Hi Phil: Part of the overall goal of getting RGBA background images to work is to get the "A" (alpha channel) component of that to work properly, and so a preliminary goal should be to implement uniformly colored semitransparent backgrounds correctly.
Some current semitransparent background issues we should be aware of: * Device driver issues: Just for a single page plot where I specify a completely transparent background, there are issues for a number of our devices. Ones that apparently do work currently are the qt devices. For example, examples/c/x00c -dev pdfqt -bg 000000_0.0 -o test.pdf does produce a transparent background for that pdf as proved by the ImageMagick "display" command rendering that transparent background as a checkerboard. But the same experiment with pdfcairo shows an opaque white background. I believe that should be considered to be a bug in the cairo devices since the fundamental principle should be if a user specifies a semitransparent background (as above), that is exactly what PLplot drivers should supply rather than substituting something else in the belief that we know better than the user! * Interactive device issues: Even though the above experiment with pdfqt demonstrates that device follows the user wishes, it is not clear that also occurs for the qtwidget device because I could not get the desired result (a completely transparent background) to work for qtwidget. Of course, that may have nothing to do with our qt device and may be related to the following compositing issue. * Desktop compositing issues: There is an ImageMagick "display" option to actually honor the transparent background request rather than replacing it with a checkboard. I cannot remember that exact option, but when I tried it years ago, the result was the plot is displayed on top of whatever your desktop would be displaying otherwise (what I call the root desktop below). But it would be cooler still if it worked for semitransparent backgrounds (as opposed to completely transparent like in the above case), but when I last tried this there were KDE/X compositing issues that didn't allow that to work properly. I have no idea whether my KDE/X desktop software has been fixed yet in this regard or not. * Multiple page issues: I can forsee the need for two modes here for those interactive cases where pages are normally just stacked on top of the previous pages with a background superimposed at the start and between each page. If that background is completely transparent, then you should see the whole stack of previous results right back to the root desktop (whatever the desktop software would be displaying if the plot GUI was not superimposed on top), and if the transparency is dropped to say 90 per cent then you will see that stack of previous results gradually fading away as you display contents from deeper in the stack to make a really cool-looking effect. So this possibility of seeing everything in the stack modified by the transparency of the superimposed backgrounds in between should be allowed, but there should also be an interactive mode that starts fresh with each page, i.e., the root desktop is displayed, then the semitransparent background, then the plot for the single page. In sum, this whole field of supporting RGBA background images for each plot page is an extremely interesting one, but to get the "A" part of that to work correctly we need to deal with the above issues that show up even when the background "RGBA image" is a simple one that is uniformly coloured. Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
