---- On Fri, 23 Aug 2019 09:16:23 -1000 Ralf Mardorf <ralf.mardorf(a)alice-dsl.net> wrote ----
> Yes, there are pitfalls [1.1], however, if you have good luck, you might
> come relatively close to the desired result.
No, I don't think so.
I get the reasoning: the karaoke effect works by subtracting the two channels, so that anything that is equal volume and equal phase between the two channels will cancel out. We could call this a "residual." Then you think "if I subtract the residual from the total, I should get only the center material."
But the math just doesn't work.
L = left channel
R = right channel
residual = L - R
Well, here's the logical error. That's not actually a "residual" (whatever "residual" was expected to mean). It's just the Side of a M/S configuration, nothing more.
So then you think, "I have an original left channel -- subtract the residual and get some sort of remainder."
left_anti_residual = L - residual = L - (L - R) = L - L + R = R
right_anti_residual = R - (R - L) = R - R + L = L
So, hooray, you've flipped the left/right stereo image and not removed anything.
"But if you sum to mono"... then you get R + L = Mid. A mono mix, with nothing removed. (You had suggested that an effect that is deliberately phase-inverted between the two channels would cancel out in this case. Of course... but that's rather special pleading, isn't it? How often does that happen in the real world?)
OK, then, maybe the mistake was subtracting from the individual channels. Maybe you need to go to the total and subtract the residual.
Total = L + R = Mid of a M/S configuration.
Total - Residual = M - S = right channel + 6 dB
Total - neg(Residual) = M + S = left channel + 6 dB
Sum that to mono and you get M - S + M + S = 2M = the same mono mix, only 6 dB louder.
Simple sums and differences of left and right channels will, as far as I can see, never recover the "center" material.
hjh
On Aug 22 2019, Ralf Mardorf has written:
> ...
>> On the quick my guess is
>>
>>
>> Original left channel + inverted right channel = plain_left
>> Original right channel + inverted left channel = plain_right
>>
>>
>> Original left channel + inverted plain_left = middle only
>> Original right channel + inverted plain_right = middle only
> ...
> That sounds good. Unforunately it didn't work.
I went down that road once before.
"Plain" channels:
a = l - r ("side" in mid/side)
b = r - l
Then l - a = l - (l - r) = r and r - b = r - (r - l) = l, neither of which
isolates the center.
The idea of the karaoke effect is to cut in-phase material that is in
common between the channels (e.g. often lead vocals). You'd think you can
do the karaoke effect and then some subtraction would leave you with
"whatever is not the karaoke effect," but in fact, variables cancel out in
a way that you don't get anything useful. This might be the reason why you
don't find a lot of plugins for this.
hjh
Sent with AquaMail for Android
https://www.mobisystems.com/aqua-mail
I've had one of these running perfectly for near 2 years now and have always
run it with a buffer size of 64 frames/48kHz. This last week I've been too busy
with 'other things' to do any composition. However, yesterday when I tried to
run it. Jack immediately went into a cascade of Xruns with a most peculiar
error report, suggestion it couldn't find the unit.
Running Yoshimi by itself under Alsa worked fine, so I was now totally
perplexed. Eventually I noticed that Yoshimi was set to 128 frames. When I set
it to 64, it was even stranger. It ran, but on attempting to play anything it
just made a horrible noise. Further tests revealed that jack also ran correctly
with 128 frames or more.
For a while I couldn't think of anything that had changed during that week - no
upgrades or installs. Then I remembered I'd had to repair a bad solder joint in
the amplifier, and as everything is mounted in a rack, had to pull quite a bit
of stuff apart to get to it. I was quite sure I'd checked everything when
re-assembling it, but just to be sure did a quick 'lsusb' in a terminal window,
and as I thought, audio, midi, and mouse/keyboard were all on a different
internal bus, and no bus had anything else on it.
Today, while playing a CD on my 'office' machine I remembered I used to have a
USB CD player that wouldn't work when plugged into the front USB3 sockets, even
though it was identified on an internal USB2 bus, so checking the music system,
I had indeed plugged it onto one of these on the back of that machine. Once
swapped into a (black) USB2 socket it once again worked fine.
I've no idea why it had given the symptoms it did though!
--
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.
It drops out especially when I boot. This drop-out happens occasionally.
I thought my mainboard was dying, but other USB devices weren't dropping
out.
Do I need to get a replacement for my USB soundcard?
Or, is it a software issue?
You seem to think that my mainboard doesn't supply enough power to X-Fi HD.
through a USB cable.
I just plugged it into a USB 3.0 port. But, I have to test it for a while.
On Tue, Aug 20, 2019 at 9:55 AM Ralf Mardorf <ralf.mardorf(a)alice-dsl.net>
wrote:
> On Tue, 20 Aug 2019 09:35:09 +0900, crocket wrote:
> >It drops out especially when I boot. This drop-out happens
> >occasionally.
> >
> >I thought my mainboard was dying, but other USB devices weren't
> >dropping out.
> >
> >Do I need to get a replacement for my USB soundcard?
> >Or, is it a software issue?
>
> It seems to be bus powered, if so, then for testing purpose use a very
> short USB cable and test different USB ports, if possible test with an
> active USB hub, too.
>
After ca. 10 months, a new prelude for synth. As usual it took long but
is relatively short. Hope you enjoy
Archive.org: https://archive.org/details/synth_prelude_02 (includes
direct download links)
Soundcloud: https://soundcloud.com/lorenzosu/synth-prelude-n-2
Made with Yoshimi and Rosegarden.
Customary thanks to all the developers of these two music applications
as well as all those contributing to the Linux Audio ecosystem for their
work and dedication :)
For the interested and/or curious, I have also set-up a Git repository
with the Rosegarden (.rg) and Yoshimi state (.state) 'source' files for
the Synth Preludes:
https://gitlab.com/lorenzosu/synth-preludes
Lorenzo.
Hello:
I have some good audio resources up and running on Ubuntu Studio
18.04. When I searched for MIDI utilities, what I have found so far is a
bunch of them that can be downloaded and compiled (supposedly). When I
attempt that, so many of them have such cryptic error streams that it
appears that only a highly skilled Linux programmer could get them to
configure, compile, and install properly.
Does anyone know why these resources seem so difficult to get
going? Please comment.
Brian
Hey hey,
"The apprentice - hard studies" was composed for an audio game - and not
chosen :) -, the title says it all in terms of intent.
I tried to include one of Diana Duetsch's acoustic illusions in the
xylophone/marimba like pattern as a point of interest and something mildly
magical. :)
youtube.be/Tjw9VjUoKrI
On the Linux side LinuxSampler with some free orchestral samples was involved
as well as loads of plugins for processing. The orchestral samples were mixed
with a hardware synth to get something a little more realistic. The strings
were also layered with a simple sine sound to give a little more girth and
resonance" to the instruments.
Hope you enjoy it, comments are welcome.
Best wishes,
Jeanette
--
* Website: http://juliencoder.de - for summer is a state of sound
* SoundCloud: https://soundcloud.com/jeanette_c
* Youtube: https://www.youtube.com/channel/UCMS4rfGrTwz8W7jhC1Jnv7g
* GitHub: https://github.com/jeanette-c
* Twitter: https://twitter.com/jeanette_c_s
Just call out my name, and I will be there... <3
(Britney Spears)