Korg Forums

[Hedegaard]

[SanderXpander]

@Hedegaard
You should pick a decent pair of headphones then because there's a very definite difference.

I agree by the way that it needs to be done with Oasys programs ported to the Kronos, I was under the impression that Leo had done it this way.

Whether you can hear a difference or not has nothing to do however with "how you're feeling one day", that's like saying that on some days, you can't tell the difference between red and green. Comparative analysis by ear is a perfectly valid (and in fact the only useful) way of comparing audio examples.

[Akos Janca]

Dear Leo and Dany,

Thank you for the information, but I got really confused.
So finally you say: 11 & 13 are OASYS, 12 & 14 are KRONOS.

For me 12 & 14 were "warmer/fuller/richer" somehow, but the difference was very small. So does it mean that Kronos is the one that sounds better? I simply can't decide the question myself based on these examples. I would make "more scientific" comparisons but at the moment I don't have the time, unfortunately. Until now I can only say - according to Dany - that both OASYS and Kronos are outstanding, head to head, in sonic capabilities and quality.

(PS. Hi Hedegaard, thanks but I'm not really back. It was an interesting topic. )
_________________
AkosJanca.com | Facebook | YouTube | SoundCloud | FuturePlant || OASYS-demos | nanoROCK demo | Korg Band in the Garage demo | Kronos Demo 1 | Kronos Demo 2

[maphill]

I agree with almost all of the points that Heregaard (sp?) and McHale mentioned. And as a DSP Engineer who spends his time writing algorithms in MATLAB, I'm equipped to do a more scientific comparison.

Disclosure: I believe the electronics are the same with regard to sound quality. And I believe that any changes in the algorithms themselves would be optimizations that negligibly affect sound quality and add capabilities (4 to 8 velocity switch points, etc.). Therefore, my hypothesis is that the differences in "sound quality" are actually patch differences, auditioning system differences, and subjectivity.

However, I'd like to approach this as if I had no clue, with an open mind, and prove it.

But, to do this, I need a patient volunteer with an OASYS that can help me make a few recordings here and there over the next few weeks. It won't take more than a few minutes at a time.

Thanks,
Mark
_________________
Korg KRONOS 73, Trinity Pro, Monotron,
Roland FA-06 and Fantom 6, Roland TDK-15
Big Knob, Sonar (Previously, OASYS 76)

[michelkeijzers]

[SanderXpander]

Btw, I don't disagree at all that computers and DSPs are entirely deterministic. I don't think computers are "voodoo boxes" or something where anything can happen and the position of the moon influences the sound and whatever other nonsense.

My supposition (purely based on the premise that identical patches were used, which, it now turns out, might not be the case) is that they must have modified the algorithms somewhat as part of the optimization. And listening to the posted examples, to be honest, it sounds to me unlikely that the differences are purely due to patch tweaks.

[maphill]

[SanderXpander]

I know it's not very scientific, but likely candidates (judging from the samples and the basic technical knowledge of the machines) seem 16-bit vs 24-bit recording and the fx. Even if Leo recorded both at 16-bit, it seems the code would have to be modified to even incorporate the 24-bit possibility on the Kronos. Also, since the fx are now processed on the second cpu core, some code must've been modified there.

In other words, I don't think there'll be much difference with the most basic form of sample playback. I'd like to help you test, but I don't have an Oasys nor a Kronos right now. Hopefully a Kronos later this month.

[McHale]

A good test would be for an OASYS owner to play a patch under every engine. Send the .pcg of those patches to a KRONOS owner (or load it into a KRONOS if they happen to have both) and play the exact same thing with the KRONOS conversion of the patches. It MAY sound exactly the same but shouldn't be surprising if it doesn't. The KRONOS OS has been updated. I wouldn't be surprised if the exact same patches on an OASYS sound different on the same OASYS under different OS versions.

The next question would be, if the OASYS and KRONOS sound different with the same .pcg file, can the KRONOS be tweaked to sound exactly like the OASYS. If so, it's a moot point IMHO.

-Mc
_________________
Current Korg Gear: KRONOS 88 (4GB), M50-73 (PS mod), RADIAS-73, Electribe MX, Triton Pro (MOSS, SCSI, CF, 64MB RAM), SQ-64, DVP-1, MEX-8000, MR-1, KAOSSilator, nanoKey, nanoKontrol, 3x nanoPad 2, 3x DS1H, 7x PS1, FC7 (yes Korg, NOT Yamaha).

[shawnhar]

If you're serious about proving this one way or the other, it is also important to make this test as a proper scientific double blind.

If there truly is a difference, multiple different listeners should be able to identify which recording is which, significantly > 50% of the time, and without knowing in advance which one they are listening to.

If the test is not blind, listening to two versions doesn't really prove anything.

[maphill]

[SanderXpander]

I don"t see why it would have to be double blind, but blind testing is a good idea, yes. That was part of leo's test as well, by the way. To be honest, after so many comments about the Oasys still being "better", I had thought that samples 12 and 14 were from the Oasys, as they sound better to me. I was positively surprised that he later said they were from Kronos.

[maphill]

[McHale]

[SanderXpander]

I would definitely not be able to tell one from the other without a direct comparison. And even then, that's assuming any differences that might show up resemble the ones in the previous sound examples.
I'll never have an Oasys and I will have a Kronos, so it doesn't really affect my buying habits. It's just interesting to me if a real difference can be found because on the surface most engines would seem to be copies of each other.