Re: Maximum image size?

In article <46854131.1958591476@xxxxxxxxxxxxxxxxxxx>, Aandi Inston
wrote:
There may be implementation dependent limits. I would not be surprised
to find limits as low as 64K.

I have seen (and puzzled over, and then had to explain to
clients, then badgered the Programming Director to PLEASE fix the
problem, but it was in a 3rd-party library) a graphics engine that fell
over at the (2^32-1) limit pixel count in any direction. Things would
start to plot at negative pixel offsets.
That system (the 3rd-party bug) appeared to originate before
1988, and it was still affecting our work in 1998 - we were locked into
using that library for so many things that we just had to come up with
ingenious cheats to get round it.

Well, Helge's assertion makes sense to me. And Andi's suggestion
that various implementations might have different restrictions also
makes sense. But it begs a question - what likelihood is there of
different implementations clearly stating what their limits are?
The only one that I've managed to find clearly stated is the
Adobe system's restriction to 200 inches in either direction. Which I
find limiting : 200 inches at 1:200 scale is not sufficient to plot more
than 4000ft of data ; I routinely have to do plots of 6000ft plus of
data. So the awkward fix-ups continue (in software and with sticky
tape).
So, does anyone know of a PostScript implementation that does not
have an image-size limit this low?

(As a side-line - does anyone know the reasons Adobe came up with
this size. My guess was that it's something to do with the size of
prints used on bill-board posters and that sort of thing.)

--
Aidan Karley,
Aberdeen, Scotland
Written at Tue, 10 Jul 2007 23:44 +0100, but posted later.
09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

.

Relevant Pages

  • Re: Multiplying Big Numbers gives a negative output
    ... echo $totalValue ... Most digital numeric calculations will have some limits, ... actual implementations do have limits - but these limits are ... digits, but some versions behave significantly better than that ...
    (comp.unix.shell)
  • Re: switch() Statement Question
    ... mandatory for exceeding an implementation's limits. ... I'm sure that most implementations try, but I suspect that some ...
    (comp.lang.c)
  • Re: Why no folder manipulation in Std C
    ... implementations. ... limits that it describes is that the one program must reach them. ... connects those limits to the one program, and contains no wording that ... limits have a relevance to an implementation's conformance. ...
    (comp.std.c)
  • Re: contiguity of arrays
    ... >> implementations pf C code like this works exactly as you ... and abort the problem if those limits are violated. ... > Name one compiler enforces such limits. ... There may or may not be such a compiler. ...
    (comp.lang.c)