MP3 Bitrates also Jitter
 

Seeing Stimpy's post where he mentions 320 Kbit MP3s has got me
wondering.....

I recently re-recorded my 'Top 20' CDs/CD rips at 256 Kbit due to now having
the space to be able to do it and they sound absolutely fine - good enough
to fool the 'Golden Ears Brigade' representative, hereabouts, a few nights
ago, as per my recent post.

I seem to remember someone somewhere saying that the highest rate MP3s can
be spitchy and wonder if anyone here can confirm or deny this? I haven't
heard any such thing myself on the 320s I've tried, but I don't want to put
a lot of effort into 'upping the rate' if there's a possibility I'll regret
it in the future. (I also can't hear much difference myself, but that means
nothing....!! ;-)

Also - playing MP3s from a computer's hard drive - does this remove/reduce
any problems from Jitter? Where is the sound coming from when you play an
MP3 thus - is it read into RAM entirely, if so I would have thought would
remove the possibility of jitter, but would jitter have been already
recorded into the MP3??? Is there any possibility of jitter being 'ironed
into' a recording and then getting 'jittered again' when it's played??

(Although, once again, I can't say I've ever heard 'Jitter' myself - I'm
confused and it, er, gives me the jitters...!!??!! :-)

MP3 Bitrates also Jitter
 

Keith G wrote:

Also - playing MP3s from a computer's hard drive - does this remove/reduce
any problems from Jitter? Where is the sound coming from when you play an
MP3 thus - is it read into RAM entirely, if so I would have thought would
remove the possibility of jitter, but would jitter have been already
recorded into the MP3??? Is there any possibility of jitter being 'ironed
into' a recording and then getting 'jittered again' when it's played??


If you have gone from a CD and taken it direct to mp3, then you won't
add jitter (or remove any), aduming it was all digital, and wasn't
recorded from the analog signal from a CD source.

As to if its read into ram and then pushed at the dac in the sound card,
thats down to the software, and the avalable ram. I suspect it will read
from the disk as required and convert on the fly, but I don't know for
sure. I could take a look at the source for mplayer and see what that does.

Jitter will be introduced by the dac, it doesn't exist as such in the
digital domain, only when the samples are converted to analog.

If you recorded from a LP, then you will have introduced jitter then.

Just how much, is down to the sound card, I would guess it depends on
many things, the stability of the supply to the sound card I would guess
is important, and how regulated the supplies on the card from that
point. I would suspect that the power drawn by the various bits of a
computer vary quite a bit, though I don;t know how much current moving a
head on a disk takes, I would guess rotating the platter is a constant
load though.

I would guess Arni is the best source of this sort of information.


(Although, once again, I can't say I've ever heard 'Jitter' myself - I'm
confused and it, er, gives me the jitters...!!??!! :-)


I wonder if anyone has actually done tests interoducing various amounts
of jitter to a DAC, and seeing just what the audible effects are.
Similar to the measurements of the effect of various harmonics.

I expect it has been done, any references anyone ?

--
Nick

MP3 Bitrates also Jitter
 

Keith G wrote:
Seeing Stimpy's post where he mentions 320 Kbit MP3s has got me
wondering.....

I recently re-recorded my 'Top 20' CDs/CD rips at 256 Kbit due to now
having the space to be able to do it and they sound absolutely fine -
good enough to fool the 'Golden Ears Brigade' representative,
hereabouts, a few nights ago, as per my recent post.

I seem to remember someone somewhere saying that the highest rate
MP3s can be spitchy and wonder if anyone here can confirm or deny
this? I haven't heard any such thing myself on the 320s I've tried,
but I don't want to put a lot of effort into 'upping the rate' if
there's a possibility I'll regret it in the future. (I also can't
hear much difference myself, but that means nothing....!! ;-)

All I can say is I listened very carefully to various encoding methods and
bitrates before deciding on 320kbps MP3 18 months or so back. If I'd have
heard anything specifically negative about 320 I'd have used 256. Not quite
sure what you mean by 'spitchy' but I don't hear anything unwelcome in any
of my 320s. I did think of sticking with 'raw' WAV but I genuinely
couldn't tell the difference between a selction of WAV rips and 320kbps MP3s
of the same CDs.

Having said all that there was, to me, virtually no audible difference
between 256 and 320kbps MP3. I went with 320 merely because I thought it
made sense to go with the highest bit rate I could. After all, disk space
is all but free these days and will only get cheaper


Also - playing MP3s from a computer's hard drive - does this
remove/reduce any problems from Jitter? Where is the sound coming
from when you play an MP3 thus - is it read into RAM entirely, if so
I would have thought would remove the possibility of jitter, but
would jitter have been already recorded into the MP3??? Is there any
possibility of jitter being 'ironed into' a recording and then
getting 'jittered again' when it's played??

(Although, once again, I can't say I've ever heard 'Jitter' myself -
I'm confused and it, er, gives me the jitters...!!??!! :-)

For what it's worth, as of today I've got around 28,000 tracks on my MP3
server - almost all ripped from original CDs at 320 or bootlegs with a known
providence - and I don't notice any unwelcome artifacts or effects.

Like Keith, I'm not sure what 'jitter' is supposed to be so maybe I wouldn't
notice it even if I heard it :-) Ultimately, it sounds fine to me and
that's all that matters really!

MP3 Bitrates also Jitter
 

"Nick Gorham" wrote in message


If you recorded from a LP, then you will have introduced jitter then.

Just how much, is down to the sound card, I would guess it depends on
many things, the stability of the supply to the sound card I would
guess is important, and how regulated the supplies on the card from
that point.

Virtually every sound card, even the sub-$20 cheapies have their own
regulated power supply(s).

I would guess Arny is the best source of this sort of information.

Thank you. I've tried to measure a lot of jitter over the past 6 years, and
rarely found enough to hear. You're spot on with your concerns over power
supply quality, in that most of the jitter I've found was at the power line
frequency or some harmonic of it. I've got a Delta 1010 that has measurable
but not audible jitter that is obviously due to a well-known problem with a
certain cap in the interface box drying out.

I wonder if anyone has actually done tests introducing various
amounts of jitter to a DAC, and seeing just what the audible effects
are. Similar to the measurements of the effect of various harmonics.

Been there done that several ways. Here are some examples of my handiwork
that you can download and listen to for yourself:

http://www.pcabx.com/technical/jitter_power/index.htm

The easiest way to add jitter that I know involves a little creative work
with Audition/CEP.

MP3 Bitrates also Jitter
 

Keith G wrote:

Also - playing MP3s from a computer's hard drive - does this remove/reduce
any problems from Jitter?

It does nothing whatsoever to jitter. the only source of jitter in your
output will be the oscillator on your soundcard.

MP3 Bitrates also Jitter
 

good enough to fool the 'Golden Ears Brigade' representative

IMHO, at least with classical music, the processing of MP3 makes the music
fatiguing and unsatisfactory when listening for any period of time ( 30
mins) even if you can't 'instantly' tell which is which. I've been using
..flac and .ace for classical rips.

I seem to remember someone somewhere saying that the highest rate MP3s can
be spitchy and wonder if anyone here can confirm or deny this?

Surely rather depends on the encoder? LAME's 'insane' preset seems fine on
pop/rock. That said, since joining the ranks of the iPoders I think (so far)
I prefer AAC to MP3 at 320.

MP3 Bitrates also Jitter
 

Hi,

In message , Colin Soames
writes
good enough to fool the 'Golden Ears Brigade' representative

IMHO, at least with classical music, the processing of MP3 makes the music
fatiguing and unsatisfactory when listening for any period of time ( 30
mins) even if you can't 'instantly' tell which is which. I've been using
.flac and .ace for classical rips.

Interesting to hear that - I have similar experience, but I can't really
put my finger on why. I tend to get 'tired' of listening to mp3 files
much sooner than uncompressed audio, even at fairly high bitrates. The
problem is that it's hard to be objective about it, as it takes a while
to discern a pattern. Over time though, I have noticed that I find it
easier to listen to uncompressed music for extended periods, even if I
can't tell any difference in an ABX test.

The compressed music I've encoded recently is either WMA9, VBR 98%
quality setting or Monkey's audio, and I haven't noticed these to be
fatiguing, but with the older stuff I encoded in mp3 format (with LAME
at 256/320 kbps) I tend to turn it off (or lower the volume) sooner than
with wav files from the same source, on the same playback system. I
really don't know if this effect is 'real' but it does seem consistent.
It doesn't seem to happen with Dolby Digital material, but I don't use
that on the same music system, so can't draw any conclusions.

It makes me wonder if compressed music stimulates the ear/brain
differently to uncompressed music, but that's pure speculation. My guess
is that it would be extremely difficult to test accurately because of
the extended time frames. Many people don't seem to have any problem
using mp3 players for hours at a time, so perhaps it's just me!

--
Regards,
Glenn Booth

MP3 Bitrates also Jitter
 

"Nick Gorham" wrote


If you have gone from a CD and taken it direct to mp3, then you won't add
jitter (or remove any), aduming it was all digital, and wasn't recorded
from the analog signal from a CD source.

As to if its read into ram and then pushed at the dac in the sound card,
thats down to the software, and the avalable ram. I suspect it will read
from the disk as required and convert on the fly, but I don't know for
sure. I could take a look at the source for mplayer and see what that
does.

Jitter will be introduced by the dac, it doesn't exist as such in the
digital domain, only when the samples are converted to analog.

If you recorded from a LP, then you will have introduced jitter then.


OK, thanks for that. I am creating MP3s from both LP and CD and, so far,
haven't heard anything to cause me any concern. There is no soundcard
(motherboard sound only) and I only use WMP9 to play them back (valve amp,
of course) and they sound perfectly fine to me. I sometimes think they sound
a little bit better played in Sound Forge though???

MP3 Bitrates also Jitter
 

"Stimpy" wrote


All I can say is I listened very carefully to various encoding methods and
bitrates before deciding on 320kbps MP3 18 months or so back. If I'd have
heard anything specifically negative about 320 I'd have used 256. Not
quite
sure what you mean by 'spitchy' but I don't hear anything unwelcome in any
of my 320s. I did think of sticking with 'raw' WAV but I genuinely
couldn't tell the difference between a selction of WAV rips and 320kbps
MP3s
of the same CDs.

Having said all that there was, to me, virtually no audible difference
between 256 and 320kbps MP3. I went with 320 merely because I thought it
made sense to go with the highest bit rate I could. After all, disk space
is all but free these days and will only get cheaper


OK, that's quite reassuring - like I say, I'm damned if I can hear the
difference!

I guess I can rip them again in the future if I feel the need to - I'll
leave them at 256 for now.

MP3 Bitrates also Jitter
 

Colin Soames wrote:
good enough to fool the 'Golden Ears Brigade' representative


IMHO, at least with classical music, the processing of MP3 makes the music
fatiguing and unsatisfactory when listening for any period of time ( 30
mins) even if you can't 'instantly' tell which is which. I've been using
.flac and .ace for classical rips.

I always found mp3s less stressful to listen to, perhaps partly because
my audio processing is odd and I seem to hear 'everything' without being
able to process it all and seperate it. mp3 often removes the 'chaff' so
perhaps in my case the processing I lack is made up for by the mp3
lossiness.

Interesting how people process audio in different ways.

Sometimes I wonder if its why a high number of aspies are 'hifi
obsessives' since weird audio processing seems to be common among us.