home *** CD-ROM | disk | FTP | other *** search
- Subject: Offscreen GX and Hi-res Printing
- Sent: 9/3/96 5:14 PM
- Received: 9/3/96 5:14 PM
- From: msandber@Sigma4.com (Martin Sandberg)
- Reply-To: ODF-Interest@CILabs.ORG
- To: OpenDoc Development Framework Discussion List
-
-
- I have two somewhat related questions. First, does anyone have any
- insights into imaging using QuickDrawQX into ODF's offscreen bitmaps? How
- should I go about connecting QDGX with one? We really need to do offscreen
- rendering for redraw speed and general cleanliness and GX has such _useful_
- capabilities <G>. Has anyone had any experience with this? Or
- suggestions?
- As I understand using QDGX in ODF, I would normally (after testing
- for the presence of QDGX, of course) just build an FW_CGraphicsContextGX in
- the Frame's Draw method each time Draw is called and pass that to a bunch
- of custom Render methods that take it as a parameter. Or is there a better
- way? And how does the QDGX GC relate to the many different types of
- GraphicsContexts that are derived from FW_CGraphicContext? Are there cases
- where, say, an FW_CViewContext (or any of the others in FWContxt.h) has
- needed capabilities not found in the GX context (which does NOT descend
- from the same class??) ?
-
- Second question: last weekend (Labor Day is a work day, right?) I
- was investigating how ODF handles high-resolution output devices like laser
- printers and ink jets. Screens are not, after all, the only output
- devices..
- I was reviewing the way the ODF private render system works (in
- SLRender.cpp) when something struck me. All of the shapes that are
- rendered thru this system are reduced to QuickDraw resolution (i.e., short
- coordinatess where 1 represents a point or 1/72 inch). Doesn't this mean
- that printing, also, is limited in its resolution on, at least, the Mac
- side? (I didn't check on the Windoze code)
- This seems to me to indicate that all the effort of using ODF
- fixed-point coordinates for graphics is just wasted when it finally comes
- to putting dots on a screen or on paper. I understand this on the screen
- side since, after all, a screen has limited resolution. However, what
- about print output? Did I miss something about when printing happens that
- does use the fractional-point portion of the coordinate information?
- I understand that ODF is always going to be a thin graphics layer.
- I accept that (grumble, grumble <G>). I had assumed that this meant that
- fancy graphics shapes (e.g., rotated text) were our problem. However, I
- had assumed that what was we saw in the code was actually used, e.g., in
- the case of FW_Fixed coordinates. So why does ODF throw away the result of
- all the extra work involved in fixed-point math when it could be used if,
- e.g., QuickDrawGX is available for printing? How do I explain to my users
- that I let them set a 0.5 point pen size but printed out a 1 point wide
- line instead on a 600 dpi laser printer?
- It looks to me as though ODF's imaging model (at least on the Mac
- side) is really intended mostly for screen display. Is this a correct
- assumption? Do I have to write my own set of renderers if I wish to
- support high resolution output devices?
-
- Thanks,
- Martin