8 Aug
2014
8 Aug
'14
3:24 a.m.
Rolling by on randomize came a Me and My Cronies jam/joke from years ago:
http://www.restivo.org/blog/podpress_trac/web/558/0/Not_OK_Computer.ogg
And I was struck by how much PHASEX sounds like a real analog synth, like an ARP 2600 or
similar, and so much more real than any other software synths I've used.
It sounds so... raw, uncontrolled, well, ANALOG. Most software simulations sound more or
less authentic, but all so much more "tame", for want of a better term. But
PHASEX always sounded to me (and felt, as I was playing with it) that at any moment it
could do something crazy like throw a DC offset, to into an uncontrollable oscillation, or
blow up my speakers, etc.
I don't like being at a loss for precise, engineering terms, or understanding WHY
something is, so I'm asking any of the DSP'ers here who might also have looked at
(and understood) PHASEX's source.
Any ideas what is so different about PHASEX, and what might be this quality of it's
sound I could be trying to describe?
-ken
8 Aug
8 Aug
8:59 a.m.
I agree that Phasex is a better virtual Linux synth, but it still is far
away from analog sound. IIRC another good virtual Linux synth is Calf
Monosynth. I compared the virtual Linux synth outputs using HDSPe AIO
outputs, ADA8000 outputs and EWX 24/96 outputs with an Oberheim
Matrix-1000 connected to the same analog mixer. The same for a virtual
digital synth, my DX7 does sound classes better than Hexter does.
[rocketmouse@archlinux ~]$ aplay -l
**** List of PLAYBACK Hardware Devices ****
card 0: HDSPMx579bcc [RME AIO_579bcc], device 0: RME AIO [RME AIO]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 1: EWX2496 [TerraTec EWX24/96], device 0: ICE1712 multi [ICE1712 multi]
Subdevices: 1/1
Subdevice #0: subdevice #0
card 2: EWX2496_1 [TerraTec EWX24/96], device 0: ICE1712 multi [ICE1712 multi]
Subdevices: 1/1
Subdevice #0: subdevice #0
I guess Phasex and Yoshimi are the most used virtual Linux synth on my
machine, OTOH in more than 10 years using Linux only at home, I never
really finished an audio production at home, resp. I never released a
production. MIDI jitter, xruns, sound quality, a bad RME driver are
issues you don't experience with good stand alone gear.
10:57 a.m.
On Thu, Aug 07, 2014 at 06:24:41PM -0700, Ken Restivo wrote:
Rolling by on randomize came a Me and My Cronies
jam/joke from years ago:
http://www.restivo.org/blog/podpress_trac/web/558/0/Not_OK_Computer.ogg
nice!
And I was struck by how much PHASEX sounds like a real
analog synth, like an ARP 2600 or similar, and so much more real than any other software
synths I've used.
It sounds so... raw, uncontrolled, well, ANALOG. Most software simulations sound more or
less authentic, but all so much more "tame", for want of a better term. But
PHASEX always sounded to me (and felt, as I was playing with it) that at any moment it
could do something crazy like throw a DC offset, to into an uncontrollable oscillation, or
blow up my speakers, etc.
I don't like being at a loss for precise, engineering terms, or understanding WHY
something is, so I'm asking any of the DSP'ers here who might also have looked at
(and understood) PHASEX's source.
Any ideas what is so different about PHASEX, and what might be this quality of it's
sound I could be trying to describe?
Perhaps it is not about a particolar DSP but probably you are using
one of the follow wavetables for one or more oscillators:
<https://github.com/williamweston/phasex/tree/v0.14.97/samples>
Of course they are usable with any sampler/tracker/etc.
Here is a simple test with SoX:
for smp in phasex/samples/*.raw; do
play -t raw -r 48000 -c 1 -e float -b 32 "${smp}" \
gain -9 speed 18 repeat 1000 fade h 0.05 1 0.2
done
12:22 p.m.
On Thu, Aug 07, 2014 at 06:24:41PM -0700, Ken Restivo wrote:
It sounds so... raw, uncontrolled, well, ANALOG. Most
software simulations sound
more or less authentic, but all so much more "tame", for want of a better
term.
But PHASEX always sounded to me (and felt, as I was playing with it) that at any
moment it could do something crazy like throw a DC offset, to into an uncontrollable
oscillation, or blow up my speakers, etc.
You have been lucky then. If your sound card and amp are DC coupled,
Phasex *will* blow up your speakers. On my first test it produced a
DC offset 30 dB above the audio signal level.
Whatever qualities it may have, this is crappy.
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)
12:55 p.m.
New subject: [Bulk] Re: Why does PHASEX sound so damn good?
On Fri, 2014-08-08 at 10:22 +0000, Fons Adriaensen wrote:
On Thu, Aug 07, 2014 at 06:24:41PM -0700, Ken Restivo
wrote:
I can't remember such an issue, with the versions of Phasex I used.
It sounds so... raw, uncontrolled, well, ANALOG.
Most software simulations sound
more or less authentic, but all so much more "tame", for want of a better
term.
But PHASEX always sounded to me (and felt, as I was playing with it) that at any
moment it could do something crazy like throw a DC offset, to into an uncontrollable
oscillation, or blow up my speakers, etc.
You have been lucky then. If your sound card and amp are DC coupled,
Phasex *will* blow up your speakers. On my first test it produced a
DC offset 30 dB above the audio signal level.
Whatever qualities it may have, this is crappy.
9 Aug
9 Aug
1:47 a.m.
On Fri, Aug 08, 2014 at 10:22:36AM +0000, Fons Adriaensen wrote:
On Thu, Aug 07, 2014 at 06:24:41PM -0700, Ken Restivo
wrote:
Ah, thanks, I'd forgotten that! I did this peice years ago in PHASEX,
and tried to correct the DC offset, but wasn't able to completely:
http://www.restivo.org/blog/podpress_trac/web/317/0/airbrush-0.1.ogg
(WARNING: might blow your speakers as described above)
It looks like this:
http://bace.s3.amazonaws.com/dcoffset.jpg
It sounds so... raw, uncontrolled, well, ANALOG.
Most software simulations sound
more or less authentic, but all so much more "tame", for want of a better
term.
But PHASEX always sounded to me (and felt, as I was playing with it) that at any
moment it could do something crazy like throw a DC offset, to into an uncontrollable
oscillation, or blow up my speakers, etc.
You have been lucky then. If your sound card and amp are DC coupled,
Phasex *will* blow up your speakers. On my first test it produced a
DC offset 30 dB above the audio signal level.
Whatever qualities it may have, this is crappy.
I guess so. But maybe it's dangerousness is what gives it its sound.
Will Alexander once described Keith Emerson's Moog as having "no
padded cell technology". Meaning, it was capable of destroying amps,
PA systems, expensive mixing boards, huge stadium-sized house sound
systems, etc. He treated the thing like a loaded weapon when plugging
it into stuff. Maybe this is what he meant by that.
-ken
11:46 a.m.
On Fri, Aug 08, 2014 at 04:47:26PM -0700, Ken Restivo wrote:
You can't hear DC. It may drive your speakers into some nice distortion
before the funny smell appears, but there are less dangerous ways to
achieve the same.
It looks like this:
http://bace.s3.amazonaws.com/dcoffset.jpg
That's a very mild case, probably harmless. What I got, just by
selecting one of the presets ('classic pad' or something similar)
was a DC offset 30 times as big as the audio signal. In terms of
power that close to a 1000:1 ratio.
Whatever
qualities it may have, this is crappy.
I guess so. But maybe it's dangerousness is what gives it its sound. Will Alexander once described Keith Emerson's Moog
as having "no
padded cell technology". Meaning, it was capable of destroying amps,
PA systems, expensive mixing boards, huge stadium-sized house sound
systems, etc. He treated the thing like a loaded weapon when plugging
it into stuff. Maybe this is what he meant by that.
That can be said of any synth in fact. And I'm pretty sure that
ELP's PA system was not DC-coupled - the mixers they used had
transformer inputs.
DC offsets occur naturally when using phase modulated oscillators
even if their basic waveform is DC-free. It's part of the way PM
synthesis works, and you want the DC in the modulation inputs. But
once the VCO outputs enter the other parts of the audio path it
should be removed. It's trivially simple to do that.
--
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)
3:11 p.m.
On Saturday 09 August 2014 05:46:03 Fons Adriaensen did opine
And Gene did reply:
Whose iron can be quite well saturated when dealing with a DC level at
their inputs. That will have a sound all its own that I've never found to
be at all pleasant due to the IM distortion it causes.
On Fri, Aug 08, 2014 at 04:47:26PM -0700, Ken Restivo
wrote:
You are being too kind, Ken. Has this list gotten so genteel you can't
use more descriptive terms?
It looks like this:
http://bace.s3.amazonaws.com/dcoffset.jpg
That's a very mild case, probably harmless. What I got, just by
selecting one of the presets ('classic pad' or something similar)
was a DC offset 30 times as big as the audio signal. In terms of
power that close to a 1000:1 ratio.
> > Whatever qualities it may have, this is crappy.
I guess so. But maybe it's dangerousness is what gives it its sound.
You can't hear DC. It may drive your speakers into some nice distortion
before the funny smell appears, but there are less dangerous ways to
achieve the same.
Will Alexander once described Keith Emerson's
Moog as having "no
padded cell technology". Meaning, it was capable of destroying amps,
PA systems, expensive mixing boards, huge stadium-sized house sound
systems, etc. He treated the thing like a loaded weapon when plugging
it into stuff. Maybe this is what he meant by that.
That can be said of any synth in fact. And I'm pretty sure that
ELP's PA system was not DC-coupled - the mixers they used had
transformer inputs. DC offsets occur naturally when using phase modulated
oscillators
even if their basic waveform is DC-free. It's part of the way PM
synthesis works, and you want the DC in the modulation inputs. But
once the VCO outputs enter the other parts of the audio path it
should be removed. It's trivially simple to do that.
The idea is simple, properly doing it in the presence of front office bean
counters who don't understand why it should be done, is not. But you knew
that...
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
6:07 p.m.
On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene Heskett wrote:
If you refer to the 'crappy', I wrote that, not Ken.
To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png>
All I needed to get that was loading the 'mellow-pad' patch from
sys-patches and hit a C-Maj chord. Output level as indicated by
jkmeter was +9.5 dB FS. Which means that all that would remain
of this after DA conversion was full scale DC for a second or so.
Mellow, indeed.
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)
> >
Whatever qualities it may have, this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you can't
use more descriptive terms? 
7:38 p.m.
On Sat, 9 Aug 2014 16:07:58 +0000
Fons Adriaensen <fons(a)linuxaudio.org> wrote:
On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene Heskett
wrote:
If you refer to the 'crappy', I wrote that, not Ken.
To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png>
All I needed to get that was loading the 'mellow-pad' patch from
sys-patches and hit a C-Maj chord. Output level as indicated by
jkmeter was +9.5 dB FS. Which means that all that would remain
of this after DA conversion was full scale DC for a second or so.
Mellow, indeed.
Ciao,
Ouch!
I've had a few tries with phasex and it sounded interesting, but never got on
with it because of information overload - too many controls all visible at
once :(
--
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.
> >
Whatever qualities it may have, this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you can't
use more descriptive terms?
10 Aug
10 Aug
11:44 a.m.
On Sat, Aug 09, 2014 at 06:38:39PM +0100, Will Godfrey wrote:
I've had a few tries with phasex and it sounded
interesting, but never got on
with it because of information overload - too many controls all visible at
once :(
The problem here is not the number of controls, but the lack of
visual hints normally provided by layout, sizes and colors.
All the controls look the same and are in a single line in each
module so the only way to know which is which is to read the
labels. It's a pity because obviously some work went into the
GUI - it was just not based on ergonomic principles.
Another strange thing is that signal flow is right to left.
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)
9 Aug
9 Aug
10:46 p.m.
On Saturday 09 August 2014 12:07:58 Fons Adriaensen did opine
And Gene did reply:
On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene Heskett
wrote:
If you refer to the 'crappy', I wrote that, not Ken.
So I miss-read the quoting depth, again. :(
> >
Whatever qualities it may have, this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you
can't use more descriptive terms? To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png>
Good grief, and that thing was allowed out the door? It wouldn't have
made it on my watch.
All I needed to get that was loading the 'mellow-pad' patch from
sys-patches and hit a C-Maj chord. Output level as indicated by
jkmeter was +9.5 dB FS. Which means that all that would remain
of this after DA conversion was full scale DC for a second or so.
Mellow, indeed.
Ciao,
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
11:08 p.m.
On Sat, 9 Aug 2014 16:46:53 -0400
Gene Heskett <gheskett(a)wdtv.com> wrote:
On Saturday 09 August 2014 12:07:58 Fons Adriaensen
did opine
And Gene did reply:
Hmmm.
Just had a look on github no development for 2 years :(
--
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.
On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene
Heskett wrote:
If you refer to the 'crappy', I wrote that, not Ken.
So I miss-read the quoting depth, again. :(
> >
Whatever qualities it may have, this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you
can't use more descriptive terms? To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png>
Good grief, and that thing was allowed out the door? It wouldn't have
made it on my watch.
All I needed to get that was loading the 'mellow-pad' patch from
sys-patches and hit a C-Maj chord. Output level as indicated by
jkmeter was +9.5 dB FS. Which means that all that would remain
of this after DA conversion was full scale DC for a second or so.
Mellow, indeed.
Ciao,
Cheers, Gene Heskett 
10 Aug
10 Aug
1:23 a.m.
On Sat, Aug 9, 2014 at 4:46 PM, Gene Heskett <gheskett(a)wdtv.com> wrote:
On Saturday 09 August 2014 12:07:58 Fons Adriaensen
did opine
And Gene did reply:
open source is well known for making do without windows; it tends to do
without doors also.
On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene
Heskett wrote:
If you refer to the 'crappy', I wrote that, not Ken.
So I miss-read the quoting depth, again. :(
> >
Whatever qualities it may have, this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you
can't use more descriptive terms? To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png>
Good grief, and that thing was allowed out the door? It wouldn't have
made it on my watch.
4:47 a.m.
On Saturday 09 August 2014 19:23:40 Paul Davis did opine
And Gene did reply:
On Sat, Aug 9, 2014 at 4:46 PM, Gene Heskett
<gheskett(a)wdtv.com> wrote:
Chuckle. Well said Paul.
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
On Saturday 09 August 2014 12:07:58 Fons
Adriaensen did opine
And Gene did reply:
open source is well known for making do without windows; it tends to do
without doors also. On Sat, Aug 09, 2014 at 09:11:15AM -0400, Gene
Heskett wrote:
So I miss-read the quoting depth, again. :(
> > > Whatever qualities it may have,
this is crappy.
You are being too kind, Ken. Has this list gotten so genteel you
can't use more descriptive terms?
If you refer to the 'crappy', I wrote that, not Ken. To see what I mean:
<http://kokkinizita.linuxaudio.org/linuxaudio/phasex-mellow-pad.png
>
Good grief, and that thing was allowed out the door? It wouldn't
have made it on my watch.
3555
days inactive
3557
days old
linux-audio-user@lists.linuxaudio.org
14 comments
7 participants
participants (7)
-
Fons Adriaensen -
Gene Heskett -
Ken Restivo -
Paul Davis -
Ralf Mardorf -
Tito Latini -
Will Godfrey