I think some took the side car route; that also made sense given the likely duplication of data across multiple frames. If you do end up copying the files be sure to use the copy raw pixels path to avoid unnecessary compression cycles.
I believe that we still have a branch with the compressed attribute which could serve well for this purpose. Piotr On 12 November 2014 20:14, Larry Gritz <[email protected]> wrote: > This issue has come up for us as well, and one problem is that we don't > know the set of IDs used until the image is done rendering, and by then, > it's too late to add the id<->name mapping to the header. Not sure if the > right approach is to then copy the file to add the right header data (which > may be huge anyway), or to have a "sidecar file" explaining the mapping. > How did you guys end up doing it? > > > On Nov 12, 2014, at 6:30 PM, Peter Hillman <[email protected]> wrote: > > It doesn't specify that the channel has to be an integer, but then I > suppose if you had less than 2048 integer IDs you could use a half to save > space in the decoded image. There's also no convention for assigning a > human-readable name to an ID number, though Florian did propose a scheme: > http://lists.nongnu.org/archive/html/openexr-devel/2011-04/msg00007.html > > > -- > Larry Gritz > [email protected] > > > >
_______________________________________________ Openexr-devel mailing list [email protected] https://lists.nongnu.org/mailman/listinfo/openexr-devel
