Results 1 to 40 of 40

Thread: this sucks

  1. #1
    Member
    Join Date
    Jul 2003
    Location
    Home of the 2010 Olympics
    Posts
    92

    this sucks

    i just wonder how many people are just fed up with the "toughness" and lack of speed of this new update. personally, i'm not even going to bother spending electricity $$$ on this junk. its a huge mistake going from a fast, small set to this huge, slow monster. at least they should have weaned us into this slow work, but no, just shocked us into this and here's the fallout. its been fun...buh bye till next time.

  2. #2
    OCworkbench Stats Ho
    Join Date
    Jan 2003
    Posts
    519
    I do not think this is the slowest we have had, I remember 1 protein where I repainted the house while the 1st Gen was cruching
    I am not a Stats Ho, it is just more satisfying to see that my numbers are better than yours.

  3. #3
    Senior Member wirthi's Avatar
    Join Date
    Apr 2002
    Location
    Pasching.AT.EU
    Posts
    820
    What's so bad about a slow protein? I know: you get less points in stats. But if that's the only reason you do distributed computing ... (hell, I know, it IS one of the reasons I do DC, but not the only one)

    It is important for the project to do proteins of different style. Some of them are smaller, some are bigger. Wait some time and there will be a smaller one.

  4. #4
    25/25Mbit is nearly enough :p pointwood's Avatar
    Join Date
    Dec 2001
    Location
    Denmark
    Posts
    831
    We'll get faster ones again

    Also remember that we're all crunching on the same protein so no one get an advantage.
    Pointwood
    Jabber ID: pointwood@jabber.shd.dk
    irc.arstechnica.com, #distributed

  5. #5
    Junior Member Trog's Avatar
    Join Date
    Sep 2003
    Location
    Bedfordshire UK
    Posts
    5
    In the interest of keeping every one happy, and providing the entertainment that makes folding fun (rather than just worthwhile).

    How about putting a function into the scoring equation on the lines of *(current protein size/average protein size)

    So that the points given approximate to the computer effort expended, rather than the number of folds done regardless of how easy/difficult they were to do.
    Athlon XP3000, ABIT NF7-S, Swiftech MC 462, 512Mb Kingston PC3500, 5x WD SE HDD's = 1.16Tb, ATI A in W Radeon 8500DV, Win TV Nova T, DSL-504, Win 2K.

  6. #6
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137
    when people drop out on slow proteins it gives us tortoises time to overtake the hares

    its been this way since DF started, i like to think of it as a positive "feature" that differentiates DF from the other DC projects.

    i like it this way, makes for more strategy than other DC projects
    Use the right tool for the right job!

  7. #7
    Member
    Join Date
    Jul 2003
    Location
    Home of the 2010 Olympics
    Posts
    92
    i don't care. electricity isn't free here. this is so slow, i'll easily make up any ground i lose(laughs) on the next update.

  8. #8
    Ancient Programmer Paratima's Avatar
    Join Date
    Dec 2001
    Location
    West Central Florida
    Posts
    3,296
    Good!
    HOME: A physical construct for keeping rain off your computers.

  9. #9
    Originally posted by N.V.M.
    i don't care. electricity isn't free here. this is so slow, i'll easily make up any ground i lose(laughs) on the next update.
    So what your saying is, Screw the science just give us meanless fast protiens to fold. It may give you points, but it doesn't further our understanding of real world protien folding if we don't fold real world protiens.

    I just wish this was in spring, I'd get my students folding this protien with the screen saver, we study cytochrome C as part of our biochem course.

    and I don't know anywhere that electricity is free, My computers stay on if they are running DF or not. Just DF just seems the most worth while of the project out there and no matter what the speed of my point gain, i know it's helping out the project.

  10. #10
    Member
    Join Date
    Jul 2003
    Location
    Home of the 2010 Olympics
    Posts
    92
    my workhorse comp stays on 24/7 too, but my farm machines(8 in total) for DF will not be running.Because yes, points are #1, the science is #2. thats the way it is.

  11. #11
    25/25Mbit is nearly enough :p pointwood's Avatar
    Join Date
    Dec 2001
    Location
    Denmark
    Posts
    831
    Originally posted by N.V.M.
    i'll easily make up any ground i lose(laughs) on the next update.
    No you wont

    At least not if those above you have about as many Ghz on this project as you do and continue to crunch on the slow proteins

    Furthermore, you don't know how fast or slow the next protein will be - maybe the next 10 proteins will be about as slow as this one

    Anyway, it's your choice and it's others opinion doesn't matter. The most important thing is that it's fun
    Pointwood
    Jabber ID: pointwood@jabber.shd.dk
    irc.arstechnica.com, #distributed

  12. #12
    Member
    Join Date
    Oct 2002
    Location
    southeastern North Carolina
    Posts
    66

    Talking yeah, but....

    I know for a fact that this is the only DC project that TRULY interests me...
    nothing else out there seems worth the time and effort..

    it's electricity [ indirectly ], as it relates to A/C !! .. THAT gets expensive !!!
    a couple hundred watt-hrs for the PCs is no big deal...

    seeing my #s accruing at 1/3 the rate of a few days ago-- that's tough !!

    [ havent figured if I'm a stats ho, pig, or simply an obsessive individual requiring instant gratification all the time !!!!!!!!!!!!!!



    " All that's necessary for the forces of evil to win in the world is for enough good men to do nothing."-
    Edmund Burke

    " Crunch Away! But, play nice .."

    --RagingSteveK's mom


  13. #13
    Member lemonsqzz's Avatar
    Join Date
    Sep 2002
    Location
    Montain View, CA
    Posts
    97
    As long as the babes are still here.. so am I !!...

    I think this protein will last longer than any relationship I've ever had


  14. #14
    25/25Mbit is nearly enough :p pointwood's Avatar
    Join Date
    Dec 2001
    Location
    Denmark
    Posts
    831
    Originally posted by lemonsqzz
    As long as the babes are still here.. so am I !!...
    Pointwood
    Jabber ID: pointwood@jabber.shd.dk
    irc.arstechnica.com, #distributed

  15. #15
    Senior Member wirthi's Avatar
    Join Date
    Apr 2002
    Location
    Pasching.AT.EU
    Posts
    820
    Originally posted by N.V.M.
    ... Because yes, points are #1, the science is #2. ...
    Your oppinion. Not mine.

  16. #16
    Junior Member Trog's Avatar
    Join Date
    Sep 2003
    Location
    Bedfordshire UK
    Posts
    5
    It does not matter why people fold just so long as they do.

    It would not matter to the science, if Howard set it up so that you got a picture of Elena in her swimsuit as wallpaper as the bonus for doing X folds in a day.
    Or if you then had to do twice as much, to make it go away again.


    So regardless of what you think of people who just fold for the fun of the points race.
    Remember that if they leave because their fun factor has been reduced it is still the science that suffers. So their complaints need to be listened too, as them leaving has the same negative effect regardless of why they go.


    Also......
    On the stats are fun front I have always thought that the competition would be more open/fun. If there was a limit on how many people were allowed in any one team.
    So that if you wanted to be able to say I am in a top 10 team you had to earn your place by doing a lot of folding. Rather than being member 200 odd of some huge great team.
    Last edited by Trog; 11-05-2003 at 01:13 PM.
    Athlon XP3000, ABIT NF7-S, Swiftech MC 462, 512Mb Kingston PC3500, 5x WD SE HDD's = 1.16Tb, ATI A in W Radeon 8500DV, Win TV Nova T, DSL-504, Win 2K.

  17. #17
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137
    hmmm... points per capita, interesting idea
    Use the right tool for the right job!

  18. #18
    HCW DF Team Leader rstarr's Avatar
    Join Date
    Dec 2001
    Location
    Fort Knox, KY
    Posts
    82
    Originally posted by FoBoT
    hmmm... points per capita, interesting idea
    I like that idea also.
    Folding 24/7 for a cause
    www.hardCOREware.net
    HCW DF Team!

  19. #19
    Originally posted by FoBoT
    hmmm... points per capita, interesting idea
    (FoBot, please do not take the following personally. It is just my thoughts on that comment and the overall picture and not directed at you.)


    Really? Then do you also think that racism, segregation and caste systems are also an interesting idea. Choice my friend, it is our choice to be part of this project. It is our choice which team we care to be a part of. It is our choice, those of us who do not control massive amounts of machine in our work environment, to expend or not expend hard earned dollars or whatever you monetary equivalent may be on more machines to increase your production.

    No, on the other hand that idea does not even warrant a passing thought. Friends group together to form teams. Those friends bring in other friends who in turn bring still others. Well, in a perfect scenario that is how it should work. I honestly believe that back in I believe it was February of last year when this project was born that is how things were happening. The numbers of participants grew almost exponentially. All was well in the world of DF. What happened to change all of that?

    Well, not much through the life of Phase I. Things ran most efficiently. But wait; here came the end of Phase I and movement onward to Phase II. System lockups, mysterious crashes, irretrievable lost production, bad releases of clients, failing servers, and a myriad of other problems suddenly beset the project. Some participants hung loose and kept hoping that tomorrow would bring that pretty pink bubble back to the program. Some still await its return. Others just plain got fed up with the waiting and moved on.

    Where does that leave us today? In my humble opinion, we are but a very trimmed down group of dedicated participants, with yet a sprinkling of others who yet may move along. Has the DF project suffered for what has transpired since the start of Phase II? One would have to be a fool not to know that it has. Phase I was the easy stuff. It was a lot like pre-school, fun and games. Now, moving somewhat rapidly ahead, we are faced with the first look at algebra and trigonometry, beasts unlike we had ever seen before. In Phase I we spent most of our time in daylight. Here in Phase II all is surrounded by darkness. We tread through uncharted forests.

    Put the elite into groups of elite and cast the others where them fall in respect to the elite? No, the non-elite need the elite as the elite need the non-elite. One feeds off of the other, all for the good of the project. The good of the project... is that not what we are here for? In the first week of the project there was no plan for teams. If memory serves me correctly it was one of my teammates that contacted Howard and requested same, as well as some kind of point system to make participation as interesting as other projects. Both of those wishes were granted within a day and we have them still today. During beta testing of Phase II a new point system was requested due to the change in the way work was to be done, and was granted.

    Folks, we are not dealing with any cast in stone givens anymore. The playbook has been revamped. We have been given and now it our turn to give by continuing to work on the project, not picking up our ball and going home because we cannot get our own way. Leave the childishness to the children. I suspect that the greatest part of us are no longer children. We should not act like we are. Yes, I have posted a couple of random pot shots at Howard and Elena. I do hope they were not taken out of the context in which they were meant... nothing more than just a friendly jab or attempt to get someone's attention.

    We wanted teams. We got teams. We wanted points. We got points. DF wants production. Are you going to help give that to them? An astonishing discover that could lead to the saving of lived by reducing cholesterol in arteries was announced yesterday. Some day that discovery may save my life, given that I recently found out that I have some extensive blockage. Is it possibly that some day, N number of Phases down the DF road that our work may help to uncover some other discovery that will save lives? I believe that is likely and I will work diligently to help get to that day, will you?

  20. #20
    Member
    Join Date
    Jul 2003
    Location
    Home of the 2010 Olympics
    Posts
    92
    probably, but not necessarily through this project.

  21. #21
    Senior Member
    Join Date
    Jul 2002
    Location
    Kodiak, Alaska
    Posts
    432
    We're doing basic research at the moment.. When we're through, we'll have helped Howard (or Dr. Feldman), Dr. Hogue (FeedB0B0), and Elena design a program that can be used by the large medical corporations to design drugs with the correct shape to stop a cellular process from happening or to encourage it.. The hystamine blocker ads on television are trying to show how high tech they are.. and show special shaped hystamine receiver sites on certain cells. Along comes the hystamine blocker which fits into most/all of the hystamine receiver sites on the cells and suddenly the person is temporarily immune from hayfever.

    The medical corporations will be able to put together a large Lunix cluster, find a particular shape they need to deal with the medical problem - and feed that to the DF on the Lunix cluster and get close enough to the right shape to use other approaches (I didn't pay close enough attention to the description of that stage) to create a working drug to help cure a problem.

    And we'll be able to tell people that we played a part in creating that program. (For the first few years, they'll probably go out of their way to show how high tech they are.. and mention it's creation had roots in the DF algorithms..) Granted, we'll probably only point out the impressive drugs/cures.. and not the common things like perhaps an anti-garlic scent halitosis tablet..
    www.thegenomecollective.com
    Borging.. it's not just an addiction. It's...

  22. #22
    Senior Member
    Join Date
    Mar 2002
    Location
    MI, U.S.
    Posts
    697
    Did you just say "Lunix"?



    "If you fail to adjust your notion of fairness to the reality of the Universe, you will probably not be happy."

    -- Originally posted by Paratima

  23. #23
    Member
    Join Date
    Oct 2002
    Location
    southeastern North Carolina
    Posts
    66
    Originally posted by MgKnight
    (FoBot, please do not take the following personally. It is just my thoughts on that comment and the overall picture and not directed at you.)

    [snip.. no idea what the editted remarks were all about]

    We wanted teams. We got teams. We wanted points. We got points. DF wants production. Are you going to help give that to them? An astonishing discover that could lead to the saving of lived by reducing cholesterol in arteries was announced yesterday. Some day that discovery may save my life, given that I recently found out that I have some extensive blockage. Is it possibly that some day, N number of Phases down the DF road that our work may help to uncover some other discovery that will save lives? I believe that is likely and I will work diligently to help get to that day, will you?
    given the way I binge on potato chips- and other aspects of my diet/lifestyle, I'm counting on some breakthru's, implicitly...

    dont like my point accumulation-- too much work for too little acknowledgement; DO like my best RMSD, thus far.. this scheme is interesting...
    " All that's necessary for the forces of evil to win in the world is for enough good men to do nothing."-
    Edmund Burke

    " Crunch Away! But, play nice .."

    --RagingSteveK's mom


  24. #24
    I guess a fair way might be to do Generation Points * CPU-TIME where cpu time is the actual amount of time spent by the program (e.g. getThreadTime() on NT, times() in Unix or similar method calls). This is protein independent, and should be relatively fair, since a user with 1 machine will still be getting half the points as a user with 2 equivalent machines (on average). It should also make the differences between slow proteins and fast proteins less pronounced.

    EDIT:

    Of course, it does a problem that a user with 1 machine which is twice as fast will on average be penalized (getting the same as a user with 1 machine at half the speed)).

    I guess then that people just want some scaling factor based on protein time (i.e. points get multiplied by the time it takes to do 250 gens on some base machine (or average over several machines)).
    Last edited by m0ti; 11-07-2003 at 08:18 AM.
    Team Anandtech DF!

  25. #25
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137
    i just said it was an interesting notion, one that i hadn't heard suggested for a DC project before.

    i don't understand the references to choice and all that stuff. i am a pretty simple guy in some respects

    howard has the choice to leave the stats as they are or change them. i totally trust his judgement, i have been a DF participant from the very early stages and i like the way howard thinks and considers input from the participants and then makes decisions and moves on.

    that is , i like howards style concerning running DF and i respect his judgement as to what is best for the project.

    that being said, i like the way the DF points/stats are now, i don't want them to change

    if i didn't care about the science at all, i would do whatever DC project had the best stats for me. i am glad the client actually does something useful, in my opinion there are currectly some other popular DC projects whose clients actually accomplish no work (sometimes referred to as "guess a number projects")

    i am very hungry as i haven't eaten lunch yet today, so i will now go eat some food. have a nice day!


    Use the right tool for the right job!

  26. #26
    Minister of Propaganda Fozzie's Avatar
    Join Date
    Jul 2003
    Location
    Bristol,UK
    Posts
    3,609

    Exclamation Interesting

    I'm pretty new to DF and only been around for a few changeovers.

    We've had small fast ones and slower large ones.

    The slower ones suit me 'cos there is less to upload. However I am a stats whore at heart and it pains me to see so little movement for the effort.

    I thought about suggesting a function for the scoring based on protein size but TBH like FoBoT I think Howard et al are doign a fine job and working hard enough without pandering to the whims of all the stats whores among us.

    Going on production and where we are we have 11 weeks more on this protein to hit 5 billion. So unless people are willing to lose out 10 million points or more (depending on their output) they should keep going.

    Also the more people who pull out and do other DC projects while this one is "less rewarding" will extend that time for completion of this protein.

    Suits me fine I'll creep up on anyone who isn't producing, DF has my full support be it a fast or slow protein.

    At the end of the day it's the team (Free DC rocks), the good cause for this science and the stats that keeps me on this one.
    Alas poor Borg, I knew it Horatio



    http://www.butlersurvey.com/

  27. #27
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137

    Re: Interesting

    Originally posted by Fozzie
    ... I'll creep up on anyone who isn't producing, DF has my full support be it a fast or slow protein.
    try the stealth mouse method
    Use the right tool for the right job!

  28. #28
    Senior Member
    Join Date
    Apr 2002
    Location
    Santa Barbara CA
    Posts
    355
    In phase I there really wasn't a scoring system in place, it was just a simple one fold one point setup. At the time Howard saw no reason to complicate things by adding a difficulty factor for the various proteins to scoring. But now with phase II there is already a scoring algorithm so it seems like it would be fairly easy to add something that would scale the score according to the length of the protein.

    Having said that, I actually like the mix of fast and slow proteins and the resulting affect that has on production. I think that it makes it more interesting.

    And I am not just saying that because I had a hard drive on my farm die on the day of the changeover that has knocked out 8 2GHz+ processors. Though since I haven't got around to get things back up yet I am relieved that this is a slow protein.

  29. #29
    Junior Member Trog's Avatar
    Join Date
    Sep 2003
    Location
    Bedfordshire UK
    Posts
    5
    Originally posted by FoBoT
    when people drop out on slow proteins it gives us tortoises time to overtake the hares

    FoBoT = No 4 in the stats with ~ 300,000,000 points.


    Is that jet or rocket propelled tortoises?
    Athlon XP3000, ABIT NF7-S, Swiftech MC 462, 512Mb Kingston PC3500, 5x WD SE HDD's = 1.16Tb, ATI A in W Radeon 8500DV, Win TV Nova T, DSL-504, Win 2K.

  30. #30
    I guess the only "fair" way would be to normalize by the progress of the entire project for the protein. That's because it's the only way to test it over the hardware that people are actually using.

    As in:

    For each person the number of points is saved. (Same formula as today).

    Once the protein is complete the scaling factor is calculated:
    time for protein / total points generated by all participants for this protein.

    Then each person has their number of points multiplied by that scaling factor.

    Of course, it would be possible to have a "current scaling factor" which gets updated every hour and which gives a "current estimated scaled points" which should be a pretty good estimate for the number of scaled points people are generating (mind you, of course, there are always last minute dumpings which will decrease the factor in the last couple of days before the changeover).

    This scoring system means that you get scored based on how much you contribute relatively. That means that for each protein you effectively get a number of points reflecting your relative contribution to the total production. This is, IMO, intrinsically fair (if someone thinks otherwise, please let me know why).
    Last edited by m0ti; 11-13-2003 at 05:19 AM.
    Team Anandtech DF!

  31. #31
    I think the most fair thing to do would be to scale everyone's score by zero. In this way, the scoring is independent of protein length or complexity, and new users have a chance to compete with long-time runners. Most importantly, everyone will be in first place, so everyone is a winner! I also think this would resolve a lot of the confusion about generations and scoring in general...
    Howard Feldman

  32. #32

  33. #33
    25/25Mbit is nearly enough :p pointwood's Avatar
    Join Date
    Dec 2001
    Location
    Denmark
    Posts
    831
    Pointwood
    Jabber ID: pointwood@jabber.shd.dk
    irc.arstechnica.com, #distributed

  34. #34
    Administrator Dyyryath's Avatar
    Join Date
    Dec 2001
    Location
    North Carolina
    Posts
    1,850
    "Yeah, I'm really kicking ass in Distributed Folding. I'm in a 6145-way for 1st place..."
    "So utterly at variance is destiny with all the little plans of men." - H.G. Wells

  35. #35
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137
    Originally posted by Trog
    FoBoT = No 4 in the stats with ~ 300,000,000 points.


    Is that jet or rocket propelled tortoises?
    its just a HERD of tortoises

    drumming tortoises
    Use the right tool for the right job!

  36. #36
    Fixer of Broken Things FoBoT's Avatar
    Join Date
    Dec 2001
    Location
    Holden MO
    Posts
    2,137
    Originally posted by Brian the Fist
    I think the most fair thing to do would be to scale everyone's score by zero. In this way, the scoring is independent of protein length or complexity, and new users have a chance to compete with long-time runners. Most importantly, everyone will be in first place, so everyone is a winner! I also think this would resolve a lot of the confusion about generations and scoring in general...


    i think the science guy is making fun of us moderate stats whores!

    i am getting a lawyer and contacting the ASPCSPCA or whoever is in charge of sueing people for unfairness in DC projects, there ought to be a law!!








    Use the right tool for the right job!

  37. #37
    Ancient Programmer Paratima's Avatar
    Join Date
    Dec 2001
    Location
    West Central Florida
    Posts
    3,296
    FoBoT, you wouldn't know a moderate stats whore if one hit you with a haddock!

    Now quit yer whining & get back to folding!!
    HOME: A physical construct for keeping rain off your computers.

  38. #38
    Originally posted by Brian the Fist
    I think the most fair thing to do would be to scale everyone's score by zero. In this way, the scoring is independent of protein length or complexity, and new users have a chance to compete with long-time runners. Most importantly, everyone will be in first place, so everyone is a winner! I also think this would resolve a lot of the confusion about generations and scoring in general...
    I agree. It's also the most trivial to implement, which is always a bonus.


    P.S.

    I really don't care how the stats are calculated, personally. I just think it's a shame if we're losing crunchers due to it.
    Team Anandtech DF!

  39. #39
    Free-DC Semi-retire gopher_yarrowzoo's Avatar
    Join Date
    Mar 2002
    Location
    Santa Ana, CA
    Posts
    3,980
    Originally posted by Paratima
    FoBoT, you wouldn't know a moderate stats whore if one hit you with a haddock!

    Now quit yer whining & get back to folding!!
    Nice one Paratima, yup Fobot wouldn't know one if it hit him from a great height even
    Semi-retired from Free-DC...
    I have some time to help.....
    I need a new laptop

  40. #40
    Ancient Programmer Paratima's Avatar
    Join Date
    Dec 2001
    Location
    West Central Florida
    Posts
    3,296
    Ah, FoBoT's OK. Just slightly

    The more I think about this scoring stuff, the more I come to the conclusion: "What's the big deal here anyway?" It works fine.
    HOME: A physical construct for keeping rain off your computers.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •