Reaktor like Rack extension designer

[avasopht]

As the title says, a graphical rack extension for the code averse.

Technically it's possible for any developer to create this technology. Uhe has something like this for their plugins.

[selig]

First, it can't be as a rack extension any time soon. SDK, NDA, ETC…

And yet there are so many that get into Reaktor and quickly find how dang difficult it can be. Like with Reaktor, there will be a very small group who actually build anything (my guess). You, me, and Exode (who all have made very similar suggestions over the years - imagine that!) are probably among the few who would truly appreciate this move, and I'm probably even less interested since I began developing REs. 

Now why would I be at all negative about an idea that I've also expressed great interest in seeing? Because you're basically suggesting putting an entire app (Reaktor) into a single device in Reason. That's a LOT of programming hours there for something that again will only appeal to a small minority of the user base. And can you imagine the support issues for such a complex (by Reason standards) device?

In the end I have concluded that, especially since the introduction of REs, there simply aren't the resources available at Props Central for such a huge undertaking - and they are clearly looking in other directions at present so this idea doesn't even come close to fitting their current marketing directions AFAICT. 

I've also concluded that a souped up Combinator is probably the closest (and possibly most useful to more folks) we'll come to this vision any time soon. 

My 2 cents, FWIW.

[Exowildebeest]

As I understood it, they've already made a step towards this with IDT.

It would be a larger step, but not a totally impossible or totally unlikely one, for Propellerheads to develop a Reaktor-like interface with elements that integrate and are derived from the RE SDK, ultimately giving (relatively) casual users the opportunity to make their own devices in a way similar to Reaktor.

However... Propellerheads currently has the hipsters prioritized, at least in their marketing, not the nerds.

So while I think it's not out of the question, I think it's unlikely, unless the now all-powerful marketing department decides that there IS money to be made among nerds

[Exowildebeest]

@Selig: no there's no capacity at Propellerheads to do this... Because they're all working on Discover

Indeed a Combinator 2.0 would be more realistic. Hoping for that in Reason 9!

[avasopht]

Well I meant more in the sense of providing something like IDT. Available only to developers, spitting out C++ code based on your design which can then be compiled by the SDK.

Possible for any developer willing to develop it, and like IDT you would probably want to take a share of the sales, or maybe just an up front fee of sorts.

But you are right, a more advanced combinator, perhaps with a per-voice routing feature would suffice. In fact it most certainly would

[selig]

Well I meant more in the sense of providing something like IDT. Available only to developers, spitting out C++ code based on your design which can then be compiled by the SDK.

Possible for any developer willing to develop it, and like IDT you would probably want to take a share of the sales, or maybe just an up front fee of sorts.

But you are right, a more advanced combinator, perhaps with a per-voice routing feature would suffice. In fact it most certainly would

What then would be difference between your idea and IDT then? Or to put it another way, what do you want from IDT that it's not already able to do? 

IDT is a lot closer to the RE SDK, and as such was simple to develop. This idea of yours couldn't be an RE (yet, at least), so it would have to be a native device like the EMI or similar, and developed 100% from scratch. And I think they are in the process of getting away from introducing new native devices, especially big ones like this - but that's pure speculation on my part, and the truth could be 180° different.

[avasopht]

What I mean is, the Reaktor like editor would be a user application that spits out the ensemble as c++ code that is then compiled into a regular rack extension.

Compared to the IDT, this allows the creation of custom DSP algorithms featuring per voice processing without the need to learn scripting or coding.


It's possible to build this right now and as mentioned, Uhe already have their own in house tool that does something quite similar (though is imagine has no graphical editor).

[selig]

What I mean is, the Reaktor like editor would be a user application that spits out the ensemble as c++ code that is then compiled into a regular rack extension. Compared to the IDT, this allows the creation of custom DSP algorithms featuring per voice processing without the need to learn scripting or coding. It's possible to build this right now and as mentioned, Uhe already have their own in house tool that does something quite similar (though is imagine has no graphical editor).

It's not possible to build this as an RE, unless there's something you know that I don't know (or that I SHOULD know, but don't!). 

Besides, you still need to learn a LOT of scripting and coding with Reaktor, just not the same languages that REs use - it's quite an intense learning curve IMO. The difference is quite minimal between IDT and Reaktor in many ways, with the scripting being at a similar level as Core in Reaktor. Building custom interfaces in Reaktor is in some ways even more complex than IDT!

My points are two fold. The first is that many folks get into Reaktor and never end up finishing anything useful (just like many Reason users never use CV or build a complex Combinator). The second is that the Props just invested a lot in REs and then IDT - it's unlikely they will now do almost the same thing a THIRD time over to develop yet one more way to create REs for Reason. 

I just don't see the market demand nor the return on investment, as much as I'd like to have that feature myself (mainly for fun!). But I'm also a hard core Reaktor user and build complex Combinators quite often, so I'm not in the mainstream here for sure!

[avasopht]

Selig, you're probably misunderstanding what I'm saying. What I'm saying is absolutely possible, there's not one part that is an issue.

An application as the editor that spits out SDK compatible code.

The RE would not be editable. It would behave just like any other rack extension because all the Reaktor like exporter does is turn the ensemble into raw code.

As to whether many would use it, I would imagine users are more likely to go the graphical route than learning c++, and this type of tool is possible for any developer to create right now.

This is done all the time, such as c++ to JavaScript converters, or flo (similar to Reaktor but geared to traditional coding problems) to c++.

I think even Faust does c++ exports.

[Lizard]

I think I know what avasopht is is saying.  Not an Re itself but an application that creates code for compiling but it would be much more visual like Reaktor is.  Simply dragging an stereo pair or audio creates the code for you.  Making the connections and routings between componenents then creates the code on how they are chained and so on.  If you were designing a synth you could drag in various filters and LFO's and so on.  Is this what you are thinking? 

[avasopht]

Yes that's exactly what I meant.

[avasopht]

Wait I've just realised what went wrong, probably because I wrote this on my phone the OP says "graphical rack extension", should have said, " graphical rack extension designer ".

They would still have to design the GUI using the SDK provided system.

[eusti]

I think I know what avasopht is is saying.  Not an Re itself but an application that creates code for compiling but it would be much more visual like Reaktor is.  Simply dragging an stereo pair or audio creates the code for you.  Making the connections and routings between componenents then creates the code on how they are chained and so on.  If you were designing a synth you could drag in various filters and LFO's and so on.  Is this what you are thinking? 

So, like an RE-building app?

D.

[avasopht]

Yes like an RE building app for non coders.

[selig]

Selig, you're probably misunderstanding what I'm saying. What I'm saying is absolutely possible, there's not one part that is an issue. An application as the editor that spits out SDK compatible code. The RE would not be editable. It would behave just like any other rack extension because all the Reaktor like exporter does is turn the ensemble into raw code. As to whether many would use it, I would imagine users are more likely to go the graphical route than learning c++, and this type of tool is possible for any developer to create right now. This is done all the time, such as c++ to JavaScript converters, or flo (similar to Reaktor but geared to traditional coding problems) to c++. I think even Faust does c++ exports.

Yes I did misunderstand, sort about that. So it's even more complex than I originally thought. An entire new app built from the ground up to design REs (the third!). And this one would be even MORE complex than the current development tools. This version is more like what I suggested years ago in the exuberance of my youth (!) because it's based on a separate app with a drag/drop mentality for non-coders.

That being said, my comments about marketing, support,  and resources apply even more then, and the number of folks that would actually use such a tool drop further IMO!

Again, love the idea, but don't think there's a chance this would ever happen. I'm not even so sure about an advanced Combinator happening any time soon! 

[Lizard]

I have always thought it would be great for the "non-coding" types like myself a tool that like say Microsoft's Visual Products allowed you to create Visual Rack devices.  Have an empty device in a window and just drag stuff to it.  Like Reaktor.  I dreamt of this well before Re was a reality for expanding the rack.  But like Selig says it is a huge undertaking and if were to happen i'd kind of lean on thinking it would already be happening.  For this to be hugely acceptable I think it would have to be designed so that programmers were allowed to expand it with their own types of plug-ins for you to get a large variety of the things it would be capable of.  I do love the idea but for now.... I'm keeping it Real. 

[avasopht]

I hear you on the smaller market and the complexity still of anyone's got a year of coding to spare, it could be a worthwhile project for the intrepid coder.

[selig]

I hear you on the smaller market and the complexity still of anyone's got a year of coding to spare, it could be a worthwhile project for the intrepid coder.

Sorry about sounding so much like a Negative Nellie in your thread! Like I've mention, I'm personally all for the idea, or at least I WAS more into it before I dove into RE development. Just wanted to be sure you knew I wasn't trying to be an ass here (sometimes I don't have to try - it just comes naturally!).

[avasopht]

Sorry about sounding so much like a Negative Nellie in your thread! Like I've mention, I'm personally all for the idea, or at least I WAS more into it before I dove into RE development. Just wanted to be sure you knew I wasn't trying to be an ass here (sometimes I don't have to try - it just comes naturally!).

It's okay, it was clear there was a bit of miscommunication going on

[plaamook]

I've posted this before but I like posting about this so here I go again...

I think that a good solution to this is small, quality, possibly half rack devices that can be chucked into Combi 2.0. Like a half rack sampler that really just plays a sample and pitches it, ADSR, sample start, done. You want an LFO, add one. Etc. Blanks of knobs, sliders, an XY pad, whatever. The smaller the better. There have already been movements in this direction with some RE's. Of course there are things that would be needed to fill the line up that may not be possible yet. But to my mind the power of Reason gets kicked in the nuts by all sorts of dumb limitations. Like Thor's formant filter. Why not just a formant filter? Or knobs that just don't go that far. Or knobs that just can't be automated. But we have the back of the rack! It's like they practically built a Reaktor-esque program and then thought "who the hell would want to do anything really crazy with it? They'll just want refills and presets. They'll want to make pop music." and that was that.

I think the question I have is what's missing in Reason as it is that prevents it from functioning more or less like Reaktor or MAx and how do we get it in there? I don't get deep into Reaktor cause I'm lazy so I can't say really, I just see how it works, then lazy me spends probably more time trying to get Reason to do what it can't!

Never mind...

[buddard]

You might want to look into Gen for Max MSP. I haven't tried it myself, but apparently it can generate AU/VST code, which you could probably port to RE with a reasonable effort. You would have to redo all the plumbing and UI, of course, but at least the DSP code could be generated...

Another interesting piece of technology is libpd for Pure Data. It's an embeddable version of Pure Data (which is the open source basis for Max MSP) that lets you to run pd patches on iOS and Android, for example. Libpd support in the RE SDK would be really interesting, I think! Obviously it would have to add some limitations -- pd patches would probably have to be built into the RE rather than provided by the end user, for example.

[Flandersh]

I have for long thought that a few additional modules in Reason could make it far more useful when programming combinators from a core level. Modules such as a unit delay, basical mathematical operators for all kind of signals and possibly some array modules. This can possibly be made by a RE programmer and sold as a RE, and will with all the other kinds of module existing in Reason make it possible to use the combinators in a broader way.

Another alternative is as buddard says to use the Gen for Max/MSP for creation of code and so port the code to the RE format. And maybe it is even a future possibility to have the RE format added as an export option in Gen.

[avasopht]

I have for long thought that a few additional modules in Reason could make it far more useful when programming combinators from a core level. Modules such as a unit delay, basical mathematical operators for all kind of signals and possibly some array modules. This can possibly be made by a RE programmer and sold as a RE, and will with all the other kinds of module existing in Reason make it possible to use the combinators in a broader way.

Another alternative is as buddard says to use the Gen for Max/MSP for creation of code and so port the code to the RE format. And maybe it is even a future possibility to have the RE format added as an export option in Gen.

Don't we already have mathematical operators in the shop? If not, what operators specifically?

[Flandersh]

I have for long thought that a few additional modules in Reason could make it far more useful when programming combinators from a core level. Modules such as a unit delay, basical mathematical operators for all kind of signals and possibly some array modules. This can possibly be made by a RE programmer and sold as a RE, and will with all the other kinds of module existing in Reason make it possible to use the combinators in a broader way.

Another alternative is as buddard says to use the Gen for Max/MSP for creation of code and so port the code to the RE format. And maybe it is even a future possibility to have the RE format added as an export option in Gen.

Don't we already have mathematical operators in the shop? If not, what operators specifically?

That is possible, I was just sketching what is really needed to be on pair with Reaktor and that is modules similar to what Reaktor Core has or Gen for Max/MSP has.

[selig]

Just to keep the conversation going, and to show one idea I had a while back that was along these lines, I give you the ReRack: a Thor like device that can run 500 series sized modules in a modular synth or FX environment. This idea is combined with a super-combinator that allows front panel design. The ReRack does the DSP, the Combi does the GUIs. Shown here is the ReRack device (of which multiples can be combined) showing modules taken from Reason devices) (not shown, the programmer or "modulation matrix".):