Yesterday I was programming a modular and to speed up the things a little I've assigned the foundamental parameters to my masterkeyboard (I was using a Korg Z1).
I've noticed that the controllers aren't very precise, and the refresh a bit slow. This sounds like: if I make a sweep (for example) assigning a knob of my keyboard to the cutoff frequence, I obtain a "stepped" glide, until I move the knob very slowly.
I'm wondering if this effect can be caused by an "hardware" limitation or by something else and if there's a method to empruve the refresh rate.
Does anybody have experience on this?
refresh of contrroller and precision
Lima,
one standard thing to try:
instead of automating the cutoff knob directly,
try using constant val> 6dB/oct. lowpass filter>modulation input of filter.
for start you could try setting filter cutoff to 0 or very low, modulation amount to max, and cutoff of lowpass to almost 0 (arrows and page up and down keys come handy for this).
the lowpass will act as an integrator, adding 'inertia' to modulation...
one standard thing to try:
instead of automating the cutoff knob directly,
try using constant val> 6dB/oct. lowpass filter>modulation input of filter.
for start you could try setting filter cutoff to 0 or very low, modulation amount to max, and cutoff of lowpass to almost 0 (arrows and page up and down keys come handy for this).
the lowpass will act as an integrator, adding 'inertia' to modulation...
just because you cannot imagine something that doesn't exclude it from reality.
Hi Michu! Thanks al lot for your trick, I will try it this evening. (it seems very interesting to do a lot of other services where some inertia is needed 
)
However I was wondering about how to increment the refresh rate instead how to correct my patches. Have you ever got any problem such the one I tell in my previous post? Firstly I'm interessed to know if it is a problem of my system or if it's related to the midi managing of the scope; then if it's a problem of my pc (for example if it's too slow) I start searching for a faster machine.
)However I was wondering about how to increment the refresh rate instead how to correct my patches. Have you ever got any problem such the one I tell in my previous post? Firstly I'm interessed to know if it is a problem of my system or if it's related to the midi managing of the scope; then if it's a problem of my pc (for example if it's too slow) I start searching for a faster machine.
Welcome to the dawning of a new empire
Lima, given that control change messages have 7 bit resolution, which equates to 128 steps, stepping is inevitable when controling something like filter cutoff covering whole audible spectrum...
even if your controller uses full resolution, and some cheaper ones do not (i believe that your Korg is ok tho')
_________________
<font size=-1>[ This Message was edited by: Michu on 2005-08-18 04:33 ]</font>
even if your controller uses full resolution, and some cheaper ones do not (i believe that your Korg is ok tho')
_________________
<font size=-1>[ This Message was edited by: Michu on 2005-08-18 04:33 ]</font>
Possibly the stepping is caused by high CPU loads, then lowering CPU usage helps, or restoring SFP's priority in Task Manager if you start it in Low Priority.
While Michu's trick works great to smoothen and enhance specific parameters like filter cutoff, mine covers whole of SFP MIDI. My trick does not enhance the resolution though, which remains at 128 steps...
While Michu's trick works great to smoothen and enhance specific parameters like filter cutoff, mine covers whole of SFP MIDI. My trick does not enhance the resolution though, which remains at 128 steps...
more has been done with less
https://soundcloud.com/at0m-studio
https://soundcloud.com/at0m-studio
Thanks my friends for your help 
@ Michu: I understand what you mean, and I agree with you that the resolution is only 1/128 of the whole intervall, so if I would slowly control a filter cutoff in the whole spectum, I necessarily hear the step between 2 consecutive steps. This is a midi limitation due the midi stndards, and I accept it. The problem is related to the "realtime response". I'll try to explain it using an example: if I take one filter and assign the cut-off knob to my korg and if I choose that the modulation intensity is the maximun range, two different things happens:
- if I turn gently (but not extremely slowly) the knob I hear a "fluid" transition. I see the knob on the monitor that follows my movement on the masterkeyboard.
- else if I turn the knob fastly the transition isn't fluid (stepped) and the virtual knob on the monitor jumps from the initial point to the final point without passing thru the intermediate values.
@ At0m: I'm not enought smart with computers to change such delicate parameters, so I never changed the priority settings. Your idea is very interesting, and I would try it.
How can I do it? There's something to know to avoid any futher problem? (for example stability or freezings problems, conflict with other applications etc..).
I've noticed that the windows refresh is also very slow in some cases (for example when I've got a lot of windows open: routing windows+stm2448+some effects etc..). Do you think that an higher priority can improve this also?
@ Michu: I understand what you mean, and I agree with you that the resolution is only 1/128 of the whole intervall, so if I would slowly control a filter cutoff in the whole spectum, I necessarily hear the step between 2 consecutive steps. This is a midi limitation due the midi stndards, and I accept it.
The problem is related to the "realtime response". I'll try to explain it using an example: if I take one filter and assign the cut-off knob to my korg and if I choose that the modulation intensity is the maximun range, two different things happens:- if I turn gently (but not extremely slowly) the knob I hear a "fluid" transition. I see the knob on the monitor that follows my movement on the masterkeyboard.
- else if I turn the knob fastly the transition isn't fluid (stepped) and the virtual knob on the monitor jumps from the initial point to the final point without passing thru the intermediate values.
@ At0m: I'm not enought smart with computers to change such delicate parameters, so I never changed the priority settings. Your idea is very interesting, and I would try it.
How can I do it? There's something to know to avoid any futher problem? (for example stability or freezings problems, conflict with other applications etc..).
I've noticed that the windows refresh is also very slow in some cases (for example when I've got a lot of windows open: routing windows+stm2448+some effects etc..). Do you think that an higher priority can improve this also?