Re: Edhel finds DOS 3.3 speed-up (was DOS 3.3 versions)

[nparker@cie-2.uoregon.edu (Neil Parker)]

In article <5d8e4f$jpi@login.freenet.columbus.oh.us>
dalloff@freenet.columbus.oh.us (Dave Althoff) writes:
>A caution regarding the Disk Speed-Up Tip:
>
>I seem to recall reading about this on a bit of Beagle Bros. "junk mail"
>many years ago.  If memory serves (it often doesn't...), this elimination
>of the start-up delay supposedly works fine with everything *except*
>text-file manipulation.  Apparently, when writing to text files, the delay
>is needed to avoid corrupting data.

If this is true, it confirms my suspicions that the speed-up isn't reliable.
But regardless of whether or not memory serves, I have to second the
caution.

>I have not experimented with it.  I wonder if perhaps Pronto-DOS
>automatically sets and resets this byte as it runs...or has a different
>start-up delay routine built in for doing text files...

I seriously doubt it.  Most fast DOS packages work by an entirely different
method--they bypass much of the excessive one-byte-at-a-time data copying
that happens inside the DOS file manager.  All that data copying is a
significant speed bottleneck, and bypassing it produces a *major* speed
improvement--far more than you can get by fiddling with the hardware
start-up delay.

               - Neil Parker
-- 
Neil Parker, nparker@{cie-2,cie}.uoregon.edu, http://cie-2.uoregon.edu/~nparker
     "I was going to be a neo-deconstructivist but Mom wouldn't let me."
                                     -- Calvin and Hobbes, 7/13/1995
Unsolicited commercial e-mail to my address will be discarded unread.