Writing program - summarizes diacom data - input request

[Kevin Vannorsdel]

 My brother is a mechanical engineer at John Deere---  he rants and raves about
how great spread sheets are...   I tell him I can crunch 1.5 MILLION  6 bit
words of data in a few seconds with my custom program written in C---   he
doesn't seem a bit impressed!---  go figure.

I'm all for compiled code but for some-  it is simpler to click and
drag...etc.   Something like... "To each his own..."

________________________________________________
  Kevin Vannorsdel     IBM Arm Electronics Development
    408-256-6492                Tie 276-6492     kv at us.ibm.com



owner-diy_efi at efi332.eng.ohio-state.edu on 10/22/98 08:45:32 AM
Please respond to diy_efi at efi332.eng.ohio-state.edu
To: diy_efi at efi332.eng.ohio-state.edu
cc:
Subject: RE: Writing program - summarizes diacom data - input request



THe program uses 4 files (6500samples each) and produces the output in
about 30 seconds (that is including giving it the name of the file to
read in).  My program would have results before you even finished
importing  the file into access.  It takes me about as long to write
the program as it would to import into access, then import into excel,
and product the graphs.  Right now I have spent 2 hours on the
program, which is probably how long I would have to mess with things
in access to get it to the same place.  And mostly right now I am
dealing with the block learn data for each cell, so I need ranges.  I
can write a program to do this sort of stuff as fast as most people
can setup a spreadsheet to do the same thing, and if I have to process
more data a second time then the program takes less repeat setup.

So what sort of manipulations do you do on the data?

My program is in Quickbasic and I will put the source code on my web
site when I get it done (and an executable).

You also need to remember that I am processing 25000 samples, which
would take a considerable time to read into access, export, read into
excel and all of that.  I would not be writing it if I was only trying
to deal with the normal 200 samples.

     Roger

On Thu, 22 Oct 1998, Kurek, Larry wrote:

> Why would you have to create a program to do this? I just import the
> Diacom data into Access, and group it by RPM, then I import that into
> Excel and do some analysis including graphing and multiple regressions
> to identify relevant data. The way I set it up, was to do multiple
> Diacom runs in one file, so I can compare and contrast them based on
> changes I made between runs...
>
> TTYL!
>
> Larry
> >
> > Sample output is at:
> >
> > http://www.hit.net/~rah/blm.txt
> >
> >     Roger
>