200e

I own 2 261e's, and have noticed that they may jitter A LOT when you 
input a voltage for pitch. This happens more with certain types of 
devices (a lot with the 225e and roland MPU-101, almost none with my 
Serge TKB).
Any of you guys experiencing similar problems? Any suggested solutions?


cuari7

It sounds like what ever you are using as a MIDI controller sends 
something the 225e and MPU-101 do not like.  However, straight CV from 
the TKB is ok.  Do you have a Darf, Arf or 266e?  You should try 
sending pitch CV from some other devices to see if the problems follow 
the device sending the pitch data or the module receiving the data. 



--- In 200e@yahoogroups.com, "cuari7" <cuari7@...> wrote:
>
> I own 2 261e's, and have noticed that they may jitter A LOT when you 
> input a voltage for pitch. This happens more with certain types of 
> devices (a lot with the 225e and roland MPU-101, almost none with my 
> Serge TKB).
> Any of you guys experiencing similar problems? Any suggested 
solutions?

> 
> 
> cuari7
>

Cuari7,

This may not solve your MIDI problem, but, then again, it might....
Many of the 'boutique' midi products from Deopfer and others (in my
system at least) have not played very well with many of the other
items in my system.  What I am suggesting is a "I can't figure out
exactly what it is, but this fixes it" solution.

A MIDI SOLUTIONS THRU BOX takes an incoming midi signal and 'corrects'
it, making it easier for the midi input to correctly read the signal.
 The box has one input and two outputs, and has solved many problems
in my system when there doesn't seem to be anything technically wrong,
but the midi stream is speaking Esperanto....

gary



--- In 200e@yahoogroups.com, "cuari7" <cuari7@...> wrote:

>
> I own 2 261e's, and have noticed that they may jitter A LOT when you 
> input a voltage for pitch. This happens more with certain types of 
> devices (a lot with the 225e and roland MPU-101, almost none with my 
> Serge TKB).
> Any of you guys experiencing similar problems? Any suggested solutions?
> 
> 
> cuari7
>

the answer to this probably lies in yr definition of 'jitter'. what
exactly are you noticing?

the 225 is an honest creature; it will instantly and faithfully render
the incoming 8 bits of MIDI as voltage (no built-in interpolation).
compared to 'analog' (i.e. perceptually continuous) control sources
this may sound unstisfactorily blippy, snappy, clicky or whatever. the
225e's CV outputs can be arbitrarily shaped and smoothed by using a
255 lag processor, or a 249 or 250 with the right settings.

-eb



--- In 200e@yahoogroups.com, "mritenburg" <mritenburg@...> wrote:

>
> It sounds like what ever you are using as a MIDI controller sends 
> something the 225e and MPU-101 do not like.  However, straight CV from 
> the TKB is ok.  Do you have a Darf, Arf or 266e?  You should try 
> sending pitch CV from some other devices to see if the problems follow 
> the device sending the pitch data or the module receiving the data. 
> 
> 
> 
> --- In 200e@yahoogroups.com, "cuari7" <cuari7@> wrote:
> >
> > I own 2 261e's, and have noticed that they may jitter A LOT when you 
> > input a voltage for pitch. This happens more with certain types of 
> > devices (a lot with the 225e and roland MPU-101, almost none with my 
> > Serge TKB).
> > Any of you guys experiencing similar problems? Any suggested 
> solutions?
> > 
> > 
> > cuari7
> >
>

"ezra.buchla" <ezra.buchla@...> wrote:
>
> the 225 is an honest creature; it will instantly and faithfully render
> the incoming 8 bits of MIDI as voltage (no built-in interpolation).

The MIDI spec for the pitch bend is 14 bit, which results in a pretty
smooth cv out - as demonstrated by the Continuum Fingerboard, which
can output clean and smooth glisses, even through a typical midi-to-cv
convertor (Kento, Frostwave, etc...)

One would assume that the 225e should be able to output this
resolution;  perhaps it is a simply midi cc# assignment issue, since
not all of the cc resolutions are as fine as the pitch bend...?

gary

yes, pitchbend is the exception, essentially using two sequential bytes per message.

this translates to pitch resolution  of approximately 1 cent if you stick to the GM 
recommended range of +/- 2 semitones. of course, the 200e oscillators can be configured 
to respond to something like +/- 3.5 octaves, in which case the stepping between values 
is quite noticeable indeed. and of course many commercial MIDI controllers are quite lazy 
in their PB output. sweep the pitch bend on an m-audio keyboard and you will certainly 
hear jittering and zippering no matter how fast the D-A conversion on the other end, 
unless some sort of interpolation is applied.

my point was simply that the observed jitter is almost certainly being produced by the 225 
and is a precise reflection of incoming messages; it is demonstrably not related to the 
261's CV sampling rate or any other property of that module.

we have considered implementing built-in interpolation to the 225's outputs as a firmware 
update, which for many users would obviate the need for a 255 to smooth out controller 
values. of course, this creates a use-case assumption which other users might find 
inconvenient...

i'm largely unfamiliar with the specific design of the frostwave or other MIDI-CV converters 
available; i'd be curious to know what kind of smoothing is being performed by these 
devices.

eb


--- In 200e@yahoogroups.com, "Gary Chang" <gchang@...> wrote:

>
> "ezra.buchla" <ezra.buchla@> wrote:
> >
> > the 225 is an honest creature; it will instantly and faithfully render
> > the incoming 8 bits of MIDI as voltage (no built-in interpolation).
> 
> The MIDI spec for the pitch bend is 14 bit, which results in a pretty
> smooth cv out - as demonstrated by the Continuum Fingerboard, which
> can output clean and smooth glisses, even through a typical midi-to-cv
> convertor (Kento, Frostwave, etc...)
> 
> One would assume that the 225e should be able to output this
> resolution;  perhaps it is a simply midi cc# assignment issue, since
> not all of the cc resolutions are as fine as the pitch bend...?
> 
> gary
>

"ezra.buchla" <ezra.buchla@...> wrote:
>
> yes, pitchbend is the exception, essentially using two sequential
bytes per message.
> 
> this translates to pitch resolution  of approximately 1 cent if you
stick to the GM 
> recommended range of +/- 2 semitones. of course, the 200e
oscillators can be configured 
> to respond to something like +/- 3.5 octaves, in which case the
stepping between values 
> is quite noticeable indeed. and of course many commercial MIDI
controllers are quite lazy 
> in their PB output. sweep the pitch bend on an m-audio keyboard and
you will certainly 
> hear jittering and zippering no matter how fast the D-A conversion
on the other end, 
> unless some sort of interpolation is applied.
> 
> my point was simply that the observed jitter is almost certainly
being produced by the 225 
> and is a precise reflection of incoming messages; it is demonstrably
not related to the 
> 261's CV sampling rate or any other property of that module.
> 
Ezra,

You are probably right - the jitter is most likely founded in the midi
source.

I just want to point out that there exist very sophisticated MIDI
controllers such as the Continuum Fingerboard (www.Hakenaudio.com)
that, utilizing typical midi to cv controllers, can output smooth
glissandos via midi (note and pitchbend information) up to 36
semitones (using an Encore Expressionist midi to cv converter).

gary

"Gary Chang" <gchang@...> wrote:
>utilizing typical midi to cv controllers, 

sorry, I meant to say midi to cv converters...

Hi, guys.

I hadn't checked this thread in a week! Thanx for all the responses.


>>>ezra.buchla" <ezra.buchla@...> wrote:
>
> the answer to this probably lies in yr definition of 'jitter'. what
> exactly are you noticing?<<<<


Hi, Ezra. What I meant by "jitter" is basically a nervous, constant, 
erratic change in pitch when playing a sustained note (almost like 
you're modulating the pitch of the oscillator with a random source). 
I could describe it as the buzzing of a mosquito. This seemed to be 
a more obvious problem with the 259e, and Jim Aikin pointed it out 
in his review of it on Keyboard magazine. Putting the voltage 
through a 255 does NOT fix the problem. Bummer, because This keeps 
me from getting any satisfactory use of the 225e as a MIDI-CV 
converter (I play the 200e exclusively with my TKB). Another thing 
I've noticed is that passing the TKB voltages through the 210e or 
the 256e will have the same effect! It's like the voltages 
get "tainted" by these devices.

Any suggestions?



cuari7

well, i have a couple of thoughts:

- maybe you answered this already (sorry), but are you sending a voltage
directly from the 225 (if so, is it the internal bus or the external cv
out?) or are you passing it through a 210 or 256? my guess is the latter but
i don't remember it mentioned...

- i just tested a 225 connected (with a patch cord) to the pitch cv of a 261
and was able to observe some jitter at certain frequencies close to the bit
transitions of the A->D converter. these jitters, at the largest, spanned
less than 0.2 cents, and were only visible on an oscilliscope. i could not
audibly detect them even by mixing the 261 output against an accurate test
oscillator and observing the beating patterns, which normally is a good way
to reveal the audible magnitude of such effects.

- using the internal bus ameliorated even these inaudible jitters. i am sure
that they were caused by a normal level of interference noise in the
patchcord or elsewhere.

- passing the cv through a 210, however, produced larger jitters that i
could audibly detect, again mostly at certain frequencies close to the bit
transition points. this is a result of more numerous and less accurate A->D
conversions in the 210, and it is certainly a problem.

- we should be able to fix this problem with a filter in the software, as
we've already done in the 259 (to fix the jitter jim aiken noticed in his
review), the 261 (because we had learned from the 259), and other modules
where e.g. noisy potentiometers posed a problem.

- there will always be some small loss of informational resolution when
doing lots of A->D->A conversions, but it needn't interfere with one's
listening experience or prevent you from using any kind of fine microtonal
pitch divisions, and the values certainly needn't leap around at random.

so anyway expect a significant improvement in the next 210 and 256 firmware
update. thanks for noticing, sorry for the annoyance...

and welcome to the digital world...

eb



On 6/11/06, cuari7 <cuari7@comcast.net> wrote:
>
>   Hi, guys.
>
> I hadn't checked this thread in a week! Thanx for all the responses.
>
>
> >>>ezra.buchla" <ezra.buchla@...> wrote:
> >
> > the answer to this probably lies in yr definition of 'jitter'. what
> > exactly are you noticing?<<<<
>
> Hi, Ezra. What I meant by "jitter" is basically a nervous, constant,
> erratic change in pitch when playing a sustained note (almost like
> you're modulating the pitch of the oscillator with a random source).
> I could describe it as the buzzing of a mosquito. This seemed to be
> a more obvious problem with the 259e, and Jim Aikin pointed it out
> in his review of it on Keyboard magazine. Putting the voltage
> through a 255 does NOT fix the problem. Bummer, because This keeps
> me from getting any satisfactory use of the 225e as a MIDI-CV
> converter (I play the 200e exclusively with my TKB). Another thing
> I've noticed is that passing the TKB voltages through the 210e or
> the 256e will have the same effect! It's like the voltages
> get "tainted" by these devices.
>
> Any suggestions?
>
> cuari7
>
>  
>


[Non-text portions of this message have been removed]

thanks gary

since this thread started i've experimented with sending double-precision
pitchbend data at high speeds from max to the 225 with fine results,
although i can't call it 'continuous' exactly. i can't for the life of me
think of any way to acheive better results on the controller side within the
MIDI spec without multiplexing data from a couple of channels (which would
require specialized logic in the converter.)

my curiousity now concerns what kind of built-in interpolation (if any) is
to be found in devices like the Expressionist. their published specs only
say that they acheive "smoothness" using high resolution D-A converters (16
bits, same as the 225.) perhaps someone on he forum knows?

i've never been satisfied with the resolution of MIDI, including pitchbend,
in any kind of environment, and it was gratifying to see at NAMM some larger
companies like ableton finally stepping up and supoprting higher rez
controllers. (with MAX it has long been possible to use powered ethernet
devices as controllers to acheive 32-bit float precision...)

often in max patches i find myself cascading two or more MIDI faders for
coarse and fine control of sensitive parameters like loop points. this
approach is also viable in the 200e...

we built the 225 to provide support for the MIDI platform, while recognizing
its limitations. (some folks may recall that don actually invented his own,
variable resolution standard in response to MIDI when it first came out,
recognizing its outdatedness even then; unfortunately WIMP never really
caught on...!)

of course, if you really want smooth pitchbend, you can always rig up a
banana-jack volume pedal and run 10 volts into it. this is the majick of
modulars.

- eb

On 6/11/06, Gary Chang <gchang@calarts.edu> wrote:
>
>   "Gary Chang" <gchang@...> wrote:
> >utilizing typical midi to cv controllers,
>
> sorry, I meant to say midi to cv converters...
>
>  
>


[Non-text portions of this message have been removed]

I have a 225e with firmware version 3.  I'm wondering if the 
controller busses on the 225e use the additional bit resolution for 
midi controller data at all?

I have a minimoog voyager which I am using as a controller keyboard.  
You can set the modwheel on it to transmit 7 bits of controller data 
or 14.  

Whether I have it at 14 or 7 the stepping from the 225e is the same 
and pretty dramatic. It suggests the 225e is just not responding to 
the additional resolution. 

Hopefully a simple firmware update will allow the 225e to use the 14 
bit resolution transmitted by many midi controllers these days?

Chris

"wiardmodular" <wiardmodular@...> wrote:
>
> I have a 225e with firmware version 3.  I'm wondering if the 
> controller busses on the 225e use the additional bit resolution for 
> midi controller data at all?
> 
> I have a minimoog voyager which I am using as a controller keyboard.  
> You can set the modwheel on it to transmit 7 bits of controller data 
> or 14.  
> 
> Whether I have it at 14 or 7 the stepping from the 225e is the same 
> and pretty dramatic. It suggests the 225e is just not responding to 
> the additional resolution. 
> 
> Hopefully a simple firmware update will allow the 225e to use the 14 
> bit resolution transmitted by many midi controllers these days?
> 
> Chris
>
hmmm - this might be the issue in a nutshell...

gary

chris,

you're right! i thought we had 14 bits for pitchbend only, but on closer
inspection this is not the case. sorry to have passed on my misconception.

we will certainly attempt something to address this in the next 225 firmware
update.

just to reiterate, i am still quite certain that the jitter observed at the
origin of this thread was produced by a 210 or 256. a firmware solution to
that problem has been formulated.

e

On 6/13/06, wiardmodular <wiardmodular@hotmail.com> wrote:
>
>   I have a 225e with firmware version 3. I'm wondering if the
> controller busses on the 225e use the additional bit resolution for
> midi controller data at all?
>
> I have a minimoog voyager which I am using as a controller keyboard.
> You can set the modwheel on it to transmit 7 bits of controller data
> or 14.
>
> Whether I have it at 14 or 7 the stepping from the 225e is the same
> and pretty dramatic. It suggests the 225e is just not responding to
> the additional resolution.
>
> Hopefully a simple firmware update will allow the 225e to use the 14
> bit resolution transmitted by many midi controllers these days?
>
> Chris
>
>  
>


[Non-text portions of this message have been removed]