Topic: Korg MR1, WTF??? SOS

[newblue]

I just emailed the Product Support Manager for KORG on this.  He's been pretty helpful to me so far, he sent me a DC plug and cable for the MR-1000 from a bad AC adapter so that I could make a power cable for an SLA battery (FOC).  Here's what I wrote,  (I'll post his response)...

James,

Hi!  Thanks for trying anyway.  I know you are busy, but I have one more question to ask you.  Since the MR-1 has debuted in the US, I know of several folks that have purchased one and are using them in the field.  There has been some discussion regarding the noise floor and apperance of the wave, either DSD > PCM @ 24/192 or direct PCM @ 24/192.  It seems that there is a very steep increase in the noise floor for the MR-1 in both scenarios.  Here's a website with some graphs that have been generated for both situations (as well as some comparisons).

http://www.unipv.it/webcib/res_techtest_uk.html

I was wondering if you might be able to shed some light on this?  And do you know if the MR-1000 will exhibit similar characteristics.  I suspect that there is some issues with the noise shaping that causing this, or that there is noise being generated through the power supply (either internal or external).

Thanks James!

Cheers!

Jack

I asked him about the algorithms that were used in AudioGate for the DSD conversion and here's what he said....

In looking at all of the materials for Audio Gate, I see no outside licensing with regards to the algorithms. That tells me that our engineers designed all of the algorithms proprietarily.

[it-goes-to-eleven]

"In looking at all of the materials for Audio Gate, I see no outside licensing with regards to the algorithms. That tells me that our engineers designed all of the algorithms proprietarily."

DSD is extremely proprietary...  If they developed their own conversion as suggested, wouldn't they need all of the proprietary dsd details?  I'm thinking they bought a library from someone that contains the secret sauce.  That's not to say that they didn't develop some aspects of the pcm conversion but I doubt the magic beans have been spilled.. Not sure that it matters..

[newblue]

Here's the response I received...


Hi Jack,

No problem. I answer what I can.

On this question, I’m really not sure; no one here in the US distributor is. These questions go a little beyond us. I’m an operational support chap for example, and these are more design and engineering questions.

I touched base with Product Management here to see if any information was available on this question was available in that channel, but no. However, he has informed me that both he and Korg Japan are observing forums such as the one that you mention, the URL you provided included. This is actually typical, since it’s a good way to keep our ear to the ground on what people are doing with our products. And while there’s no comparison of the noise floor yet, should there be a study or release of info any this nature, it will either be posted to the list(s) or disseminated back through Product Management to us here is Support.

Until then, I’m afraid we have no further information. Hopefully that will change in the near future.

Best,

James

I think that the longest "I don't know" I've ever seen.   

[newblue]

"In looking at all of the materials for Audio Gate, I see no outside licensing with regards to the algorithms. That tells me that our engineers designed all of the algorithms proprietarily."

DSD is extremely proprietary...  If they developed their own conversion as suggested, wouldn't they need all of the proprietary dsd details?  I'm thinking they bought a library from someone that contains the secret sauce.  That's not to say that they didn't develop some aspects of the pcm conversion but I doubt the magic beans have been spilled.. Not sure that it matters..

I don't think DSD is proprietary however SACD is.  That's why all those SACD mastering programs are several thousand $$$.

Those graphs look like the LPF isn't working properly or something generating a lot of high freq noise.  Can a leaky power supply do that?

[jerrythek]

Hello folks:

I have not been "hiding", but until I get answers from Korg Inc. I couldn't really help. Now I can.

The issue of the noise is due to the nature of the Sigma-Delta noise-shaping that happens with the technologies we (and others) use. The MR recorders use the Texas Instruments PCM1804 A/D converter - a highly respected component. You can refer to its data sheet (page 12, 17) for more information on the structure and characteristics of practical sigma-delta modulators.

http://focus.ti.com/lit/ds/symlink/pcm1804.pdf

The anti-aliasing filter used in AudioGate for 2.8MHz DSD -> 192kHz PCM conversion is similar to the filters used in other company's DSD->PCM conversion software already in the market. These are designed to keep the information loss to a minimum rather than to reduce the out-band noise generated by sigma-delta modulators.

But we recognized the issue you are experiencing, and have designed 2 new filters for the conversion based on other methods, which will reduce those out-band noise. They will be included in the next version of AudioGate (ver.1.0.1) which will be released in April along with the support for Windows Vista.

We appreciate that you want and need to make these conversions and just ask for you to be a little patient while we finish implementing and testing the software.

We have also just finished a new OS for the MR-1 (1.01) which will get posted to korg.com tomorrow. It fixes a small bug which caused the System Clock to not always keep accurate time.

I know it's not related to this discussion, but this is as good a place to let you know.

Thanks for your patience - we are listening and watching. But again, many times I need to check with Japan to get answers for you.

Regards,

Jerry

[jerrythek]

DSD is proprietary, and we have been working with the original inventor/developer with his full cooperation/license. The same professor (Dr. Yoshio Yamasaki) who licensed it to Sony/Phillips. So we have ALL the magic beans, as you say.

:-)

Regards,

Jerry

"In looking at all of the materials for Audio Gate, I see no outside licensing with regards to the algorithms. That tells me that our engineers designed all of the algorithms proprietarily."

DSD is extremely proprietary...  If they developed their own conversion as suggested, wouldn't they need all of the proprietary dsd details?  I'm thinking they bought a library from someone that contains the secret sauce.  That's not to say that they didn't develop some aspects of the pcm conversion but I doubt the magic beans have been spilled.. Not sure that it matters..

[newblue]

So the MR-1 has the PCM1804 and the MR-1000 has the PCM4202?

[Brian]

edit: nevermind...need more coffee....

[jerrythek]

So the MR-1 has the PCM1804 and the MR-1000 has the PCM4202?

Yes, that's correct.

Jerry

[newblue]

Thanks Jerry!  As always your expertise and attention here is always appreciated.

[Nick's Picks]

Nice Jerry.
Thank you for lurking!

make note of the battery charging issues (lack of notification while off), if you would.

[Nick's Picks]

DSD is proprietary, and we have been working with the original inventor/developer with his full cooperation/license. The same professor (Dr. Yoshio Yamasaki) who licensed it to Sony/Phillips. So we have ALL the magic beans, as you say.

:-)

Regards,

Jerry

IMO, this HUGE info.  you can't really say that the software or the hardware is 1/2 assing anything compared to other DSD products (and since there really are none in this hand held portable market).

Jerry, i've got a question.
in your (or Korgs) opinion, what is the best way to actually listen to the DSD masters we record, in their raw form (no PCM conversion at all) ?
DSD>CD sound super sweet to me, but id rather listen to just straight dope DSD.

[jerrythek]

cts (and since there really are none in this hand held portable market).

Jerry, i've got a question.
in your (or Korgs) opinion, what is the best way to actually listen to the DSD masters we record, in their raw form (no PCM conversion at all) ?
DSD>CD sound super sweet to me, but id rather listen to just straight dope DSD.

Hi Nick,

That easy - there is no doubt that listening to true DSD is the best possible listening format. The market for distribution etc. is not ready for it, but simply from an audio standpoint the DSD format is far superior.

Regards,

Jerry

[jerrythek]

Thanks Jerry!  As always your expertise and attention here is always appreciated.

LOL - my expertise is something that is being aquired each day - this is new for most of us, myself included!

 

But I'm glad to try to help where I can, and it's important to me and Korg to listen, learn and help you folks out. You're obviously REALLY into this which is exciting for us. Thanks!

Regards,

Jerry

[Nick's Picks]

I think you missed my question Jerry.
I was asking what is the best way to actually listen to these DSD masters?
in your own opinion.  Since we just can't burn SACDs, what other, if any...alternatives are there?