Path: utzoo!utgpu!water!watmath!clyde!att!rutgers!mailrus!purdue!decwrl!labrea!polya!rokicki
From: rokicki@polya.Stanford.EDU (Tomas G. Rokicki)
Newsgroups: comp.text
Subject: Re: Laserjet Questions
Message-ID: <3959@polya.Stanford.EDU>
Date: 14 Sep 88 07:57:44 GMT
References: <405@ucrmath.UUCP> <101870003@hpcvlx.HP.COM>
Reply-To: rokicki@polya.Stanford.EDU (Tomas G. Rokicki)
Organization: Stanford University
Lines: 26

>Something you should be aware of with tricks like this is that it could 
>make adding support for other printers (even HP PCL printers) much more
>difficult.  For example, if you sent data formatted as described above to
>a DeskJet, it would probably take several hours to print a page.

On the other hand, using tricks like this can speed up output by more
than a factor of two.  On the DeskJet, adjusting your margins often can
speed things up quite a bit, but you will have to empirically determine
how often.  It's not obvious; it certainly isn't every 50 pins, although
that might not be a bad approximation.

But all implementations are not equal.  Take the NEC P6 and the Epson LQ.
The NEC P6 emulates the Epson LQ.  But if you try horizontal white-space
positioning on the Epson LQ, it goes crazy and takes forever to print;
the NEC just glides on, very fast and very smooth.  (The new LQ's now
handle it better, too.)  The solution?  Test your software as carefully
as possible.  It's always nice to supply a `stupid' driver along with
the optimized one; the stupid one should always work, but more slowly
than the other.  Especially when trying tricks with the Epson FX series.

>	Brian Cripe

So when can I put more useable graphics RAM in my DeskJet?  I only want
16K more . . . and of course cartridge RAM doesn't work for graphics.

-tom