2 Feb
2015
2 Feb
'15
6:40 a.m.
Something has been bugging me...
Ardour is doing 32-bit FP math to handle samples internally.
And yet, when I change the volume on the master fader, even if I
compensate for the volume change on my RME Multiface's physical volume
knob for the headphone monitor, I'd almost swear I can hear some sort of
change in the audio. I'm not using any post-fader plugins. I'm
probably just fooling myself, but is it possible to change the perceived
audio by just changing the output level over a 4-5db range when you've
got a precision like Ardour is using, and you're listening through
decent 24-bit converters?
In a related question (which was the real reason I wanted to know), what
kind of levels should we be shooting for in the master output anyway?
Of course, in the old analog tape world, they used to shoot for zero,
maybe a little hotter if you wanted some saturation. In digital, they
say mix to about -3 or -4db. And mastering houses seem to want more
headroom than that (-6 or 7db), even though in priciple, with 24-bit
resolution, they could just adjust it themselves without changing it
timbrally (true?). Why do they care what your maximum level is as long
as it's 24-bit and it's not clipping?
Commercial CD's these days seem to be mastered with the idea that if the
level ever fell below -1db, that would be a sin against God.
Suppose I'm just trying to produce a demo that can be burned to CD,
uploaded to a streaming service, or given to a mastering house. Is
there a different ideal level for those various uses?
I suppose the question I'm wondering is, if you've got it just the way
you like it at some particular master fader setting, is anything harmed
by making it a little hotter before you export, especially if you're
going to be putting a track on CD where it will be only 16-bit? Or
should you just keep it where it is, even if you're wasting some headroom?
--
+ Brent A. Busby + "We've all heard that a million monkeys
+ Sr. UNIX Systems Admin + banging on a million typewriters will
+ University of Chicago + eventually reproduce the entire works of
+ James Franck Institute + Shakespeare. Now, thanks to the Internet,
+ Materials Research Ctr + we know this is not true." -Robert Wilensky
2 Feb
2 Feb
8:05 a.m.
On Sun, 1 Feb 2015, Brent Busby wrote:
Ardour is doing 32-bit FP math to handle samples
internally.
And yet, when I change the volume on the master fader, even if I
Which master fader? The one in Ardour or the one in ALSA?
compensate for the volume change on my RME
Multiface's physical volume
knob for the headphone monitor, I'd almost swear I can hear some sort of
change in the audio. I'm not using any post-fader plugins. I'm
probably just fooling myself, but is it possible to change the perceived
audio by just changing the output level over a 4-5db range when you've
got a precision like Ardour is using, and you're listening through
decent 24-bit converters?
Ardour's steps (around 0db) are very small, sometimes as many as 4 step
for each .1db. (ctl/alt/mouse wheel or send MIDI pitch wheel in MC mode)
ALSA's step are a lot grainier. (my ice1712 with mudita24 is .5db steps
and Qasmixer gives 1db steps)
In a related question (which was the real reason I
wanted to know), what
kind of levels should we be shooting for in the master output anyway?
I have heard -6 to -3.
Of course, in the old analog tape world, they used to
shoot for zero,
maybe a little hotter if you wanted some saturation. In digital, they
With FP, it is possible to go over 0 :) but not suggested.
say mix to about -3 or -4db. And mastering houses
seem to want more
headroom than that (-6 or 7db), even though in priciple, with 24-bit
resolution, they could just adjust it themselves without changing it
timbrally (true?). Why do they care what your maximum level is as long
as it's 24-bit and it's not clipping?
6db down is one bit of 24... to go to 16 for a CD means cutting off bits
on the bottom (maybe adding some noise) so -6 is safe.
Commercial CD's these days seem to be mastered
with the idea that if the
level ever fell below -1db, that would be a sin against God.
This depends on your audience. This is not always the end listener, but
may be the broadcaster or distributor or whoever is paying for it.
Dynamics are nice IMO.
I suppose the question I'm wondering is, if
you've got it just the way
you like it at some particular master fader setting, is anything harmed
by making it a little hotter before you export, especially if you're
going to be putting a track on CD where it will be only 16-bit? Or
should you just keep it where it is, even if you're wasting some headroom?
Only 16 bit? High quality tape with noise reduction is 13 bit or so.
Try for yourself three mixes, one at -1 and one at -3 and one at -6.
Listen for the difference. Run them on more than one player at a time and
switch them with jack or a mute key and see if the slightly higher level
is worth it.
--
Len Ovens
www.ovenwerks.net
3:30 p.m.
Len Ovens <len(a)ovenwerks.net> writes:
Which master fader? The one in Ardour or the one in
ALSA?
Sorry...Ardour's.
Ardour's steps (around 0db) are very small,
sometimes as many as 4
step for each .1db. (ctl/alt/mouse wheel or send MIDI pitch wheel in
MC mode) ALSA's step are a lot grainier. (my ice1712 with mudita24 is
.5db steps and Qasmixer gives 1db steps)
No, I'm not meaning that I can hear stairsteps as the volume changes. I
was meaning that I thought I noticed timbral changes somehow. I
experimented around with it more, and I'm pretty sure now I was just
falling for Fletcher-Munson effect -- my perception of EQ was changing
at different sound pressure levels.
6db down is one bit of 24... to go to 16 for a CD
means cutting off
bits on the bottom (maybe adding some noise) so -6 is safe.
Wow, I never would have imagined that.
So, I could mix to -6db in Ardour and still use most of the headroom on
a CD? Is there any advantage to keeping your peak level that low on the
master in the digital world? Or, put another way, if I *could* get the
whole track to behave itself and not go above -1db at loudest peak, is
there a reason not to?
--
+ Brent A. Busby + "We've all heard that a million monkeys
+ Sr. UNIX Systems Admin + banging on a million typewriters will
+ University of Chicago + eventually reproduce the entire works of
+ James Franck Institute + Shakespeare. Now, thanks to the Internet,
+ Materials Research Ctr + we know this is not true." -Robert Wilensky
3:41 p.m.
On Mon, Feb 2, 2015 at 2:30 PM, Brent Busby <brent(a)keycorner.org> wrote:
So, I could mix to -6db in Ardour and still use most
of the headroom on
a CD? Is there any advantage to keeping your peak level that low on the
master in the digital world? Or, put another way, if I *could* get the
whole track to behave itself and not go above -1db at loudest peak, is
there a reason not to?
Throughout the digital signal flow (JACK, JACK clients) assuming a
"somewhat" normal signal level, there is no perceivable degradation of
the signal. Traditionally degradation of sound was caused by lack of
bits to represent sound: signal-noise-ratio is proportional to number
of bits used. Floating point removes this cause of degradation.
Soundcards play integer streams: not floating point, so the
floating-point samples are converted to integers. During DAC stage the
number of bits used is important: so a very low signal level will use
less bits, causing a worse SNR (signal-noise-ratio).
In short: mix with enough headroom in Ardour, ensure that the
master-output level in Ardour is happy - ideally somewhere between -20
and -6dB. When that signal arrives at the DAC, it will not
significantly impact on SNR. Then use an analog volume fader to change
the speaker-volume.
Do *not* do this:
-Good Ardour, analog dials on speakers turned up to max, and digitally
(Ardour master fader) regulate the volume (eg -60dB through DAC causes
much larger SNR).
Hope that helps, -Harry
--
http://www.openavproductions.com
4:03 p.m.
On Mon, 2 Feb 2015 14:41:44 +0000, Harry van Haaren wrote:
In short: mix with enough headroom in Ardour, ensure
that the
master-output level in Ardour is happy - ideally somewhere between -20
and -6dB. When that signal arrives at the DAC, it will not
significantly impact on SNR. Then use an analog volume fader to change
the speaker-volume.
You don't want to give away a demo that has got such a low level.
I would stay below 0.0 dBFS to ensure that I wasn't sleeping during
mastering and the 0.0 dBFS by accident were above 0.0 when recording
the master, IOW around I would try to have peaks between a little
bit below -1.5 dBFS and -1.5 dBFS, if possible even - 0.5 dBFS. Sure for
live recording to much headroom is better than to less headroom.
Assumed the meters suffer from the -3.0 vs 0.0 dBFS sine wave vs square
wave issue, then I would replace -0.5 dBFS with -3.5 dBFS for the
highest peak. Again, the headroom for private, professional, live
recording etc. is something else, but if you give away a demo, don't
expect the people who are listening to adjust the amp's volume. I'm not
talking about the kind of the used compression, just about the peaks,
for a demo you will give away.
3 Feb
3 Feb
10:58 p.m.
On Mon, Feb 02, 2015 at 02:41:44PM +0000, Harry van Haaren wrote:
In short: mix with enough headroom in Ardour, ensure
that the
master-output level in Ardour is happy - ideally somewhere between -20
and -6dB. When that signal arrives at the DAC, it will not
significantly impact on SNR. Then use an analog volume fader to change
the speaker-volume.
Right.
Signal to quantisation noise ratios are not a concern in
most cases even with 16 bits - other factors will dominate
the result. Some numbers to think about:
16 bits gives you a dynamic range of more than 97 dB, that
is unweighted and without dithering. Assume your average
RMS level is 15 dB below maximum, and that the listener
will adjust her volume to 95 dB SPL (that is quite loud).
Quantisation noise will be at 12 dB SPL. That is below
the ambient noise level in most places. With dithering
on the final 16-bit result it will even be better.
Second example: You're recording an artist who produces
80 dB(A) SPL, and the self-noise of your mic is 12 dB(A),
a typical value for a good quality studio condenser mic.
Then the S/N ratio of the recording can't be better than
68 dB, the equivalent of around 11 bits.
Ciao,
--
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
11:13 p.m.
On Tue, 3 Feb 2015 21:58:32 +0000
Fons Adriaensen <fons(a)linuxaudio.org> wrote:
On Mon, Feb 02, 2015 at 02:41:44PM +0000, Harry van
Haaren wrote:
There is a really good video explaining this very clearly but in very
understandable terms here:
http://xiph.org/video/vid2.shtml
--
Will J Godfrey
http://www.musically.me.uk
Say you have a poem and I have a tune.
Exchange them and we can both have a poem, a tune, and a song.
In short: mix with enough headroom in Ardour,
ensure that the
master-output level in Ardour is happy - ideally somewhere between -20
and -6dB. When that signal arrives at the DAC, it will not
significantly impact on SNR. Then use an analog volume fader to change
the speaker-volume.
Right.
Signal to quantisation noise ratios are not a concern in
most cases even with 16 bits - other factors will dominate
the result. Some numbers to think about:
16 bits gives you a dynamic range of more than 97 dB, that
is unweighted and without dithering. Assume your average
RMS level is 15 dB below maximum, and that the listener
will adjust her volume to 95 dB SPL (that is quite loud).
Quantisation noise will be at 12 dB SPL. That is below
the ambient noise level in most places. With dithering
on the final 16-bit result it will even be better.
Second example: You're recording an artist who produces
80 dB(A) SPL, and the self-noise of your mic is 12 dB(A),
a typical value for a good quality studio condenser mic.
Then the S/N ratio of the recording can't be better than
68 dB, the equivalent of around 11 bits.
Ciao,
4 Feb
4 Feb
9:42 a.m.
2015-02-03 23:13 GMT+01:00 Will Godfrey <willgodfrey(a)musically.me.uk>uk>:
There is a really good video explaining this very clearly but in very
understandable terms here:
http://xiph.org/video/vid2.shtml
Thanks Will that was a really great video!
Regards,
Robert
2 Feb
2 Feb
7:18 p.m.
On Mon, 02 Feb 2015 08:30:27 -0600, Brent Busby wrote:
No, I'm not meaning that I can hear stairsteps as
the volume changes.
I was meaning that I thought I noticed timbral changes somehow. I
experimented around with it more, and I'm pretty sure now I was just
falling for Fletcher-Munson effect -- my perception of EQ was changing
at different sound pressure levels.
Or did very low levels miss dithering for your mastering? JFTR I've got
loss of sound quality issues when using Linux, that I don't have when
using professional or even consumer digital hardware or even 4-track
analog tapes with Dolby C. IMO there's something broken, at least on all
Linux machines I used and I'm still using. However, at some point you
need dithering, so perhaps that's the culprit for your masterings?!
3 Feb
3 Feb
11:42 a.m.
New subject: master levels (and dithering)
Quoting Ralf Mardorf <ralf.mardorf(a)alice-dsl.net>et>:
JFTR I've got
loss of sound quality issues when using Linux, that I don't have when
using professional or even consumer digital hardware or even 4-track
analog tapes with Dolby C. IMO there's something broken, at least on all
Linux machines I used and I'm still using.
If I record an ALESIS D4 cymbal with Linux, the result
is week. If I
record it with a consumer DAT recorder, there's no audible difference
to the original sound. Even a 4-track cassette player with Dolby C
still sounds better, than the Linux recording (tested with TerraTec and
RME cards).
Linux dithering is a PITA, it adds annoying audible
noise, no stand
alone consumer gear, no proprietary DAW and especially no professional
gear comes with such audible dithering noise.
In the same vein you claim to be able to hear the difference between a
properly encoded MP3 and the original WAV but refuse to do an ABX test
to see if you really can tell them apart.
If your ears are so golden you can hear the linux dithering noise you
should be able to ace an ABX test with ease.
Take an undithered file, dither it in linux and in some professional
gear and see if you really can ABX them.
- Peder
1:19 p.m.
New subject: master levels (and dithering)
Peder,
are you aware about the zero-copy issue?
If you mix e.g. all your Qtractor tracks and master them to a stereo
Qtractor track, this digital mix does cause a loss of sound quality.
If you should be that hearing-impaired, that you even are unable to
notice this, then take a look at spectrograms. Likely there's a very
old thread somewhere in the Qtractor mailing list archive.
It's not a Qtractor issue, it's a jackd issue.
Linux audio is full of pitfalls.
Peder, any hints for this problem:
On Mon, 02 Feb 2015 22:25:49 +0100, Atte wrote:
I had a fairly unpleasant experience last weekend. A
band (indie/folk)
in which I play and for whom I also did/does some recordings + mixes
for, were played on national danish radio on the national chart. The
unpleasant part was that our song sounded really wimpy compared to the
other tracks.
ABX test? Is Atte the most bad audio engineer on this planet?
Whitewashing vs reality. Perhaps Atte is able to find out how many of
the other bands used Linux too.
Never ever the problem was that the other bands used more compression.
And I don't think that Atte is the most bad audio engineer. I suspect
that he's a victim of Linux audio pitfalls. Next time Atte should
compare his mastering with recordings of other bands, instead of
trusting the Linux audio community.
It's possible to do good recordings with Linux, but you need to care
about much stuff, no audio engineer likes to care about and you have to
check the master 20 times before you give away a demo.
That I mentioned recordings of relatively good cymbal samples from a
drum sample player has got a reason.
Regards,
Ralf
7:58 p.m.
New subject: master levels (and dithering)
On 02/03/2015 01:19 PM, Ralf Mardorf wrote:
I suspect
that he's a victim of Linux audio pitfalls.
Hardly :-)
As stated in other post the recording was done on zoom r24 and mixed in
reaper/wine. I'm not posting since this was done with linux (cause it's
hardly), merely as a reply to the master level discussion...
Next time Atte should
compare his mastering with recordings of other bands, instead of
trusting the Linux audio community.
Ahem. How do you know who I trust?
It's possible to do good recordings with Linux,
but you need to care
about much stuff, no audio engineer likes to care about
...which is why I use reaper and renoise as my DAWs. Heretic some say,
focus on music I'd say...
--
Atte
http://atte.dk http://a773.dk
10:30 p.m.
New subject: master levels (and dithering)
Ralf, we (readers of this list) are used to your occasional
fits, to hearing for the zillionth time about your troubles
with Brauner, and that only two of the eight ADAT channels
of your sound card are usable.
What you have been writing recently is different. Gratituous
drivel about how bad Linux is for audio, without any hint of
evidence and without any reasoning behind it.
Some examples from the last days:
JFTR I've got loss of sound quality issues when
using Linux,
that I don't have when using professional or even consumer
digital hardware or even 4-track analog tapes with Dolby C.
IMO there's something broken, at least on all Linux machines
I used and I'm still using.
If I record an ALESIS D4 cymbal with Linux, the result
is week.
If I record it with a consumer DAT recorder, there's no audible
difference to the original sound. Even a 4-track cassette player
with Dolby C still sounds better, than the Linux recording
(tested with TerraTec and RME cards).
Linux dithering is a PITA, it adds annoying audible
noise, no
stand alone consumer gear, no proprietary DAW and especially
no professional gear comes with such audible dithering noise.
Is there usable dithering available for Linux?
If you mix e.g. all your Qtractor tracks and master
them to a
stereo Qtractor track, this digital mix does cause a loss of
sound quality. If you should be that hearing-impaired, that
you even are unable to notice this, then take a look at
spectrograms. Likely there's a very old thread somewhere in
the Qtractor mailing list archive. It's not a Qtractor issue,
it's a jackd issue.
What's wrong with you ?
--
FA
A world of exhaustive, reliable metadata would be an utopia.
It's also a pipe-dream, founded on self-delusion, nerd hubris
and hysterically inflated market opportunities. (Cory Doctorow)
11:01 p.m.
New subject: master levels (and dithering)
On Tue, Feb 3, 2015, at 09:51 PM, Paul Davis wrote:
On Tue, Feb 3, 2015 at 4:30 PM, Fons Adriaensen
<fons(a)linuxaudio.org>
wrote:
I posted a gratuitously snarky reply to something of Ralf's a few months
ago, he objected to it, and I realised I had been too personal.
Since then I've been reading his stuff with more care and, although I do
think he has always posted far far too much, his posts have generally
been informed by a basic soundness of experience.
Fons is right, something changed. But what?
Chris
What's wrong with you ?
ralf madorf is the only individual who has a kill filter in my gmail
configuration. that's my only observation.
11:19 p.m.
New subject: master levels (and dithering)
On Tue, Feb 3, 2015 at 10:01 PM, Chris Cannam
<cannam(a)all-day-breakfast.com> wrote:
something changed. But what?
In my opinion:
A statement like "Linux audio is full of pitfalls" has no constructive
element in it. It diminishes the subject, and does so in a generalized
and sweeping manner. This type of comment is not going to motivate
improvement: it has only a negatively impacts the community.
To be more precise, it is a negative observation of an individual: an
individual who is entitled to their opinion. However posting a
personal opinion as a fact in a public discussion is neither
constructive or relevant.
On a positive note, I welcome input from any individual on matters
where they have knowledge and experience. With that I will politely
request that time is taken when writing to list, in order to be
understanding to all those in the community, and write a single
coherent email that adds to the constructive resolution of the
discussion / debate at hand.
Cheers, -Harry
11:31 p.m.
New subject: master levels (and dithering)
On Tue, Feb 3, 2015, at 10:19 PM, Harry van Haaren wrote:
A statement like "Linux audio is full of
pitfalls" has no constructive
element in it. It diminishes the subject, and does so in a generalized
and sweeping manner.
In isolation it would, but Linux audio *is* full of pitfalls, and
there's nothing wrong with saying so in a sentence within a longer
email.
I see that your objection is to posts that dismiss something without
giving any route to fix it, and I get that that doesn't help and it
would be really nice if we could all think again before posting such
things and maybe give them a bit more substance.
But I think what Fons is pointing out is different -- he's quoting posts
that have started saying (effectively) "the Linux version of this is
worse" without explaining what it was that was worse, or why. That's a
strange thing to start doing, and it suggests maybe something has gone
wrong that has made you (Ralf) reconsider your perception of things. It
would be good to know what that was.
Chris
11:33 p.m.
New subject: master levels (and dithering)
On Tue, Feb 3, 2015 at 5:19 PM, Harry van Haaren <harryhaaren(a)gmail.com>
wrote:
On Tue, Feb 3, 2015 at 10:01 PM, Chris Cannam
<cannam(a)all-day-breakfast.com> wrote:
stuff like "linux dithering is a PITA" is just utterly ignorant assholery,
unjustifiable, indefensible, ungrounded, meaningless, jibberish.
those of you who choose not to ignore his posts --- what have you EVER
gotten from reading them?
something changed. But what?
In my opinion:
A statement like "Linux audio is full of pitfalls" has no constructive
element in it. It diminishes the subject, and does so in a generalized
and sweeping manner. This type of comment is not going to motivate
improvement: it has only a negatively impacts the community.
11:57 p.m.
New subject: master levels (and dithering)
On Tue, Feb 3, 2015, at 10:33 PM, Paul Davis wrote:
stuff like "linux dithering is a PITA" is
just utterly ignorant
That is what Fons was remarking upon. It isn't something I would expect
Ralf to have said, in the past. The fact that he does is interesting.
those of you who choose not to ignore his posts ---
what have you EVER
gotten from reading them?
An impression of the experience of a particular, quirky, enthusiast.
But here we are, talking about Ralf while Ralf is here. He does read
criticism -- for example he started posting a bit less after someone
complained about the volume a year or so ago.
Chris
4 Feb
4 Feb
6:17 a.m.
New subject: master levels (and dithering)
Am 03.02.2015 um 23:33 schrieb Paul Davis:
On Tue, Feb 3, 2015 at 5:19 PM, Harry van Haaren
<harryhaaren(a)gmail.com <mailto:harryhaaren@gmail.com>> wrote:
On Tue, Feb 3, 2015 at 10:01 PM, Chris Cannam
<cannam(a)all-day-breakfast.com
<mailto:cannam@all-day-breakfast.com>> wrote:
Between all the lines Ralf wrote, I found enough stuff to make me laugh.
Therefore it's worse for me to read them all, even the bitter ones.
regards
hermann
something changed. But what?
In my opinion:
A statement like "Linux audio is full of pitfalls" has no constructive
element in it. It diminishes the subject, and does so in a generalized
and sweeping manner. This type of comment is not going to motivate
improvement: it has only a negatively impacts the community.
stuff like "linux dithering is a PITA" is just utterly ignorant
assholery, unjustifiable, indefensible, ungrounded, meaningless,
jibberish.
those of you who choose not to ignore his posts --- what have you
EVER gotten from reading them?
4:58 p.m.
New subject: master levels (and dithering)
On Tue, 3 Feb 2015 17:33:18 -0500, Paul Davis wrote:
those of you who choose not to ignore his posts ---
what have you EVER
gotten from reading them?
Perhaps nothing. OTOH maybe I helped a few people with less audio
engineering experiences than Gene, Fons, a few others and I've got.
Since Fons doesn't live very long in Germany, I even could help him to
find a good supplier for semiconductors in Germany. Sure, after a
while somebody else would have give him this hint or he would have
found out himself were to get smidgens of semiconductors for a fair
price from a customer-friendly retailer.
Who cares? You killfiled me, a hand full of people dislike me, but
another hand full of people like me :p.
5 Feb
5 Feb
12:33 a.m.
New subject: master levels (and dithering)
On 05/02/15 02:58, Ralf Mardorf wrote:
On Tue, 3 Feb 2015 17:33:18 -0500, Paul Davis wrote:
> those of you who choose not to ignore his posts --- what have you EVER
> gotten from reading them?
> ...
>
> Who cares? You killfiled me, a hand full of people dislike me, but
> another hand full of people like me :p.
I don't dislike or like you Ralf.
On several occasions you have posted
information that was very helpful, and others provided me with small
amusements.
If only there was a "baseless opinions" filter such as the previously
mentioned one where you can tell the difference between a high bitrate
lossy file but refuse an ABX test. Personally I can't pick an ogg -q7
file from flac.
Best wishes
Roger
4 Feb
4 Feb
4:11 p.m.
New subject: master levels (and dithering)
On Tue, 03 Feb 2015 22:01:01 +0000, Chris Cannam wrote:
On Tue, Feb 3, 2015, at 09:51 PM, Paul Davis wrote:
My apologies to you Chris,
I'll take a timeout.
Regards,
Ralf
On Tue, Feb 3, 2015 at 4:30 PM, Fons Adriaensen
<fons(a)linuxaudio.org>
wrote:
I posted a gratuitously snarky reply to something of Ralf's a few
months ago, he objected to it, and I realised I had been too personal.
Since then I've been reading his stuff with more care and, although I
do think he has always posted far far too much, his posts have
generally been informed by a basic soundness of experience.
Fons is right, something changed. But what? What's wrong with you ?
ralf madorf is the only individual who has a kill filter in my gmail
configuration. that's my only observation. 
9:21 a.m.
New subject: master levels (and dithering)
On 03.02.2015 22:51, Paul Davis wrote:
ralf madorf is the only individual who has a kill
filter in my gmail
configuration. that's my only observation.
Similar here, plus he's the only person I ever saw mentioned as reason
why people unsubscribed. Several times, on IRC.
To be fair, I caused one person to leave, but it didn't take me hundreds
of emails ...
--
Thorsten Wilms
thorwil's design for free software:
http://thorwil.wordpress.com/
6:23 p.m.
New subject: master levels (and dithering)
On Tue, 3 Feb 2015 16:51:18 -0500, Paul Davis wrote:
On Tue, Feb 3, 2015 at 4:30 PM, Fons Adriaensen
<fons(a)linuxaudio.org>
wrote:
At least I never ever participate on exploitation. Have you ever
done something very important for Amazone? Just to get money?
Paul?
C'mon!
Talk to the community :)!
Have I ever done something that unethical in my live as you did?
Please, don't be shy, point out what I've done that was even nearly
that rack rent, evil as anything that is "normal" for you guy :).
Paul, call me names again and then take a look into the mirror?
:D
What's wrong with you ?
ralf madorf is the only individual who has a kill filter in my gmail
configuration. that's my only observation.
6:56 p.m.
New subject: master levels (and dithering)
Between all the lines Ralf wrote, I found enough stuff
to make me laugh.
Therefore it's worse for me to read them all, even the bitter ones.
That's what I mean, funny eh?
16:11
My apologies to you Chris,
I'll take a timeout.
Regards,
Ralf
16:58
. . . Who cares? You killfiled me, a hand full of
people dislike me, but
another hand full of people like me :p.
18:23
Please, don't be shy, point out what I've done
that was even nearly
that rack rent, evil as anything that is "normal" for you guy.
seems like
the timeout is between 16:11 <-> 16:58
. . . to be continued . . . :-D
sorry Ralf, cant resist
7:40 p.m.
New subject: master levels (and dithering)
Is there some REALLY good reason this has to be made personal?? Like towards anybody??
I cannot imagine what some poor noob who just downloaded a Linux audio distro, is looking
for help, and has found this thread, would be thinking!
~ Russell
8 p.m.
New subject: master levels (and dithering)
Am 04.02.2015 um 19:40 schrieb Russell Hanaghan:
Is there some REALLY good reason this has to be made
personal?? Like towards anybody??
I cannot imagine what some poor noob who just downloaded a Linux audio distro, is looking
for help, and has found this thread, would be thinking!
~ Russell
Sorry, I really tried, but, unfortunately, I can't resist.
May I'm ralfified :-D :-D
(duck and run)
8:17 p.m.
New subject: master levels (and dithering)
Quoting Russell Hanaghan <hanaghan.osaudio(a)gmail.com>om>:
Is there some REALLY good reason this has to be made
personal?? Like
towards anybody??
I cannot imagine what some poor noob who just downloaded a Linux
audio distro, is looking for help, and has found this thread, would
be thinking!
Since the person making the most posts here on LAU seems to be 100%
anti-Linux I would guess the noob who found this thread would reformat
his hard drive immediately and install an OS that has four blue
rectangles in the logo instead.
Some people tell me that OS sounds marvelous!
- Peder
7:31 a.m.
New subject: master levels (and dithering)
On 03/02/15 at 09:30pm, Fons Adriaensen wrote:
Ralf, we (readers of this list) are used to your
occasional
fits, to hearing for the zillionth time about your troubles
with Brauner, and that only two of the eight ADAT channels
of your sound card are usable.
What you have been writing recently is different. Gratituous
drivel about how bad Linux is for audio, without any hint of
evidence and without any reasoning behind it.
Some examples from the last days:
If he tells you he'll have to kill you ;-)
--
«My mama said to get things done
You'd better not mess with Major Tom»
JFTR I've got loss of sound quality issues
when using Linux,
that I don't have when using professional or even consumer
digital hardware or even 4-track analog tapes with Dolby C.
IMO there's something broken, at least on all Linux machines
I used and I'm still using.
If I record an ALESIS D4 cymbal with Linux, the
result is week.
If I record it with a consumer DAT recorder, there's no audible
difference to the original sound. Even a 4-track cassette player
with Dolby C still sounds better, than the Linux recording
(tested with TerraTec and RME cards).
Linux dithering is a PITA, it adds annoying
audible noise, no
stand alone consumer gear, no proprietary DAW and especially
no professional gear comes with such audible dithering noise.
Is there usable dithering available for Linux?
If you mix e.g. all your Qtractor tracks and
master them to a
stereo Qtractor track, this digital mix does cause a loss of
sound quality. If you should be that hearing-impaired, that
you even are unable to notice this, then take a look at
spectrograms. Likely there's a very old thread somewhere in
the Qtractor mailing list archive. It's not a Qtractor issue,
it's a jackd issue.
What's wrong with you ?
8 p.m.
New subject: master levels (and dithering)
Quoting Ralf Mardorf <ralf.mardorf(a)alice-dsl.net>et>:
Ralf,
are you aware about the zero-copy issue?
If you mix e.g. all your Qtractor tracks and master them to a stereo
Qtractor track, this digital mix does cause a loss of sound quality.
A loss of sound quality compared to what? Your mint condition vinyl
copy of "Der Ring des Nibelungen"?
Have you had someone do a blind test on you to see if you can tell the
difference between the individual tracks and the stereo mix?
Please do and report the results.
Peder, any hints for this problem:
On Mon, 02 Feb 2015 22:25:49 +0100, Atte wrote:
> I had a fairly unpleasant experience last weekend. A band (indie/folk)
> in which I play and for whom I also did/does some recordings + mixes
> for, were played on national danish radio on the national chart. The
> unpleasant part was that our song sounded really wimpy compared to the
> other tracks.
Yes Ralf, and the answer is not "Use Windows or OSX". I'm sure even
you understand that.
ABX test? Is Atte the most bad audio engineer on this
planet?
I never said anything about ABXing Atte's wimpy sounding recordings.
What good would that do?
I suggested you show us your golden ears and that you can tell the
difference between a properly encoded MP3 and the original WAV. Since
you say you can even hear the dithering noise you should have no
problem acing an MP3/WAV ABX.
Never ever the problem was that the other bands used
more compression.
And I don't think that Atte is the most bad audio engineer. I suspect
that he's a victim of Linux audio pitfalls. Next time Atte should
compare his mastering with recordings of other bands, instead of
trusting the Linux audio community.
If he had only used Windows or OSX like all the other bands I'm sure
his music would have been selling in the millions.
Some people say it's impossible to make bad sounding music in Windows
or in OSX.
Only in Linux.
And I'm not so sure about OSX since it's foundation is BSD and that's
almost like Linux.
That I mentioned recordings of relatively good cymbal
samples from a
drum sample player has got a reason.
Atte was recording using a Zoom R24. Are you saying that runs Linux?
- Peder
3 Feb
3 Feb
12:05 a.m.
On Mon, 2 Feb 2015, Brent Busby wrote:
Len Ovens <len(a)ovenwerks.net> writes:
I do not know if there is something in ardour to do this, but if you
import the final wav file into Audacity, it is very easy. Once you have
imported the track, there is an "Effect" dropdown menu and the first
effect is called amplify. If you select that, the default amplification
will be the amount needed to set the highest peek to 0db.
However, as I (and Ralph) meantioned in passing and Atte went into much
more and better detail. It does depend on the audience. What is
technically best may not be what is best for your application. If what you
are recording will be played along side other people's work, you pretty
much need to follow the same guidlines as they have with reguards to
levels and compression. What sounds best to most people is what they are
used to hearing. So mixdown to -6 peeks or so for your private master to
archive, then tweek it for the finished product to what is commercially
viable in your sittuation. After doing all the tweeks, you find you wish
to change the mix to work with the tweeking :)
This speaking from a strictly Amateur POV. I have a lot to learn about
mixdown and mastering and so in mastering I find the need for remixing to
make my mastering work. Lots to learn.
--
Len Ovens
www.ovenwerks.net
6db down is one bit of 24... to go to 16 for a CD
means cutting off
bits on the bottom (maybe adding some noise) so -6 is safe.
Wow, I never would have imagined that.
So, I could mix to -6db in Ardour and still use most of the headroom on
a CD? Is there any advantage to keeping your peak level that low on the
master in the digital world? Or, put another way, if I *could* get the
whole track to behave itself and not go above -1db at loudest peak, is
there a reason not to? 
12:10 a.m.
On Mon, 2 Feb 2015 15:03:01 -0800 (PST)
Len Ovens <len(a)ovenwerks.net> wrote:
Both Audacity and Ardour have a normalization function which will let you
amplify a track (or selection) to some selectable level. It's quite handy.
--
======================================================================
Joe Hartley - UNIX/network Consultant - jh(a)brainiac.com
Without deviation from the norm, "progress" is not possible. - FZappa
So, I could
mix to -6db in Ardour and still use most of the headroom on
a CD? Is there any advantage to keeping your peak level that low on the
master in the digital world? Or, put another way, if I *could* get the
whole track to behave itself and not go above -1db at loudest peak, is
there a reason not to?
I do not know if there is something in ardour to do this, but if you
import the final wav file into Audacity, it is very easy. Once you have
imported the track, there is an "Effect" dropdown menu and the first
effect is called amplify. If you select that, the default amplification
will be the amount needed to set the highest peek to 0db.
2 Feb
2 Feb
10:25 p.m.
On 02/02/2015 08:03 AM, Len Ovens wrote:
This depends on your audience. This is not always the
end listener, but
may be the broadcaster or distributor or whoever is paying for it.
Dynamics are nice IMO.
I had a fairly unpleasant experience last weekend. A band (indie/folk)
in which I play and for whom I also did/does some recordings + mixes
for, were played on national danish radio on the national chart. The
unpleasant part was that our song sounded really wimpy compared to the
other tracks. I recorded the whole stream and imported it in software to
investigate. It turned out that most of the wimpy-ness comes from us
being much lower (same peak levels, but not so heavily compressed) than
the other tracks.
So in theory or utopia, we should refrain from compressing our audio so
hard. But in reality, the loudness war is lost.
NB: This weekend were our third week on the chart. Now I squashed our
master accordingly and send it to them to play instead of the old, wimpy
one, should we be fortunate enough to still be on the list next week...
--
Atte
http://atte.dk http://a773.dk
3 Feb
3 Feb
12:09 a.m.
Atte <atte(a)youmail.dk> writes:
I had a fairly unpleasant experience last weekend. A
band (indie/folk)
in which I play and for whom I also did/does some recordings + mixes
for, were played on national danish radio on the national chart. The
unpleasant part was that our song sounded really wimpy compared to the
other tracks. I recorded the whole stream and imported it in software to
investigate. It turned out that most of the wimpy-ness comes from us
being much lower (same peak levels, but not so heavily compressed) than
the other tracks.
So in theory or utopia, we should refrain from compressing our audio so
hard. But in reality, the loudness war is lost.
NB: This weekend were our third week on the chart. Now I squashed our
master accordingly and send it to them to play instead of the old, wimpy
one, should we be fortunate enough to still be on the list next week...
A certain amount of compression is nice to make everything sound big,
but when it gets to the point that everything in the mix is exactly the
same loud volume, it's just gone too far.
Go ahead and compress though, certainly!
--
+ Brent A. Busby + "We've all heard that a million monkeys
+ Sr. UNIX Systems Admin + banging on a million typewriters will
+ University of Chicago + eventually reproduce the entire works of
+ James Franck Institute + Shakespeare. Now, thanks to the Internet,
+ Materials Research Ctr + we know this is not true." -Robert Wilensky
5:44 a.m.
New subject: Dithering - Was: master levels
On Mon, 02 Feb 2015 22:25:49 +0100, Atte wrote:
I had a fairly unpleasant experience last weekend. A
band (indie/folk)
in which I play and for whom I also did/does some recordings + mixes
for, were played on national danish radio on the national chart. The
unpleasant part was that our song sounded really wimpy compared to the
other tracks.
I don't think that it has to do with the loudness war. How well do you
chose Linux EQs? Most of them are completely unusable. IMO
Fons' parametric EQ is ok.
If I record an ALESIS D4 cymbal with Linux, the result is week. If I
record it with a consumer DAT recorder, there's no audible difference
to the original sound. Even a 4-track cassette player with Dolby C
still sounds better, than the Linux recording (tested with TerraTec and
RME cards).
Assumed it should be missing dithering, then what dithering to use?
Linux dithering is a PITA, it adds annoying audible noise, no stand
alone consumer gear, no proprietary DAW and especially no professional
gear comes with such audible dithering noise.
Is there usable dithering available for Linux?
7:46 a.m.
New subject: Dithering - Was: master levels
On Monday 02 February 2015 23:44:50 Ralf Mardorf did opine
And Gene did reply:
On Mon, 02 Feb 2015 22:25:49 +0100, Atte wrote:
First, that cymbal (and the snare drums and banjos all have harmonics that
are well into the ultrasonic range. Our ears, generally are their own
brickwall filter. I say generally because ears do vary, and when I walk
into a store that has an ultrasonic motion detector for a burglar alarm, I
don't waste a lot of time, not because I can hear it, but because my ears
are feeling it and hurting, and I'm now 80 yo, with a right ear thats not
even up to telephone bandwidth anymore!
Part of the specs for a DAT recorder, in order for it to use the
trademarked DAT logo, will cause it have a brickwall filter to prevent
digital aliasing, which will otherwise be translated down into the audible
by the digitizer. Even at my age I can hear the aliasing and it is not
conducive to enjoying the crash of the cymbal.
There are 2 ways to stop this, first being the brickwall filter, some sort
of Chebychev (or better) 5 to 7 stages deep thats pretty flat to 19 or 20
kilohertz, but is 30 db down by 30 kilohertz. That is quite a stack of
parts, taking up board real estate and both cost lots of sheckles. So you
will not find that, at least to a usable degree in any sub $300 audio
card. Facts of life.
The other way is to raise the digitization rate, commonly 44 or 48
kilohertz IIRC, to something in the 200-250 kilohertz range so that there
is essentially no upper sideband as it moved upward in frequency as the
sample range is raised, by a factor of 2x the sample frequency increase.
Digitally this high speed sample is then down sampled in a shift register
by mathematically averaging the last 4 or 8 samples by just adding them
and throwing away the LSB for every 2 samples added, usually by a right
shift that shifts the LSB's out to nowhere, leaving a net gain of one, and
that digital sample is then taken at a more sane rate, maybe 96 kilohertz,
and fed to the media for storage.
Aliasing is a cast iron bitch, because once it has managed to get into
your data stream, there is no known way to remove it
I had a fairly unpleasant experience last weekend.
A band (indie/folk)
in which I play and for whom I also did/does some recordings + mixes
for, were played on national danish radio on the national chart. The
unpleasant part was that our song sounded really wimpy compared to the
other tracks.
I don't think that it has to do with the loudness war. How well do you
chose Linux EQs? Most of them are completely unusable. IMO
Fons' parametric EQ is ok.
If I record an ALESIS D4 cymbal with Linux, the result is week. If I
record it with a consumer DAT recorder, there's no audible difference
to the original sound. Even a 4-track cassette player with Dolby C
still sounds better, than the Linux recording (tested with TerraTec and
RME cards). Assumed it should be missing dithering, then what
dithering to use?
Linux dithering is a PITA, it adds annoying audible noise, no stand
alone consumer gear, no proprietary DAW and especially no professional
gear comes with such audible dithering noise.
Is there usable dithering available for Linux?
Dithering, to be useful must be applied to the reference, or to the
signal itself, in an amount that is not more than about half the value of
the least significant bit. In a 24 bit system, that amount of dither is
less than the thermal noise. If that 24 bit output is fed to a 32 bit
word, which 3/4ths of the word do you use? Often there will be a random
noise fed into the lSByte, and the 3 upper bytes are the signal.
But this solution is not used in any product I know of. But there are
many I have no experience with too.
Now lets see what Fons A. has to say, we often disagree. ;-)
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>
US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS
3:35 p.m.
New subject: OT: Dithering - Was: master levels
On Tue, 3 Feb 2015 01:46:43 -0500, Gene Heskett wrote:
ultrasonic motion detector
radar vs ultrashall
In Germany radar is common. That's one reason why radar warning
receiver are useless in Germany. Usually they receive radar from
Star Trek like automatically opening doors. The other reason is, that
police in Germany usually does use laser instead of radar nowadays.
1:11 p.m.
New subject: Dithering - Was: master levels
On 02/03/2015 05:44 AM, Ralf Mardorf wrote:
I don't think that it has to do with the loudness
war. How well do you
chose Linux EQs? Most of them are completely unusable. IMO
Fons' parametric EQ is ok.
This particular track was recorded on a zoom r24 and mixed in
reaper/wine. So no linux EQ involved :-)
It's not gear, it's not EQ. It's partly a bad recording (of mainly the
drums including a crappy kit), partly my ability as a producer (don't
think I'm *that* bad, though), but mostly a too low gain. I played alot
with the recording I made of the entire chart. To mee it's pretty clear
that the main problem is us being lower than the surrounding tracks.
--
Atte
http://atte.dk http://a773.dk
1:22 p.m.
New subject: Dithering - Was: master levels
On Tue, 03 Feb 2015 13:11:31 +0100, Atte wrote:
On 02/03/2015 05:44 AM, Ralf Mardorf wrote:
Oops :D.
I don't think that it has to do with the
loudness war. How well do
you chose Linux EQs? Most of them are completely unusable. IMO
Fons' parametric EQ is ok.
This particular track was recorded on a zoom r24 and mixed in
reaper/wine. So no linux EQ involved :-)
It's not gear, it's not EQ. It's partly a bad recording (of mainly the
drums including a crappy kit), partly my ability as a producer (don't
think I'm *that* bad, though), but mostly a too low gain. I played alot
with the recording I made of the entire chart. To mee it's pretty clear
that the main problem is us being lower than the surrounding tracks. 
5 Feb
5 Feb
10:07 a.m.
New subject: Dithering - Was: master levels
On Tue, 03 Feb 2015 13:11:31 +0100
Atte <atte(a)youmail.dk> wrote:
This particular track was recorded on a zoom r24 and
mixed in
reaper/wine. So no linux EQ involved :-)
It's not gear, it's not EQ. It's partly a bad recording (of mainly the
drums including a crappy kit), partly my ability as a producer (don't
think I'm *that* bad, though), but mostly a too low gain. I played
alot with the recording I made of the entire chart. To mee it's
pretty clear that the main problem is us being lower than the
surrounding tracks.
If you use reaper I'd recommend checking out either the included
JS/Eventhorizon, or the standalone VST version as the last plugin on
the master channel. Lower the threshold until the RMS level of the
master hoovers around -12 to -16dB FS. Leave the peaks at -0.2 or so.
Very transparent on the music I've tested it with, seems similar in
loudness to many CDs, but still not overly compressed (of course
depending on what you've done to it earlier in the chain :)
--
Joakim
12:44 p.m.
New subject: Dithering - Was: master levels
On 02/05/2015 10:07 AM, Joakim Hernberg wrote:
If you use reaper I'd recommend checking out
either the included
JS/Eventhorizon, or the standalone VST version as the last plugin on
the master channel.
Thanks, I'll look into that. Limiter no6 seems to work quite well here...
--
Atte
http://atte.dk http://a773.dk
2 Feb
2 Feb
11:29 a.m.
On Sun, 01 Feb 2015 23:40:45 -0600, Brent Busby wrote:
In digital, they say mix to about -3 or -4db.
"RMS: For the case in which the RMS value of a full-scale square wave is
designated 0 dBFS, all possible dBFS measurements are negative numbers.
A sine wave could not exist at a larger RMS value than −3 dBFS without
clipping, by this convention." -
http://en.wikipedia.org/wiki/DBFS#RMS_levels
I can't say anything about CD quality and the need to use a margin as
close as possible to 0 dBFS, but for DAT quality there is no
need to be close to 0 dBFS, OTOH it also doesn't cause audible issues,
if you master up to 0.5 dBFS. Headroom doesn't harm, OTOH if a demo
recording isn't as loud as other demo recordings are, this might be a
disadvantage, since not everybody knows how to use an amp's volume
control.
4:06 p.m.
On Sun, 01 Feb 2015 23:40:45 -0600
Brent Busby <brent(a)keycorner.org> wrote:
Of course, in the old analog tape world, they used to
shoot for zero,
maybe a little hotter if you wanted some saturation. In digital, they
say mix to about -3 or -4db. And mastering houses seem to want more
headroom than that (-6 or 7db), even though in priciple, with 24-bit
resolution, they could just adjust it themselves without changing it
timbrally (true?). Why do they care what your maximum level is as long
as it's 24-bit and it's not clipping?
A side note:
Check the type of meter Ardour uses for the master channel. I wondered
why my Ardour masters always seemed quiet, then realized that the master
bus used a different meter type (K20? I can't remember) than all the other
tracks. When I changed to the same meter type (Peak, I think) I saw that I was
actually mixing to a lower value than I thought.
--
======================================================================
Joe Hartley - UNIX/network Consultant - jh(a)brainiac.com
Without deviation from the norm, "progress" is not possible. - FZappa
4:25 p.m.
On Mon, 2 Feb 2015 10:06:29 -0500, Joe Hartley wrote:
On Sun, 01 Feb 2015 23:40:45 -0600
Brent Busby <brent(a)keycorner.org> wrote:
IIUC the OP wants to know, how much the highest peak in dBFS should be
for a demo he wants to give away.
^^^^
The sound quality doesn't suffer from headroom, so -10 dBFS and -6 dBFS
are ok and even lower is still ok, but if you give away the demo even a
peak of 0.0 dBFS wouldn't harm the recording. What kind of compression
you are using, e.g. Bob Katz style, loudness war etc. is something
else. And sure, for recording much headroom is wanted and that
headroom usually doesn't make the sound quality less good. IIUC this
(how to record, what compression to use) wasn't the question.
Use a dBFS meter, ensure that a 0 dBFS sine wave is 0 dBFS and ry to
get the highest peak as close to 0.0 dBFS as possible for a demo you'll
give away.
Of course, in the old analog tape world, they
used to shoot for
zero, maybe a little hotter if you wanted some saturation. In
digital, they say mix to about -3 or -4db. And mastering houses
seem to want more headroom than that (-6 or 7db), even though in
priciple, with 24-bit resolution, they could just adjust it
themselves without changing it timbrally (true?). Why do they care
what your maximum level is as long as it's 24-bit and it's not
clipping?
A side note:
Check the type of meter Ardour uses for the master channel. I
wondered why my Ardour masters always seemed quiet, then realized
that the master bus used a different meter type (K20? I can't
remember) than all the other tracks. When I changed to the same
meter type (Peak, I think) I saw that I was actually mixing to a
lower value than I thought.
3528
days inactive
3531
days old
linux-audio-user@lists.linuxaudio.org
45 comments
19 participants
participants (19)
-
Atte -
Brent Busby -
Chris Cannam -
Fons Adriaensen -
Gene Heskett -
Harry van Haaren -
Hermann Meyer -
Joakim Hernberg -
Joe Hartley -
Len Ovens -
Paul Davis -
Peder Hedlund -
Raffaele Morelli -
Ralf Mardorf -
Robert Jonsson -
Roger -
Russell Hanaghan -
Thorsten Wilms -
Will Godfrey