Please account for the AEG ADSR factor when performing polyphony testing. I think some are accounting for this. Sladjan has mentioned playing detached notes - not legato notes.

The problem with legato is that as you lift one note to play the next - the release will continue to sound and terminate sometime after you lift off the key. The amplitude envelope is programmed to keep sounding the element after you lift a key.

Either add space equal to just above the release time between notes or go in and reprogram your envelope to have zero release time (instant sound cut-off). I'm not sure how the AWM2 "Init" performance works from memory. Maybe the release is already 0.

... or if you're going to use the overlap - just keep in mind the release.

This doesn't map out all the "gotcha's" - but it may help eliminate one.

Jason wrote:
Please account for the AEG ADSR factor when performing polyphony testing. I think some are accounting for this. Sladjan has mentioned playing detached notes - not legato notes.

The problem here IMO is not legato because if you have 64 elements active, one should be possible to hit two notes at the same time, at least for the first time. Further playing as legato would probably cut notes off, but even if you hit two notes once only, you will hear just one note, a single element instead of the expected two.

Sladjan wrote:
The problem here IMO is not legato because if you have 64 elements active, one should be possible to hit two notes at the same time, at least for the first time. Further playing as legato would probably cut notes off, but even if you hit two notes once only, you will hear just one note, a single element instead of the expected two.

Yes

One of the big problems. With Stated/Claimed Polyphony.
Is there does not seem to be. A scientific way to Prove/Disprove it.(4 the user lets say).
Now I am in no way inferring Yamaha has made any bogus Claims.
But someone may find and prove it is 126 and 1/2 polyphony. LOL
More logical thought. Makes me lean toward. Elements are Not exact Numbers. In Poly.
or the routing/Hardware/software. Take a Little out/Adds to???
There are so many possible lines of thought
I do know That. if We Users Could prove a true measurement of polyphony.
A certain Company Which Starts. with R. Would have their ass sued to the hilt. *256 poly my ass.
Of course. They always CTA/CYA. With. Well it depends on how You apply it. And they state that Upfront.
-------
Now The other thread about poly. Was answered by Bad Mr.
Why He ignored this one???
But this thread. Has some solid results. That need to be answered by Yamaha.
Just to end it. * The used M6/7/8 market has taken a huge hit in the last 3 weeks. Part Namm But .....
When left unresolved. Things like this Shake customers up.
And Prior to paying out a lot of money people do read this board.
There are potential buyers out there. That Are also waiting for an answer to this thread/Post.
I know because one ask me about it.

I would personally not use the keyboard keys to test polyphony. Previous testing I used a DAW to invoke MIDI note on/off messages.

The problem with the switches in keys is that you start to introduce a possible point of failure. If you're monitoring MIDI output from the keyboard as you press switches - you can be reasonably assured that the class of failure involving the switch itself did not happen. You would see a double MIDI note on (on/off/on) or more on/off/on/off/on "switch bounce" activity. Korg infamously had this problem in all of its early production run Kronos keybeds. Made for a lousy workstation and I think they "faked" a fix with software to let some of the hardware pass while also fixing the root cause in the keybed itself.

At any rate - I would feel better doing a test with an external MIDI message directed to Montage's MIDI-in to ensure there's just one MIDI-note-ON message per intended note and remove physically playing the keyboard even if striking two keys is easy-as-pie.

I'm not saying Yamaha has the lousy Korg issue (there's no evidence of that) - just that I believe it's a better test using MIDI messages instead of pressing buttons that eventually generate MIDI messages.

You're likely going to replicate the same results (because I somewhat trust no funny business with the switches). Even so, I'd shy away from the potential of human or device error that is inherent in mashing keys.

Just tested with DAW sending MIDI. Same deal.

Jason wrote:
Korg infamously had this problem in all of its early production run Kronos keybeds. Made for a lousy workstation and I think they "faked" a fix with software to let some of the hardware pass while also fixing the root cause in the keybed itself.

I'm not saying Yamaha has the lousy Korg issue (there's no evidence of that) - just that I believe it's a better test using MIDI messages instead of pressing buttons that eventually generate MIDI messages.

Hi Jason,
I don't want to play Korgs lawyer here, but I don't think that there is a need to make comparisons with other companies in the way you did.
As a matter of fact, it is not true that ALL keybeds had that problem you mentioned. I have a friend of that first version, and he doesn't have this problem and never had. That problem appeared only on some weighted keys, not on the 61 keybed which is similar to what we have in the Montage 6. And it really doesn't matter how they solved the problem if it works and people don't complain about strange feel or keybed behavior. If you don't know what they did, let us not suppose and speculate.
I'm not oversensitive about it, but lousy is a word which I could apply to quite some issues and solutions implemented on the Montage and unlike the keybed issue on the Kronos, those issues are on ALL Montage keyboards because they are of a software nature.
On top of all that, It would be highly unlikely that all keybeds would erratically send on/off messages. In fact, I don't really understand how you imagined that? In that case, we would hear double notes, just in the same way as some owners of the early Kronos models did.

@Sladjan:

I said in my post I doubt Yamaha has this issue. I've played many early Kronos keyboards and frequently noticed the double-strike issue. That was a lousy flaw to ship because it destroys your performance. And the issue was wide-spread not isolated to one or two units. All of that, as already mentioned, is in the past for Korg because they've corrected it. You may not see that failure as lousy - that's fine - opinions are subjective.

I point that out because to give an example of a major manufacturer having hardware-level failure of the keybed. The way to 100% ensure absolutely no chance of this class of failure is to use MIDI. I am simply stating that rigor can reduce the likelihood, however slim, of clouding results from other spurious sources of error.

Analogy: If you want to be 100% sure of keeping your hair dry walking outside - wear a shower cap. It may never rain - but if your test requires dry hair -- the extra insurance of using the shower cap will ensure no wet hair. You may look outside and say your area is in drought conditions for a long time -- but the instruction for a shower cap covers all bases. This is what is referred to as process/experimental rigor. These are lab tests so I suggest steps to ensure introducing the least amount of outside/unnecessary "moving parts".

There was a question about how to best perform reproducible tests so I give my feedback. Thanks for the question/comment.

I like Yamaha and own their gear however they are notorious for overstating, exaggerating and embellishing on features to sell their products.They love using big words and big promises in advertisements too. If there's a 10% chance under prime conditions that you'll get 128 poly then they will spec. out as 128. It might be technically accurate but practically it's not true. We need regulations on "practical specs" for the end user.

Anyone remember the EX5? Every other function was followed by that infamous DSP overload message.

I thought that the Montage and the Genos had this next generation processor with future expansion capabilities. It seems that the 128 limitation after 30 or 40 years will never be overcome. The Montage can do 128 AWM plus 128 FM and the Genos can do 128 internal plus 128 expansion but still it's only 128 per engine. In the year 2018 shouldn't be getting 512 and 1024 poly?

@David @Bad MR
First. If there is only a 10% chance of it ever working. At full 128 polly. It can Never be technically or legally accurate.
It has to work. 9 out of ten times. And have Legal Statement that. Full 128 Polyphony. Has a 10% tolerance.
No such statement exists. No disclaimers of any kind exist. To my knowledge.
And since both of the tests so far. Show it Failing 100% of the time. Then.
Bad MR./Yamaha has to answer this post.
Their avoidance of this post. Is starting to concern me.
As they answered another Current post on Polly. While avoiding this one.
(Which is showing up in the used market. Very plain to see. Down $500 to $600 in the last 3 weeks.)
Part of this is Namm. But the Kronos. Is only down $300.00 to $400.00.
And the majority of people expect a Kronos upgrade.(6 years. Is a long time for a keyboard.)

And Hence. Their dilema(Yamaha).
It may well be that any way of answering this post. Is a Catch 22. For Yamaha.
So if the test that uses midi switches. Is accepted as Concrete Fact. Poly Testing.
Then The Montage Does not have True 128 Polyphony. On the AWM2 side.(no one has tested the FM side yet?).
If true. Yamaha Will never admit it. Because. They would have to do a massive Recall/Refund. So do not hold your breath.
If this testing. Is a 'False positive'. Then Yamaha Will have to prove it wrong.
And it needs done ASAP.

And Yes this is 2018. Montage was brought out 2 years ago. So higher than 128 Poly per side.
Was and Could have been an expectation.
Now if announced in 2017/2018. To get excited about it. Would have taken 256 Poly per side.
Personally. I could never use that kind of poly up. But I am sure there are some that could even go past it.
And Maybe That is 1 of the many reasons. Why the Montage was Rushed to market.

Genos. Man I was so excited about it. Until I Realised. Not only was it. Also rushed to market.
But. The absolutely Stupid! Paranoia about editing sounds on an arranger. Was still there.
Korg has been. Very successful selling both arrangers that You can edit sounds on. And there workstation line.
The 2 worlds. Rarely meet. And the Poly. Limit Should be higher. 256 per. Again. This is 2018!

>>TANGENT<<

Tom, are you a Yamaha shill? :p

>>BACK ON TOPIC<<

Has anyone else tried this test?

I'm not arguing about anything to do with any of this, it's just a simple test. Post your results.

Perhaps don't talk about it here at all and instead just post the results of your actual experience, regarding polyphony.

...........................≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥≥PEACE≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤≤..........................

.....................vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvRESULTSvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv.....................