On 2007-08-17 20:10-0400 Hazen Babcock wrote: > > On Aug 16, 2007, at 9:41 PM, Alan W. Irwin wrote: > >> On 2007-08-08 14:27-0700 Alan W. Irwin wrote: >> >> (4) Text clipping. Hazen's fix ended up costing huge amounts of computer >> time (10x - 100x) for his software stack so he commented it out. I would >> like to verify the computer time problem for my own software stack so I >> suggest this fix should be implemented as a driver option. My software >> stack is the latest/greatest cairo and pango. From the release notes >> there >> have been a number of efficiency improvements in the latest releases so I >> may find that pscairo does fast text clipping for my software stack. Or >> it >> may not.... > > Revision 7812 contains a text clipping option for the cairo drivers, called > text_clipping. Off, the default, is text_clipping=0, On is text_clipping=1. I > was measuring the time by looking at how long it took xcairo to render a > complicated plot like example 9 page 1. It behaves a little unexpectedly in > that you get the message "Key <Return> to finish" and a blank window almost > immediately. Then if you wait a few (10s?) of seconds the plot is finally > rendered.
For Revision 7812 built for the latest pango/cairo stacks, clipping slows down pscairo by a factor of 4.3. [EMAIL PROTECTED]> time ./x09c -dev pscairo -o test.ps real 0m0.491s user 0m0.435s sys 0m0.029s [EMAIL PROTECTED]> time ./x09c -dev pscairo -o test.ps -drvopt text_clipping=1 real 0m2.120s user 0m1.889s sys 0m0.196s The factor of 4.3 does seem to be excessive for the relatively small (at least compared to a normal printed page) amount of text we have for example 9. So this inefficiency may be worth reporting to the cairo developers to see if they would like to address the issue. (I imagine text clipping is not used very often for normal printed results so they may not have worried about this inefficiency up to now.) Considering the current inefficiency of text clipping for the cairo library, I think the way you have dealt with this issue is the best compromise. So that only leaves the 3D labelling issue (issue 3) to be dealt with to achieve what I consider to be maturity for this driver. At that point I would recommend building the cairo devices by default and putting a pscairo test stanza into ctest. 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); PLplot scientific plotting software package (plplot.org); 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 __________________________ ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
