Hi,
On Wed, 2006-04-05 at 07:58 -0400, Ales Hvezda wrote:
> Hi Mike,
>
>
> > I have been working quietly in the background on cleaning up the
> >postscript output for gschem. So far, with my local copy, on my test
> >`schematic' the output Postscript file shrinks by about 50%. I have all
> >of the basic drawing primitives working, a prolog from a file is copied
> >into the output, and am about 30% of the way through the final
> >primitive, text. (Unicode support, overbar support on postscript fonts,
> >and postscript multiline text support.) I few questions have come up
>
> I can't wait to play with your changes. :-)
>
I am nearly there.
>
> >while re-coding the output that I am seeking clarifications for.
> >
> > 1) What units is the line width expressed in? Is it in mils,
> >postscript native points, or in some other unit?
>
> It is in mils.
Thanks, that's in the output now.
>
>
> >
> > 2) I may have to look into this some more, but the dots in the
> >`center' and `phantom' line styles look a little anemic, and this seems
> >to be because the dot is drawn at half the line width. I am not sure if
> >the effect is because of rounding at the printer or because of design.
> > Right not, I set it up so that the dots are rendered at the full line
> >size for the printer. Does anyone object? (I did see code in there
> >that messed with the dot size, when I replicated this code in the output
> >postscript, the results looked anemic, plus, the old output also looked
> >pretty thin on my monitor.)
>
>
> Yeah, it if looks anemic, go ahead and leave in your fix and
> I'll take a look at it. Patrick is really the one who implemented these
> line types so he might have something more constructive to say, but I'm
> fine with the change if it makes the output look better.
>
I added more test lines to my schematic, with various widths, and indeed
there must have been some rounding going on with Ghostsript on my
monitor. Probably integer vs. real numbers as Postscript allows both,
and I have don't totally understand the typing rules, yet. The original
was correct, I have returned the calculation, making sure that the
division is by `2.0' to force a real number calculation, which makes
perfect sense after re-reading the postscript documentation.
>
>
> >
> > 3) What units the the text size in? Is it in units of 10mils, or
> >points?
>
> It is in points. There might be a fudge factor somewhere in there
> to make the rendered text match the postscript output.
I have added the appropriate scaling conversion to the text output
routine. I have set up the co-ordinates on the printer to correspond
directly to mils, so I convert points to mils then use that number to
scale the font. We can add a fudge factor later if this does not work
out.
>
>
> Let me know if you have any other questions.
>
> -Ales
>
--
--------------------------------------------------
Mike Jarabek
FPGA/ASIC Designer
http://www.istop.com/~mjarabek
--------------------------------------------------
