does on 8101
 
I got an 8102 using bigadv. It's getting 418K PPD (13% complete), up from 252K on 8101.
 
This is good news! I was beginning to think that 4P's were going to be left with only 8101 and that it might not be quite worthwhile to build one now... but the preliminary news of the 8102 sounds promising! Guess my 4P plans are back on the table.
 
Carbon_Rod said:
This is good news! I was beginning to think that 4P's were going to be left with only 8101 and that it might not be quite worthwhile to build one now... but the preliminary news of the 8102 sounds promising! Guess my 4P plans are back on the table.
Click to expand...

Bear in mind these are still beta projects - they can adjust the points at any time.

Does anyone have any SR-2 tpf's for these yet?
 
402blownstroker said:
What is the machine setup and TPF? These new project have yet to impress me....
Click to expand...

I am currently folding another 8102 (33% complete). Following info is from HFM.net.

TPF = 11m 55s
Credit = 348,157
PPD = 420,710

This is on the rig in my sig, 4x 6168 @ 1.9 GHz, Tyan S8812 mobo.
 
Anyone else getting any of these new ones? I've got one 48 core on bigadv and one on bigbeta trying to get them but just getting 8101's
 
I got 2 8102 wu's over the weekend, but since then, all I have gotten is 8101 wu's.
 


Getting 8102s only on 6180 rig.
Mixed 8101/8102 on 6166 rig #1.
8101 only on 6166 rig #2.
Running -bigbeta on all three.
 
I have one of my 2P and one 4P box getting 8102's every time now. All my other 2P and 4Ps are still getting 8101's.
 
One of my 2p ([email protected]) is crunching 8102 @ TFp=17:03 ==> PPD=245738 ... while the another works on 8101 (X5649@4ghz) with TFP=21:54 ==> PPD = 169464
 
Nope no 8102s for me. :( yet.... Usually takes a month or so before I get new wu's
 
8101...8101...8101............ you get the idea..
I think the 8102's are by special invitation only.:rolleyes:
 
I don't think they have ENOUGH new BA WUs for everyone .. so whoever connects to AS at right time would get one :D ... I got only 1 so far .. and it seems pretty good. Don't know what would the TPF would be on 4p rigs .... Could be 8-9minutes/frame on 6172?
 
Last 10 WU: 8101/6901 x 6/8101 x3

Haven't seen one of those tasty 8102s yet, running 4P 6176SE @ 230/16 with BB, U10.10 v6.34 client.
 
3rd 8102 WU on my 6166HE @2.0Ghz TPF 11.55 for 338,224 points. on SLOW DSL
I have got to upgrade to FIOS asap.

edit"
finally got a 8101 on my 4P 6180, have only been getting smp on that machine for last 3 days
 
Last edited:
You must log in or register to reply here.
Back
Top