New Kyma "supercomputer"

[Tau]

While S|C strives to bring us or much awaited XITE, Symbolic Sounds updates their Capybara following ths same lines: 15x as much power, in a smaller, portable form:

http://www.symbolicsound.com/cgi-bin/bi ... s/Pacarana


Decisions, decisions

T

[kensuguro]

wow... that's seriously awesome. I've always been drooling at their technology.
$3000 for the 5 times their earlier "starter kit" eh.. that's insane. $4500 for the top dog ain't so bad either. Now I seriously have to try to find a rational for xite.. It's shocking they're in the same price group.

lol, why's the dude from Lucasart's Grim Fandango holding the box?? what's his name..... "Manny".

[dawman]

I am sure it will be awesome as the Capybara's.
Carla Scaletti is worshipped like a Goddess from all who own one. She even answers the phone herself !!
It would be an awesome companion w/ an XITE-1.
It's strengths are realtime audio mulching as in the band http://austere.org/ used it for.
It does do analog emulations very well also, but it seems to lack the developers for that.
But I am glad to see yet another DSP Platform rising to the call.

In the last decade I have never, ever seen one sell used, so that's a testament to it's support and use.

I really wish I wouldn't have seen the link now.

[Tau]

man... it's what I've been thinking. I was almost getting a Capy a couple of years back, but it was very big, and bringing one to Portugal (with added shipping and taxes) made it quite expensive. Then they announced Solaris HW, the XITE-1, and I've been on hold since then. Now this is really awesome.

I still have some doubts about it, eg this one doesn't have audio output as the capy did, and I'm not sure if one can do with the onboard audio card on a laptop, or even with the xite to output audio. But it will surely introduce some latency to the process, more than the old system for sure.

The XITE will be a great platform to run all synthy things (I'm thinking Solaris, SpaceF and Adern), plus it's a great digital mixer and sound processor- well we know what it does. But Kyma is made specifically for real time audio manipulation (samplewise), so they seem to complement each other perfectly. Plus, the XITE is Express card and Kyma is Firewire, so no problems in having both on a laptop - although 5000 EUR is quite a lot for most people, I guess (me included ). Time to start selling the old hardware...

But you have to admit: I saw some cryptic announcements in Symbolic Sounds site for the last month or so, suggesting the arrival of a new machine. And when they finally revealed it, it was already shipping. Good work!

[dawman]

They actually made an announcement a while back, but they worded it very cautiously.
You are sick......a laptop w/ Pacana and XITE-1 %*&) ??
I am also sick though.
All you need is a wireless one of these. http://www.keyboardmag.com/story.asp?storycode=4526

The old Capy's sound awesome and when used w/ Ableton LIve was an incredible axe.
I can imagine a DJ using both of these w/ a capable sequencer. Now that Picana is out the need for Ableton wouldn't be necessary I think, but rather a well matured MIDI sequencer like a hardware model or Cubase Essential, which also looks like they are going to DSP based solutions. Their little controllers look usable, but the Yamaha KX keyboard series are designed perfectly for Steinberg apps.

NAMM is going to be fun. I haven't been there for years, but even during a bad economy there seems to be so many great advancements, soft and hard.

[darkrezin]

Kyma is pretty great.. a friend of mine has the old base system and even that is amazing, even though you have to process most things offline instead of realtime because the dsp count is so low. The new system looks like a great deal.

But it is really not that intuitive - you need to put in a LOT of learning time. If you can invest the time though it lets you do things not possible with anything else.

[astroman]

my intentionally provocative bottomline would be
invest the same amount of time (or cash) into a Scope system as in a SymolicSound one and there will be nothing left to wish
you might get your results significantly earlier in Scope, tho

cheers, Tom

[Zer]

seems to be a system for modular sound engineers. I bet it`s more complicated than reaktor or other modular systems, but has its own strengths when compared to other devices like it.

[kensuguro]

ya, I guess it's sort of like max/msp.. full of potential, but a pain to use. (beause it takes so much energy) Xite is a much quicker out of the box solution.

[Tau]

Well, I'm all in for the XITE (when it's done), and no doubt Scope is extremely powerful. On top of that, it's something I have already learned, with the help of the PlanetZ dwellers, and I'm quite confortable using it, and happy with the results. In terms of traditional synthesis, I believe that nothing comes close to Solaris or the Modular featuring our beloved 3rd party modules, and the FX+ processors available for SFP are extremely musical and easy to use.

I have never touched a Kyma system, only saw it in videos and heard some music and examples that used it, and it seems interesting enough. However, I believe that its strongest point is with sample manipulation, and I tend to see it (maybe wrongly) as a sort of super-sampler. It does come with 1 or 2GB of RAM all for itself, and I believe it will perform its best in terms of resynthesis and time/frequency manipulation of recorded sounds or even full tracks - (like timestretching and pitchshifting) something that is not Scope's strongest point, and has to be dealt with by the sequencer, or other native apps. I would really welcome the opportunity to get my hands on it (without having to pay first, of course )

Anyway, please excuse my gear-disease. I already do a lot with what I have here, and there's so much more to learn that these sort of discussions may seem to be on the superfluous side. For sure, nobody "needs" a Kyma, and nobody "needs" an XITE either

Cheers,

T

[braincell]

This seems like direct competition for the Xite-1 although I realize it is not exactly the same. I don't think many people would be getting both since the cost of either of them is outlandish.

May the best box win!

[kensuguro]

well, cost of either the new kyma or xite-1 is not so bad compared to a full kyma system.

[darkrezin]

Its strength is definitely granular/additive/resynthesis. It's probably not that practical to use for 'conventional' sampling tasks, but if you are especially creative-minded and dedicated (to the uber-geek level) with sample-based sound design it's nice.

The sound is also verrry digital. Not in a bad sense - it sounds ridiculously high quality because it has great algorithms. But don't expect it to do everything like fat analog type sounds. Its main use is in film sound design.. most of the musicians I've seen using it *ahem* may not have the best sense of perspective. I've heard that the Continuum fingerboard gives some crazy possibilities with it though, if you can somehow keep hold of your musical direction when using the crazy hyper-digital functions.

btw here's a good info thread about using kyma:

http://www.gearslutz.com/board/electron ... dular.html

[siriusbliss]

This seems like direct competition for the Xite-1 although I realize it is not exactly the same. I don't think many people would be getting both since the cost of either of them is outlandish.

May the best box win!

They are both distinct paradigms - IMO not a real direct comparison can be made between the two other than you can do 'sound design' in either system.

Greg

[darkrezin]

One of the coolest things about this box is that it supports some firewire/USB audio interfaces like MOTU, TC Konnect, M-audio etc (I'm guessing ones that work as generic firewire/usb audio devices). Just in terms of an engineering angle that's a pretty crazy achievement IMHO, even though it might have been a good idea to have digital i/o like ADAT/AES etc. Even though the converters in the Capybara are quite legendary (the new box does not have them), they probably added a large amount to the cost.

[petal]

my intentionally provocative bottomline would be
invest the same amount of time (or cash) into a Scope system as in a SymolicSound one and there will be nothing left to wish
you might get your results significantly earlier in Scope, tho

cheers, Tom

I don't think you are right. Kyma is a different beast than XITE/SCOPE. XITE/SCOPE is all about synths, effects and mixing, meant for the musician. KYMA is (as has been mentioned) a lot like Max/MSP and Pure Data and is primarily aimed at sounddesigners. They are two completely different beasts, that both are good at different tasks, and they actually supplements each other perfectly - In my opinion they are NOT in direct competition.

I for one would be very interested in exactly what KYMA does more than what XITE does, but that's because I have a SCOPE system already and don't expect XITE to become stronger in the audio/sample manipulation department than it is now - it's simply not the focus of XITE/SCOPE.

[dawman]

http://www.acm.org/crossroads/dayinlife ... letti.html

I have admired her old analog modular music, and of course Kyma for a long time.
She actually spoke on the phone with me ten years ago for about a half an hour explaining why Scope would better suit my needs, and how Kyma could expand those Horizons further.
Even then I still waited another 5 years before I talked our management into downsizing.

Kyma is a steep curve for sure, but there are so many ready to use examples from a user base that's just like our beloved PlanetZ & Scope.
I have spoke of Kyma here many times but nobody ever replied much, but it can be used live in very unique ways.

The DJ I worked with did some amazing stuff by sending audio into a Grain Cloud and Morphing into another tune all over the top of Ableton Live's continuos groove. It floored me actually as I never heard anything like it. But this new design is an amazing instrument, and the audio quality is stunning. No munchkinizing or any artifacts of the source material.

I can pretty much guarentee that anyone using Scope cards would die for one of these.
Just go the Symbolic Sound website and listen to simple drum tracks, and then imagine the possibilities of finished tracks, and live mutations. An analog synth processed by Kyma X and a Capybara turn it into distorted reality.

What really sucks is I am looking at an empty 1U slot right now and I wish Tau wouldn't have reminded me of an old dream I had. 4 large is 6 weeks of Spartan Boot Camp for me.

[astroman]

...I don't think you are right. Kyma is a different beast than XITE/SCOPE. XITE/SCOPE is all about synths, effects and mixing, meant for the musician. KYMA is (as has been mentioned) a lot like Max/MSP and Pure Data and is primarily aimed at sounddesigners. They are two completely different beasts, that both are good at different tasks, ...

let me return this estimation...
before I made the frivolous statement, I quickly browsed a comparison someone had written about Kyma, Max and the Nord Modular... to have at least some background, fake tho
it's actually much closer to the Scope (library/graphic) paradigm than to a programming language like Max.

you CAN do sample and granular stuff in Scope, and you actually CAN custom code whatever you like.
Write a killer processing with VisualDSP and SonicCore would probably be last to prevent themselves from a marketing coop with you...
the point is, that it's extremely difficult.
People starting with Scope SDK who don't get along with it in a reasonable amount of time on their own can forget right about it - if you 'have to' ask a question, you don't qualify.
SDK really IS kid's stuff compared to an assembly language developement system.
I for one am not among such talented math and bit gurus - one got to accept one's limits
But since I've seen quite a number of developement systems, at least I qualify to estimate efficiency and learning curve, somehow...

Sorry for being a bit long-winded, but the topic would probably be adressed in this context anyway.
Such stuff (as on Kyma) doesn't exist on Scope because it is impossible, but because noone has written it.
Maybe noone had the talent, maybe noone considered it profitable...
But for sure NOT because the developement system is limited or not extensible.
SDK cannot help you in this domain, it's (intentionally) a high-level tool.

Admittedly there are some aspects where limited calculation power or memory bandwidth of the current system can become showstoppers, but Xite adresses much of that. How good it does... we'll have to see.
Actually some posts in this thread reminded me on the (in)famous ...the grass is always greener... proverb.

And the 'need for better tools', next generation-I-dunno-what syndrome seems to be the tec plague of our times
I once bought a (relatively) simple bass... not bad, but well... should have more options...
So I replaced the pickups by a pair of humbuckers, added a double 4-stage switch to dial each combination of coils (16!) plus blend the 2 PUs continously...

Do I have to mention I never really got happy with it ?
let alone I never found a matching amp, cabinet, etc bla, bla...
Today I have a Precision bass copy, 1 PU, 1 Volume, 1 Tone. And that's about it.
Tuned a few parts, picked the right strings, found a nice preamp (that doesn't even have an EQ) and I miss nothing - instead I play the instrument, for hours if time allows... and I love it's tone and response.
Maybe it's not the world's best bass sound, but then I'm not Victor Wooten either
At least the instrument has it's own sound signature and it's motivating, that counts, imho.

Recently someone mentioned Dave Smith's latest Monosynth.
You CAN do that in Scope as well - if you limit yourself to mono
There is a ton of undiscovered or overlooked MONO stuff in Scope, because features and poly usually get the focus, don't they ?

cheers, Tom

[darkrezin]

In theory yes I agree Tom (actually not about the Dave Smith Mopho because it's fundamentally not something you could ever do in Scope - it's analog, well DCO with analog filter anyway). But the thing is it's a big commercial risk to write the granular/resynth algos as good as the ones in Kyma because it's a LOT of work to get it as good, and even if you did it you would have to try and compete with something that is extremely well established in this very niche market.

So it's effectively moot saying it *could* be done on Scope - the fact is that it *can't* be done on Scope at this time, and personally I'd be surprised if the sheer depth of sample-manipulation and additive/resynthesis that is possible in Kyma ever becomes possible on Scope (inc. X-ite). It really is a very deep and complex system with many years of dedicated development (bizarrely by a husband/wife team) but as already mentioned, maybe not that easy to extract musical use unless you put in a lot of time investment.

[petal]

Tom, just because you might be able to program xite to do what the KYMA does, does not mean that you actually can do it on it now - get it My point is simply, that out of the box the two system has different focuses, that actually complement each other. Each system excels over the other in it's own area.

Please, as much as I love Scope and what it does, it's not the holy grail in every possible department