Re: Risc PC Sound quality

In message <6559dbfd4e.druck@xxxxxxxxxxxxxxxx>
druck <news@xxxxxxxxxxxxxxxx> wrote:

On 5 Jul 2007 Simon Smith <simon_smith_news@xxxxxxxxx> wrote:
In order to compare [MP3 and CD] I once tried ripping a track
from a CD and converting to WAV, and comparing that against the
highest-quality MP3 I could make of the same track. The most noticeable
difference was in the song balance, actually. The WAV had a guitar twanging
away quietly in the background, but you could hear it if you listened for
it. On the MP3, it came right into the foreground. Quite intrusive. Really
noticeably changed the balance of the track. Not having much in the way of
finely-tuned musical senses, I couldn't really say it sounded /worse/, but
it was certainly different from the reasonably faithful WAV version.

Was it encoded with a dumb encoder (as most RISC OS ones are for
speed) or one which uses a psychoacoustic model. The latter whilst
giving better results in most cases, does rely on people's audio
perception being close to the norm, which is not always the case.

---druck

To be honest I can't remember which encoders. It wasn't a very scientific
test, and it was quite a long time ago. I know it included Lame 3.91 at
various settings. Maybe one or two other Lame versions as well. And I think
I used at least one other encoding program too. Heaven knows what. I had PCs
and Acorns handy at the time, so I think I tried CLI programs on both
platforms. The object was mainly to get a qualitative feel for how the whole
process worked. If I had produced a file with no perceptible differences I'd
have tried a few more CDs in different musical styles to see if they were
affected. Had they all sounded OK, I'd have planted my flag in the 'sounds
fine to me' camp and blithely burned mp3s whenever I felt like it without
worrying too much about the technical intricacies. As it was the differences
were immediately apparent, which told me straight away roughly what to
expect.

I'm quite sure the 'best' version was using a psychoacoustic model because
it was apparent in the output; that compression technique assumes the ear
picks up higher frequencies at the expense of lower frequencies, and pays
more attention to compressing the higher frequencies faithfully. For this
track it brought the weaker but higher-pitched guitar twangs into the
foreground as I reported. Doubtless there are other subtleties to that
compression system too.

--
Simon Smith

When emailing me, please use my preferred email address, which is on my web
site at http://www.simon-smith.org
.

Relevant Pages

  • Re: 2 audio clips for your esteemed perusal
    ... Does the Rockbox firmware let you record to .wav instead of .mp3? ... Even the H2's plug-in power driving the Sony mic might yield some good results versus the crappy H2 internal mics. ... can make things sound thin and distant, but compression and limiting can ...
    (rec.music.makers.guitar.jazz)
  • Re: Mp3 room tone recorded on 320kb/s - creating sounds like garbled water noises
    ... "oleg kaizerman" wrote ... ... OP is describing sound like the MP3 compression is doing its job. ... No issues in sound quality when recording to flac / wav mono or wav poly ...
    (rec.arts.movies.production.sound)
  • Re: How to compress a music wav file
    ... This way you have the compression on an mp3 and the ... embeddability ... to delete pictures and the wav file. ...
    (microsoft.public.powerpoint)
  • Re: ProShow Gold Problem - Solved
    ... results by converting mp3 to wav first. ... the intermediate format issue -- have to experiment with that. ... It probably depends on the compression ...
    (rec.photo.digital)
  • Re: OT Ipods
    ... >>> Analogue tape always had a distinct limitation to how well it could record ... >>> There is absolutely nothing inherent to Fraunhauffer .mp3 compression ... >>> the file size without affecting the sound. ...
    (rec.sport.golf)