Re: [dev] Re: [farbfeld] announce

From: FRIGN <>
Date: Wed, 18 Nov 2015 10:29:49 +0100

On Tue, 17 Nov 2015 18:05:43 -0500
Matthew of Boswell <> wrote:

Hey Matthew,

> It's not "better"; it sucks less. There's a huge difference.
> "better" is a matter of opinion. PPM has a lot of features / alternate
> formats / modes of data representation / endian choices. It supports
> binary 256-color, maximum. Anything higher can be represented in ascii.
> PPM was designed to handle anything and is complex enough to warrant
> having its own library to parse.
> farbfeld has one format/data representation/endian choice. You write one
> set of code that can handle all farbfeld images. It's a good
> intermediate format. Just write converter front-ends for all other
> image formats, and the farbfeld client/consumer code stays much the
> same.

you nailed it, nothing to add! Some attendees asked me at slcon2 about
why I came up with farbfeld when ppm is available.
But the point is to finally provide a simple format you don't need a
finite state machine for to read it in.
I've been working with libpng for 1.5 days now just to find a way to
read in 16-Bit per sample PNG-data while not breaking other aspects
of the program.
libpng is a big mess.



Received on Wed Nov 18 2015 - 10:29:49 CET

This archive was generated by hypermail 2.3.0 : Wed Nov 18 2015 - 10:36:09 CET