Re: speed

To: CALC-TI@LISTS.PPP.TI.COM
Subject: Re: speed
From: Scott Trimmer <strimmer@RIFT.FOO.NET>
Date: Thu, 19 Jun 1997 10:37:13 -0700
Comments: To: E S B <staysail@JUNO.COM>
In-Reply-To: <19970617.210855.8710.2.Staysail@juno.com>
Reply-To: Scott Trimmer <strimmer@RIFT.FOO.NET>

Recently, a friend and I were playing w/ his 85 and my 83.  We typed in
the same exact programs in.  One looped using a while or FOR (I can't
remember) loop and went to
250.  The other used the lbl and goto functions.  On the 83, the one w/
the labels and goto went very slowly, and on the 85 it went very fast.  On
the 83, the for went much faster than the 85.  Is there a good reason for
this?  I didn't think to check the memory, but it was the comparison
between the two types of loops  we were testing.  MORAL of the story:  On
the 85, use labels and gotos and on the 83 use the premade loops if speed
it what you are shooting for.
        -Scott
=;)


On Tue, 17 Jun 1997, E S B wrote:

> This speed testing is an interesting idea.  I once wrote a stopwatch
> program for my 82 that was pretty accurate.  A friend of mine copied it
> to an 85 one day just to kill time.  Weeks later, when he had discovered
> ZShell and he had very little free mem on his calc he noticed that this
> exact same program wasn't accurate anymore.  Memory definitely has an
> effect on speed.
>
> May I suggest that people post a standardized test result so this will be
> all the more meaningful.  Let's all use this program:
>
> For(X,1,400)
> Disp X
> End
>
> It's a simple program that's not too fast to time and I think is on all
> the calcs.  May I suggest we all post our results like this:
>
> Calc      Time        Rom     Mem Free    (get the mem free from your mem
> menu)
> -----------------------------------------------------
> 82        21 secs     19.0         28688      <---That's my $0.02
>
>
> Erik B.
> Staysail@Juno.com
>
> P.S. I'm thinking about starting an ASH programs webpage soon.  Any
> suggestions?
>
>
> On Tue, 17 Jun 1997 11:29:42 -0400 Mark G Malley <markgm@JUNO.COM>
> writes:
> >What would be a cool thing to do is test the actual speed of the
> >calculators, but with different conditions.  Maybe one with the memory
> >cleared, one with it half full, and one with it almost filled.  The
> >reason for this is because the RAM is the hard drive as well!   It
> >would be interesting to see the results if all things were equal
> >except for the type of calculator.
> >-MarkGM@Juno.com
> >
> >
> >>I did this test on my 83 and it took about 16 seconds.
> >>
> >
>

References:

Re: speed

From: E S B <staysail@JUNO.COM>