Korg Forums

Posted: **Wed Feb 16, 2011 3:05 pm**

ozy wrote:Releasing it only to smart customers?

And how do you define "smart"?

This is simple. Does the customer debate esoteric points on synth forums? If yes, release nothing to this individual.
(this leaves me out!)

Posted: **Wed Feb 16, 2011 3:15 pm**

Zeroesque wrote:Does the customer debate esoteric points ?

Posted: **Wed Feb 16, 2011 3:18 pm**

Re: ozy's comment

I want to suggest ideas for Kronos that can be achieved in reality.

Not big changes - but easily conceivable useful improvements based on existing functions of the Korg operating system.

(PS. A picture again... No comment. What has this place become...)

Posted: **Wed Feb 16, 2011 3:28 pm**

Akos Janca wrote:Re: ozy's comment

I want to suggest ideas for Kronos that can be achieved in reality.

Not big changes - but easily conceivable useful improvements based on existing functions of the Korg operating system.

What's possibly missing in Kronos? It it has the same sampling and sample editing tools as previous Korg workstations and can match unique sample editing functions found in other workstations (Motif, Fantom, ASR, V-Synth, MPC, MV)...I would say nothing is missing. I am not a programmer, but I speculate that these tools can be achieved in Kronos as it appears to be more powerful than these mentioned workstations......

Posted: **Wed Feb 16, 2011 4:16 pm**

StephenKay wrote:
robinkle wrote:How many is a bunch? Did you use complex sounds?
I recall Stephen Kay naming MOD-7 a CPU hog in the Challenger thread.
Doesn't sound like it now. Thanks.
Well, that depends on how you use it. If you've got a double MOD-7 E. Piano, then each note is 2 voices. If you are playing a typical E.Piano part, with a sustain pedal, then you could easily be using up to half of the CPU of the entire machine, on one track (playing with both hands, some notes overlapping: 12 voices x 2 = 24 MOD-7 voices). Then, put a MOD-7 pad sound with long release on another track. Let's say it's a single, and you play 6-note chords. Due to the long release, every time you release the chord and play the next one, they overlap. so that's 12 more voices on and off. Now you've used 75% of the available CPU on two tracks.

I'm not saying that this is bad, that's just the way it works, and you will always have limitations with resources on computer DAWs as well. Part of using the tools is learning how to work within the limitations. If you want to use lots of MOD-7 sounds in a sequence, then you may have to use workarounds, such as dumping the tracks to audio once you have the track finished, thereby freeing up all those voices to be used somewhere else (which is what I did).

I'm with you. I'm sure I will meet the barrier with MOD-7s polyphony, so I would need to do workarounds, like making audio tracks. You could say it solves the "problem", but not quite. It sounds like a plan B. I hope it doesn't make the work-flow too complicated though.

One thing I would like to know though, does complexity have of the patch have any effect on the polyphony?
Like if you compare a INIT sound with just a pure sine wave and no modulation, to a complex sound with tons of modulation and waves.
Is it a minor difference? Huge difference? moderate?

Thanks Steven.

Posted: **Wed Feb 16, 2011 5:04 pm**

robinkle wrote:One thing I would like to know though, does complexity have of the patch have any effect on the polyphony?
Like if you compare a INIT sound with just a pure sine wave and no modulation, to a complex sound with tons of modulation and waves.
Is it a minor difference? Huge difference? moderate?

From my understanding, the complexity of the patch (assuming a simple single MOD-7 compared to a "complex" single MOD-7, with the same unison settings) is not a large factor (or any factor) in the polyphony, because the engine is the engine - and all of the parameters and LFOs are there, whether they are being used or not. But I could be wrong about that....

Posted: **Wed Feb 16, 2011 5:44 pm**

StephenKay wrote:
robinkle wrote:One thing I would like to know though, does complexity have of the patch have any effect on the polyphony?
Like if you compare a INIT sound with just a pure sine wave and no modulation, to a complex sound with tons of modulation and waves.
Is it a minor difference? Huge difference? moderate?
From my understanding, the complexity of the patch (assuming a simple single MOD-7 compared to a "complex" single MOD-7, with the same unison settings) is not a large factor (or any factor) in the polyphony, because the engine is the engine - and all of the parameters and LFOs are there, whether they are being used or not. But I could be wrong about that....

If you are right it would be a relief, because I would know the polyphony wouldn't drop no matter what I programed (except for unison and dual engines etc)
On the other hand, if you are right, I think what is not in use within the engine parameters shouldn't be taking resources in the processes in the CPU, because that way we could increase polyphony(on simpler sounds).
From what I understand, you didn't notice much difference in polyphony between patches, if any difference at all.

A dynamic engine which limits the CPU usage to the current parameters that are being used and excludes the rest, could be a new task to the Korg programers. This is how other manufactureres do on their VA synths, dynamic voicing.
Would be interesting to hear what the forum members from Korg would say to that.

Posted: **Wed Feb 16, 2011 6:06 pm**

Bachus wrote:
So what do you think is missing from this lovely instrument

kronos being a Software Synth

Posted: **Wed Feb 16, 2011 6:42 pm**

robinkle wrote:A dynamic engine which limits the CPU usage to the current parameters that are being used and excludes the rest, could be a new task to the Korg programers. This is how other manufactureres do on their VA synths, dynamic voicing.
Would be interesting to hear what the forum members from Korg would say to that.

I'd say: read the OASYS manual (or the KRONOS when it's released)! Your questions are answered there. For the MOD-7, for instance:

"If the PCM Oscillator is not being used, you can disable it to save processing power. You can do this in two ways..."

"The Mode also affects the required processing power. Triangle, Square, and Sine + Waveshaper need the most power, and using them may result in slightly lower polyphony. Off requires no processing power at all."

The allocation system in the OASYS and KRONOS allows multiple voices (and effects) with wildly different algorithms and computational requirements to coexist sharing common resources, and to steal voices from one another when necessary - without CPU overs, dropouts, or the other things you might expect from normal computer-based soft-synths. It's one of the parts of the system of which we're particularly proud.

- Dan

Posted: **Wed Feb 16, 2011 6:47 pm**

robinkle wrote:
danatkorg wrote:
danatkorg wrote: In general, our voice count specs are fairly conservative; I don't think you'll be seeing it drop to 30 voices. When the KRONOS comes out, you can check this yourself on the Performance Meters page.
A follow-up on this: I just tried a bunch of factory MOD-7 sounds, and with all of them I was able to get at least 48 voices.

Hope this helps,

Dan
How many is a bunch? Did you use complex sounds?

About a dozen, and yes.

robinkle wrote:I recall Steven Kay naming MOD-7 a CPU hog in the Challenger thread.
Doesn't sound like it now. Thanks.

Everything's relative. As Stephen notes, stacking two of them will halve the polyphony again. And, the AL-1 gets twice as many voices, HD-1 about three times as many, PolysixEX and CX-3 about four times as many...

Posted: **Wed Feb 16, 2011 7:20 pm**

@ akos

Exactly.

I wanted to point out that what was discussed in the "velocity" thread is NOT a easily added feature.

Or maybe you didn't carefully read the thread before linking to it.

Asking for the impossibile is wrong.

Asking for the impossible as if it was not just possible, but trivial, is even more wrong.

As for the picture: they are actors in costumes, akos, not real devils.

Posted: **Wed Feb 16, 2011 7:51 pm**

ozy wrote:@ akos

Exactly.

I wanted to point out that what was discussed in the "velocity" thread is NOT a easily added feature.

Or maybe you didn't carefully read the thread before linking to it.

Asking for the impossibile is wrong.

Asking for the impossible as if it was not just possible, but trivial, is even more wrong.

As for the picture: they are actors in costumes, akos, not real devils.

Maybe I missed something from that thread. I still believe what I'm suggesting is possible to implement: I don't want only to select from predefined curves but I want to set (= "draw") my own. Why would it be impossible?

PS. I doesn't matter if actors or not. What matters is the content, the "message". (Yes, I know you wanted to joke... etc., please don't repeat that.)

Posted: **Thu Feb 17, 2011 11:00 am**

Akos Janca wrote:I don't want only to select from predefined curves but I want to set (= "draw") my own. Why would it be impossible?

Please read the velocity thread carefully, there's some very interesting consideration about the futility and limits of manipulating curves in the midi domain (which is the only possible "customization" which can be done done by a simple menu).

Mental energy was spent there, so pls let's not reivnent the wheel on a daily basis.

Of course your request is comprehensible, but we are speaking of a 3 or 4 grands keyboard,

and we are in the "Korg, give me the most at start since I know you won't upgrade later [

]" phase,

so it's better to set the record straight: whatever must be done in the OS [and REAL curve customization must be done in the OS, not through midi], opening it to customization, must be done NOW.

Akos Janca wrote:PS. I doesn't matter if actors or not. What matters is the content, the "message". (Yes, I know you wanted to joke... etc., please don't repeat that.)

The "message" is: exoterism is a freak pantomime for nutjobs. Do you disagree?

Posted: **Thu Feb 17, 2011 12:54 pm**

ozy wrote:Please read the velocity thread carefully, there's some very interesting consideration about the futility and limits of manipulating curves in the midi domain (which is the only possible "customization" which can be done done by a simple menu).

Yes, I understand there are limits.

I don't want to repeat my idea just to add: that probably would make many users happier. Maybe a decision maker feature for some to buy Kronos (instead of Yamaha TruePianos

).

Great things are often the very simple ones. If customization is too complicated for the average musician then nobody will use it (= customer dissatisfaction). So yes, a simple but effective menu would do.

What you suggest, unfortunately, won't happen now because it's too big change in the system. I don't think Korg have resources to achieve that this time. They want to release Kronos soon. I'm afraid your idea will be preserved for future development.

Posted: **Thu Feb 17, 2011 12:59 pm**

Akos Janca wrote:I'm afraid your idea will be preserved for future development.

where have I heard this before?

Korg Forums

Whats missing in KRONOS..

Re: Whats missing in KRONOS..