The Persistence of Vision Raytracer (POV-Ray).
This is the legacy Bug Tracking System for the POV-Ray project. Bugs listed here are being migrated to our github issue tracker. Please refer to that for new reports or updates to existing ones on this system.
Attached to Project: POV-Ray
Opened by Jim Holsenback - 2010-05-21
Last edited by Jim Holsenback - 2010-09-26
Opened by Jim Holsenback - 2010-05-21
Last edited by Jim Holsenback - 2010-09-26
FS#114 - Mosaic Preview not displaying properly
Mosaic preview display didn’t work as expected, given these command line options: +sp64 +ep16. The preview was solid colored instead of the coarse preview that you’d expect.
I’ve tested a fix to unix/disp_sdl.cpp from clipka and it appears to work.
fix submitted as change #4990
It seems to work (with source from #5066) but I have the following observations:
1. the previewed mozaic is one pixel short (at least on linux): at iteration with size x, we can still see the border (one pixel width) of the iteration with size*2. (and from starting grid, the black layout): is it an intended effect ?
(see attached picture)
2. There is a side effect on the reported statistics (or actually ?) for the number of threads used to perform the rendering.
With +sp64 +ep16 +H1025 +W1025:
POV-Ray finished
With +H1025 +W1025:
Render Time:
POV-Ray finished
With +sp256 +ep16 +H1025 +W1025:
Render Time:
With +sp32 +ep16 +H1025 +W1025:
Render Time:
POV-Ray finished
So it seems the number of working Threads are accumulated/multiplied by the number of steps on the mozaic.
(and maybe +sp256 is maxed out to a lower value)
As the threads used for the mozaic are probably ended before actually rendering the remaining picture, it's a bit of a lie: there probably was never 72 simultaneous threads alive...
(My default +WT is 18, YMMV)
Corrected 1 pixel error with change #5099
Preview is now fine with beta 39.
(from sources via perforce)
I would like to suggest to close.