Why is Steinberg working so hard to kill off VST2 when it works so well?

[Vertion]

Darn.

[Topcheese]

The code for it is free and open-source, so good luck trying to slow down that train.

Only if your plugin release is GPL, right? That won't work for most commercial developers.

Darn.

[Deep Purple]

[Bump1]

Steinberg didn't have to throw VST2 away.

They DID...if they wanted to keep bleeding the consumer public, of course.

[vasyan]

Steinberg is behaving as if it were the world police and from what I've seen, that develops aversity among the community. Add in that VST3 suffers from the same problems that VST2 does, it's hard to see why people and companies would force themselves to follow suit.

//Daniel

In my opinion, VST3 is not yet very relevant among developers. Nevertheless, so far no one has canceled the time-tested VST2, which defacto remains the standard for many developers, and praise to those manufacturers of DAW that support VST2!
And with VST3, everything is much worse, as say in Russia - "позвольте кинуть камень в огород Штенберга" (let me throw a stone at kitchen garden of Steinberg).
Much needs improvement, as well as porting SDK to other programming languages, for example, the Pascal Object, which I wrote in the another branch.

[BlueprintInc]

C++ is the relevant language in audio industries. Don't bet they'd put any effort in to port to anything else.

[vasyan]

C++ is the relevant language in audio industries. Don't bet they'd put any effort in to port to anything else.

No one bothers to do this.
Work on C++ is just a fashion inspired by the consumer society. But like us Punks do not support it

[quikquak]

No one bothers to do this.
Work on C++ is just a fashion inspired by the consumer society. But like us Punks do not support it

Other languages try their best, but they are just not as fast (yet) as C++ at algorithmic computations. C# and other Jits score well, but my own experiments a couple of years ago showed it to be about 10 times slower than the native code of C++, which was disappointing but not surprising.I’ve heard good things about Java ‘hotspots’ but the whole idea of Java gives me the shivers!

[imrae]

Other languages try their best, but they are just not as fast (yet) as C++ at algorithmic computations.

FORTRAN says "HELLO^M"

[Vokbuz]

It is always possible to create a bridge between C++ interfaces and <your favourite language> implementation, like you need to do with Audio Unit v3 (and partially with v2) if your implementation is not in Objective C.

[aciddose]

Can we get a rapper rapping about wrappers wrapping wrappers wearing a bow tie inside a gift box?

https://youtu.be/XbGs_qK2PQA

[quikquak]

Other languages try their best, but they are just not as fast (yet) as C++ at algorithmic computations.

FORTRAN says "HELLO^M"

I was talking about modern JITs as I later explain.

[Vokbuz]

wrappers wrapping wrappers

"Никогда такого не было и вот опять" (c)

Oh, come on. Most VST2 implementations already uses C to <your favourite language> wrappers. VST2 SDK has C to C++ wrapper. If we put Symbiosis on top of it, here you have "wrappers wrapping wrappers". There is also C to C++ wrapper for AU.

[aciddose]

Of course, that's the thing. The interface is really just the skin and they try to make it out to be the meat and bones; it isn't anything but the surface layer, far less than 1%. The wrappers and in-between interfaces are all the veins and nerves that provide the interface between the meat and bones and the VST/AU/whatever skin. Skin is just another wrapper wrapped around a wrapper wrapped around wrappers wrapping wrappers.

[bibz1st]

Is it me or does the Danish socket look a little surprised