800 Points per Day - Why I Really Like Double Gromacs

LeonardoLeonardo Wake up and smell the glaciersEagle River, Alaska Icrontian
edited February 2006 in Folding@Home
Some of us were bemoaning the fact that Stanford University recently completed their run of QMD protein models. There was consternation that points production had gone in the tank. NOT SO! Some of the Double Gromacs being fielded now are very lucrative point producers. My system one is working on two units, one of which is 1112 points. System 4, the D820 dual core, is crunching two Double Gromacs - a p2361 and a p2393...


for a production of 807 points per day! Holy cow. That's even better than a QMD!

QMD threads here and here.

Comments

  • dragonV8dragonV8 not here much New
    edited February 2006
    They are very impressive numbers, Leo. Great stuff.:thumbsup:
  • QeldromaQeldroma Arid ZoneAh Member
    edited February 2006
    :: Can see QCH2002 changing his sig to "Chicks Dig Guys with BIG Double Gromacs!!" ::
  • DanGDanG I AM CANADIAN Icrontian
    edited February 2006
    I wish I could get more of those 1112 pointers. I've got a dual xeon box begging for them, and my 820D.
  • csimoncsimon Acadiana Icrontian
    edited February 2006
    I haven't had one yet ...what flag are you guys using? I need to change flags anyway the old one is getting worn.
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited February 2006
    1) "-advmethods"
    2) in client.cfg file, select "yes" for "bigpackets"

    that simple

    There is a big variety of Double Gromacs protein models being sent out. The 1100-point work units aren't necessarily the most lucrative, but are the biggest.

    The most productive, point-wise, that any of my computers have processed - even better than QMDs, are the Double Gromacs shown for the bottom two F@H clients in the EMIII thumbnail I posted above.

    If Stanford can keep these coming, I don't think I'll miss the QMDs.
  • BudBud Chesterfield, Va
    edited February 2006
    i agree with leo. im notcing a huge jump now
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited February 2006
    Chris, I have no idea how efficiently the Double Gromacs are processed on AMD CPUs. I'm assuming you are looking to try them on an AMD dual core?
  • QCHQCH Ancient Guru Chicago Area - USA Icrontian
    edited February 2006
    Qeldroma wrote:
    :: Can see QCH2002 changing his sig to "Chicks Dig Guys with BIG Double Gromacs!!" ::
    I'm working on it.... :bigggrin: Surgery costs so much though!!!
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited February 2006
    That 1112-point Double Gromacs is not a big point producer. On my HT P4 rig, System 1, I've got the 1112-pointer and a regular Gromacs folding. The 1112 (first client in the EMIII screenshot) is producing 128PPD. The regular Gromacs is producing 168PPD (second client in screenshot).
  • LeonardoLeonardo Wake up and smell the glaciers Eagle River, Alaska Icrontian
    edited February 2006
    That 1112-point Double Gromacs is not a big point producer.
    Well, that statement was certainly premature!

    Currently System in HT mode is pulling 449 points per day with the 1112 DC. I guess when I measured it earlier, there was a lot going on with the computer.
  • csimoncsimon Acadiana Icrontian
    edited February 2006
    I'll set the -advmethods flag and we'll see what happens.
    I'm supposed to be getting a new system here at work real soon. This dual xeon 2.66 is ok but I'm sure whatever I get will honk right past it production wise.
  • lemonlimelemonlime Canada Member
    edited February 2006
    Wonder if these are going out to AMD users as well? I switched off timeless a few weeks ago for sh*ts and giggles, but picked up nothing but 100-point gromacs :hrm:

    Might have to try again :)
Sign In or Register to comment.