NRPN question
-
AdmiralQuality AdmiralQuality https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=83902
- Banned
- 6657 posts since 10 Oct, 2005 from Toronto, Canada
-
- KVRian
- 960 posts since 27 Jun, 2009 from Germany
It's a pity the standard did not define it the other way around.
If the optional LSB was the first message, then it would be safe to act only on the MSB and track the LSB (if it was sent). Now we have to deal with this unclear situation...
Ok. Enough moaning. Back to problem solving. Does anybody know whether a controller which sends MSB+LSB _always_ does that, or whether there is a chance that it occasionally omits the LSB if it is zero (or has not changed).
If one could rely on that, one could detect at "learn" time, whether both parts are recieved, and store the information, and then either wait for the LSB (if we know that we will get it) or not (if at learn time, there was no LSB).
Could that be a viable option???
I made some experiments with synthedit, and it copes well with all situations, even if you randomly send MSB only or MSB+LSB. And SE for sure does not simply ignore the LSB, since you can only achieve controller values of 1 when both MSB and LSB go to 127. I wonder how Jeff is doing it...??
Cheers,
Martin
If the optional LSB was the first message, then it would be safe to act only on the MSB and track the LSB (if it was sent). Now we have to deal with this unclear situation...
Ok. Enough moaning. Back to problem solving. Does anybody know whether a controller which sends MSB+LSB _always_ does that, or whether there is a chance that it occasionally omits the LSB if it is zero (or has not changed).
If one could rely on that, one could detect at "learn" time, whether both parts are recieved, and store the information, and then either wait for the LSB (if we know that we will get it) or not (if at learn time, there was no LSB).
Could that be a viable option???
I made some experiments with synthedit, and it copes well with all situations, even if you randomly send MSB only or MSB+LSB. And SE for sure does not simply ignore the LSB, since you can only achieve controller values of 1 when both MSB and LSB go to 127. I wonder how Jeff is doing it...??
Cheers,
Martin
-
AdmiralQuality AdmiralQuality https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=83902
- Banned
- 6657 posts since 10 Oct, 2005 from Toronto, Canada
Just curious Martin, what controller do you use that sends the high res control changes? I've never owned one.
It does seem the host would have a lot of power to mess this up too so definitely test in as many hosts as you can. Why they had to half-interpret MIDI instead of just streaming the raw bytes to the plug-in, I'll never understand. It breaks most long Sysex messages as they won't fit in the event structure.
If you'd like, email me some .mid of the high res control changes and I'd be happy to take a look at them and tell you what I make of them. aq (AT) admiralquality.com
It does seem the host would have a lot of power to mess this up too so definitely test in as many hosts as you can. Why they had to half-interpret MIDI instead of just streaming the raw bytes to the plug-in, I'll never understand. It breaks most long Sysex messages as they won't fit in the event structure.
If you'd like, email me some .mid of the high res control changes and I'd be happy to take a look at them and tell you what I make of them. aq (AT) admiralquality.com
-
- KVRian
- 960 posts since 27 Jun, 2009 from Germany
I am using TB MIdi Stuff on my ipad (connected via the io Dock). TB Midi Stuff (and there are others around) allows you to set up conrtrollers and define what they are sending. So, for testing I defined two sliders, which send the same NRPN, one of them with fine resolution, the other one only the MSB.AdmiralQuality wrote:Just curious Martin, what controller do you use that sends the high res control changes? I've never owned one.
It does seem the host would have a lot of power to mess this up too so definitely test in as many hosts as you can. Why they had to half-interpret MIDI instead of just streaming the raw bytes to the plug-in, I'll never understand. It breaks most long Sysex messages as they won't fit in the event structure.
If you'd like, email me some .mid of the high res control changes and I'd be happy to take a look at them and tell you what I make of them. aq (AT) admiralquality.com
I think it is also available for iPhone (in case you have one). Otherwise, I can try to record a MIDI file with some slider movements and send to you.
Cheers,
Martin
-
AdmiralQuality AdmiralQuality https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=83902
- Banned
- 6657 posts since 10 Oct, 2005 from Toronto, Canada
Ah, no. (I have an Android.) I prefer controllers I can play without having to look at them. Old fashioned, I know.martin_l wrote:I am using TB MIdi Stuff on my ipad (connected via the io Dock). TB Midi Stuff (and there are others around) allows you to set up conrtrollers and define what they are sending. So, for testing I defined two sliders, which send the same NRPN, one of them with fine resolution, the other one only the MSB.AdmiralQuality wrote:Just curious Martin, what controller do you use that sends the high res control changes? I've never owned one.
It does seem the host would have a lot of power to mess this up too so definitely test in as many hosts as you can. Why they had to half-interpret MIDI instead of just streaming the raw bytes to the plug-in, I'll never understand. It breaks most long Sysex messages as they won't fit in the event structure.
If you'd like, email me some .mid of the high res control changes and I'd be happy to take a look at them and tell you what I make of them. aq (AT) admiralquality.com
I think it is also available for iPhone (in case you have one). Otherwise, I can try to record a MIDI file with some slider movements and send to you.
Cheers,
Martin
Anyone use any 14 bit hardware MIDI controllers? What do you like?
