perry@madnix.UUCP (Perry Kivolowitz) (07/04/89)
We recently received April's Amiga mail (deliver in late June, not bad)
and read with great interest an article by Bryce in which he alluded to
a method for increasing the speed of many PostScript printers.
He mentioned that PostScript printers with revision 38 or later of the
PostScript interpreter should be able to run their ordinary serial con-
nections at speeds as high as 57,600 baud (or 6 times faster than the
standard setting of 9600 baud).
We tried doing what Bryce suggested (it took a little doings in inter-
active PostScript) and cheerily set our Apple LaserWriter II NTX to the
maximum rate.
We then created a mountlist entry called LW: for one of the ports on a
Dual Serial Board. Specifying LW: in Professional Page sent the data to
to the LaserWriter at a substantially accelerated rate.
We did some performance timings in Pro Page:
PP PostScript dump to SER: at 9600 baud: approximately 4 minutes.
vs.
PP PostScript dump to RAM: approximately 40 seconds.
Copy of PostScript dump from RAM: to LW: took an additional 20 seconds.
The performance given by this set up for average PostScript dumps from
Professional Page is now quite quite livable.
The article by Bryce suggested that 19,200 baud would be the limit that
one could drive the standard Amiga serial port for this same purpose.
There are two possible reasons: First, occasionally PostScript printers
do send back status and error information to the host. At speeds higher
than 19,200 baud with multitasking it is unlikely that the Amiga will
actually receive data back from the printer correctly. Second, it is not
possible to set SER: to a higher rate which is acceptable to the printer
(SER: can be set to MIDI but that won't match an acceptable PostScript
baud rate).
Commercial Plug: Thanks to Bryce we've seen yet another unlooked for be-
nifit from having an ASDG Dual Serial Board. It is nice.
If people are interested, I will post instructions for setting a version
38 or later PostScript printer to higher serial rates for use with eith-
er the standard Amiga serial port or an ASDG Dual Serial port.
NOTE: The first production run of DSB's used 6 MHz serial controllers
running at 4.9 MHz. We discovered that this combination did not allow
boards of this production run to hit 57.6KBaud accurately enough (they
actually gave 51200 baud when asked for 57600 baud). With these boards,
the maximum rate that you can set your PostScript printer to is 38400
baud. 76800 baud would have worked fine but PostScript printers don't go
that high.
As a result of this all future production runs will have an 8MHz serial
controller running at approximately 7.4 Mhz. This provides accurate tim-
ing at all standard baud rates including 57600 baud on up to a new high
(and reliable) setting of 115200 baud.
--
Perry Kivolowitz, ASDG Inc.
ARPA: madnix!perry@cs.wisc.edu {uunet|ncoast}!marque!
UUCP: {harvard|rutgers|ucbvax}!uwvax!astroatc!nicmad!madnix!perry
CIS: 76004,1765 (what was that about ``giggling teenagers''?) limonce@pilot.njin.net (Tom Limoncelli) (07/07/89)
It is my understanding (this was publicly mentioned at a JAUG meeting
by CBM people at least a year ago, and definitely before 1.3 was
shipping) that 1.4 would include a new driver for the serial and
parallel ports.
The reason (then stated) for the serial.device loseing chars at
high-speeds and for not performing at all at very high-speeds is NOT
the fault of the chips, but is instead a problem with the software;
and the drivers in 1.4 would be completely written from scratch to use
less CPU time and provide better functionality.
Assuming that the ASDG serial port software was written using a better
technique from the start, does this mean that we can expect 1.4 to use
a similar technique which would result in similar speeds?
Just wondering,
-Tom
--
Tom Limoncelli -- tlimonce@drunivac.Bitnet -- limonce@pilot.njin.net
Drew University -- Box 1060, Madison, NJ -- 201-408-5389
Standard Disclaimer: I am not the mouth-piece of Drew University
"DEC's All-In-1 isn't completely useless, but it's a nice attempt."