What are your thoughts on Symbolic Sounds Kyma 7 in 2019?

[Neon Breath]

I would consider a software only version, the hardware is too much hassle...

That would be a clever move from them, as I'm sure many artists & sound designers would finally dive into Kyma for good and buy the software only without all that cumbersome hardware gear.

While I don’t agree with the view posted earlier that Pacarana is merely an elaborate dongle, it does serve that purpose as well. If Kyma was released as a software only version, many artists and sound designers would finally be able to steal it, and there goes the entire business that Symbolic has built up over decades.

But wouldn’t be more profitable to sell 1000 copies of a Kyma software at let’say just for fun at $350, than selling 25 Kyma systems at $3000?

I don’t disagree with you though, in other words they are prisoner of their own hardware system.

[Tj Shredder]

Jazz Mutant might make more money with less work just with the iPad version of Lemur...; - )
I think porting a grown Application like Kyma from a dedicated DSP to a general purpose CPU would be a nightmare...

[V0RT3X]

Anyone know what kind of processors are in the Paca and Pacarana? They list the system power by Capstones, but I've never heard of this before, is it a custom chip design or are they using something similar to UAD and Access Virus ?

[Neon Breath]

I've read in forums some Kyma users themselves asking Carla and Kurt for a lightweight system of Kyma, for portable/road purposes or quick prototyping & testing sessions without the hassle of the heavyweight gear. So it's something not only the non-Kyma users ask...

I think they would win, for the most part. But it's probably not a model implemented into their philosophy (which is totally fine if that's the way they want to go)

[deastman]

I would consider a software only version, the hardware is too much hassle...

That would be a clever move from them, as I'm sure many artists & sound designers would finally dive into Kyma for good and buy the software only without all that cumbersome hardware gear.

While I don’t agree with the view posted earlier that Pacarana is merely an elaborate dongle, it does serve that purpose as well. If Kyma was released as a software only version, many artists and sound designers would finally be able to steal it, and there goes the entire business that Symbolic has built up over decades.

But wouldn’t be more profitable to sell 1000 copies of a Kyma software at let’say just for fun at $350, than selling 25 Kyma systems at $3000?

I don’t disagree with you though, in other words they are prisoner of their own hardware system.

Sure, if it worked out that way. More likely you’d have tons of people complaining about how it costs too much and there should be an iOS version for $2, and it is too confusing and doesn’t work like other programs and it’s too cpu heavy and there should be a demo version and ilok sucks and challenge/response sucks and serial numbers suck and Synthedit sounds better and they’d pay for it instead of using a crack if they’d only support retina displays.

[justin3am]

I know Justin3am bought a system after the Pacarana was released. I’m hoping he’ll chime in here... it would be good to get the perspective of someone who actually owns one.

Actually I ended up passing my pre-order spot to someone else. I did work on a project with a friend who uses two Pacaranas. That experience kinda left me thinking that Kyma might not be for me.

The thing about Kyma is that you can get stunning results which really would be difficult/impossible to get from other tools. But it's also really difficult and time-intensive to get those results from Kyma. The detailed spectral morphing examples I've heard, weren't made by loading two samples and then moving a 'morph' slider, it was more like working with Alchemy's spectral editor. There are of course generic tools to speed up the process but if you have a specific result in mind, you may need to write the tools you need in smalltalk.

The impression that I got was that I would need to spend a few years with Kyma, to get where I wanted to go. I decided to move in a different direction but every now and then I get the urge to pickup a Pacarana to give it a try. I would probably need to quit my day job if were going to get the most from it though.

[whyterabbyt]

Anyone know what kind of processors are in the Paca and Pacarana? They list the system power by Capstones, but I've never heard of this before, is it a custom chip design or are they using something similar to UAD and Access Virus ?

The older Capybara systems were 56K series (press release indicates the expansion cards had a 56002 on them), and Ive seen indications the Paca and so on are based on later processors in the same family, which would make sense.

[Tj Shredder]

The 56k series of processors had been realy a boost compared to a 68000... That was the time...; - )
Does anybody know what processors are used in the UAD boxes?

[whyterabbyt]

The 56k series of processors had been realy a boost compared to a 68000... That was the time...; - )
Does anybody know what processors are used in the UAD boxes?

SHARCs.

[Tj Shredder]

Yeah SHARc, it was a real boost when PPCs ruled...; - )
I found this discussion from 2015 interesting:
https://www.gearslutz.com/board/music-c ... eback.html
Already 15 years ago I thought that the advancement in speed of general purpose CPUs is faster than that of dedicated DSPs. Now its 4 years later than that discussion...
The time a programmer needs to learn to program the new only two times faster DSP chip would give him a 3 times faster CPU that would allow to code immediately while the speed is still slower. In the end the ease of implementation of the software is more important than the speed of a given technology... The plug-ins that have been made 15 years ago still run on a modern computer with a speed which was completely out of reach back then...

[whyterabbyt]

Yeah SHARc, it was a real boost when PPCs ruled...; - )
I found this discussion from 2015 interesting:
https://www.gearslutz.com/board/music-c ... eback.html
Already 15 years ago I thought that the advancement in speed of general purpose CPUs is faster than that of dedicated DSPs. Now its 4 years later than that discussion...
The time a programmer needs to learn to program the new only two times faster DSP chip would give him a 3 times faster CPU that would allow to code immediately while the speed is still slower. In the end the ease of implementation of the software is more important than the speed of a given technology... The plug-ins that have been made 15 years ago still run on a modern computer with a speed which was completely out of reach back then...

Yeah, I think there was a time up until around a decade or so ago at which going DSP gave a company a few year's lead time before implementation on a general-purpose CPU became affordable. But the rate of DSP processor 'improvment' cycles couldnt keep up with those for CPUs and that time closed until it was only a couple of years but it was enough to keep a 'premium' market afloat, especially with boards that ran multiple parallel DSPs. Its closed further since then, though, and faster, and its predominantly focussed on the CPU now.

What's interesting now in the 'boutique' space is that digital hardware is still going strong, and thriving even, but embedded low-power CPUs (mostly ARMs, but also things like Teensys and Propellors) have more or less replaced embedded DSPs.

[Tj Shredder]

There are boxes running Linux inside which you can fill yourself with Pd patches like the Organelle. This is a more universal version of something like the Kyma system. You get the advantages of a preparable seperate hardware box and the free programmability of a computer together...
Wait another 5 years and that organelle kind of toy is as performant as a Paca...
You could have started to learn it already 20 years ago...
And if you want it smaller than an Organelle, just take a smartphone and run MobMuPlat on it...; - )

[Neon Breath]

Some sound demos
https://soundcloud.com/roland-kuit/sets/kyma-x

I have to admit it, these examples are pretty weak and not very impressive, for sounds created on a $3000 system. I'm sure the artist had fun creating them but sound wise there's nothing mind blowing in those sounds nor anything special in term of characteristics that could not be done with other tools around. I know Kyma is a very capable machine no doubt about that, it can do crazy stuff but I know too that it takes alot, I mean alot of time to achieve them. By experience I know a simple granulation could takes lot of efforts & time to set it up while you could grab Crusher-X and granulate massive grain clouds in no time. FFT processing can be difficult too, although it does sound very good especially the resysnthesis. But nowadays, there are other options so I stay on my position, it's a great tool among many others now. Kyma is a legacy from an another era.

[Neon Breath]

Edit-double. Sorry about that.

[whyterabbyt]

There are boxes running Linux inside which you can fill yourself with Pd patches like the Organelle.

Yeah, Ive got a Qubit Nebulae module which is exactly that (plus various Pis and a Bela as bare systems)

As I say, though, ideally Im waiting for a good graphical frontend for building CSound instruments; CSound has the breadth that I think an 'alt-Kyma' would need. All eyes on whether Cabbage Studio goes lower level...