Re: [PARPORT] Slowness of using Parallel Devices

Grant R. Guenther (
Sat, 16 May 1998 08:45:30 -0400 (EDT)

> you know how everything slows down when data is written to a SparQ... or
> the scanner is being used... on the parallel port ?... i was just
> wondering if anything can be done to avert this... maybe by manipulation
> through IRQ's... i'm talking about how when data is written to a parallel
> port IDE drive... then the mouse moves in 'jumps'... and the moving
> windows around the desktop is slow... that sorta stuff..

Longer term readers of this mailing list will remember the problems that
the old "ep" driver had. ep (and ez before it) had the epat and pd logic
all in a single module. They also attempted to use the interrupt. When
it worked, things were fine. But probably 25-30% of systems did not work,
and system lockups, corrupt disks and other problems were frequent
topics here.

Furthermore, of all the PARIDE adapters, only EPAT has the wiring to
support an interrupt. The documentation for the EPAT is just plain
wrong about how their interrupt works, and as you probably realise, not
all parallel ports work the same way.

You may also have observed the ongoing troubles reported here with interrupt
handling in the lp driver.

And of course, there are the problems with all the people who have their
parallel port and sound card interrupts set to the same IRQ.

For the time being, I've decided to go for reliablity. As for the future,
who knows ...

In the meantime, if you want better interactive performance, at the expense
of PARIDE transfer rates, enable the "nice" option when you load pd.

Grant R. Guenther

-- To unsubscribe, send mail to: --
-- with the single word "unsubscribe" in the body of the message. --

This archive was generated by hypermail 2.0b3 on Wed 30 Dec 1998 - 10:17:45 EST