Loading ...
Sorry, an error occurred while loading the content.

Re: [libertybasic] 24/7/8+(!) crunching & simulating all of Life

Expand Messages
  • Robert M. Hosang
    Hi Foon, I m sorry, I thought you meant external I/O. I m reading from one large file into an array, and appending to each of 4 files once a complete loop.
    Message 1 of 7 , Feb 1, 2003
    • 0 Attachment
      Hi Foon,

      I'm sorry, I thought you meant external I/O. I'm reading from one
      large file into an array, and appending to each of 4 files once a
      complete loop. I've tried to keep the displayed text down to a minimum -
      about 4 short lines once a loop.

      I'm sorry to say that I don't know how to setup a ram disk. It
      might help with ver.2, which maxes out all of 256meg ram that I have. I
      would appreciate knowing how to do that.

      Number wise LB's numeric capabilities seem to be good - and
      pretty fast. I've tried floating point tests which show complete accuracy
      to at least 100 decimal places (sci notation).

      As far as my thinking aloud about an analysis of natural
      selection, my idea is that with so much of our own arbitrary intervention
      within our own species, that trying to do something like that would seem
      to be a necessity for us to make social decisions, policies and laws
      which could be used to guide our own 'common sense' hundreds of years in
      the future. I believe that one measure is actually cause of death.
      Another might be our knowledge of our own own genome. I think in the last
      century or so we've seen that cancer is a leading cause of death. I
      wonder if this was so when the light bulb was first invented? Or how the
      cancer rate in places like China and Africa compare with our own
      'enlightened' nation? Well, this is just speculation on my part right
      now, but I'm interested in the topic, and will try to learn more about it
      in the future. And I thank you very much for your thoughts.


      Bob Hosang
      rmhosang@...

      On Fri, 31 Jan 2003 22:20:39 -0700 "Foon" <foon@...> writes:
      > Bob,
      >
      > I'm not sure what you meant when you said that you're not looping
      > any I/O but that you are writing to disk. If you're doing ANY disk
      > I/O in that loop (and writing to disk is an I/O function), then why
      > not try a ram disk? You just set up the disk (and tell your
      > customers to do the same thing, specifying the disk letter to the
      > pgm), copy the files to it toward the beginning of the program, then
      > copy them back to HD just before the pgm ends.
      >
      > Also, if you're doing ANY screen I/O (which is a notorious speed
      > hog in comparison to pure number crunching) [like progress status
      > updates!], you should cut out most if not all of it and you should
      > see a significant speed-up.
      >
      > Another thing: I'm not sure if this is still needed or not, what
      > with integrated math co-processors in P3/P4s, but you might benefit
      > from using a math-co-processor-specific .DLL or API which supports
      > optimized math functions. I may be over my head here: it could be
      > that the compiler/interpreter would have to support generation of
      > such code natively. Carl, can you make any sense of what I'm
      > suggesting? :)
      >
      > As to spit-balling the idea of a natural selection simulator...
      > good grief! I mean, what aspect of NatSel would you be wanting to
      > model? Would it be meant to be predictive, or statically
      > analytical? Per species, or what? And why do it in the first
      > place (meaning, what is the market for such a system)? You'd have
      > to take geography, cyclical weather patterns, long-term climatic and
      > geologic changes, and dependent/competitive species populations into
      > account (all of the food chain). And how could you anticipate which
      > odd mutations might be not only viable but advantageous without all
      > these (and more) variables? Sounds like you'd need full genomes
      > for... well, everything. hahah Or maybe I misunderstand what you
      > meant. (I hope so.)
      >
      > Good luck,
      > - Foon
      >
      >
      >
      > To unsubscribe from this group, send an email to:
      > libertybasic-unsubscribe@egroups.com
      >
      >
      >
      > Your use of Yahoo! Groups is subject to
      > http://docs.yahoo.com/info/terms/
      >
      >
      >
      >

      ________________________________________________________________
      Sign Up for Juno Platinum Internet Access Today
      Only $9.95 per month!
      Visit www.juno.com
    • Jerry Muelver <jerry@allmyfaqs.com>
      ... from one ... minimum - ... It ... have. I ... Copy & paste the following into your config.sys DEVICE=C: WINDOWS HIMEM.SYS DEVICE=C: WINDOWS EMM386.EXE
      Message 2 of 7 , Feb 1, 2003
      • 0 Attachment
        --- In libertybasic@yahoogroups.com, "Robert M. Hosang"
        <rmhosang@j...> wrote:
        > Hi Foon,
        >
        > I'm sorry, I thought you meant external I/O. I'm reading
        from one
        > large file into an array, and appending to each of 4 files once a
        > complete loop. I've tried to keep the displayed text down to a
        minimum -
        > about 4 short lines once a loop.
        >
        > I'm sorry to say that I don't know how to setup a ram disk.
        It
        > might help with ver.2, which maxes out all of 256meg ram that I
        have. I
        > would appreciate knowing how to do that.
        >

        Copy & paste the following into your config.sys

        DEVICE=C:\WINDOWS\HIMEM.SYS
        DEVICE=C:\WINDOWS\EMM386.EXE NOEMS
        DOS=HIGH,UMB
        DEVICEhigh=C:\WINDOWS\RAMdrive.sys 5120 /E

        See http://dslnuts.com/ramdrive.shtml
        http://www.cyberwizardpit.com/article3.htm

        ---- jerry
      • bbjen@hotPOP.com
        ... Here s how to do it, but it probably is not the answer because it has to utilize part of your existing RAM. Unfortunately, you can t create additional
        Message 3 of 7 , Feb 1, 2003
        • 0 Attachment
          At 08:15 PM 2/1/03 -0500, Bob Hosang wrote:
          > I'm sorry to say that I don't know how to setup a ram disk. It
          >might help with ver.2, which maxes out all of 256meg ram that I have. I
          >would appreciate knowing how to do that.

          Here's how to do it, but it probably is not the answer because it has to
          utilize part of your existing RAM. Unfortunately, you can't create
          additional memory. That takes a bigger chip.

          To set up a ramdisk::
          in C:\Config.sys enter the lines:

          DEVICE=C:\WINDOWS\HIMEM.SYS
          DEVICE=C:\WINDOWS\RAMDRIVE.SYS 4096/e

          For example, 4096 gives 4 MB of ram to the ramdisk.

          The file on my system (win98) looks like:

          DEVICE=C:\WINDOWS\HIMEM.SYS
          DEVICE=C:\WINDOWS\EMM386.EXE
          DEVICE=C:\WINDOWS\RAMDRIVE.SYS 4096/e

          >
          > Number wise LB's numeric capabilities seem to be good - and
          >pretty fast. I've tried floating point tests which show complete accuracy
          >to at least 100 decimal places (sci notation).

          I am VERY interested in how you verified such great accuracy! Any chance
          for a demo or a snippet?

          Bill J.

          ---
          [This E-mail was scanned for viruses by http://www.intouchmi.com]
        • Foon
          Bob, I see others have given you advice on the RAM disk issue. I was just about to tell you that if you re already short on memory that approach won t help.
          Message 4 of 7 , Feb 1, 2003
          • 0 Attachment
            Bob,

            I see others have given you advice on the RAM disk issue. I was just about to tell you that if you're already short on memory that approach won't help.

            As to generating math co-processor specific code, the benefit isn't necessarily accuracy, but rather speed. Again, I'm not sure if you need to do this, or even can with LB. Hoping Carl or someone else can clarify this for us.

            I know what you mean about a NatSel model being of general value, but I *really* think it's a bit too ambitious given the piecemeal state of current scientific knowledge. We'll need a MUCH better understanding of all aspects of genetic and environmental systems before such a model would be even partially valid. IMHO.

            All the best,
            - Foon
          • Robert M. Hosang
            Thanks for the info, Jerry Bob On Sun, 02 Feb 2003 01:31:33 -0000 Jerry Muelver ...
            Message 5 of 7 , Feb 1, 2003
            • 0 Attachment
              Thanks for the info, Jerry

              Bob

              On Sun, 02 Feb 2003 01:31:33 -0000 "Jerry Muelver <jerry@...>"
              <jerry@...> writes:
              > --- In libertybasic@yahoogroups.com, "Robert M. Hosang"
              > <rmhosang@j...> wrote:
              > > Hi Foon,
              > >
              > > I'm sorry, I thought you meant external I/O. I'm reading
              > from one
              > > large file into an array, and appending to each of 4 files once a
              > > complete loop. I've tried to keep the displayed text down to a
              > minimum -
              > > about 4 short lines once a loop.
              > >
              > > I'm sorry to say that I don't know how to setup a ram
              > disk.
              > It
              > > might help with ver.2, which maxes out all of 256meg ram that I
              > have. I
              > > would appreciate knowing how to do that.
              > >
              >
              > Copy & paste the following into your config.sys
              >
              > DEVICE=C:\WINDOWS\HIMEM.SYS
              > DEVICE=C:\WINDOWS\EMM386.EXE NOEMS
              > DOS=HIGH,UMB
              > DEVICEhigh=C:\WINDOWS\RAMdrive.sys 5120 /E
              >
              > See http://dslnuts.com/ramdrive.shtml
              > http://www.cyberwizardpit.com/article3.htm
              >
              > ---- jerry
              >
              >
              > To unsubscribe from this group, send an email to:
              > libertybasic-unsubscribe@egroups.com
              >
              >
              >
              > Your use of Yahoo! Groups is subject to
              > http://docs.yahoo.com/info/terms/
              >
              >
              >
              >

              ________________________________________________________________
              Sign Up for Juno Platinum Internet Access Today
              Only $9.95 per month!
              Visit www.juno.com
            • Robert M. Hosang
              Foon, I think you re right about the present state of the info available, but it doesn t hurt to try to think ahead a little. Bob ...
              Message 6 of 7 , Feb 1, 2003
              • 0 Attachment
                Foon,

                I think you're right about the present state of the info
                available, but it doesn't hurt to try to think ahead a little.

                Bob

                On Sat, 1 Feb 2003 19:51:53 -0700 "Foon" <foon@...> writes:
                > Bob,
                >
                > I see others have given you advice on the RAM disk issue. I was
                > just about to tell you that if you're already short on memory that
                > approach won't help.
                >
                > As to generating math co-processor specific code, the benefit
                > isn't necessarily accuracy, but rather speed. Again, I'm not sure
                > if you need to do this, or even can with LB. Hoping Carl or someone
                > else can clarify this for us.
                >
                > I know what you mean about a NatSel model being of general value,
                > but I *really* think it's a bit too ambitious given the piecemeal
                > state of current scientific knowledge. We'll need a MUCH better
                > understanding of all aspects of genetic and environmental systems
                > before such a model would be even partially valid. IMHO.
                >
                > All the best,
                > - Foon
                >
                >
                >
                > To unsubscribe from this group, send an email to:
                > libertybasic-unsubscribe@egroups.com
                >
                >
                >
                > Your use of Yahoo! Groups is subject to
                > http://docs.yahoo.com/info/terms/
                >
                >
                >
                >

                ________________________________________________________________
                Sign Up for Juno Platinum Internet Access Today
                Only $9.95 per month!
                Visit www.juno.com
              Your message has been successfully submitted and would be delivered to recipients shortly.