This Sucks

scottscott Medina, Ohio Icrontian
edited December 2006 in Folding@Home
With the 13 clients I have running here at home 10 of them are the suckie 5-way patty melts.

27ghz and it is only producing 1300 ppd. So if anybody was wondering why my points are down, this is the reason. I have had at least 7 of these for the past few weeks.

I fold'em as I get'em. It's for the science. ( It still sucks )

I have all boxen set to bigpackets=yes

Should I be using the -advmethods flag ?


Scott
EM.jpg 115.6K
«1

Comments

  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited November 2006
    If you are using the latest console client, I believe it's 5.04, -advmethods is on by default in the client configuration. If you need to confirm, just open your console with only the flag -configonly, select 'yes' for change advanced options, and you will see if advmethods is on. But then, I don't think it hurts to set it also in the start up properties.

    Yes, the 2124s and 2125s really bite the big one. I'll be glad when that series ends.
  • DonutDonut Maine New
    edited November 2006
    Scott,

    I've reconfigured 1/2 my clients (12 total) with the -advmethods flag, and 1/2 without. It doesn't seem to make a difference either way. (they've been a couple of weeks this way now)
  • ins4n17yins4n17y Cabanatuan City, Philippines Icrontian
    edited November 2006
    ok ok ok everyones saying the 2124 and 2125 proteins suck but considering how large they are don't they give you bonus points for such a large computation? i mean hey, someone asnwer, truth be told here, do we get extra points for those large WU's?

    by the way, i guess chicks can all learn to dig a guy with a large gromac.
  • ThraxThrax 🐌 Austin, TX Icrontian
    edited November 2006
    The points given for each work unit are determined by a variety of things: Processing time, importance to the project, amount of those work units in the wild to be crunched, etcetera.

    Processing time is probably the least important indicator of earned points.
  • ins4n17yins4n17y Cabanatuan City, Philippines Icrontian
    edited November 2006
    Panties wrote:
    The points given for each work unit are determined by a variety of things: Processing time, importance to the project, amount of those work units in the wild to be crunched, etcetera.

    Processing time is probably the least important indicator of earned points.

    thanks [strike]thrax[/strike] Panties. thats good to know.
  • the_technocratthe_technocrat IC-MotY1 Indy Icrontian
    edited November 2006
    scott wrote:
    With the 13 clients I have running here at home 10 of them are the suckie 5-way patty melts.

    27ghz and it is only producing 1300 ppd. So if anybody was wondering why my points are down, this is the reason. I have had at least 7 of these for the past few weeks.

    I fold'em as I get'em. It's for the science. ( It still sucks )

    I have all boxen set to bigpackets=yes

    Should I be using the -advmethods flag ?


    Scott

    Tell me about it. I think I have over 100 boxes on the melts. With the exception of yesterday, I'm getting abused on ppd. There's nothing you can do about it though, and everyone is getting shafted the same way. Overall it hasn't affected my personal ranking in the project, so I'd say tht is everyone is sucking it up at the same rate, it really doesn't mean anything as far as competition is concerned. Of course, it will take longer to get a higher point total, but it will also take longer for teams to pull ahead of others, so...meh.
  • profdlpprofdlp The Holy City Of Westlake, Ohio
    edited November 2006
    Panties wrote:
    The points given for each work unit are determined by a variety of things: Processing time, importance to the project, amount of those work units in the wild to be crunched, etcetera.

    Processing time is probably the least important indicator of earned points.
    Which is a shame, since "importance to the project, amount of those work units in the wild to be crunched, etcetera" are all out of the end users control and solely at the discretion of Stanford. The only thing we can control is the "processing time", which can be accomplished by using faster computers.

    I'm willing to take whatever WU Stanford thinks is most promising for the research, but they ought to take into account the impact on morale when the same dozen computers you had a month ago are now turning in a third fewer points.
  • scottscott Medina, Ohio Icrontian
    edited November 2006
    profdlp wrote:

    I'm willing to take whatever WU Stanford thinks is most promising for the research, but they ought to take into account the impact on morale when the same dozen computers you had a month ago are now turning in a third fewer points.


    My Point exactly !

    In the last few months I have added a few machines to try and keep my points up, but they just keep going down. ( Good for the science,Hard on the morale )

    On a side note:

    Technocrat
    You dirty rotten scoundrel....after folding for 3 1/2 years I make it to within a breath of the "TOP TEN" of Short Media's elite folders and you come along and mow me down.. :(

    But honestly, I am really glad you are on our team !!! awesome folding dude !!


    Scott
    exoc.jpg 133.2K
  • ins4n17yins4n17y Cabanatuan City, Philippines Icrontian
    edited November 2006
    scott wrote:
    My Point exactly !

    In the last few months I have added a few machines to try and keep my points up, but they just keep going down. ( Good for the science,Hard on the morale )

    On a side note:

    Technocrat
    You dirty rotten scoundrel....after folding for 3 1/2 years I make it to within a breath of the "TOP TEN" of Short Media's elite folders and you come along and mow me down.. :(

    But honestly, I am really glad you are on our team !!! awesome folding dude !!

    Scott

    yeah i'm all in for the points but i won't spend too much time on it. i'm just going to build another athlon xp rig to fold in some extra numbers....and keep it that way :) if i make it to the top 100 or something that'd be great but i'm not so thrilled about the points.....i mean i'm practically folding melts all the time anyways.
  • the_technocratthe_technocrat IC-MotY1 Indy Icrontian
    edited December 2006
    scott wrote:
    Technocrat
    You dirty rotten scoundrel....after folding for 3 1/2 years I make it to within a breath of the "TOP TEN" of Short Media's elite folders and you come along and mow me down.. :(

    But honestly, I am really glad you are on our team !!! awesome folding dude !!


    Scott

    sorry man! You've got quite a while though - http://folding.extremeoverclocking.com/user_future.php?u=195375&f=48
  • scottscott Medina, Ohio Icrontian
    edited December 2006
    the_technocrat wrote:
    sorry man!

    No problem !


    the_technocrat wrote:
    You've got quite a while though - http://folding.extremeoverclocking.com/user_future.php?u=195375&f=48

    Damn ! it was 53 days yesterday and only 48 days today :rolleyes: You are gaining faster than I thought....must pedal faster...more squirrels....


    Scott
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited December 2006
    Yeah, same thing's happening with me. I've had my scope centered on Prof, but Mr. Lovely (Sledgehammer) will knock me out of the way before I get a clear shot.

    But that's what makes this so fun....
  • profdlpprofdlp The Holy City Of Westlake, Ohio
    edited December 2006
    Leonardo wrote:
    ...I've had my scope centered on Prof, but Mr. Lovely (Sledgehammer) will knock me out of the way before I get a clear shot...
    You buzzards can fight it out over my poor eviscerated carcass...
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited December 2006
    Awesome photograph! That's one baaad bird!


    Title: Leo Removed from His Kill by Sledgehammer
  • csimoncsimon Acadiana Icrontian
    edited December 2006
    Maybe the issue is dead but as far as the ppd is concerned ...it's all relative. Everyone else is getting the same thing so no one is being disadvantaged here. We've all got low ppd respectively so it shouldn't be a concern to anyone.

    Also ...with the release of the GPU and SMP beta clients out of the focus I think we will start to see larger faster wu's soon. I think that the focus is being drawn back to the standard clients that we run typically. This is however mostly assumption on my part but I'll stand behind it unless otherwise disproven. :fold:

    Let's get on with the folding shall we? :D:csimon:
  • scottscott Medina, Ohio Icrontian
    edited December 2006
    csimon wrote:
    Maybe the issue is dead but as far as the ppd is concerned ...it's all relative. Everyone else is getting the same thing so no one is being disadvantaged here. We've all got low ppd respectively so it shouldn't be a concern to anyone.

    I do realize this and you are 100% correct.

    Sorry for the rant. I have been having one of those weeks and really needed to whine about something. I did not mean to start a points dispute thread.
    <s> I want the QMD'S Back </s> ... whoops...did I say that out loud ? :)

    csimon wrote:

    Let's get on with the folding shall we? :D:csimon:


    I never stopped , nor can I see stopping in the foreseeable future. :fold::fold:

    Scott
  • csimoncsimon Acadiana Icrontian
    edited December 2006
    scott wrote:
    I never stopped , nor can I see stopping in the foreseeable future. :fold::fold:

    Scott
    Oh I know it ...I didn't mean to come off as a lecturer or anything. Heck I feel the same way from time to time. Sometimes I need someone to knock perspective back into me.

    :bigggrin: This has always been the "little team that could" ...it's a historical fact!:headbange
  • muddocktormuddocktor
    edited December 2006
    It's because of these streaks we get with the crappy melts coming in almost exclusively have changed my mind about bonus points lately. I think that Stanford needs to dump the bonus points for all wu's now because when they send out a whole load of the crappy ones it just discourages people from folding and also makes it tempting to dump the melts when the assignment server finally starts assigning to a "good points" server, like 142, 162 or 128. I fold for the science, but lately I've shut down some rigs for a while when Stanford's assignment server goes into a 5 way melt punky spasm and stays stuck on server 160 sending those duds out. I have noticed that the assignment server has been acting pretty weird at times lately and won't send work from certain servers for a long time and then all at once it will get stuck and send oodles of wu's out from a server you haven't seen anything from for a while. If they do away with the bonus on the wu's, then the 5 way melts wouldn't suck as bad on points production (although I do think they are undervalued for the amount of time they take compared to other normal wu's).
  • Ultra-NexusUltra-Nexus Buenos Aires, ARG
    edited December 2006
    I fixed the 5way melts "problem" by shifting my 2 AMD X2 boxes to SMP clients.... besides its a good excuse to start learning a bit of Linux :D

    From 400PPD (4 console clients/windows/5way melts) I am at 1500-1700PPD (2 SMP clients/linux64/587 pointers) with the same hardware.

    Anyone with any dedicated dual core (with 64bit support) should change to SMP for the sake of SM´s PPD :D
  • TimTim Southwest PA Icrontian
    edited December 2006
    I've got a 2124 melt going on my main system right now (Barton 2500+ overclocked to 2.2 Ghz), and it's saying 15 seconds per frame. On a 20,000 frame work unit.

    I wonder why the melts fold slowly. Are they very complex, was the work packet designed poorly, or something else? Task Manager says FAH is using barely 6 MB of memory right now.

    We'll get through them sooner or later. I think I read here that 70,000 melt work units were made to be sent out.
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited December 2006
    15 seconds per frame
    That's impossible. You might want to reinstall your monitoring app.
  • the_technocratthe_technocrat IC-MotY1 Indy Icrontian
    edited December 2006
    For anyone with a dual-core who's still complaining about ppd: get the SMP client NOW! Looks like a big boost!

    :rarr::rarr::rarr:

    SM6

    http://folding.stanford.edu/FAQ-SMP.html
    HOW TO RUN THE FAH/SMP BETA CLIENT

    This is a beta release and we expect that there will be several bugs, flaws, problems, etc. To minimize problems, we have been testing the client and cores extensively in house and they run well there. However, it's our experience that running in the controlled setup in our lab and running "out in the wild" are very different situations.

    As in the use of any beta software, please make sure to back up your hard drive, and do not run this client on any machine which cannot tolerate even the slightest instability or problems.

    There are two steps:

    1. Download the client file from the Folding@Home web site download page. Note that we currently only support the console client for 2 platforms: Mac OSX and 64-bit Linux. We are working on a 32-bit Linux client. Due to the nature of our code, porting to Windows is considerably more challenging and we are still looking into the best way to complete this port.

    2. Install the program.
    For Mac OSX: untar the files and then copy the mpiexec file to the ~/Library/Folding@Home directory and then run the fah5 binary. More detailed instructions are at the bottom of this page.

    For Linux: untar the files and then run the fah5 binary. For those unfamiliar with running FAH on linux, see the FAH wiki entry for the basics on running FAH on Linux (or the FINSTALL code tutorial on the wiki). More detailed instructions are at the bottom of this page.

    edit: except win32 users, of course. But hey, you're looking to switch over, right? :)
  • csimoncsimon Acadiana Icrontian
    edited December 2006
    the_technocrat wrote:
    For anyone with a dual-core who's still complaining about ppd: get the SMP client NOW! Looks like a big boost!

    :rarr::rarr::rarr:

    SM6

    http://folding.stanford.edu/FAQ-SMP.html



    edit: except win32 users, of course. But hey, you're looking to switch over, right? :)

    "Note that we currently only support the console client for 2 platforms: Mac OSX and 64-bit Linux."

    There are some faster big pointers in the works for the Windows client ...just don't know how soon they'll appear yet.
  • KentigernKentigern Milton Keynes UK
    edited December 2006
    I'm now on third one of these 2124, each one has been really slow (see pic) is this normal?
    emmirco.jpg
  • SPIKE09SPIKE09 Scatland
    edited December 2006
    Unfortunately yes, Dan "the numptie" Ensign seem to have given us the folding community some really slow WU's not using the SSE optimizations to the limit.:rant:
  • KentigernKentigern Milton Keynes UK
    edited December 2006
    Thanks Spike :) - thought I was doing something wrong, being a newbie at folding
  • csimoncsimon Acadiana Icrontian
    edited December 2006
    Kentigern wrote:
    I'm now on third one of these 2124, each one has been really slow (see pic) is this normal?
    emmirco.jpg

    Open your fahlog.txt and be sure that the "Extra SSE boost OK" line is there at the beginning where the client is started. If not then restart the client and it should appear. If it's already there they you're going full speed ahead unless you aren't configured to run 100% cpu.
  • KentigernKentigern Milton Keynes UK
    edited December 2006
    If it's already there
    It is
    you're going full speed ahead unless you aren't configured to run 100% cpu
    Where do I find this info?

    Task manager shows at FAH at 50 when I'm doing other work as well as folding.
  • csimoncsimon Acadiana Icrontian
    edited December 2006
    It's in the client.cfg. It defaults to 100% and it's there unless you changed it during the configuration process. You can check it or change the configuration at any time using a shortcut to the client and adding the -configonly flag in the shortcut. You can even do it without turning off the client but once you reconfigure you can only get it to take effect by restarting the client.

    So how many instances (clients) do you have folding on that cpu? Just that one or do you have two (one for each thread)?
  • KentigernKentigern Milton Keynes UK
    edited December 2006
    So how many instances (clients) do you have folding on that cpu?
    I only installed FAH once
    Just that one or do you have two (one for each thread)?

    not being hardware savvy - only just found out that I have hyper-threading

    These pics are of the two client cfg that I found.

    This one is from the EM3
    clientcfg.jpg


    This one was inside the FAH folder
    fahcfg.jpg


    If I can run more than one (console version) without overheating my pc then I will.

    The CPU fan that I have - is the one that came in the Intel Retail boxed processor Intel(R) Pentium(R) 4 CPU 3.20GHz 3.25 GHz

    Current RAM = 512 MB WinXP ver 2002 SP2
Sign In or Register to comment.