I don't understand this VST 3 thing. Every DAW seems to have embraced VST, but very few of them seem to support VST 3. Like, what went wrong...?

ive never heard of vst3?! :s

--------------------
www.soundcloud.com/skerrick

There's a general description here:

http://www.steinberg.net/en/company/technologies/vst3.html

But Reaper doesn't support it, nor does Sonar, or Ableton...Cubase and Studio One are the only things that I've come across so far that do.

My guess is that it takes more to implement and developers are keeping costs down by not doing so.

--------------------
http://soundcloud.com/richie-royale
http://www.mixcrate.com/richieroyale

Which is a shame, because the benefits to CPU usage and sidechainability are worthy.

--------------------
Composer;
www.ogonline.org

Is the difference mainly side-chaining? I use Sonar, and I can do that with the standard plugins already... What else is there?

--------------------
www.thediplomatz.com

I remember being at the trade show when Steinberg launched VST3. There were several third-party developers there who were extremely cross about it, saying that it offered very little that couldn't already be done with VST2.4, but that because the SDK was completely different, it risked destroying the huge community of freeware and shareware developers.

My guess is that it's a waiting game on both sides at the moment. Few plug-in developers feel the need to port existing products to VST3 while there are not many hosts that support it, and few host developers feel the need to embrace VST3 while there are no VST3-only plug-ins.

VST3 was developed just to annoy the dolts on the Steinberg forum who'd been demanding side-chaining for decades because they couldn't work out how to do it themselves.
Apparently side-chaining is the Gok Wan's underpants in Cubase land.

--------------------
Success is round the corner. It's also round the bend.

Edited by tex (05/02/13 03:52 PM)

Quote tex:

---

VST3 was developed just to annoy the dolts on the Steinberg forum who'd been demanding side-chaining for decades because they couldn't work out how to do it themselves.
Apparently side-chaining is the Gok Wan's underpants in Cubase land.

---

VST3 is very useful. Its extreme laziness that so many companies haven't implemented it yet.

Quote johnny h:

---

VST3 is very useful. Its extreme laziness that so many companies haven't implemented it yet.

---

One of the major benefits is the fact that despite loading a bunch of plug-ins in session, the CPU hit only occurs if the plug-in has to process something.

I suspect that in respects of groundbreaking tech. Steinberg are so far ahead, that other companies make up by giving away softsamplers...

--------------------
C.LYDE
http://soundcloud.com/c-lyde

Quote Sam Inglis:

---

I remember being at the trade show when Steinberg launched VST3. There were several third-party developers there who were extremely cross about it, saying that it offered very little that couldn't already be done with VST2.4, but that because the SDK was completely different, it risked destroying the huge community of freeware and shareware developers.

My guess is that it's a waiting game on both sides at the moment. Few plug-in developers feel the need to port existing products to VST3 while there are not many hosts that support it, and few host developers feel the need to embrace VST3 while there are no VST3-only plug-ins.

---

So why weren't they doing it with VST2.4? There are definite tech limitations with 2.4., the sad part is that none of these developers care about user CPU budget... "let them buy faster PCs" ay-ay?

--------------------
C.LYDE
http://soundcloud.com/c-lyde

Quote johnny h:

---

Quote tex:

---

VST3 was developed just to annoy the dolts on the Steinberg forum who'd been demanding side-chaining for decades because they couldn't work out how to do it themselves.
Apparently side-chaining is the Gok Wan's underpants in Cubase land.

---

VST3 is very useful. Its extreme laziness that so many companies haven't implemented it yet.

---

I'm surprised you didn't find a way to blame it on Windows...

Quote The_Big_Piano_Player:

---

Is the difference mainly side-chaining? I use Sonar, and I can do that with the standard plugins already... What else is there?

---

Well i use FL Studio; on the standard onboard mixer, you get the channel you want sidechained, and on the channel youre sidechaining to - theres a little box on the bottom of every channel and you right click it (when the synth or pad or whatever youre sidechaining is selected) and choose "sidechain to this track" but you can also do it quite easily with 'fruity limiter' as well.. i didnt know it was something that was difficult/a hassle to do..?

--------------------
www.soundcloud.com/skerrick

Quote Skerrick:

---

i didnt know it was something that was difficult/a hassle to do..?

---

These guys are talking about programming - not just utilizing side-chaining in a DAW (which is a doddle in most DAWs, including Cubase itself).

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Can you expand on that Elf...? Is it not possible to code a VST plugin with sidechain support in 2.4...?

From trawling the forums I'm certainly getting the impression that the 2.4 SDK is the one to go with for any budding developers. I was curious about it because I came across a VST3-only plugin the other day (Vocalign Pro 4 running under Studio One...like the look of Studio One actually) although they've discontinued PC support entirely with the latest version. Maybe 'cos no-one bought their VST3.

Quote Skerrick:

---

I use FL Studio

---

Incidentally that's one of the few DAWs with VST3 support.

I can't comment on the programming aspects (it's a lot of years since I sat totally confused for a week on a 'Java for COBOL programmers' course! And I still don't get it...), but *using* side-chaining in Cubase is just a case of clicking one button. Most of the Waves plugs have side-chaining, for example, but I have no idea whether they are VST 2 or 3 under the covers.

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote The Elf:

---

I can't comment on the programming aspects (it's a lot of years since I sat totally confused for a week on a 'Java for COBOL programmers' course! And I still don't get it...)

---

(It's taken me a long time to get my head round OOP after being raised in the procedural school, but it's finally starting to make sense, partly thanks to this. VST programming is still pretty impenetrable so far though - and I have a sneaking suspicion the developers like it that way...)

Ha ha!

It beats me that we had perfectly usable programming laguages that looked like English way, way back.. and then the newer OO languages came along with their impenetrable mess of brackets and short-hand! It made machine code seem friendly!!

Even experienced C and Java programmers never tire of telling me how difficult it is to interpret and de-bug OO code. And this is 'progress'?

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote The Elf:

---

I can't comment on the programming aspects (it's a lot of years since I sat totally confused for a week on a 'Java for COBOL programmers' course! And I still don't get it...), but *using* side-chaining in Cubase is just a case of clicking one button. Most of the Waves plugs have side-chaining, for example, but I have no idea whether they are VST 2 or 3 under the covers.

---

If it's any consolation, Mr Elf, I still program in COBOL for a living, and have done for 25 years. (and yes, I've had 25 years of people telling me how dead COBOL is).

--------------------
www.thediplomatz.com

Quote BJG145:

---

VST programming is still pretty impenetrable so far though - and I have a sneaking suspicion the developers like it that way...

---

If you think VST is bad, you should try looking at DirectX. I always thought that Steinberg invented VST because they were being forced to support the PC and they couldn't get their heads around the existing DirectX standards (which Cakewalk and others had been using for a few years before VST came out).

James.

--------------------
JRP Music - Audio Mastering and Restoration.
http://www.jrpmusic.net

Quote The Elf:

---

I can't comment on the programming aspects (it's a lot of years since I sat totally confused for a week on a 'Java for COBOL programmers' course! And I still don't get it...), but *using* side-chaining in Cubase is just a case of clicking one button. Most of the Waves plugs have side-chaining, for example, but I have no idea whether they are VST 2 or 3 under the covers.

---

Waves works as vst3 under the covers. That very easy sidechain button is nowhere to be found on hosts which do not support vst3!

Does anyone know if ableton 9 supports vst3? It's been far too long and their built in sidechain equipped compressor really isn't good enough.

Quote johnny h:

---

Does anyone know if ableton 9 supports vst3?

---

Nope.

I'm still pretty confused about this sidechaining thing. For instance in the SOS review for Cytomic The Glue it notes that:

Quote:

---

The side-chain EQ is a gentle 6dB/octave filter with a cutoff frequency that can be set from zero to 2000Hz, and it can be applied to the internal signal or an external side-chain signal. When you press the external side-chain button, the plug-in reports one or two extra inputs — depending on whether it is set up for mono or stereo processing — making it possible to use any signal to control the compressor. Unfortunately, the VST3 standard is not yet supported, so the success of setting up an external side-chain very much depends on the DAW host and its ability to route audio in a flexible way.

---

...I guess that means that non-VST3 plugins can be written to support sidechaining in hosts that allow it. Would I be right in thinking that Waves compressors don't allow sidechaining in non-VST3 hosts, even though some other plugin compressors do...? It all seems a bit of a mess...

Quote BJG145:

---

Quote johnny h:

---

Does anyone know if ableton 9 supports vst3?

---

Nope.

I'm still pretty confused about this sidechaining thing. For instance in the SOS review for <a href="/sos/nov10/articles/the-glue.htm" target="_blank">Cytomic The Glue</a> it notes that:

Quote:

---

The side-chain EQ is a gentle 6dB/octave filter with a cutoff frequency that can be set from zero to 2000Hz, and it can be applied to the internal signal or an external side-chain signal. When you press the external side-chain button, the plug-in reports one or two extra inputs — depending on whether it is set up for mono or stereo processing — making it possible to use any signal to control the compressor. Unfortunately, the VST3 standard is not yet supported, so the success of setting up an external side-chain very much depends on the DAW host and its ability to route audio in a flexible way.

---

...I guess that means that non-VST3 plugins can be written to support sidechaining in hosts that allow it. Would I be right in thinking that Waves compressors don't allow sidechaining in non-VST3 hosts, even though some other plugin compressors do...? It all seems a bit of a mess...

---

There are complicated ways to do perform sidechaining with multiple inputs etc, but VST3 makes it very easy. I really don't know why audio programmers are so lazy. Its been years now and the added functionality is very useful.

audio programmers are lazy?
What next? Maybe they could download VST3 code from pirate bay and Mega instead, and share it.

--------------------
~~~ A weasel hath not such a deal of spleen as you are tossed with! www.feline1.co.uk ~~~

Quote The Elf:

---

I can't comment on the programming aspects (it's a lot of years since I sat totally confused for a week on a 'Java for COBOL programmers' course! And I still don't get it...), but *using* side-chaining in Cubase is just a case of clicking one button. Most of the Waves plugs have side-chaining, for example, but I have no idea whether they are VST 2 or 3 under the covers.

---

Or if you don't have a VST3 plugin, creating a new surround track with non-stereo input channels used as sidechain on the source channel and the... BOOM.

--------------------
Composer;
www.ogonline.org

Quote The Elf:

---

Ha ha!

It beats me that we had perfectly usable programming laguages that looked like English way, way back.. and then the newer OO languages came along with their impenetrable mess of brackets and short-hand! It made machine code seem friendly!!

Even experienced C and Java programmers never tire of telling me how difficult it is to interpret and de-bug OO code. And this is 'progress'?

---

I work as a full-time software engineer...

I can tell you that for large projects, object-oriented languages greatly *simplify* the development process. When many developers are working on a broad codebase, structure and planning of code becomes very important (I'm not going to explain how an OO design helps accomplish this as it would be like trying to write a book on WWII in a forum post ).

Also it's interesting that you contrast C and machine language as being "friendly" in comparison to OO languages (btw, Java is OO)....object oriented languages like C#, C++ and Java have syntax greatly borrowed from C and how the languages work from a procedural point of view is very similar.

I don't mean to be contrary; it's just that I work with OO languages nine hours a day (I've also worked with C, assembly language, etc.) so I see the benefits in a commercial environment.

Peter

--------------------
http://www.pfernandes.com

I'm interested in development with Haskell on a biological platform, but I don't think the Steinberg SDK is compatible.

http://www.guardian.co.uk/science/2013/jan/23/shakespeare-sonnets-encoded- dna

Quote Peter Fernandes:

---

I work as a full-time software engineer...

I can tell you that for large projects, object-oriented languages greatly *simplify* the development process.

---

Yep. The stuff you guys do these days is just massively more advanced than the simple programs that were written back in day using the likes of COBOL. In practice, developing such stuff requires sophisticated tools, which, unsurprisingly, have a fairly steep learning curve.

So yes, for someone who wants to write another accounting package - go ahead and write it in COBOL, if you must. If however you want to develop the next Camel Alchemy, then you'll probably find some more advanced tools are in order...

FWIW I found c++ (object oriented) soo much easier than old fashioned c. It made me think - in such a way that accidently introducing a bug was actually quite hard

Which is nice.

Quote The Elf:

---

Even experienced C and Java programmers never tire of telling me how difficult it is to interpret and de-bug OO code. And this is 'progress'?

---

Nonsense - they, or the original programmer's can't be doing it right. OO makes life easier by encapsulating state so that when there's a bug you know what to watch or check and allow you to partition abstraction into meaningful chunks.

By check I mean that it is often possible to check that a class of objects remains internally consistent, but this only works if the interfaces don't allow an object to be transiently inconsistent.

Where it doesn't fit so well is in code that is highly algorithmic, and a certain amount of audio programming (the audio processing kernels specifically) falls into that category.

I haven't written COBOL in donkey's, but its nice to know that having done so in the past, I will always be able to find a job.

I'm sure there are many good things about C and its derivatives, but I stand by what I said about all those horrendous brackets!

--------------------
An Eagle for an Emperor, A Kestrel for a Knave.

Quote The Elf:

---

all those horrendous brackets!

---

Ah - you should learn Perl, where there is no key or symbol on your keyboard that doesn't have some special meaning. It's quite possible to write code that resembles an explosion in a punctuation factory.