Head-Fi.org › Forums › Equipment Forums › Portable Source Gear › FLAC levels question ...
New Posts  All Forums:Forum Nav:

FLAC levels question ...

post #1 of 24
Thread Starter 
Will startt using EAC setup with FLAC and it looks like there are 8 levels available.

At one extreme (0 or 8) the files are bigger but it takes less encoding time and at the other extreme (8 or 0) the files are smaller but it takes longer to encode.

I can't help but feel that the larger file have more relevant information and is, therefore, more 'perfectly lossless' than the smaller, but also 'perfectly lossless', file.

Does selecting one specific FLAC level over another have any impact on playback SQ?

TIA
post #2 of 24
no, they are all identical in quality. the only difference is the higher levels take more time to optimize things, thus the file size is a bit smaller, but it takes longer.

http://wiki.hydrogenaudio.org/index....sked_questions
post #3 of 24
It is also important to note that while the higher compression rate (8) takes more processor time to encode, there is no difference in the processor time it takes to decode FLAC files of different compression rates. The higher compression settings are a one-time expense for smaller file sizes.
post #4 of 24
agree with other posters there, same qual, but only go past midway if you REALLY need that extra few mb storage space on your player/storage

as much as i hate to leave something at its 'default' setting , i must admit i find FLAC 5 to be best for me, especially as i am encoding on a latitude x300 laptop which isnt very powerful.

anything higher just isnt worth the small space saving vs time taken for me personally.
post #5 of 24
Flac level 5 is near identical in filesize to 8, yet it takes only a fraction of the time to encode to flac 5 vs 8. Decode speed is all the same though.

All the levels are perfectly lossless. Level 8 just spends a lot more time trying to find patterns than Level 0.
post #6 of 24
I don't use FLAC 8 because it need a lot more processor task to decode it than say FLAC 5. And this high processor usage decreases my Iriver H120 battery life.
I realised that the file size difference between FLAC 5 and 8 don't worth the extra battery needed. So, FLAC 5 is my choice for now ...
post #7 of 24
Quote:
Originally Posted by epaludo View Post
I don't use FLAC 8 because it need a lot more processor task to decode it than say FLAC 5.
I'm sorry, but this is simply not true. The processor time needed to decode a FLAC file is completely independent of the compression level used.
post #8 of 24
Quote:
Originally Posted by epaludo View Post
I don't use FLAC 8 because it need a lot more processor task to decode it than say FLAC 5. And this high processor usage decreases my Iriver H120 battery life.
I realised that the file size difference between FLAC 5 and 8 don't worth the extra battery needed. So, FLAC 5 is my choice for now ...
Quote:
Originally Posted by Redo View Post
Flac level 5 is near identical in filesize to 8, yet it takes only a fraction of the time to encode to flac 5 vs 8. Decode speed is all the same though.

All the levels are perfectly lossless. Level 8 just spends a lot more time trying to find patterns than Level 0.
Ok so if the decode speed is identical for all compression levels, why would certain levels decrease battery life?? Is more processing power needed to decode higher compressions?

EDIT: Answer:
Quote:
Originally Posted by SysteX View Post
I'm sorry, but this is simply not true. The processor time needed to decode a FLAC file is completely independent of the compression level used.
So battery life will not be (negatively) affected if one decides to use higher compression levels.
post #9 of 24
Quote:
Originally Posted by dissembled View Post
So battery life will not be affected if one decides to use higher compression levels?
Correct. Without going into too much detail about the specifics of the codec, a good oversimplification is that during compression, many different algorithms are tested to try to model the data. The higher the compression level, the longer it searches for a good algorithm--usually resulting in a more efficient modeling of the data (hence a smaller file size). When the compressed file is written, that algorithm is written into the file as well--to decode, simply apply the algorithm. All the algorithms, regardless of compression level, are stable and relatively computationally cheap; all the work is done upfront during encoding trying to find that ideal algorithm.
post #10 of 24
Quote:
Originally Posted by SysteX View Post
Correct. Without going into too much detail about the specifics of the codec, a good oversimplification is that during compression, many different algorithms are tested to try to model the data. The higher the compression level, the longer it searches for a good algorithm--usually resulting in a more efficient modeling of the data (hence a smaller file size). When the compressed file is written, that algorithm is written into the file as well--to decode, simply apply the algorithm. All the algorithms, regardless of compression level, are stable and relatively computationally cheap; all the work is done upfront during encoding trying to find that ideal algorithm.
So theoretically, wouldn't a higher compression size increase battery life due less hard drive spinup?
post #11 of 24
Quote:
Originally Posted by dissembled View Post
So theoretically, wouldn't a higher compression size increase battery life due less hard drive spinup?
Yes, theoretically. I don't know how much savings you'd actually see in practice though. In my experience, the difference between compression level 5 (the default) and 8 is anywhere from 20MB to a just a couple of MB (EDIT: for an entire album). My guess is that in reality, you wouldn't ever be able to tell the difference in battery life--especially when there are so many other small things that you can do for which you will see a real battery life savings (turning down the LCD brightness, not using the screen as much, conditioning the battery properly when you first get the player, etc.).
post #12 of 24
From what i've read before, the player needs less job to decompress FLAC 0 than FLAC 8. This way, FLAC 8 consumes more energy to be decompressed than FLAC 0. That is not correct ... ?
post #13 of 24
With the Cowon i7 8gb allowing 20hrs of FLAC playback, I've been thinking of ripping my CD collection to FLAC 2.

FLAC 2 is only about 5% larger than FLAC 8, and no floating point calculations are required to decode FLAC 0-2 - so FLAC 2 should be easier on the portable player's battery power without taking up too much more space. But I haven't tested this hypothesis yet.
post #14 of 24
From what I understand, no matter what compression level is used, decoding involves only integer arithmetic.
post #15 of 24
Quote:
Originally Posted by SysteX View Post
From what I understand, no matter what compression level is used, decoding involves only integer arithmetic.
If so, then definitely decoding task is independent from the compression level and did not have influence on the battery life. This could clear up lot of people minds.
Will run a test when the time to run my player battery flat arrives. This is probably the easiest way to kill this doubt ...
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Portable Source Gear
Head-Fi.org › Forums › Equipment Forums › Portable Source Gear › FLAC levels question ...