Rockbox.org home
release
dev builds
extras
themes manual
wiki
device status forums
mailing lists
IRC bugs
patches
dev guide



Rockbox mail archive

Subject: Rockbox Sound Quality

Rockbox Sound Quality

From: Douglas Anthony Cooper <dysmedia_at_gmail.com>
Date: Thu, 11 Sep 2008 23:37:01 -0500

On Thursday 11 September 2008 20:43, Douglas Anthony Cooper wrote:

>> Another interesting point: I get *superb* sound quality via the
>> headphone jack of the iPod 5G when I boot into Rockbox. The iPod's
>> native OS sounds considerably worse: less clarity; narrow soundstage,
>> etc. An article I just read

>Where, please?

The test is here: http://forums.rockbox.org/index.php?topic=6139.0

>> suggested this shouldn't be the case:
>> graphed outputs from both systems are identical.

>What are the "graphs"? If you mean analogue sound waveform output, then
>identical means identical. If you mean spectrum-analyser-type frequency vs.
>amplitude charts (sometimes likened to graphic equaliser settings) then this
>can mean anything, really.

They're frequency response graphs, using Rightmark Audio Analyzer

>> So what's going on here? These are mostly AAC files at 160 kbps, and MP3's
>> at approx. 192 (VBR). I'm pretty sure it's not the placebo effect -- the
>> difference is pretty obvious.

>There's no reason why two different codecs (in this case, the "dec" part)
>can't produce noticeably different results from the same input (I assume you
>*are* playing the exact same files when comparing the two systems, yes?)

Of course.

>I'm not familiar with the original firmware on the iPod 5G - can you play a
>lossless (eg: WAV) file and see (hear?) if you notice any difference between
>the two, thereby identifying whether it's the codec algorithm which is
>causing the distinction?

If you look at the page I refer to, you'll see that he provided for
this possible
discrepancy: he tested the codecs with identical compressed files, and managed
to produce precisely the same results: which is to say, no difference
between the
graph produced on rockbox and the graph produced by the same file decoded via
the original firmware.

Am I missing something here?

(Yes, I know the pitfalls of the placebo effect, and I haven't
performed a rigorous
double-blind test -- but as I say: the differences are not subtle.
And I haven't
tried lossless yet -- it's not really of interest to me, although it
would probably
interest people here a fair bit.)
-------------------------------------------------------------------
List admin: http://cool.haxx.se/cgi-bin/mailman/listinfo/rockbox
FAQ: http://www.rockbox.org/twiki/bin/view/Main/GeneralFAQ
Etiquette: http://www.rockbox.org/mail/etiquette.html
Received on 2008-09-12


Page was last modified "Jan 10 2012" The Rockbox Crew
aaa