Korg Forums

[Broadwave]

Hi all, first post so please be gentle with me.

Plucked up the courage to order the Kronos 61 from Bonners Music here in the UK. Their initial price is £2599, but they have told me that if it's officially higher, they'll honour the quoted price, and if it's lower, I'll be refunded the difference... I'll also get a pair of Behrringer Truth Monitors (not that I'll need or use them, but a nice freebie all the same), although delivery may not be for another 8 weeks!

OK, one question I have is regarding the AL-1 Oscillators. From what I gather from the Kronos manual is that they are not truly free running like other VA synths, but the start point for the waves could be set to random - Can any Oasys owners assure me that, apart from the start point issue, the osc's still behave in an analogue fashion and sound good?

Thanks.

[Megakazbek]

As you said you know yourself that initial phase of the oscillators can be set to random. So what more clarification do you need regarding this question?

As for oscillators not being "true" free-running: running oscillator code when you can't possibly hear it's output is a waste of resources without any advantage. So it's actually a good thing that they are not free-running, because that saves you some CPU for other tracks.

[ozy]

digital, virtually modeled "free-running" is substantially meaningless

does the VA model real discrete components' behaviour? Yes? Really?

Oh, yes? Does it model two separate power supplies for the oscillators, or just one power supply?

Oh, it doesn't even take the issue into account... I see.

[X-Trade]

I think you'll probably find that the MS-20Ex oscillators are free running, or at least somehow pretend to be.
If you're after a distinctively analog sound that's the engine I'd be playing with.

Indeed the oscillators don't actually need to be 'running' to be 'free running' As long as you know what pitch an oscillator last played and when it stopped playing it you can extrapolate what phase position it would be at any given point in time (i.e when you trigger it again), so performance isn't as much of an issue. However in Korg's 'ground up' philosophy for AL-1 this probably didn't figure into their view of things...


In reality random is almost as good as free-running, unless you want the phasing between two oscillators in your mono lead to appear to be continuous.
Otherwise, free-running oscillators are only really noticeable when you have small amounts of polyphony, e.g. for monophonic things or when the number of voices is limited to four or perhaps six.
_________________
Current Gear: Kronos 61, RADIAS-R, Volca Bass, ESX-1, microKorg, MS2000B, R3, Kaossilator Pro +, MiniKP, AX3000B, nanoKontrol, nanoPad MK II,
Other Mfgrs: Moog Sub37, Roland Boutique JX03, Novation MiniNova, Akai APC40, MOTU MIDI TimePiece 2, ART Pro VLA, Focusrite Saffire Pro 40.
Past Gear: Korg Karma, TR61, Poly800, EA-1, ER-1, ES-1, Kawai K1, Novation ReMote37SL, Boss GT-6B
Software: NI Komplete 10 Ultimate, Arturia V Collection, Ableton Live 9. Apple OSX El Capitan on 15" MacBook Pro

[jimknopf]

You can all tell me what you want, and I acknowldge the limited effects of free running versus random, but I still prefer the orginal analog concept of free running oscillators anytime anyhwere.

I didn't know AL-1 misses it, and I'm glad MS-20 has it implemented in some way.

[Megakazbek]

[Broadwave]

[orpheus2006]

I'm happy the AL-1 will not be free running, but remains in the unit.
_________________
www.soundcloud.com/orpheus2006
.............................................................
Korg M3-88 w/ EXB-Radias, Yamaha Motif Rack w/ PLG150-AN & PLG150-DR, Novation A-Station, RME Multiface II Audio Interface, Thinkpad T60 Notebook w/ Sonar X3 and various VSTi, Event TR8 monitors, Beyerdynamic DT-770 headphones

[danatkorg]

Analog synths have a fixed number of oscillators which are always running. Simple digital systems may work this way, too. In complex, dynamically allocated digital systems, on the other hand, oscillators are created and deleted as necessary. This includes not just the OASYS and KRONOS but also, for instance, the Virus Ti and Waldorf Q.

On such systems, the standard way to emulate free-running oscillators is to set the phase randomly for each note-on. From the Virus Ti manual, for example:

"Off: The phase angle (see glossary) of each oscillator is reset to a random value at the start of each note. This closely emulates the "free-running" oscillators found in real analogue synthesizers."

Along these lines, the AL-1 includes phase controls for both oscillators, including a Random setting.
_________________
Dan Phillips
Manager of Product Development, Korg R&D
Personal website: www.danphillips.com
For technical support, please contact your Korg Distributor: http://www.korg.co.jp/English/Distributors/
Regretfully, I cannot offer technical support directly.
If you need to contact me for purposes other than technical support, please do not send PMs; instead, send email to dan@korgrd.com

[jimknopf]

I always thought it might be one of the reasons that my Access TI was missing some (tiny) characteristics of real analog synths, besides fine tuned oscillator drift (which is thankfully available since the last update), overtone behavior and the like.

But it makes sense to me that it could be a waste of resources in powerful VA systems with a lot to calculate anyway to have oscillators running through virtually.

[Kevin Nolan]

And of course real analogue systems sound warmer not just because of oscillator variances such as oscillators' phase and tuning; but on similar variances within all of the systems components and modules - variance in LFO rate, wave shape inaccuracies and variance from transistor to transistor and chip to chip, various harmonic (desirable) distortions, envelope ADSR variances and so on. One of the biggest failings of all current VA's is the almost total inability to apply pseudo-random variance to a host of such parameters.

Although I don't own a Virus, my understanding is that their latest OS offers the ability to apply such variance to several, user closable, synth parameters, offering a closer approximation to a real analogue system (but still falling far short). It would be an excellent update to AL-1, MS20EX and PolySixEX if a similar choice of random-seeding could be applied to many synth parameters and not just oscillator phase.

A very useful one would be envelope attach time. On my Jupiter 6, for example, there is some minute variance on one or more of the voice envelope attack time. It is noticeable on fast arpeggios with fast attack times, where, quite randomly, some notes sound a little softer or with less attack - but where on the next pass that voice will have its fast attach - I'm sure it's down to some clocking / syncing issues - but it makes the arpeggiator sound more organic, more alive, more rhythmic, than a standard VS 'perfect every time' arpeggiator. It has drawn my attention to such issues on my other analogue synths - they are all engendered with such character; and it'd be great to see enhancements to the likes of AL-1 in this regard.

Note also that MOD-7, with 6 oscillators per voice all with random phase setting, is an fabulous synth engine for creating very convincing analogue synth sounds (you don't have to use MOD-7 as an FM synth - it's also a thick, fat VA!!).

Kevin

[danatkorg]

[Kevin Nolan]

Hi Dan -

Yes - absolutely – an important point you’ve flagged this before; and as an OASYS user I can vouch for this to prospective Kronos users - AL-1 (and other Kronos synth engines) use extremely low aliasing oscillators. In practice this means that you can create far more authentic analogue emulations – as just one example, lead sounds that can reach the upper most octaves and beyond like a real analogue synth and where the character of the sound remains clear, crisp and devoid of typical aliasing artifacts that dog even some of the best known VA's. Overall, it gives AL-1 a very strong sound basis for the whole gauntlet of analogue emulations.

Also worth flagging again (I’ve flagged this before as has Dan and Daz on the OASYS forum) is the fact that in AL-1’s 'double saw' oscillator, the phase between the two sawtooth waves can be modulated w.r.t each other; and when the modulator is an LFO it gives AL-1 a really gutsy vintage/raw analogue feel (and so can do microKorg sounds very well too). I exploit that on some of my AL-1 sounds I've pointed before that can be downloaded free from my web site (http://www.knect.ie/OASYS.html).

Finally, Dan - thanks for the extended detail on the 'analogue' feature of PolySix and MS20 - that innocent looking feature of those already great synths passed me by and definitely needs more careful consideration and use.


As an opinion, AL-1, MS20EX and PolysixEX are all _VERY_ strong. I own a MonoPoly and am a dab hand at playing it live - especially its arpeggiator (in a college 'battle of the band's' night some years ago I played a psychedelic piece on my MonoPoly and both it and I survived water and hay dropped from the stage gantry, while tomato ketchup form the audience still proudly resides between a few keys!!). But such a tank is the MonoPoly it survived and is still 100% operational today. But I have to say, and I', usually reticent to say this - when I first played the OASYS PolysixEX aerpeggiator live (with unison mode doubling up the oscillators) I could not believe my ears - it sounded, played and felt uncannily like a MonoPoly - I was very surprised that it really did sound like the real deal. So many emulations do not sound like their original s- but Korg's circuit modeling is amazing and the PolysixEX is a fabulous emulation. Likewise, the MS20EX is raw, sharp, cutting and edgy just like the original which I don’t own but have experience on. So hats off to Korg - the three VA engines on Kronos / OASYS are each distinct, full of character (stunning in quality) and complement each other. Korg have truly ‘nailed’ software modeling of synthesizers, and Kronos is a class act in this regard.

Finally, I love the self-resonance of the MS20EX filter in particular. With LFO modulation, you get the full gauntlet of Jarre-like synth effects - MS20 is particularly excellent in that respect.

A lot of praise I'm dishing out here, but they are all deserving of it.

Cheers!
Kevin.

[Broadwave]

Thanks for all the replies... I think I can rest easy

I didn't notice that there was a dual saw wave that can be modulated... that would be the equivalent of up to 8 saw waves in one patch then (using two AL-1 engines) - nice!

I'm not betting on it, but hoping that Korg will make the MonPoly available as an upgrade in the future, as it's one of my favourite plug-ins. I guess we can all speculate as to where the Kronos will go... or not, but at the moment it's still the most powerful workstation available and I'm happy about that.
_________________
Synth DIY Projects

[Kevin Nolan]

Also - you also have 'polyphonic unison' mode which simply layers the entire program from 2 - 16 times, so you can make absolutely massive voices. Polyphonic unison mode is very smooth, and also provides 'thickness' and 'stereo' spread' features - so you could create a program that uses well in excess of 100 oscillators on just one note. Think of it - all of Kronos's synthesizer power on just one note.

In my AL-1 bank there are some patches called CS80Bass and UnisonSynth (or names like that) that stack about 64 oscillators per voice. These programs only work well when you play one note in the lower part of the keyboard, but boy are they absolutely gigantic. Don't expect much else from your Kronos when you play them, but they are by far the fattest, thickest and most gargantuan sounds I've heard from a synth in a long time. Be careful with your speakers - while the sounds don't inherently distort, they are pushing the synth engine to the limits and you could blow a woofer.

If you're into layering oscillators for really fat sounds, Kronos and OASYS are for you.

Kevin.