KVR Audio

zvenx · Post by **zvenx** » Wed Jan 31, 2018 2:32 pm

EvilDragon wrote:Yes, which makes it even crazier that Steinberg wouldn't want to adopt that way of doing sidechaining even with VST2. No, they had to tout VST3 instead.

unless of course they considered it a technical work around.
Maybe they considered vst3 a technically better solution, or the internet knows better again?
rsp

EvilDragon · Post by **EvilDragon** » Wed Jan 31, 2018 2:33 pm

Many developers disagree that VST3 is technically better conceived compared to VST2...

zvenx · Post by **zvenx** » Wed Jan 31, 2018 2:35 pm

EvilDragon wrote:Many developers disagree that VST3 is technically better conceived compared to VST2...

And why are they necessarily right?
and don't they have a motivation to do so? (no need to recode in/for vst3 format)

rsp

EvilDragon · Post by **EvilDragon** » Wed Jan 31, 2018 2:38 pm

Why would Steinberg be necessarily right?

They are no saints. There's a word that perfectly explains VST3, and any developer can agree: OVER-ENGINEERED.

LawrenceF · Post by **LawrenceF** » Wed Jan 31, 2018 2:56 pm

They're "right" because they own the spec and can do whatever they want with it, regardless of what anyone else thinks about that.

We really need to get over it. People keep saying it could be better, and I don't doubt that for a micro second, but I live in the real world where we don't get to tell other people what to do with their own intellectual property, we just take it as it is, make use of it, or create something better.

The critiques are fair imo, but it won't change anything, it will still be exactly what it is.

Romantique Tp · Post by **Romantique Tp** » Wed Jan 31, 2018 3:00 pm

imrae wrote:Developers are expecting Cubase to drop VST2 support one day and are trying to be ready for that.

People have been repeating this nonsense ever since Cubase dropped 32 bit plugin support which is not comparable to supporting VST2 at all.

VST3 is simply more future proof and will continue to be improved by Steinberg, and there are several other benefits not mentioned in this thread.

32 bit plugin support was removed because it caused crashes and instability. It was always a temporary solution, and it was there just so you could import old projects and change the plugins to new 64 bit alternatives (I think this was even stated in their documentation at some point). You were never supposed to work on big projects while using 32 bit plugins with the internal bit bridge, they wanted to help you move on.

Most of us had already completely moved on to 64 bit plugins or used 32 bit plugins with jBridge (which works like a sandbox). This forced the less tech savvy users who were complaining about Cubase being unstable to do the same.

VST2 support is nothing like this. Not only are VST2 plugins still widely used, VST2 support is tightly integrated into the DAW and is very stable. There's no technical or financial benefit to removing it

-

EvilDragon wrote:which didn't catch on too well even after all these years

Almost every effect plugin released recently supports VST3, and VST3 instruments are starting to show up too.

VST3 isn't only about side-chaining. Steinberg didn't even advertise that as an advantage over VST2 when they announced the VST3 SDK:
https://web.archive.org/web/20080129212 ... 73ab0.html

imrae · Post by **imrae** » Wed Jan 31, 2018 6:50 pm

Romantique Tp wrote:
imrae wrote:Developers are expecting Cubase to drop VST2 support one day and are trying to be ready for that.
People have been repeating this nonsense ever since Cubase dropped 32 bit plugin support which is not comparable to supporting VST2 at all.

VST3 is simply more future proof and...

Ok then, what threat is it that developers need to "future proof" against?

xphen0m · Post by **xphen0m** » Wed Jan 31, 2018 7:19 pm

I almost never use VST3. If I'm having a problem with a plugin, seriously, 9 times out of 10, it's solved by using the VST2 version.

So now? I don't even install VST3 plugins, always VST2.

Cyoon · Post by **Cyoon** » Wed Jan 31, 2018 7:32 pm

i'd like vst4 please

Or a new DX

swatwork · Post by **swatwork** » Wed Jan 31, 2018 8:24 pm

Romantique Tp wrote:Almost every effect plugin released recently supports VST3

This would suggest otherwise: http://www.kvraudio.com/plugins/vst-plu ... cts/newest

aciddose · Post by **aciddose** » Wed Jan 31, 2018 8:54 pm

Grizzellda wrote:Hey out there,

So help me understand something here, if I am not getting something about this - well I would surely re-align my thinking...but, hey waddaya guys think?

The VST 3 "standard", it clutters up my plug list...what is it for actually, any ideas?

I found VST 2.4 to be quite reasonable...am I just an old curmudgeon?

There is only one reason the authors developed VST3: it is not possible to utilize the interface without a license.

VST2 is a C interface implemented via a data structure containing various fixed offsets to various data as well as a small number of core function pointers (abstract message, process, set parameter, get parameter) which were a part of VST1. Most functionality is routed via the abstract message procedure rather than expanding the core structure and many data members are obsolete.

It is possible to use open-source headers (see vestige) to implement VST2 either with a GPL license or no license at all (negotiable with the author.) This is because VST2 is an abstract interface. It does not fall under copyright protection due to the fact it is merely a description of a fact (the way plug-ins communicate) much like telephone numbers, names and addresses or the black/white/hot/neutral/earth mains wiring used nearly worldwide.

VST3 is defined by its implementation and does not provide an abstract interface. This means you must utilize the "official" implementation source-code, which means you must agree to the author's license.

This is the only reason.

chk071 · Post by **chk071** » Wed Jan 31, 2018 8:57 pm

You mean VST3 only opposed to VST2 and VST3. Not sure what the OP meant in this case. As far as i'm concerned, many plugins come in both formats these days.

aciddose · Post by **aciddose** » Wed Jan 31, 2018 9:19 pm

Hanlon's razor: "Never attribute to malice that which is adequately explained by stupidity."

It is also entirely possible they made the completely insane decision to write bug-prone VST3 code dependent upon the C++ ABI without making any of the core improvements that should have been made to VST2 as an abstract C interface; where improving VST2 would have taken at most a few more lines of source-code they decided to write tens of completely new source files for VST3 at the same time as increasing the workload of developers required to utilize the interface rather than decreasing it.

I've tried programming after a pint too. Not recommended!

1 pint (568ml) stout @ 9.9%
Estimated BAC: 0.076
Ballmer Peak BAC: 0.1337

Grizzellda · Post by **Grizzellda** » Wed Jan 31, 2018 11:43 pm

EvilDragon wrote:Many developers disagree that VST3 is technically better conceived compared to VST2...

Thank You.

LawrenceF wrote:They're "right" because they own the spec and can do whatever they want with it, regardless of what anyone else thinks about that.

Thank You.

aciddose wrote:There is only one reason the authors developed VST3: it is not possible to utilize the interface without a license.

VST2 is a C interface implemented via a data structure containing various fixed offsets to various data as well as a small number of core function pointers (abstract message, process, set parameter, get parameter) which were a part of VST1. Most functionality is routed via the abstract message procedure rather than expanding the core structure and many data members are obsolete.

It is possible to use open-source headers ([url=<span class="skimlinks-unlinked">https://github.com/falkTX/Carla/blob/ma ... tx.h</span>]see vestige[/url]) to implement VST2 either with a GPL license or no license at all (negotiable with the author.) This is because VST2 is an abstract interface. It does not fall under copyright protection due to the fact it is merely a description of a fact (the way plug-ins communicate) much like telephone numbers, names and addresses or the black/white/hot/neutral/earth mains wiring used nearly worldwide.

VST3 is defined by its implementation and does not provide an abstract interface. This means you must utilize the "official" implementation source-code, which means you must agree to the author's license.

This is the only reason.

Oh, come on aciddose, why don'tcha tell us what you really think!

Hey, thanks guys, thanks alot, some quality perspective here.

jancivil · Post by **jancivil** » Thu Feb 01, 2018 12:48 am

VST 3.5 has 'Note Expression' where you can apply various things MIDI to single notes within a part or event/clip which contains more than one note. Has to be a VST 3 instrument and I have not used any so far. But it seems compelling. I almost purchased Cubase 6.5 to get it.

I connect to the VST 3 version of VE Pro because this provides up to 48 MIDI ports. Rather than one which is what people that don't use VE Pro (or Gigastudio back in the day) are looking at. For orchestral composers this is pretty big.

So to reduce it to sidechaining, otherwise of no use is just another one of these statements at teh KVRs where someone doesn't mind not knowing before they assert their disposition in the form of a fact.

So, VST 3...