[flac-dev] Retuning compression levels

[lvqcl]

Martijn van Beurden wrote:

>> And I also suggested to consider a different setting for -7 preset:
>> -7 == -l 12 -b 4096 -m -r 6 -A tukey(0.5);partial_tukey(2)
>> see <http://www.hydrogenaud.io/forums/index.php?s=&showtopic=106545&view=findpost&p=871797>
>>
>> But it will decrease decoding speed for this preset.
>
> Yes, that is another thing to consider. My proposal keeps the
> decoding speed for all presets the same, but I don't think there
> is a problem in shifting things a little bit more. The above
> setting would make the decoding of -7 as CPU-intensive as -8 is
> now. Seeing that all FLAC-capable hardware that I know of has
> been decoding -8 without problems for years, there's probably no
> harm done with changing this too.
>
> Any thoughts?

 From <http://www.rockbox.org/wiki/CodecPerformanceComparison>
it seems that the decoding of -8 is about 6% slower than -5. Sometimes
the difference is bigger but doesn't exceed 10%.

The only exception is "Freescale i.MX21 (454MHz ARM926EJ?-S, Fuze+)":
43x vs 35x realtime. But the decoding time is so small (4 sec vs 5 sec)
that it can be some measurement error.