The true story of a product flaw nightmare.

[risome]

Read this its probably the MAudio thing
http://www.analogindustries.com/blog/en ... 9879147926

[VicDiesel]

I have an Axiom 25 which I returned because the pads were misfiring. Sweetwater sent me a new one which still had the same problem. Since I don't use the pads I let it slide, but the other day I noticed it misfiring notes. I'm not amused.

Victor.

[Bronto Scorpio]

I don't think the problem is your controller. Your controller does'nt know wich plugins / presets you are playing. And the problem seems to happen only with a few plugins / presets.

[AdmiralQuality]

The Poly-Ana demo I tried a while back also had many presets with this problem.

I don't think so. It may have presets that aren't set the way you like, but Poly-Ana (when set to Unison mode) supports both legato envelope triggering (or not) and legato portamento triggering (or not). It will also give you a choice between LOW, HIGH, and LAST note priority. With these settings, you can exactly duplicate the note assignment and triggering functionality of ANY classic synthesizer.

And if there is a problem, could you please send me an email and let me know how to reproduce it? An example MIDI file would be great for that too. But as far as I know, it works as it should, and whatever behavior you like should be selectable through the various options on Poly-Ana's performance panel (left of the keyboard).

[mcnoone]

I don't think so. It may have presets that aren't set the way you like, but Poly-Ana (when set to Unison mode) supports both legato envelope triggering (or not) and legato portamento triggering (or not). It will also give you a choice between LOW, HIGH, and LAST note priority. With these settings, you can exactly duplicate the note assignment and triggering functionality of ANY classic synthesizer.
And if there is a problem, could you please send me an email and let me know how to reproduce it? An example MIDI file would be great for that too. But as far as I know, it works as it should, and whatever behavior you like should be selectable through the various options on Poly-Ana's performance panel (left of the keyboard).

Hi.
The problem I am still having is related directly to this.

I found something out about this problem.
It seems to be related to polyphony, as was mentioned before.
I tried out this freeware synth called Symptohm-Melohman Performer Edition
by Ohm Force.
1 of the presets had this note cutoff problem.
It was a mono retriggered patch and the polyphony was set at 1.
I switched it to 2, and still the problem, then I switched it to 3 polyphony, and the problem did not happen anymore, at 3 note polyphony.
Oh well, I guess I'll never figure out why it's happening.

I can no longer try your demo as I believe it had a 15 day limit or something, but I thought you used noise bursts instead? You changed the demo of PA, so maybe I can still try.
Here's the thing...if it doesn't have the above quoted function to change polyphony, then it will not be able to fix the problem.
I think it's directly linked to a polyphony and envelope timing problem, that may be down to some programming choices in those regards, but why would it happen in Live's simpler, and instrument racks?
It's really something that needs to be looked at.
If I can still use the PolyAna demo, then I'll recreate it with the midi, and audio Live files...Garageband too, if necessary.
I will add, this problem has never happened with any U-he product, or AAS, or Korg product. So it's not with every synth.

[AdmiralQuality]

Hi.
The problem I am still having is related directly to this.

I can no longer try your demo as I believe it had a 15 day limit or something, but I thought you used noise bursts instead? You changed the demo of PA, so maybe I can still try.
Here's the thing...if it doesn't have the above quoted function to change polyphony, then it will not be able to fix the problem.
I think it's directly linked to a polyphony and envelope timing problem, that may be down to some programming choices in those regards, but why would it happen in Live's simpler, and instrument racks?
It's really something that needs to be looked at.
If I can still use the PolyAna demo, then I'll recreate it with the midi, and audio Live files...Garageband too, if necessary.
I will add, this problem has never happened with any U-he product, or AAS, or Korg product. So it's not with every synth.

The Poly-Ana demo is no longer time limited, it just makes occasional silent gaps in the audio, so give it a try and let me know. Thanks!

[mcnoone]

The Poly-Ana demo is no longer time limited, it just makes occasional silent gaps in the audio, so give it a try and let me know. Thanks!

Okay.
Thanks.
Will check it out soon.

[mcnoone]

I don't think so. It may have presets that aren't set the way you like, but Poly-Ana (when set to Unison mode) supports both legato envelope triggering (or not) and legato portamento triggering (or not). It will also give you a choice between LOW, HIGH, and LAST note priority. With these settings, you can exactly duplicate the note assignment and triggering functionality of ANY classic synthesizer.
And if there is a problem, could you please send me an email and let me know how to reproduce it? An example MIDI file would be great for that too.

Here's an mp3 playing 3 of 6 presets the note cutoff happens with.
The 3rd sound-preset on the mp3 (last part) has a corresponding midi file below it.
The midi looks normal, but the audio cuts off short on random notes throughout it.
I haven't tried any altering of the presets to change it.
I have tried with the same results on 2 different macs and 2 different M-Audio keyboard midi controllers with the same results.
It seems to happen only with mono patches, in legato, with short release times.
It happens with too many plugins and especially with Puremagnetic, and other Live racks samples.
Here's the files
3 preset note cutoff problem mp3
http://www.box.net/shared/1bud40r2q1

3rd preset midi part
http://www.box.net/shared/i5v5tcnqo8

btw...the presets should not need any altering to fix the problem.
They should play normally without changing settings.
So even if I can fix it by changing things, it's not fixing the problem, but hiding it.
I'm on a macbook pro 2.88 w/t 4gb ram.
The problem happens in every daw I tried. Live8, Garageband, and before using the Reaper demo.

[AdmiralQuality]

Interesting technique!

I think I see what's happening here, and it's caused by the high polyphony of the notes you're holding, even though the synth patch is set to UNISON (monophonic). This only happens in when note priority mode is set to LAST. Unfortunately, last note priority is what you need to make sense of this kind of playing. When switched to LOW or HIGH note priority most of your notes don't sound because you're holding a note above or below them. (Note that all classic monophonic synths were LOW or HIGH note priority. So if you tried to play this same way on a MiniMoog for example, it wouldn't come out right.)

It sounds fine with a polyphonic patch. (POLY MODE switch set to "POLY") But I'm guessing that's not what you want.

Let me take a look at the last note priority code and see if I can't improve the behavior some when it's presented with a whole bunch of simultaneously held notes.

[aciddose]

i guess it's either bad midi data or you don't know what you're doing. (you expect something else to happen, than what technically should happen and does happen)

does the same happen with xhip?

monophonic note handling is done with a list of all notes currently held. when a note is turned on, it is entered into the table, when it is released it is removed. the last entry is always the newest, and lowest/highest scanning is used for those modes. it is impossible for it to function incorrectly.

things that can "confuse" it (actually, bad midi inputs): illegal order of events in midi such as double note-on, or double note-off. note-off before note-on. zero velocity note-on (counts as note-off in midi spec).

it can not be more simple than this;
http://xhip.presetexchange.com/temp/notelogic.c

double note-on can be a problem - if the note is already sustained, and another note is played - when the release happens for that same note, which "virtual note" should be released? the double, or the original? the midi spec says this situation can not happen, it's an illegal order of events and the result if it does happen is undefined - however the recommendation is that all matching notes are turned off when a release occurs to prevent stuck notes. in my code, i also ignore any additional note-on after the initial one.

double note-off is a problem if the code doesn't handle that case. mine does. again, midi says this doesn't happen, and the behavior is undefined if it does. in my code it will just be ignored.

zero velocity note-on is valid midi, but not if the intention was a note-on action. a controller is not allowed to send velocity zero for a note-on during a keypress event, only a keyrelease. the spec says any zero velocity note-on counts as a note-off with default release velocity.

all-notes-off, all-sound-off events: the midi spec says all-sound-off is only to be used as a control function, such as a "panic" button for stuck notes. it says all-notes-off can be inserted after the final key-release event has completed (and a note-off or zero vel note-on was sent) or it can also act to replace the final note-off event. in addition to that the spec says that all-notes-off should _not_ be used to turn notes off, but must be handled as a valid signal in the case it's encountered. it acts both to inform the receiver that all events have ended and also that all sustained notes must first be turned off before further action. all-notes-off, unlike all-sound-off should not trigger any voice cuts, it should only act to naturally release all notes.

[mcnoone]

Thanks AdmiralQuality for checking into this problem, and aciddose for the extra info.
This problem actually happens with too many plugins. I'm really surprised that I couldn't find any other information about it at other forums, or google searches. At the same time there's other users telling me they had the same thing happen on their systems.
I really thinking it's definitely what aciddose was talking about though.
Thanks again for your help.

Michael

[aciddose]

well as i said i think it isn't a problem. it might just be that you're doing something and expect a different result from whatever you're doing. the project file, or a midi export would help a lot to diagnose if the issue is the order of events.

if it happens on multiple machines, multiple controllers, multiple hosts and with multiple plugins it's a sign the problem might be on your end instead. although i wouldn't be 100% surprised if it were a real bug in all that software. did you try it with xhip? i know how the code works to handle midi events in that, so i can say if it's a software bug or not.

if some plugins work fine, then it's going to be how those plugins handle that particular configuration i guess. it doesn't necessarily need to be a real "bug", it might just be different ways of handling things, both valid.