[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: grayscale vs. color

David Fanning, davidf@dfanning.com writes:

> But what I would like, very much, is for the PostScript
> device to be a true 24-bit device so that 24-bit images
> could be reproduced in good CYMK colors. Now, of course,
> you have to Color_Quan 24-bit images to get PostScript
> color.

    This is the sort of more fundamental issue I feel RSI should
address.  The whole colour mess in general should really be
sorted out.  The direct graphics colour model in particular seems
stuck in the early 80's when 8-bit colour was something new and

    It would be a real benefit to programmers if RSI made the
whole of IDL work like object graphics, with the details of
on-screen representation handled behind the scenes by default,
but with tools to interrogate the display for those that needed
to ensure that, for example, annotations and overlays really did
stand out as-plotted.

> And the other thing I could *really* use is some kind
> of decent preview image in PostScript files.

    Sharing files across platforms I find that previews seem to
choke programs more than almost anything else, but I'd like the
option to have them.  On systems with colour management installed
I'd also like to be able to embed a colour profile, particularly
for greyscale images I've carefully tailored to have just the
right amount of shadow or highlight detail.

    I've recently installed and started programming the latest
versions of several other packages which do similar things to IDL
(Labview for instrument control, Igor for 1D graphing and
analysis).  This has reminded me of all the little niggly
problems I've come to simply accept about IDL, but which are not
necessary and waste everyone's time.  I still like and use IDL a
lot, but I resent having to find workarounds because RSI are too
busy implementing new gadgets to sort out longstanding problems
with the core of the program.

    I feel better now.