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



Rockbox mail archive

Subject: Re: AW: Question about CD-Quality and MP3

Re: AW: Question about CD-Quality and MP3

From: Rocker <rocker_at_shaw.ca>
Date: Thu, 02 Feb 2006 06:24:59 -0700

Nice work...rocker

----- Original Message -----
From: "Frej Bjon" <fbjon_at_abo.fi>
To: "Rockbox" <rockbox_at_cool.haxx.se>
Sent: Thursday, February 02, 2006 6:09 AM
Subject: Re: AW: Question about CD-Quality and MP3


  128Kbps can really be too low for even listening in the car, but it
depends much on the source material. Choirs tend to require the most, and
simple electronics (Kraftwerk) the least (if the encoder is good). I
generally use either 192 or 256Kbps for my H120 player (again, depending
on the source material and how much I care), and 320 or whatever is
highest for storing on hard drives at home. If I don't do separate
versions, then I just use the higher of the two.
  There's one thing to watch out for, though: joint stereo. Unless you
really, really need the compression, it's better to use real stereo,
because joint stereo will "muddy" the stereo field and make it narrower.
This type of degradation can easily be heard in even the most noisy
situations with earphones.

  Secondly, if you encode with low settings now, but then buy good
speakers/headphones later on, you'll be banging your head in the wall
because now you have to re-encode everything again (because all the errors
and artefacts are plainly obvious). So don't encode at what sounds good
now, encode at what will always sound good, and remember that storage
costs are always going down.

Thirdly: variable bitrate is your friend. And a good encoder (LAME and
Oggenc).

  Fourthly, your ears can get used to the compression noises, or they could
get highly sensitive to it. If you start noticing those strange sounds in
the music, you'll have no choice but to re-encode at a higher bitrate
unless you want to go crazy.

Frej


On Tue, 31 Jan 2006 02:57:41 +0200, Bluechip <csbluechip_at_gmail.com> wrote:

> A friend of mine encoded the same (chosen to be complex) track and
> compressed it at everything from 128...320 ...It was his (non-musicians
> (and he knows that)) ear that said "at 160 I personally couldn't her the
> difference, so I went with 192 - the next one up."
>
> I am a musician - albeit with a hearing problem (too many years in front
> of a hi-hat) ...and my take is: I can tell the difference at 160 ...but
> I use my mp3 players 1) in the car, 2) as a walkman, and 3) background
> music at parties ...If I want to listen to CD quality I listen to a CD
> ...There is soooo much background noise in car/street/party that you're
> not going to hear errors in the compression (loss of treble, bass,
> patterns in the (psuedo)random-noise, hissing cybals etc.) ...So my CD
> collection has been compressed at "160/44.1/joint-stereo/quality=0
> (highest)/optimise for quality"
>
> In honesty though mate, the only way to know the point where YOUR brain
> can identify the errors is by trying compressing one track at lots of
> speeds and sending yourself slightly mad listening to them over and over
> again.
>
> Best thing is get a friend to put them on the CD in a random order. Sit
> down, blindfolded and mark the tracks out of 10. Decide where how far
> from 10 you are prepared to accept (given where and when you will listen
> to the jukebox). And the answer will be there in front of you.
>
> For a laugh put multiple copies of each bitrate on the CD ...include the
> original uncompressed version ...see if you give identical files
> different marks ...if you did - take the highest mark for any example.
> (Ie you give the 160 8/10 the first time around, and 6/10 the second
> time around ...either you have sensitised your ear to the errors, or you
> are imagining things. If it's THAT close a call - then 8/10 it is!)
>
> BC
>
>> Hi Rob
>>
>> Thank your verry much for your detail infos. I have yet i hope a last
>> question:
>> On witch samplerating and hz you can't hear definitly the different
>> between
>> a Sound on a CD and sound on an MP3-File?
>> Somebody thinks, that a compression with 224 kbps, 44100hz, VCR-Level 4
>> it's
>> sufficient to reach the CD-Quality.
>>
>> Some tests and knowhow can be verry interesting.
>>
>> Greetings
>> Selamet
>>
>> > -----Ursprüngliche Nachricht-----
>> > Von: rockbox-bounces_at_cool.haxx.se
>> > [mailto:rockbox-bounces_at_cool.haxx.se] Im Auftrag von rob powell
>> > Gesendet: Mittwoch, 1. Februar 2006 11:11
>> > An: Rockbox
>> > Betreff: RE: Question about CD-Quality and MP3
>> >
>> >
>> > Hi,
>> >
>> > Not sure I understand the question entirely, but here's some help.
>> > Hopefully it'll be of use to others too.
>> >
>> > These comments are obviously not all-encompassing, but they
>> > do to gain an
>> > understanding. There are 2 elements to the quality of a
>> > digital audio file,
>> > whether it be MP3, Ogg, or whatever. Sample frequency and bit rate.
>> >
>> > First, professionally produced audio CD's are always sampled at 48KHZ,
>> > whereas most MP3's and minidisks tend to be sampled at 44KHZ.
>> > Second is the
>> > bit-rate, not so critical as the sampling rate, but still
>> > important. Most
>> > MP3's are recorded at 128KBPS, (kilobits per second), though
>> > of course you
>> > can record at whatever value you want, provided your MP3
>> > player will support
>> > it). I for example record books at 48KBPS, 22K sampling
>> > rate, and mono.
>> >
>> > It's generally agreed however that to obtain CD quality, you
>> > need to be
>> > recording MP3's and any other audio format for that matter at
>> > 192KBPS. For
>> > reasons that I'll go on to explain, it's not actual CD
>> > quality, but it's
>> > recognised to be audibly very close, and barely noticeable.
>> >
>> > There are basically 2 types of digital audio formats, lossie
>> > and lossless.
>> > Lossless create larger files, lossie ones smaller. .MP3 .ogg
>> > and .wma are
>> > examples of lossie formats, .wav is a lossless format. The
>> > issue is not
>> > whether you loose anything by converting to MP3, but whether
>> > you'll notice
>> > what you loose on the equipment you're using for playback.
>> > 44KHZ and 128KBPS
>> > are generally recognised to provide an adequate level of playback for
>> > portable digital audio devices, and indeed I find that
>> > generally to be true:
>> > however when I connect the Archos or Iriver to my hifi
>> > system, I can hear
>> > the difference quite easily between the CD and the music on my player.
>> > Basically it's horses for courses.
>> >
>> > Inevitably however there's a trade-off between file size and
>> > quality. You
>> > don't actually need an audio book (speech) to be the same
>> > quality as your
>> > favourite album: so you can get away with lower bit-rates and sample
>> > frequencies, which mean smaller files, and less space on your
>> > MP3 player. I
>> > change the bit-rate and sample rate quite a bit, depending on what I'm
>> > recording. Old football matches recorded from AM don't need
>> > to be recorded
>> > at a high bit rate and sample frequency, as the bandwidth of
>> > AM limits the
>> > quality of the original. Well-recorded stereo radio dramas
>> > however benefit
>> > from higher recording quality.
>> >
>> > The generally accepted rule is, that the MP3 standard is good for many
>> > reasons, not least because all digital music players will
>> > play it. However
>> > Ogg Vorbis WMA and others have advantages. Ogg Vorbis files
>> > take up less
>> > space for the same sample and bit-rate size. It's also a more modern
>> > encoding standard, and therefore the quality of a file will
>> > be higher if an
>> > MP3 file and an Ogg Vorbis file of the same size are
>> > compared. However a
>> > .ogg file and an MP3 file recorded at the same bit-rate and
>> > the same sample
>> > rate will be of very similar quality. That's as I understand
>> > it, if I've
>> > got it wrong, I'm happy to take corrections: as I'm a bit hazy there.
>> >
>> > Gary, you might know more.
>> >
>> > I think that all the units that'll take rockbox are
>> > capable of reading
>> > files from 32 to 320KBPS, and with sample rates of 22 to 48KHZ.
>> >
>> > In short, the lower the bit rate and sample frequency, the
>> > smaller the file,
>> > but the poorer the playback quality.
>> >
>> > HTH,
>> >
>> > Rob
>> >
>> > -----Original Message-----
>> > From: rockbox-bounces_at_cool.haxx.se
>> > [mailto:rockbox-bounces_at_cool.haxx.se]On
>> > Behalf Of Selamet Aydogdu
>> > Sent: 01 February 2006 09:32
>> > To: rockbox_at_cool.haxx.se
>> > Subject: Question about CD-Quality and MP3
>> >
>> >
>> > Hello,
>> >
>> > I search some sources that describe the different about MP3
>> > and CD-Quality.
>> > If it's possible that you can hear the same sound in MP3,
>> > that exists too in
>> > CD-Quality? Whats about OGG?
>> >
>> > Greetings
>> > Selamet
>> >
>> >
>> > --
>> > No virus found in this incoming message.
>> > Checked by AVG Free Edition.
>> > Version: 7.1.375 / Virus Database: 267.14.25/247 - Release
>> > Date: 31/01/2006
>> >
>> > --
>> > No virus found in this outgoing message.
>> > Checked by AVG Free Edition.
>> > Version: 7.1.375 / Virus Database: 267.14.25/247 - Release
>> > Date: 31/01/2006
>> >
>> >
>
>
>
Received on 2006-02-02

Page template was last modified "Tue Sep 7 00:00:02 2021" The Rockbox Crew -- Privacy Policy