How important is VST3 support for you as a customer

[Big Tick]

If I understand this Note Expression stuff correctly:
- everything it does could have been done using a dedicated midi CC, whose value indicates the type of articulation to use next. This would have been more useful for people who actually *play* the instruments and need to control the articulations in realtime.
- the only benefit I can see from Note Expression, a la Steinberg, is that you can now move notes around, and the selected articulation will follow.

[trimph1]

Seems about it...

[standalone]

Oh yes, the infamous "two simple lines of code", so simple that we will never see them.

[trimph1]

Oh yes, the infamous "two simple lines of code", so simple that we will never see them.

That's because they have to push their system....

[Big Tick]

Oh yes, the infamous "two simple lines of code", so simple that we will never see them.

What are you referring to ?

[Urs]

- everything it does could have been done using a dedicated midi CC, whose value indicates the type of articulation to use next. This would have been more useful for people who actually *play* the instruments and need to control the articulations in realtime.

That would require a convention, e.g. an addendum to the standard MIDI protocol so that it can be added to instruments and hosts. Seeing how impossible it is to get things through with the MMA I think it's faster, easier and less prone to trouble to go VST3.5 - even if it takes another 3 more years to pick up, that's by far better than waiting for MMA to join in.

[Big Tick]

That would require a convention, e.g. an addendum to the standard MIDI protocol so that it can be added to instruments and hosts. Seeing how impossible it is to get things through with the MMA I think it's faster, easier and less prone to trouble to go VST3.5 - even if it takes another 3 more years to pick up, that's by far better than waiting for MMA to join in.

Yes, to make it a universally accepted protocol supported by hardware synths, I agree. But for vstis... the host could simply query the plugin for the CC control it wishes to use for note expression, and voila.... and know what ? with some clever programming on the host side, it wouldn't have been too hard to attach such CC events to the next played note, so that they would also move along with the notes in the piano roll editor...

Anyway, I guess my biggest complaint about vst3 is that it didn't need to be such a complete, complex re-write of something that had stood the test of time, and had plenty of momentum. It didn't need to put pressure on all the plugins developers to upgrade.

Having said that... we won't rewrite history. Steinberg made decisions, maybe for technical reasons beyond my understanding, maybe for marketing purposes, maybe to get a headstart over the competition... Whether we like them or not, we (myself included) can spend years whining about how things could have been done in a better way, it won't change a thing... So I guess we should move on, and find a more productive way to use our time.

So, of course at some point I will embrace vst3. Maybe I should be happy. Maybe it will downsize the landscape of plugin developers. I miss the time when we were only a handful

[Big Tick]

and one last thing - a standard is not a standard when it is owned by one company. Heck, even Microsoft, Apple and IBM are able to talk together to define standards in OASIS. Why can't we have the same with vst ?

In my opinion the needs of the end-users would be much better addressed if vst was managed by a committee involving the major host and plugins developers.

[Elektrobolt]

Like polyAT for example?

indeed - but there is much much more of them - so polyphonic panning, volume, filter settings etc etc etc etc etc
Basically as many as the plug-in shows to the OS

Sure, but synths and sequencers could do this already twenty years ago. But apart from a nice effect now and then it has never been used much.

You are muddling the point. They could not do those per note years ago. They could do them per channel. Though they COULD do Polyphonic Key Pressure (aka Aftertouch) per note. That is, the one MIDI message.

This of course raises the question, someone already back then DID see the importance of being able to do PER NOTE messages.

Just because You (obviously) never used these controllers much, does not speak for the rest of the music industry. Me personally, have used them lots, panning/balance is a dandy controller, modulation wheel is probably the most used though, especially since most keyboards have one in hardware.

Simulating a cello, violin or brass for example, or the mod wheel controlling filter parameters, the modulation wheel in Your opinion, is much unused? LOL

I wonder why MIDI, with already existing Pitch Wheel, Aftertouch and all those CC's, was later on extended with NRPN and RPN's, if those original messages weren't used much.

All the effects you mentioned can be easily created with any VSTi and even existing hardware just by using multiple MIDI channels and/or tracks if needed. So the possibilities are already there but where is new revolutionary new music created with per note effects?

Currently, those effects, as you call them, are attached to the channel, and does not move with the note.

In VST3 they are attaced to the note. This, it would seem, you have never tried. Because when you have, you would see how much faster your workflow gets. You can move or copy a note or notes and zing, you're there. No need to move the controller data to match the notes I just rearranged. Just one track per instance of something, no need to manage multiple tracks. It certainly is a game changer, mate!

Also, NE does not remove or change traditional MIDI in any way.

I don't see how VST3.5 could change the game just because you are not limited to a single polyphonic controller per MIDI channel anymore.

We are actually (and finally) "moving on". MIDI is nice, but it is far from being the protocol of protocols, though it has supported us for a long time, and will stay for a while yet.

That a plug-in can announce its own NE parameters and the host picks them up and you as a user can utilize them, is in itself a pretty cool feature.

One can trivialize VST3, like many people are do, but VST3 ultimately resolves more than one point. Some COULD have been implemented differently, sure, some already available, sure, but one here, one there, etc.

I think that in some way, the reluctance to support VST3 is simpy because of the changes needed. A valid point, of course, but it hardly invalidates VST3.

[Benutzername]

Well, I'm coming from a tracker background so I'm constantly doing thinks that go far beyond everything that you can achieve even with VST3.5. Maybe this is why I'm not impressed at all with all this 'game changing', 'revolutionary' or similar marketing BS.

[koolkeys]

and one last thing - a standard is not a standard when it is owned by one company.

I disagree. A standard can be defined by a general consensus. Would you say that ProTools is not a standard in professional studios? It's owned by a single company. But it became a standard by general consensus(a definition that even appears in any dictionary as one way of defining a standard).

Brent

[Big Tick]

Yes, by "standard" I meant "open standard".

[koolkeys]

If I understand this Note Expression stuff correctly:
- everything it does could have been done using a dedicated midi CC, whose value indicates the type of articulation to use next. This would have been more useful for people who actually *play* the instruments and need to control the articulations in realtime.
- the only benefit I can see from Note Expression, a la Steinberg, is that you can now move notes around, and the selected articulation will follow.

Actually, no. You're talking about VST Expression in it's FIRST encarnation, which is something that was supported in Cubase whether you had a VST3 plug-in or not. And many companies like VSL and many others jumped in with expression maps. VST3.5 lets you bundle expression maps with the plug-in so you don't have to find them and load them or go through any hassle of setting them up. It's all possible to package them together.

Note Expression is part of VST Expression 2.0, which requires VST3.5 to my knowledge. With Note Expression, you can have individual CC settings PER NOTE. Not articulations, but actual CC curves(essentially, though not actually using the MIDI spec to do it directly). So you could have pan per note at the same time as having expression per note and mod wheel per note, etc. So what you used to do on multiple tracks, you could now do on a single track if you wish. And that data goes with the note. You can still do things the old way, but it's like having polyphonic aftertouch on each note, and with the ability to use multiple types of controllers. So it's definitely not something that was possible before.

There is more to it, but that's it in a nutshell.

Brent

[koolkeys]

Yes, by "standard" I meant "open standard".

Well, regardless of which one we're talking about, you CAN have a standard, open or otherwise, based on general consensus.

On that note:

But is an open standard necessarily better? If somebody gets it right with a non-open standard, I don't think it is always a benefit to create an open one.

Open standards are often subject to their own set of issues. First of all, you have to get all possible parties involved or there will be a huge variance in how people implement the standard. Being "open" will of course mean that people can change it at will, invent their own features, etc. Some open standards work, but can you name one open standard that is better than a closed standard? I bet the list will be short. Even those that ARE open have somebody governing them, and they aren't truly open.

Devs have talked in the past about putting together an open plug-in standard, but I think the climb is so far uphill that it isn't likely to get far off the ground. Not that I wouldn't welcome it, but creating an open standard means even MORE change for devs around the world. You think VST3 requires some changes? Wait till there is a completely new standard, not grounded in anything previous. It will take years to even convince the general audio world that it's worth switching.

Like Steinberg or not, and like VST or not, that's really the way it is, IMO. Maybe it will happen some day, and that would be great. I just don't see it being very viable.

Look at Linux. It's open, and even has some branches that have overseeing parties. But it will never be a standard. Ever. Not as long as there are so many hands in the kitchen. Not as long as it's completely open.

Brent

[Big Tick]

First of all, you have to get all possible parties involved or there will be a huge variance in how people implement the standard. Being "open" will of course mean that people can change it at will, invent their own features, etc.

Well, I still think a panel made of representatives from Steinberg, Ableton, Cakewalk, NI, U-he, and a few others... would be better to ensure that changes to the vst "standard" are made in the smoothest way, for everybody impacted by them, i.e, end-users, host developers, and plugin developers.

Some open standards work, but can you name one open standard that is better than a closed standard? I bet the list will be short. Even those that ARE open have somebody governing them, and they aren't truly open.

Plenty major ones: HTML, XML, TCP/IP, POP, PNG, FTP, SVG, PDF....

Even the ones that originated from a closed-format, like PDF, were eventually opened. In the case of PDF, yes, Adobe is still very much behind it, but even for them, any change to the standard has to be proposed, discussed, and approved by ISO voting countries. I would have loved Steinberg to make a similar move, but as I said, history is history.

For other standards (take SVG for instance), the list of contributors makes me feel somewhat safe about any upcoming change to the format - at least these will have been discussed by people who have a broad view of all the implications.

You think VST3 requires some changes? Wait till there is a completely new standard, not grounded in anything previous.

Not if it were one based on a fork of vst 2.4....