Synth-DIY Yahoo! Groups Archives

Thread

QTKB response times

2002-12-31 by metafoetus2002 <jan-hinnerk.helms@t-onli

Hi,

for the Serge Quantizer my 1994 STS catalogue states (quote) response
time for all channels is about 8 milliseconds (unquote).

Does this mean that I'd hear pitch glitches in a situation like the
following: TKB A output is quantized by one QTKB channel, which controls a
PCO; the PCO's output is gated via a UAP; the gate signal that opens the UAP
is the same that clocks the TKB?

Further question: Is the response time equal for all QTKB channels?

If it is, is it possible to cure the assumed pitch glitch by using
one QTKB channel to delay the gate signal that is used for clocking and
gating?

Thanks in advance
and a happy & healthy new year wishes
Jan-Hinnerk Helms

P.S. Asking myself questions like this may be one reason why it takes
me so
long to follow my heart and buy a Serge ;-)

Re: QTKB response times

2003-01-01 by metafoetus2002 <jan-hinnerk.helms@t-onli

--- In SergeModular@yahoogroups.com, "metafoetus2002 <jan-
hinnerk.helms@t...>" wrote:

Show quoted textHide quoted text

> Further question: Is the response time equal for all QTKB channels?
>
> If it is, is it possible to cure the assumed pitch glitch by using
> one QTKB channel to delay the gate signal that is used for clocking and
> gating?

Sorry, it must of course read: Is it possible to cure the assumed pitch glitch by
using one QTKB channel to delay the gate signal that is used for gating (the
gate signal used for clocking the TKB should not be delayed)?

Best regards
Jan-Hinnerk

Re: QTKB response times

2003-01-03 by John Papiewski

As far as I know, the 8 ms settling time is a worst case thing for dithering.
If what you're doing is using the QTKB to tune the TKB, then you won't have to worry
about dithering. You'll set the pitch you want & it'll be in an 'easy' range for the
quantizer to interpret. The worst case is when you're quantizing some other voltage
(sample & hold for example) and it hits a voltage very close to the quantizer's
threshold... it takes a tiny interval for the quantizer to decide what to output, and
you hear this as a little hiccup. But for other intervals it's very fast ... I would
have to say faster than 8 ms.

The text of the catalog is pretty old. I think Rex did a redesign of the quantizers
a while ago.

Best,

John P.

"metafoetus2002 " wrote:

Show quoted textHide quoted text

> --- In SergeModular@yahoogroups.com, "metafoetus2002 <jan-
> hinnerk.helms@t...>" wrote:
>
> > Further question: Is the response time equal for all QTKB channels?
> >
> > If it is, is it possible to cure the assumed pitch glitch by using
> > one QTKB channel to delay the gate signal that is used for clocking and
> > gating?
>
> Sorry, it must of course read: Is it possible to cure the assumed pitch glitch by
> using one QTKB channel to delay the gate signal that is used for gating (the
> gate signal used for clocking the TKB should not be delayed)?
>
> Best regards
> Jan-Hinnerk
>
> To unsubscribe from this group, send an email to:
> SergeModular-unsubscribe@yahoogroups.com
>
> Keep on Patchin'!
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Re: QTKB response times

2003-02-02 by metafoetus2002 <jan-hinnerk.helms@t-onli

Hi John,

thanks for your answer.

This weekend I visited George Miller and his nice Serge (Blue Voice, Red
Control, TKB, all new STS panels). It was very moving to finally get my hands
on this beautiful system - thanks to George for his hospitality.

Two things puzzled me on his system:

First, comparing the unquantized and quantized signal as it controlled an
NTO through its 1V/Oct input revealed, that the QTKB's response times are
quite audible - as a short but noticeable portamento, even when using the
QTKB with the TKB as the input source.

Second I was a little bit disappointed with the scaling of the Decay and
Release times of the ADSR. As much as I appreciate a fine resolution of the
shorter Decay and Release times, I have to say that it surprised me that the
more noticeable changes in these times started occuring from 1 o'clock
upwards of the control setting, regardless of the curve switch setting.

Can anyone confirm these observations? And does anyone have experience
if the controls of the Modcan Envelope Generator 04A have the same scaling
problem?

BTW: As there is one lengthy thread in the SMOG archive regarding the
Modcan-Serge comparison, please let me state that my last question is not
meant to start a thread regarding if Modcan is copying Serge, nor am I stating
by my above question that Modcan is copying Serge or not.

In any case I have to get my own Serge. Soon.

Best regards
Jan-Hinnerk

Show quoted textHide quoted text

--- In SergeModular@yahoogroups.com, John Papiewski <
johnp299792@a...> wrote:

> As far as I know, the 8 ms settling time is a worst case thing for dithering.
> If what you're doing is using the QTKB to tune the TKB, then you won't have
> to worry about dithering. You'll set the pitch you want & it'll be in an 'easy'
> range for the quantizer to interpret. The worst case is when you're
> quantizing some other voltage (sample & hold for example) and it hits a
> voltage very close to the quantizer's threshold... it takes a tiny interval for the
> quantizer to decide what to output, and you hear this as a little hiccup. But
> for other intervals it's very fast ... I would have to say faster than 8 ms.
>
> The text of the catalog is pretty old. I think Rex did a redesign of the
> quantizers a while ago.
>
> Best,
>
> John P.
>
> "metafoetus2002 " wrote:
>
> > --- In SergeModular@yahoogroups.com, "metafoetus2002 <jan-
> > hinnerk.helms@t...>" wrote:
> >
> > > Further question: Is the response time equal for all QTKB channels?
> > >
> > > If it is, is it possible to cure the assumed pitch glitch by using
> > > one QTKB channel to delay the gate signal that is used for clocking and
> > > gating?
> >
> > Sorry, it must of course read: Is it possible to cure the assumed pitch glitch

> > using one QTKB channel to delay the gate signal that is used for gating

(the

> > gate signal used for clocking the TKB should not be delayed)?
> >
> > Best regards
> > Jan-Hinnerk
> >
> > To unsubscribe from this group, send an email to:
> > SergeModular-unsubscribe@yahoogroups.com
> >
> > Keep on Patchin'!
> >
> >
> > Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Re: QTKB response times

2003-02-02 by Les Mizzell

> Can anyone confirm these observations? And does anyone have experience
> if the controls of the Modcan Envelope Generator 04A have the same scaling
> problem?

My Modcan seems a lot smoother than this. Good control over very short
releases, but smooth control of longer releases noticeable way before the 1
o'clock position. Don't notice any particular position where there seems to
suddenly be a "jump" or "dead spot" or anything...

Les Mizzell

Re: QTKB response times

2003-02-02 by Chris Whitten

> Don't notice any particular position where there seems to
> suddenly be a "jump" or "dead spot" or anything...

I didn't notice that on my Serge ADSR either.
Either I'm not so critical or there is a problem with that module.

Re: QTKB response times

2003-02-06 by DarkStr717@aol.com

Hmmmm..... this is surprizing. Or else it's just not a factor in the music i
make. I do agree on the ADSR to a point. The fact that nothing seems to
happen till about 1o'clock is a relationship to the incoming voltage source;
or else the system you worked on may have power supply problems. Response
time on my TKB and ADSR are quiet fast in my opinion. Don't know what to say
to address your concerns. Maybe try another system.
Kind regards
John DuVal

In a message dated 2/2/03 3:49:50 PM, jan-hinnerk.helms@... writes:

<< First, comparing the unquantized and quantized signal as it controlled an
NTO through its 1V/Oct input revealed, that the QTKB's response times are
quite audible - as a short but noticeable portamento, even when using the
QTKB with the TKB as the input source.

Second I was a little bit disappointed with the scaling of the Decay and
Release times of the ADSR. As much as I appreciate a fine resolution of the
shorter Decay and Release times, I have to say that it surprised me that the
more noticeable changes in these times started occuring from 1 o'clock
upwards of the control setting, regardless of the curve switch setting. >>

Re: QTKB response times

2003-02-06 by DarkStr717@aol.com

In a message dated 2/2/03 3:49:50 PM, jan-hinnerk.helms@... writes:

<< First, comparing the unquantized and quantized signal as it controlled an
NTO through its 1V/Oct input revealed, that the QTKB's response times are
quite audible - as a short but noticeable portamento, even when using the
QTKB with the TKB as the input source.

Second I was a little bit disappointed with the scaling of the Decay and
Release times of the ADSR. As much as I appreciate a fine resolution of the
shorter Decay and Release times, I have to say that it surprised me that the
more noticeable changes in these times started occuring from 1 o'clock
upwards of the control setting, regardless of the curve switch setting. >>

Re: QTKB response times

2003-02-10 by metafoetus2002 <jan-hinnerk.helms@t-onli

--- In SergeModular@yahoogroups.com, DarkStr717@a... wrote:

Show quoted textHide quoted text

> Hmmmm..... this is surprizing. Or else it's just not a factor in the music i
> make. I do agree on the ADSR to a point. The fact that nothing seems to
> happen till about 1o'clock is a relationship to the incoming voltage source;

I am not sure what you mean with "incoming voltage source". If you mean
voltages to control the ADSR times - we had nothing plugged into these jacks
of the ADSR. Just an NTO's variable wave output patched into the audio input
of the UAP (switched to Pan mode), the ADSR's output into the VC1+2 input,
Gain turned up at around 11 o'clock, and the TKB's KP output went into the
ADSR's Gate input.

> or else the system you worked on may have power supply problems.

Guess not - everything else on this Blue Fun Station worked perfectly and as
expected.

> Response time on my TKB and ADSR are quiet fast in my opinion. Don't
> know what to say to address your concerns. Maybe try another system.

To quote you: LMAO! ;-) Do you really think that these issues will keep me
away from the Serge? No way. Not having to integrate the QTKB and the
ADSR just gives me...more panel space to fill with other modules :-))

Best regards
Jan-Hinnerk

> ---In a message dated 2/2/03 3:49:50 PM, jan-hinnerk.helms@t... writes:
>
> << First, comparing the unquantized and quantized signal as it controlled an
> NTO through its 1V/Oct input revealed, that the QTKB's response times are
> quite audible - as a short but noticeable portamento, even when using the
> QTKB with the TKB as the input source.
>
> Second I was a little bit disappointed with the scaling of the Decay and
> Release times of the ADSR. As much as I appreciate a fine resolution of the
> shorter Decay and Release times, I have to say that it surprised me that the
> more noticeable changes in these times started occuring from 1 o'clock
> upwards of the control setting, regardless of the curve switch setting. >>