A month till namm... (speculations)

Kevin Nolan · Post by **Kevin Nolan** » Sat Dec 31, 2016 3:10 pm

GregC wrote:
Other posters have asserted that Korg should have an ' open system ' with their next Kronos. Or that its ' better for Korg' .

I have been unable to persuade anyone to explain the pro's and con's of Korg doing so. To make a business case so to speak.

I think organized discussions are more interesting. We would have to be clear on what ' open platform' is or what 'open system ' is. Not everyone has the same knowledge on terminology.

Agreed. As I've argued for nearly a decade on this forum - in my view it is overlooked that "Open System" likely meant standardisation across various groups / teams in Korg as much as it was ever to mean an open system standard for 3rd party developers.

In my opinion, OASYS and Kronos could have only been possible as we know them today with Korg's attempt for an open system - so whehter ti enabled 3rd part development or not, it enabled OASYS and Kronos to actually happen.

What I mean is - the ubiquity of architecture across various synth engines, the development on a standard intel processor and reduced Linux Kernal all enabled Korg themselves to develop independent modules that coudl all nit together into the integrated system we know today - and to allow for future developments / add ons that did happen (those of us buying the original OASYS has no MOD-7, PolySix, MS20 and many other features to start off with - those were all added on later). But the point being - it was possible for Korg to add them on because of the modular / open nature to how they developed the system.

Something like a Motif or Kurzweill could not ever be so adaptable because they weren't developed in quite the same way.

And with regard to 3rd party development - even if one considers all OASYS and Kronos sales - how many units are we talking about. Perhaps 20 - 30 thousand in total?? That is not a market for 3rd party development. Who in their right mind is going to invest their time into developing a plugin for that size market, when development of a VST or AU plugins reaches perhaps 100 times the users? Kronos as an open, 3rd party development environment, is never going to be.

GregC · Post by **GregC** » Sat Dec 31, 2016 5:06 pm

Kevin Nolan wrote:
GregC wrote:
Other posters have asserted that Korg should have an ' open system ' with their next Kronos. Or that its ' better for Korg' .

I have been unable to persuade anyone to explain the pro's and con's of Korg doing so. To make a business case so to speak.

I think organized discussions are more interesting. We would have to be clear on what ' open platform' is or what 'open system ' is. Not everyone has the same knowledge on terminology.
Agreed. As I've argued for nearly a decade on this forum - in my view it is overlooked that "Open System" likely meant standardisation across various groups / teams in Korg as much as it was ever to mean an open system standard for 3rd party developers.

In my opinion, OASYS and Kronos could have only been possible as we know them today with Korg's attempt for an open system - so whehter ti enabled 3rd part development or not, it enabled OASYS and Kronos to actually happen.

What I mean is - the ubiquity of architecture across various synth engines, the development on a standard intel processor and reduced Linux Kernal all enabled Korg themselves to develop independent modules that coudl all nit together into the integrated system we know today - and to allow for future developments / add ons that did happen (those of us buying the original OASYS has no MOD-7, PolySix, MS20 and many other features to start off with - those were all added on later). But the point being - it was possible for Korg to add them on because of the modular / open nature to how they developed the system.

Something like a Motif or Kurzweill could not ever be so adaptable because they weren't developed in quite the same way.

And with regard to 3rd party development - even if one considers all OASYS and Kronos sales - how many units are we talking about. Perhaps 20 - 30 thousand in total?? That is not a market for 3rd party development. Who in their right mind is going to invest their time into developing a plugin for that size market, when development of a VST or AU plugins reaches perhaps 100 times the users? Kronos as an open, 3rd party development environment, is never going to be.

excellent post, Kevin. Thank you

That 30,000 # is close IMO, and its spread over some yrs for the Oasys and 5 yrs for the Kronos.

I think cost of development and unit selling price is absolutely key. I don't have detailed knowledge of creating a VST for the marketplace but I can hardly believe its a seat of the pants endeavor.

Post by **Sharp** » Sat Dec 31, 2016 8:04 pm

My 2 cents.
All the obstacles above only exist because KORG choose Linux. If they had of gone with an embedded version of Windows and built the OASYS platform upon that, then making the system truly open to support VST's becomes straight forward.

You lock the system down except you give the VST HOST the ability to select and load DLL files. This makes it dead simple for developers to get their VST's running with little to no development cost.

Regards
Sharp.

Bachus · Post by **Bachus** » Sat Dec 31, 2016 8:55 pm

Sharp wrote:My 2 cents.
All the obstacles above only exist because KORG choose Linux. If they had of gone with an embedded version of Windows and built the OASYS platform upon that, then making the system truly open to support VST's becomes straight forward.

You lock the system down except you give the VST HOST the ability to select and load DLL files. This makes it dead simple for developers to get their VST's running with little to no development cost.

Regards
Sharp.

But then it still was a smart choice to go for Linux. Linux can work with real time processing in its Kernel, which is a requirement for extreme low latencies. Windows still doesnt... and if Korg would want to, its yet still possible to add the required DLL's dor certain VST's.. but then i dont think Korg wants or needs 3rd party VSTs to work inside the Kronos..

There is also something like an open system, with support for korg products only, korg could develop more engines based on what they have done for the ipad.. wavestation,arp, M1, electribe drums, stuff from gadget and the volca range.. MOSS.. or just adding the filters from the King Korg.

No, the more i think about it, the more its obvious that VST support aint the way for a Korg workstation.. tough what i would like to see is a Korg based VST wrapper like VIP or NKS that makes it easy to controll a DAW and VSTs directly from a korg DAW.

GregC · Post by **GregC** » Sat Dec 31, 2016 9:03 pm

Sharp wrote:My 2 cents.
All the obstacles above only exist because KORG choose Linux. If they had of gone with an embedded version of Windows and built the OASYS platform upon that, then making the system truly open to support VST's becomes straight forward.

You lock the system down except you give the VST HOST the ability to select and load DLL files. This makes it dead simple for developers to get their VST's running with little to no development cost.

Regards
Sharp.

James, you and others have more knowledge than I do.

I figured Korg is like many other co's that work in a hardware/software environment.

In that the platform choice Korg made, also protects their IP ( Intellectual Property).

Kevin Nolan · Post by **Kevin Nolan** » Sun Jan 01, 2017 2:56 am

Sharp wrote:My 2 cents.
All the obstacles above only exist because KORG choose Linux. If they had of gone with an embedded version of Windows and built the OASYS platform upon that, then making the system truly open to support VST's becomes straight forward.

You lock the system down except you give the VST HOST the ability to select and load DLL files. This makes it dead simple for developers to get their VST's running with little to no development cost.

Regards
Sharp.

Booo - Windows - booo!!!

(I used to work for Sun Microsystems - so it's a natural "booo" for Windows

).

in earnest James - the number of layers involved means that the required latency, massive polyphony, true polyphony, realtime operating system,... would simply not have been possible on any processor available at the time.

Afterall - I challenge you to run about 100 voices of true polyphonic VST/AU plugins such as Arturia's CS80V or NI Monark on even a quad iCore7 today. I tryly think people are grossly undervaluing the job Korg did to get 100+ voices running along side effects, karma, ... (and I genuinely believe Korg would have considered if not tested that possibility too).

Wishing you all a very happy new year

.

SeedyLee · Post by **SeedyLee** » Sun Jan 01, 2017 4:32 am

Sharp wrote:My 2 cents.
All the obstacles above only exist because KORG choose Linux. If they had of gone with an embedded version of Windows and built the OASYS platform upon that, then making the system truly open to support VST's becomes straight forward.

I disagree.

Not that there's anything wrong with Windows Embedded/Mobile/CE (or whatever they call it these days, there's been so many iteration sof their embedded platform it's hard to keep up), but many VST have complex dependencies that would negate the benefits of running an embedded version of Windows in the first place. Think iLok keys, Internet authorisation, custom UI toolkits, custom installers etc. If Korg went Windows Embedded, they'd still need to support all the runtime libraries etc that support these functions - before you know it, you're running full-blown Windows.

Alternatively, Korg could supply an SDK to allow VSTs orginally for Windows/Mac to be cross-compiled for the Kronos whilst also providing access to the inbuilt cryptographic module for authorisation.

What Korg have done with Linux n the Kronos is really very impressive, and wouldn't be much more challenging with Windows Embedded. The audio engine on the Kronos actually runs as part of the kernel - combined with the RTOS Kernel extensions, I have measued the Kronos' round-trip audio latency (in and out) at under 2ms, which is incredibly impressive! From memory, the latency from a received MIDI message is around the 1.4ms mark.

Whilst a similar approach could have been taken with Windows, it would have been much harder for them to achieve without access to source code etc.

Bachus · Post by **Bachus** » Sun Jan 01, 2017 10:10 am

If you want VST´s in your instrument thee is currently a single choice available...

Here a little video of the interface...

https://www.youtube.com/watch?v=dyBuZfY61qg

Post by **Sharp** » Sun Jan 01, 2017 12:45 pm

GregC wrote:
James, you and others have more knowledge than I do.

I figured Korg is like many other co's that work in a hardware/software environment.

In that the platform choice Korg made, also protects their IP ( Intellectual Property).

Personally, I'd guess it's a cost factor. Linux is free.

It's unlike to be any IP related thing given that pretty much the entire world runs on Windows Embedded. Even your ATM machine, or your local shop barcode scanner. It's everywhere.

Regards
Sharp.

Post by **Sharp** » Sun Jan 01, 2017 1:15 pm

Kevin Nolan wrote: Booo - Windows - booo!!!

(I used to work for Sun Microsystems - so it's a natural "booo" for Windows )..

How long ago are we talking here? DOS days? Windows 3.1 ?

in earnest James - the number of layers involved means that the required latency, massive polyphony, true polyphony, realtime operating system,... would simply not have been possible on any processor available at the time.

The KRONOS itself (GEN1) was a rather low powered machine. Same too for the OASYS. At the time of their release there were CPU's available many times faster.

Afterall - I challenge you to run about 100 voices of true polyphonic VST/AU plugins such as Arturia's CS80V or NI Monark on even a quad iCore7 today. I tryly think people are grossly undervaluing the job Korg did to get 100+ voices running along side effects, karma, ... (and I genuinely believe Korg would have considered if not tested that possibility too).

There are many great VST's that require very little CPU. Poor performing VST's are not the fault of the OS. That's the developers fault.

Honestly though, nobody seems to care. Fact is Workstations are on the decline due to the DAW and VST's running Windows and OSX.

To sum it all up.....
Better to go with the flow than swim against the huge tide.

Cheers
Sharp.

Post by **Sharp** » Sun Jan 01, 2017 1:50 pm

Not that there's anything wrong with Windows Embedded/Mobile/CE (or whatever they call it these days, there's been so many iteration sof their embedded platform it's hard to keep up), but many VST have complex dependencies that would negate the benefits of running an embedded version of Windows in the first place. Think iLok keys, Internet authorisation, custom UI toolkits, custom installers etc. If Korg went Windows Embedded, they'd still need to support all the runtime libraries etc that support these functions - before you know it, you're running full-blown Windows.

I'm fine with all that or whatever way it would pan out. Given that Wersi have no problem doing this, I'm sure KORG can too.

Alternatively, Korg could supply an SDK to allow VSTs orginally for Windows/Mac to be cross-compiled for the Kronos whilst also providing access to the inbuilt cryptographic module for authorisation

True, or they could do it themselves like they already do for any Third Party EXS developer. Sort of like a VST Wrapper.

What Korg have done with Linux n the Kronos is really very impressive, and wouldn't be much more challenging with Windows Embedded. The audio engine on the Kronos actually runs as part of the kernel - combined with the RTOS Kernel extensions, I have measued the Kronos' round-trip audio latency (in and out) at under 2ms, which is incredibly impressive! From memory, the latency from a received MIDI message is around the 1.4ms mark.

Whilst a similar approach could have been taken with Windows, it would have been much harder for them to achieve without access to source code etc

Yes, it's a fantastic achievement. No doubting that. But as above, if Wersi can do it, KORG can too. I kind of see this as the direction workstations will have to take in order to survive in the future.

Regards
Sharp.

Post by **Sharp** » Sun Jan 01, 2017 2:01 pm

Bachus wrote:If you want VST´s in your instrument thee is currently a single choice available...

Here a little video of the interface...

https://www.youtube.com/watch?v=dyBuZfY61qg

That's a very good example of what I'd like to see KORG doing, only in their style.

Wersi only lack class when it comes to design.

Regards
Sharp.

Bachus · Post by **Bachus** » Sun Jan 01, 2017 4:22 pm

Sharp wrote:
Bachus wrote:If you want VST´s in your instrument thee is currently a single choice available...

Here a little video of the interface...

https://www.youtube.com/watch?v=dyBuZfY61qg
That's a very good example of what I'd like to see KORG doing, only in their style.

Wersi only lack class when it comes to design.

Regards
Sharp.

well, their design dates back to their Organ roots, which clearly is something you dislike...

What worries me more about the OAX is that it lacks certain features that i deem very important in an arranger key..
- The standard styles lack hmpfff..
- There are no multi pads..
- There are no registration memories
- etc etc etc
All in all the Arranger part is very very basic.. which for an arranger thats two times as expensive as the market leading Tyros5 is unacceptable despite the perfect VST sound...

spaceman3 · Post by **spaceman3** » Sun Jan 01, 2017 6:41 pm

To everyone who has commented on this topic.
I have heard some informative things.
You guys are too smart.
This forum has taught me much.
I hope KORG reads things on this forum.
Even KORG could get some good info, and ideals from this forum.

Post by **Sharp** » Sun Jan 01, 2017 7:07 pm

well, their design dates back to their Organ roots, which clearly is something you dislike

Yeah. I wouldn't be a fan of Organ designs at all. I grew up in the 80's so for me it was all about Synths and Samplers. The Organ for me as a kid was something you only heard during Mass in a Church.

What worries me more about the OAX is that it lacks certain features that i deem very important in an arranger key

Given it's price tag, that's quite disappointing. That being said, if it were a KORG, can you imagine the fun programmers on this forum would get up to. I bet there would be some exciting third party apps developed. I'd personally build a VST for streaming samples that loads sounds on demand in order to avoid the need for preloading everything into RAM, and running out of RAM.

Regards
Sharp.