Hi, Sorry to bother you on something potentially purposeful - it just seems like you may have meant to temporarily remove this functionality given that some of the code for it exists but is commented out. In JPEGVideoRTPSource.cpp you set the precision in the DQT segments to 0 (meaning 8 bit), even though the corresponding sink supports setting the precision via a call to the quantizationTables method. Line 384 of JPEGVideoRTPSource.cpp is as follows: // unsigned Precision = (unsigned)headerStart[resultSpecialHeaderSize + 1]; Is this a purposeful lack of support for the 16 bit quantization tables, or was it partially implemented then commented to prevent errors? Should I throw away JPEG frames with 16 bit quantization tables in my JPEGVideoSource subclass? If this was intentional would it not be better to remove support in the sink, or is that for compatibility with other receivers?
Regards Josh Hitchen
_______________________________________________ live-devel mailing list live-devel@lists.live555.com http://lists.live555.com/mailman/listinfo/live-devel
