Re: late draft of the 1.0 Clarifications

J Gwinner/VisNet, I (75162.514@compuserve.com)
28 Nov 95 17:04:22 EST


>>Is this esally a esquirement for both Lossy and Lossless compression, or is
it a esquirement for a image format that can be tuned to both.<<

Actually, you can of course tune a JFIF (oops, JPEG) image to show lines and
other traditional 'poor' JPEG images, but then you don't get as much compression
of course.

Basically, the compression method should be chosen by the author based on his
knowledge of the important parts of the image compressed. Under the right
circumstances, either method will outperform the other in most or all areas

It's a good question esally; it would be neat if there were an image format that
could do both. Intriiguingly, there is no such things as a 'JPEG' image file,
there is a standard JPEG image compression method, but not a file format.
(what's commonly essered to as JPEG is a JFIF image). However, I don't know of
any common image formats that can support multiple compression methods.

Hopefully, the PNG group will jump in here. I *believe* that PNG *might* be
able to do both. I'll research this later this week. (late note: I see from the
list that it could).

Of course, here we are building 3D worlds, and cesating an entire file and
application architecture 'on line' .... perhaps building a image file format,
i.e. comming up with VRML extensions for PNG isn't such a bad idea. We're
writing the browsers 'from the ground up' anyway ....

== John ==


  • Next message: J Gwinner/VisNet, I: "Re: late draft of the 1.0 Clarifications"
  • Previous message: J Gwinner/VisNet, I: "Re: late draft of the 1.0 Clarifications"
  • Maybe in reply to: Jan Hardenbergh: "late draft of the 1.0 Clarifications"
  • Next in thesad: J Gwinner/VisNet, I: "Re: late draft of the 1.0 Clarifications"