Actually, I checked my archives, it's 1.2.0 which introduced the additional scheme for rice coding, not 1.2.1, but that doesn't change the gist of it...<div>I remember seeing a note from Josh a long time ago, saying that those new encodings were only for 24-bit files. I has never seen these on standard 16-bit CD-rip type FLACs, but I found them in the first 24-bit file I was confronted with (a simple 24/44.1 file).</div>
<div>It may very well be that the low-effort compression doesn't bother using these.</div><div><br></div><div>--<br>Pierre-Yves Thoulon<br><br>
<br><br><div class="gmail_quote">On Mon, Feb 7, 2011 at 01:51, Brian Willoughby <span dir="ltr"><<a href="mailto:brianw@sounds.wa.com">brianw@sounds.wa.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
What is a "local implementation?" Do you mean the hardware version<br>
number?<br>
<br>
I think Pierre-Yves may be correct. There certainly were some<br>
changes to 24-bit support, and many of these problematic FLAC files<br>
are HD audio. In other words, they're not simply 16-bit 44.1 kHz CD<br>
audio converted to FLAC, but they are 24/96 or 24/192 audio in FLAC<br>
format.<br>
<br>
The only curious thing is that using flac 1.2.1 with --fast or<br>
compression level 0 is enough to make the hardware happy. In that<br>
case, are only the old Rice codings used for lower compression levels<br>
with 24-bit audio?<br>
<br>
You raise a good point, Nicholas. I would like to see manufacturers<br>
give specific information about what level of the FLAC format they<br>
support. The BDP-95 does not mention FLAC in the manual at all, and<br>
the web page only mentions FLAC twice - once in a bold heading, and<br>
again in the body of text. Neither mention of FLAC gives any details<br>
at all - they just put it in the list of formats. I suppose, in<br>
comparison, that MP3 players usually don't give any details about<br>
whether the hardware supports 320 Kb or multichannel or anything<br>
else. Perhaps we're reaching an age where nobody cares about the<br>
details.<br>
<font color="#888888"><br>
Brian Willoughby<br>
Sound Consulting<br>
</font><div class="im"><br>
<br>
On Feb 6, 2011, at 15:24, Nicholas Bower wrote:<br>
> Version 1.2.1 of the standard/spec or the local implementation?<br>
><br>
> I've not seen "FLAC 1.0/1.1 Compliant" or "FLAC 1.2 Compliant" on<br>
> the specs of hardware gear for example when FLAC is stated supported.<br>
><br>
> Just a curious on-looker.<br>
><br>
><br>
> On 7 February 2011 02:34, Pierre-Yves Thoulon<br>
> <<a href="mailto:py.thoulon@gmail.com">py.thoulon@gmail.com</a>> wrote:<br>
>> Version 1.2.1 introduced new rice coding techniques that are used by<br>
>> the reference encoder for 24 bit files. An older version of the<br>
>> decoder will have trouble with frames that use this encoding... Maybe<br>
>> that's where the strange noises come from...<br>
</div><div><div></div><div class="h5">_______________________________________________<br>
Flac mailing list<br>
<a href="mailto:Flac@xiph.org">Flac@xiph.org</a><br>
<a href="http://lists.xiph.org/mailman/listinfo/flac" target="_blank">http://lists.xiph.org/mailman/listinfo/flac</a><br>
</div></div></blockquote></div><br></div>