Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: TAK 1.0 - Final release of the new lossless codec (Read 250522 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

TAK 1.0 - Final release of the new lossless codec

Reply #175 –
TAK shows some interesting results - some data I thought I'd share 

Got some good results for 24bit files - ALL presets beat FLAC
 i.e. TAK @ 0 compression beats FLAC @ -8 compression. 

Did anyone else find any pleasant irregularities like the following?
[a href="http://img412.imageshack.us/my.php?image=24bit01jp0.jpg" target="_blank"]

Moderation: Amended inline images to linked thumbs

TAK 1.0 - Final release of the new lossless codec

Reply #176 –
Quote from: wildnewt on
Got some good results for 24bit files - ALL presets beat FLAC
 i.e. TAK 0 beats FLAC -8.  Did anyone else find any pleasant irregularities like these?
I'm confused.  Are you referring to the fact that TAK Turbo compresses better than FLAC -8?
I'm on a horse.

TAK 1.0 - Final release of the new lossless codec

Reply #178 –
Quote from: wildnewt on
Quote from: Synthetic Soul on
  Are you referring to the fact that TAK Turbo compresses better than FLAC -8?
yup 
I wouldn't call it an irregularity, as I think everyone saw similar results.

In my comparison TAK Turbo easily beats FLAC -8 using two apodisation switches as well.  I'm not sure that any number of apodisation switches could provide the same compression as TAK Turbo.

There is a lot of talk about the speed comparison between TAK and FLAC, but if you take compression into account there is no comparison - FLAC simply cannot compete in the same arena.

Of course, I still find the speed comparisons interesting, especially given the swift release of FLAC 1.1.4 following 1.1.3, and FLAC's resurgence as the faster decoder.  I look forward to further streamlining from both quarters (hmm... 'both'... 'quarters'... ermm...).
I'm on a horse.

TAK 1.0 - Final release of the new lossless codec

Reply #179 –
Quote from: Synthetic Soul on
Quote from: wildnewt on
Got some good results for 24bit files - ALL presets beat FLAC
 i.e. TAK 0 beats FLAC -8.  Did anyone else find any pleasant irregularities like these?
I'm confused.  Are you referring to the fact that TAK Turbo compresses better than FLAC -8?

Maybe I'm misunderstanding, but it looks like the irregularities he's talking about are the fact that increasing the compression level did not always result in a smaller file.  In the pure tones, Tak High compressed better than Tak Extra, which is certainly unexpected.

TAK 1.0 - Final release of the new lossless codec

Reply #180 –
Quote from: sthayashi on
Maybe I'm misunderstanding, but it looks like the irregularities he's talking about are the fact that increasing the compression level did not always result in a smaller file.  In the pure tones, Tak High compressed better than Tak Extra, which is certainly unexpected.
Yes, that's why I checked before answering, because that really is an irregularity.  However, wildnewt has already answered contrary to that.

I remember seeing a similar thing with FLAC in one of the tests I've been involved in, where (for sake of argument) -5 and -6 compressed worse that -4.  I guess it just depends on the individual sample and the settings the presets are using... and whether the sample finds a 'loophole' in the normal situation.
I'm on a horse.