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

[Tj Shredder]

Quality of coding does not help on a lame CPU... My Macbbok Air is faster and could run the same code...
I am not sure if specialized DSP chips are an advatage nowadays, but the effort which are put into graphic chips might be something. The market for graphic and general purpose chips is million times bigger than that for DSP chips. That translates also to money for developement...

[Muied Lumens]

Interesting thread. I have been contemplating on selling my Paca for over a year, but I keep coming back to it occationally (very occationally!) because it still does things that's hard to find elsewhere. Someone mentioned the granular engine - on paper, it's simpler than many VST's (but that's only before you add Capytalk scripting - every parameter can be a little program containing sequencers and non linear chaotic functions etc etc). But even in its simplicity it sounds very smooth and I have used it to realistically simulate that gentle noise you get on a quiet night far from civilisation. Also you can do audio rate FM on a sample, with another sample, or a live input... And so on. I think you get the point.

To me, it always sounds great. It's easier to use than Max and Reaktor, at least when you cobble something together quickly - because it's more forgiving (and higher level, arguably) than both of those, even when you know what you're doing. I think Kyma is fairly quick compared to other graphical software modulars, provided you know it well. But what isn't quick when you know it well? Also - it depends on what you want to do - if you want 600 delays where the times are mutually prime, you are only going to do this using scripting. Imagine the headache doing this graphically.

In the end a casual listener isn't going to be interested in how a track was made, and that's perhaps one reason why I'm losing interest. I have been improvising tracks out using more hardware lately. I haven't switched on Kyma for a couple of months, because I have to run it on the Mac, simply because the firewire card on my windows machine isn't working properly. It's a drag compared to just firing up Plogue Bidule and getting the job done. Then again I don't really need the money and I feel slightly attached to it anyhow... I really don't know.

Bottom line: it's still a sound designer's dream, provided you get along with its quirky interface.

[Ah_Dziz]

The granular stuff in the Kyma system still sounds far better and more flexible than any plugins i’ve Diddled around with (including reaktor and crusher). I’m not sure how much work it takes to make all these fun sounds, and the time might not be worth it. I’m far less impressed with the spectral manipulations these days now that my cpu doesn’t try to explode when performing complex fft processing.

[Neon Breath]

The granular stuff in the Kama sytem still sounds far better and more flexible than any plugins i’ve Diddled around with (including reaktor and crusher).

Any examples showing that? Would love to hear since I didn’t hear anything granular from Kyma that truly impressed me yet. With so many granular tools available now, I have a hard time believing it’s still far superior today. 10 years ago, maybe. But still?

[Muied Lumens]

Any examples showing that? Would love to hear since I didn’t hear anything granular from Kyma that truly impressed me yet. With so many granular tools available now, I have a hard time believing it’s still far superior today. 10 years ago, maybe. But still?

If you aren't convinced, I doubt another demo will convince you. And that's okay. I don't think anything is far superior these days. Software has come a long way and you can achieve almost anything you set your mind to with cheap or free software. Kyma isn't a be-all, end-all program, but if you're one of those nerdy types who like to pursue academic experiments and 'what if' scenarios, Kyma is likely to accommodate you. Just like Max or Supercollider will, once you know it.

I can tell you that the granular engine supplied with Reaktor dissapointed me. If I were using Reaktor I would have to roll my own, which would take time. You can think of the money you spend as something that will save you some time and effort. I bought my Paca 10 years ago, and I don't regret it. These days, it's harder to justify buying a new system. Also, I wonder if they have a new incarnation coming?

And for full disclosure, IMO one of the weak areas of Kyma is delay based effects. For that, get an Eventide H9000 instead. The stock delay doesn't have adjustable smoothing (it's a direct function of the delay time) which has frustrated me from time to time. It's not strong on midi either - but as with all modular/flexible systems, there are always workarounds.

[Ah_Dziz]

There were some demos posted earlier in this thread. I thought they showed quite nicely how well Kyma does with very dense grain clouds. Especially ones where the grains are coming out at Audio rate (which is not a place where any of my granular vsts shine). It’s not something I’d ever spend the money on, but it does seem very good in that regard. Always has.
I would imagine that with one of these boxes and a stack of good control gear, you could build quite the cool live rig, but outside of that I don’t really see the appeal when the price is taken into account.

[sonicnaan]

This is a wonderful thread. I studied music applications of kyma back in 200 and saved for a system, which I bought thwne the first Paca came out. Firstly, I gotta say that apart from the system itself, you're accessiing some of the most interesting and best sonic minds of the century with Kurt, Carla, Pete J, Vogel all keen to really help out and talk audio processing and ideas. Quite often though I did feel that the "music" of most Kyma (and modular) users tended towards academic noise.

The community was truly inspiring and wonderful to be a part of… but yes, I left Kyma in 2016, as in the end I was using this wonderful and expensive piece of gear for very basic use and for me, the physical patching, learning how to economically programm was actually getting in the way of my writing - and my granular synth patches would always run out of DSP (likely my fault for not programming economically).

What I used Kyma for in 2016
My run with Kyma in 2014 - 2016 was an amazing tool to work with - once the programming had been done, being able to perform with NO latency was incredible - re pitching my vocalist in real time, we quickly learned how to get amazing effects or natural effects.

By far my favourite thing to do was (nutshell) a spectral-analysis, then re-pitch the sine waves via Soundplane or Ipad OSC/MPE - Like a live MPE harmoniser with vocal or instrument input. I'd use 4 fingers and completely and naturally change or harmonise entire melodies by performing them with the soundplane.

My singer quickly found that you get a fantastic effect if he sung falsetto while I pushed the low octaves. You can hear us using KYMA to push the boundaries of processing vocals at the end of the film, "The Drycleaner (Chris Carr,2019)". And I'm pretty sure I felt Kyma's fingerprints all over the granular stuff in Transformers and Lucy (not to mention the voice of WallE of course).

I have not found anything that does that, that well and in such an immediate, almost viscose way since. Kyma definitely still has a place in sound design and music, but for me, it was a little too "academic" for what I was doing - but I'd love to have it back! Absolutely.

[soundmodel]

I don't want to play down someone's product, but my personal view is that the Kyma hardware is a dongle. This is probably not announced somewhere, but my perception comes from:

• Kyma is a business, businesses sometimes focus on making money
• The business is to sell an audio processor product, which appears unique and exclusive
• Using one's own hardware allows one to:
  • Decide over access to the software and disallow the usage of competing offerings
• It's not evident whether and when the hardware produces significant benefits over general-purpose computers and native plug-ins. The Capybara 320 is certainly much slower than modern general-purpose computers and the amount of memory is ridiculous by today's standards
• The hardware, being electronics, will eventually stop functioning, and when it does it means buying yet again from the same vendor to continue running patches
  Further:
• It's likely tailored a bit towards non-programmers, even if the system has that Smalltalk etc.

The hardware might still make sense from a performance point of view, but the main issue I find is the lock-in on software, which I find both suspicious and hindering the use of the algorithms. The mention about H9000 above is an indication of this. I for example thought the plug-in is a better buy than a H3000. These firms compete on offerings on their closed systems.

I think there should be a FOSS alternative, if not for other than access and longevity. With free licenses around, we could maybe also figure out, what algorithms are truly innovative and what are just closed source of "the old and usual". And if there was need for hardware, then it should run Linux or something.

Reaper did it for Pro Tools, so ...

[whyterabbyt]

I think there should be a FOSS alternative

On you go, then. Nothing has stopped you except you in the past 5 years...

viewtopic.php?t=511630

Reaper did it for Pro Tools, so ...

No it didnt. Reaper isnt FOSS.

[soundmodel]

Reaper did it for Pro Tools, so ...

No it didnt. Reaper isnt FOSS.

No, but it's still significantly more open than Pro Tools. Even if we just had a native closed Kyma, then even that'd be an improvement, IMO.

[whyterabbyt]

No, but it's still significantly more open than Pro Tools.

Nope, its exactly as closed.

[whyterabbyt]

Even if we just had a native closed Kyma, then even that'd be an improvement, IMO.

On you go then...

https://squeak.org/

[soundmodel]

No, but it's still significantly more open than Pro Tools.

Nope, its exactly as closed.

I think JSFX + Reaper is a pretty good half-Kyma, but it lacks the library. The full-timeline pitch shifting already mimicks the aspect that "the software is an instrument".

I've done some research on that mentioned thread though and as of now I think maybe 95% of the stuff on Kyma is found on papers via search engine. It just has large amount of stuff, that's all, + some creative "layered things" (which build on simple building blocks though).

Again I don't want to downplay the product, but some products do attempt to repackage well-known stuff under a different name or a flashy GUI so that it'd seem novel. The main idea is still to sell the product.

[whyterabbyt]

And? Its not the first time that you've said that something else 'could' be used to replicate Kyma. But all you're going to do is come back in 18 months time and yadda yadda about it all over again, same as youve been doing for 5,6 years.

Its your assertion that there should be an alternative, so dont talk about what should or could be done. Do.

[soundmodel]

And? Its not the first time that you've said that something else 'could' be used to replicate Kyma. But all you're going to do is come back in 18 months time and yadda yadda about it all over again, same as youve been doing for 5,6 years.

Its your assertion that there should be an alternative, so dont talk about what should or could be done. Do.

https://en.wikipedia.org/wiki/Availability_heuristic

What I mean is that there's a reason why Reaper is so similar to other DAWs for example, because there's loads of risk in being different.

Kyma is possibly baked into a variety of plug-ins at pieces, but it's all spread and therefore hard to recall.