How important is VST3 support for you as a customer

[Music Engineer]

But they're not Steinberg anymore, they're Yamaha. And I have to suspect they're the ones pulling the strategy strings now.

hmm...maybe. didn't think of that yet.

[kelldammit]

The distinction is that apparently in C6 -- at least according to the demo video linked to above -- that they're going to split controllers into separate, distinct envelopes for *each note*. Not one channel-wide controller stream that affects every note, as with MIDI (with the exception of Polyphonic-Aftertouch, which IS different for every note in MIDI as well). This is a square peg/round hole mapping that's not interchangeable between the two editor modes. And all the problems that I pointed out will come into play. You won't be able to switch from one to the other, as they're conflicting paradigms.

i understood the distinction, and get what they're doing with note expression. how a vst instrument would actually handle that makes my head spin. i'm glad i'm not a dev
but like i said, sb simply CAN'T drastically change midi leaving cubase, otherwise, they lose compatibility with the rest of the world. do you think apple will move to vst3.5 solely to deal with a few weird midi notes coming out of cubase every now and again?
my guess is that if you really want note expression, you'll just have to use steinberg instruments in steinberg hosts. i just don't see it as some huge, earth-shattering thing that mandates universal change.

k

[kelldammit]

But they're not Steinberg anymore, they're Yamaha.

suddenly, it all makes sense...they're still mad about XG midi not taking off...this is their revenge!

[AdmiralQuality]

But they're not Steinberg anymore, they're Yamaha.

suddenly, it all makes sense...they're still mad about XG midi not taking off...this is their revenge!

It just may be. I've also come up with a conspiracy theory that maybe they're just trying to kill Steinberg in the vain hope that people will start buying keyboard workstations again.

[jancivil]

with the advent of Cubase 6, this just got a lot more important to me. Primarily for instruments, in my case.
the issue of x64 is important to me for any plug at this point.

[jdieks]

steinbergs moves, policy and marketing around VST 3 were certainly not optimal, but to counterbalance a bit, i think steinberg did us all a great favour by bringing VST to life in the first place. so let's not turn this into a diffuse and general steinberg bashing fest.

peace

Hey I love VST 2.x. I love the 3.1 version of their hosts too. "Last known good version."

But they're not Steinberg anymore, they're Yamaha. And I have to suspect they're the ones pulling the strategy strings now. I can almost forgive the folks at Steinberg.

In all honesty, you're clearly biased. In my experience, steinberg products have been getting better and better since they've been bought by yamaha. Any cubase 5.x version was pretty solid while e.g. pro-tools pre-8 LE crashed like hell on windows.Or are you saying that any DAW dev is a "badd-ass" except for the ones that don't make money? that would explain a lot

Another thing, have you ever done any media composing on a deadline? DON'T having to split tracks could actually save time=money.. Isn't that what the whole bussiness is about? So the option to "keep splitting tracks" or first write for piano and then split it is just more time consuming then NOT having to first write it for piano and split it, but doing it right from the beginning.. who cares for scores anyway these days unless you're in a MAYOR production.

I'm not beeing a steinberg fanboy, which in all honesty i'm beginning to be... allthough I like pro-tools 9 and live also, but I don't get the steinberg bashing just for steinberg bashing and not wanting to see THE USERS side of the story. Tunnel vision if you ask me..

On the other hand, you're right if you just want to emulated old gear... But the fact that you "just want to emulate old gear" only confirms that you're not very open to progress

[jancivil]

sb simply CAN'T drastically change midi leaving cubase, otherwise, they lose compatibility with the rest of the world. do you think apple will move to vst3.5 solely to deal with a few weird midi notes coming out of cubase every now and again?
my guess is that if you really want note expression, you'll just have to use steinberg instruments in steinberg hosts. i just don't see it as some huge, earth-shattering thing that mandates universal change.

k

no, apple will use AU, steiny will use ASIO and not Core Audio (and will continue to wrap AUs installed on a sytems as the vst of your choice). However, if there is a demand for it in apple's perception, AU protocol could be affected. EG: Right now, VSL is seeking to implement a MAS version to use in DP, to stand with their VST3 (in terms of multiple midi ports, in place of an AU implementation that can't include it).

[aciddose]

it's nice that we all agree vst3 is trash, but can't we take a more optimistic stance towards life? where are all the fish votes?

also i think the poll is missing the "if the plugin is vst3 only, i won't buy it" option.

a far more enlightening query from vst3's perspective is: given that a plugin is only available in vst3, would you: 1) "buy it", 2) "not buy it".

i think the result of that would be worse than 10% for vst3.

i ranted about yamaha and speculated on how they're behaving, how it could sub-divide their markets and the extreme risk they'd take to force vst3 or "vst4" on the world in a thread a few months ago. one of my speculations was also that they might simply not care about the cubase/vst market given the ratio for revenue between different markets. (typical yamaha hardware instruments, vs. software) i've seen it speculated here that they might "have it in" for cubase. i wouldn't go that far, but it may not be as much of a stretch as it seems.

also remember the syncrosoft aspect of all this. yamaha now owns syncrosoft, which is now "elicenser".

[jancivil]

I know and use actual differences between vst2 and vst3 and like the latter and the fact of further development.
I'm very much an optimist, I want things to be wonderful and am in favor of more of it.

this whole objection to it is ever so KVR though.

[aciddose]

which differences matter to you though?

if you're talking about the "side-chain" feature in cubase, that's all in the host. they should've implemented that years ago in sx2 - they didn't, and massiva became energyxt and so on.

i'm actually surprised you still need something like energyxt and that modern hosts don't simply include something like it as a basic element of the host.

that isn't a fault of vst2 though, it's the hosts that just don't implement those features.

do you use the new preset management stuff? per-voice modulation? sample-accurate setParameter?

(technically, the events type "parameter" can be done in vst2 as-is as well, just nobody does it and it isn't standardized. the same could be used for per-voice modulation. the preset stuff is possible but would require vst 2.5 or something)

[contrast]

I understand that all the developers think that VST3 sucks, but there are improvements to be made and nobody else is even trying to make them. Until such time as someone else steps up to the plate, I'm all for it and would like nothing more than for it to become as mandatory as VST2 support.

[FrettedSynth]

i'm actually surprised you still need something like energyxt and that modern hosts don't simply include something like it as a basic element of the host.

IMHO this would be a very welcome feature to any host. Have been doing it for some time now, would be nice if it was built in the host.

VST3 I could really care less about.

BTW great stuff aciddose will donate if I ever have any money to spare.

Sorry for the OT

[AdmiralQuality]

i'm actually surprised you still need something like energyxt and that modern hosts don't simply include something like it as a basic element of the host.

IMHO this would be a very welcome feature to any host. Have been doing it for some time now, would be nice if it was built in the host.

VST3 I could really care less about.

BTW great stuff aciddose will donate if I ever have any money to spare.

Sorry for the OT

Guys, we've been able to sidechain in Cubase/Nuendo 3.x the whole time. You just insert a 4 channel group track, pan one stereo channel to 1+2, your other to 3+4, then insert your 4 channel sidechain plugin. Yes, it's slightly clunky but it takes all of 15 seconds to set up.

I'm amazed at some of the comments above. It seems some of you just can't believe that a big corporation would intentionally shaft you and sell you nothing but a bill of goods. There's universal consensus from the development community on this for a reason folks. Most of us aren't ok with the strategy of throwing away everything that was already well tested and working perfectly, adding a bunch of unneeded complexity for virtually no new benefit, then slapping on a new major version number and charging everybody to buy it yet again. But that's how these guys roll I'm afraid. I'm not "biased" by anything other than exactly this issue. Like I said, I love SX 3.1. I own several Steinberg products. But after this, probably never again. This is why we're so pissed off (most of us small developers are customers too, eh?)

I can just imagine the shitstorm I'd get from my customers if I ported my VST 2.4 product -- 4 and a half years of well tested code with currently zero bugs -- to VST 3. And if I then called it version 2 and forced them to buy it again? Oh man...

[kelldammit]

no, apple will use AU, steiny will use ASIO and not Core Audio (and will continue to wrap AUs installed on a sytems as the vst of your choice). However, if there is a demand for it in apple's perception, AU protocol could be affected. EG: Right now, VSL is seeking to implement a MAS version to use in DP, to stand with their VST3 (in terms of multiple midi ports, in place of an AU implementation that can't include it).

the VSL/MAS discussion actually illustrates my argument pretty well. the multi-port feature has been around for a LONG time, and only mas (later rtas) supported it until vst3 came along. au still does not. while i like that feature, i can (and have) live(d) without it fairly happily over the past several years when it just wasn't an option. i don't hear logic users screaming that they need it either...otherwise, it'd probably be there already.
as for vsl/mas...with the advent of au, many users were calling for motu to ditch mas altogether, despite this unique feature. mas was considered a "dead" protocol. although it provided some "nice touches" for a small subset of users of one host, even the majority of that one hosts' users could've cared less. so it obviously wasn't worth it to vsl to put in the "free work". now that that protocol feature fits with their own design features, it perhaps makes more sense to consider, even though most devs still don't bother with mas at all. ironically, motu's au support is just an au-MAS wrapper,anyway.

that's exactly why i have my doubts about vst3x taking off. it has benefits, sure, but they're largely just "nice touches" for a fairly small subset of users. for some devs, it's a worthwhile endeavor to pursue (aka vsl), but for most, it's just "free work" to have to do, for no real benefit. if they choose not to support it, their plugs will still work as they do now. my bet is that most users simply won't care about the difference, if they notice it all.

k

[aciddose]

Guys, we've been able to sidechain in Cubase/Nuendo 3.x the whole time. You just insert a 4 channel group track, pan one stereo channel to 1+2, your other to 3+4, then insert your 4 channel sidechain plugin. Yes, it's slightly clunky but it takes all of 15 seconds to set up.

actually i was thinking of exactly that method when i posted my comment on it. it is clunky and it does have limitations though. i think they could set up something like a group-channel, only give you drop down menus to set up the routing if they hate the "wires" system so much. you know like you'd have four inputs to the channel, the plugin might have three inputs and five outputs... you could set up which channel input maps to which plugin input, and which plugin output to which channel output or something like that.

if "complexity" is the biggest issue, you could settle on something like having a maximum bus width of 16 channels, and label them 0-F, maybe color-coded.

sort of like the way the input/output mapping already works, but applied to individual channels rather than the global asio i/o.

it wouldn't add complexity for the ordinary user because it could be hidden away behind a "advanced" checkbox, and default to the way it works now. (automatically assign 0, 1, 2, 3 for the four inputs)

...but i think the wires method is _way_ easier to use. the plugin has the ability to define busses like: stereo(0,1), mono(2), stereo(3,4), quad(5,6,7,8), and can even give them names. so why shouldn't the host put that information to use to provide named busses in the menu system, or just work exactly like energyxt does and put each bus on a pin (of course allowing you to click on the pin to break-out the bus nodes, or whatever)

also have you ever used the "routing view" in a cubase channel? why did they add that, but then not make it possible to adjust the routing?