2000.04.03 13:12 "Re: Complex Floating Point", by Frank Warmerdam

I think the Remote Sensing community is interested in getting this right for purposes of interoperability.

I'd like Frank to elaborate on why he would discourage you from putting 128 bits per sample.

Max,

My rationale was to ensure that the data was still accessable by any floating point savvy application deployed today.

However, your case for an explicit sampleformat type for complex is reasonably convincing to me. Certainly my current client has a vested interest in good support for complex data.

If the list is supportive of adding new sample format values for floating point complex, and signed integer complex values then I am happy to make the appropriate changes in libtiff, and to update some of the higher level applications I work with to honour the new tag values.

Best regards,

---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerda@home.com
light and sound - activate the windows | http://members.home.com/warmerda
and watch the world go round - Rush    | Geospatial Programmer for Rent