Pages: [1]
Author Topic: FliF  (Read 2774 times)

Cakes 35
Posts: 14520

« on: October 02, 2015, 12:34:31 PM »

another image format that seems to do a lot, but what i'm interested is the "smaller than PNG" and "outperforms PNG" part.

because, I know i've said PNG is slow as hell many many many times and have always stayed away from PNG for that reason.  and the progressive loading might be handy for just skipping the picmip resampling step for lower detail on texture load.

Though now i'm worrying about memory usage and how it behaves on the ideal Pentium 200 target  Cry

EDIT: GPLv3 sucks, so can't really integrate their reference implementation
« Last Edit: October 02, 2015, 01:02:53 PM by fromhell » Logged

asking when OA3 will be done won't get OA3 done.
Progress of OA3 currently occurs behind closed doors alone

I do not provide technical support either.

new code development on github
In the year 3000

Cakes 45
Posts: 4391

« Reply #1 on: October 02, 2015, 01:30:51 PM »

EDIT: GPLv3 sucks, so can't really integrate their reference implementation

Also, "format is not finalized and new versions may break compatibility" may be a problem.

I never want to be aggressive, offensive or ironic with my posts. If you find something offending in my posts, read them again searching for a different mood there. If you still see something bad with them, please ask me infos. I can be wrong at times, but I never want to upset anyone.
grey matter

Cakes 8
Posts: 381


« Reply #2 on: October 02, 2015, 06:16:13 PM »

  • Encoding and decoding speeds are not blazingly fast, but they are in the right ballpark

FLIF does not yet support the following features:
  • A highly optimized implementation

Why trade memory/disk for CPU cycles?

Besides the reference implementation is a C++ binary, not a C library.

This space is for rent.

Cakes 24
Posts: 584

« Reply #3 on: October 03, 2015, 01:26:14 AM »

Improving algorithms is a good thing, especially when the new ones are truly free.

But I think these guys should have added a lossy mode before publicizing this.  Supporting both modalties in a single format would make this new format a lot more compelling (to switch over to it) than just reducing file sizes by 20-35%, imho.
Pages: [1]
Jump to: