AMP spike at ending Envelope

[DIAA]

Hello there,

this is so basic that either it has already been discussed (and I just can’t find where) or I’m definitely missing something in MZX’s envelope programming.

 

I see much more flexibility than found on simple ADSR envelopes, however, in contrast to former CZ-like 8 steps rate/level programming for example, MZX’s has a different (unnatural to me) sense regarding such parameters, which is also shown in its (nice) graphical display.

 

I understand that MZX’s envelope course has been split into negative and positive halves (those funny numbers revealing a 7-bit representation resolution). Although offering a midrange zero is interesting for pitch envelopes, I can’t understand how this benefits AMP envelopes, for example. 

 

I think that a note shouldn’t be sounding before KEY ON. Hence, I don’t undertand what the INITIAL LEVEL before ATTACK means. What worries me much more, though, is the note being forced to end at midrange ZERO value, which is not zero amplitude. A note should stop once you have KEY (and PEDAL) OFF after all decays expire.  That, however, does not happen: assume a minimum (negative) sustain level, for example. If the key is held long enough to so that decay is over, a postmortem breath is heard because the envelope jumps from the minimum (negative) to midrange (zero) when the key is released…

 

In other words, you’ll hear a short sound peaking when you release the key after decay expires…

My MZ-X500 came in  firmware version 1.21, I didn’t dare to update it yet. In anyway, newer versions do not mention anything about this.

 

Since I can’t believe this is a standing bug, I'm surely misunderstanding something. Any light on this would be truly appreciated.

Thanks in advance.

[Brad Saucier]

Absolute versus Offset

Standard tones versus Hex Layer tones

 

Standard tones have editable envelope parameters that operate as offsets of Wave envelope parameters.  The user does not have access to Wave envelopes in these tones.  You will notice all values in an envelope are initially zero (unless tweaked by Casio at the user level).  A value of zero "represents" the default setting but does not reflect the default wave value (if that makes sense).  With this system, users can offset wave absolute values by adding or subtracting from them. The maximum offset is -64 or +63, since the system is operated by the MIDI controller standard of 128 steps.  

 

With hex layer tones, the user has access to wave envelopes in each layer of the tone.  Here, envelope parameters are absolute values.  What you see is what you get.  The range is 0-127, with zero being lowest level.  This is where the true synthesizer is located in the MZ-X500.  

 

Initial level is the level of sound at Key On.  Attack occurs after Key On at the rate and level programmed. 

[DIAA]

Thank you BradMZ for your prompt and clarifying reply.

Your 1st line already helps making envelope parameters and graphs meaningful. 

 

I must confess, I started using the instrument out of the box, going after clarification only when surprised by the issue above.

In this regard, I think:

a) MZX’s envelope description is not clear about such an important difference, even on searching web updated versions;

the corresponding screen on the instrument should state “offset” and not “envelope”, as those are substantially different things.

 

Your comprehensive explanation truly helps, also by raising this fundamental difference between Wave and HexLayers tones.

THANK YOU !

 

An issue still remains, however. The Forum and your expertise are valuable help. 

Let me narrow my original comments above to its final part (amplitude spike).

 

I can add screen pictures or sound files if necessary, but let try to be quicker with an example:

Assume, for example, you’re layering a piano and a string (or organ or any non-decaying wave), while playing a slow passage (notes being held for longer, eventually with a sustain pedal). While the piano naturally fades down, the string keeps sounding loud (since an AMP decay to zero is not in its preset). Ironically, this is aggravated by the wonderfully large polyphony, leading to a potentially unfitting note mix - forcing you to prematurely kill it by releasing the key/pedal. 

 

In such scenario, you might wish to easily modulate the string AMP, smoothly offsetting its amplitude down during sustain.

It’s not rare to expire all decays and an undesired spike will be heard when you finally release the key(s)/pedal because the offset jumps from its programmed silence (negative max) to its zero (midrange) default.

 

Users may have noticed this in a grand finale where notes are kept sounding until they completely fade out: a spike will sound when you finally release key(s)/pedal. 

 

The same happens with the PX-560 and PX-5S pianos, where it is particularly bothering as similar layerings are beautiful and quite often used.

 

Thanks once again.

[Brad Saucier]

The cause of the spike is due to release level being louder than decay level.  Normal tones only have release time, so you could shorten that to silence the release stage but it's not the way I would go about it.  Hex layer tones have 2 release levels that would help your situation.  Try that and see what you come up with.

[DIAA]

Yes, BradMZ, that's what happens, and it's unfortunate. Penalize non-HexLayer tones by limiting such a basic feature to hex layers is not fair, specially under such a powerful and flexible system architecture. I'm an R&D engineer myself and I feel a correction as worth and easily addressable. It's disappointing to see Casio leaving this as "a feature", as it seems to be the case. I suggest we wait and see if somebody else feels differently.