Measurements of Korg MR-2000S Professional Digital Recorder

[c1ferrari]

Cheers, Graemme

 

[zenmastering]

Yes, it does act as a mass-storage device. I use their USB driver because the transfers seem much faster with it than without. I don't use AudioGate, as I have better conversion tools that integrate well with my studio setup.

Thanks a bunch Graemme. That should make it easier to test.

On USB, does it look like a mass storage device so I can just grab the files in Windows or do I need their software to access them?

 

[zenmastering]

Here's another:

http://pinknoisemag.com/pink-papers/pink-paper-002

From my experience, an external clock can indeed sound 'better' under the following circumstance: The internal clocking of the device in question is sub-standard. Otherwise, an external clock is primarily useful for system synchronization. In that system scenario, it's always nice when the external master clock doesn't change the playback/recording quality for the worse ;->

When using an external clock, the DAC/ADC in question needs to have a sufficiently high-quality PLL/clock recovery system so that it doesn't render the whole effort meaningless.

The Behringer example cited in the Sound on Sound article is an example of either: a) A competently-designed internal clocking system or b) the DAC has other problems that clocking alone doesn't address.

https://www.soundonsound.com/techniques/does-your-studio-need-digital-master-clock

Oldie, but a goodie about master clocks. They don't improve jitter performance. Not even when applied to a Behringer from a very expensive clock.

 

[Blumlein 88]

Here's another:

http://pinknoisemag.com/pink-papers/pink-paper-002

From my experience, an external clock can indeed sound 'better' under the following circumstance: The internal clocking of the device in question is sub-standard. Otherwise, an external clock is primarily useful for system synchronization. In that system scenario, it's always nice when the external master clock doesn't change the playback/recording quality for the worse ;->

When using an external clock, the DAC/ADC in question needs to have a sufficiently high-quality PLL/clock recovery system so that it doesn't render the whole effort meaningless.

The Behringer example cited in the Sound on Sound article is an example of either: a) A competently-designed internal clocking system or b) the DAC has other problems that clocking alone doesn't address.

If you have one of the worst clocks in the history of digital devices, like the Digidesign, then external might help. The basic idea is the most accurate clock in the business once it goes over long cabling and passes through a couple PLL's is not equal to a relatively inexpensive crystal clock right next to the ADC/DAC. Some extraordinary PLL's might nearly equal the on board crystal. So the idea external high precision clocks improve jitter is nearly always not going to be true. They manage to synch up mupltiple devices well enough, and the jitter of each device is mostly down to quality of the PLL. Jitter generally isn't high enough in level to get close to being an audible problem in the first place. So incredibly precise and expensive external clocking doesn't look to be a wise use of resources for sound quality improvement.

 

[zenmastering]

All well and good, in theory. Results vary in practice. So much so, that I don't want to take either position and that was my real point: There are variables to account for before unilaterally declaring either as better or worse.

My experience is primarily studio-based where cumulative jitter can become audible (and measurable). At home I prefer USB as an interface with the DAC as the master (RME ADI-2 Pro).

If you have one of the worst clocks in the history of digital devices, like the Digidesign, then external might help. The basic idea is the most accurate clock in the business once it goes over long cabling and passes through a couple PLL's is not equal to a relatively inexpensive crystal clock right next to the ADC/DAC. Some extraordinary PLL's might nearly equal the on board crystal. So the idea external high precision clocks improve jitter is nearly always not going to be true. They manage to synch up mupltiple devices well enough, and the jitter of each device is mostly down to quality of the PLL. Jitter generally isn't high enough in level to get close to being an audible problem in the first place. So incredibly precise and expensive external clocking doesn't look to be a wise use of resources for sound quality improvement.

 

[zenmastering]

Correcting the following info I posted earlier:

You can use the SPDIF input as a clock input while audio rx/tx occurs over AES/EBU or vice versa. This is not a word-clock input, rather a SPDIF/AES frame clock.


"The only external clocking on the ADI-2 Pro is via the USB, AES or S/PDIF data streams. There is no word-clock input nor any kind of separate clock-only input."

You can switch between internal and external clock in the 'Clock' window of the Setup menu.

I am in the process of purchasing RME ADI-2 Pro. I think it has external clock input and if so, I could test them with your external clocks.

 

[Jimster480]

I really want to know the ADC performance of this device. I wonder if I could buy one and use it for measurements like Amir does.

 

[amirm]

Because it doesn't stream over USB it would be very cumbersome for computer analysis.

 

[Jimster480]

Oh it only records and saves files?

 

[amirm]

That is its intended purpose.

 

[zenmastering]

You'd need a wholly separate soundcard with S/PDIF interface to use the Korg in the way you describe. It would be cheaper (maybe not better) to use a USB audio interface such as the Focusrite Scarlett 2i2 and one of these:

https://linearaudio.nl/la-autoranger

Graemme

I really want to know the ADC performance of this device. I wonder if I could buy one and use it for measurements like Amir does.

 

[Dayz]

If you have one of the worst clocks in the history of digital devices, like the Digidesign, then external might help. The basic idea is the most accurate clock in the business once it goes over long cabling and passes through a couple PLL's is not equal to a relatively inexpensive crystal clock right next to the ADC/DAC. Some extraordinary PLL's might nearly equal the on board crystal. So the idea external high precision clocks improve jitter is nearly always not going to be true. They manage to synch up mupltiple devices well enough, and the jitter of each device is mostly down to quality of the PLL. Jitter generally isn't high enough in level to get close to being an audible problem in the first place. So incredibly precise and expensive external clocking doesn't look to be a wise use of resources for sound quality improvement.

I generally will always go with SOS and I agree here we are talking about the smallest of differences in possible sound quality. There are so many variables, it's not just external or not external clocking. And, it's not just the clock stage that influences what people and most engineers hear with higher end clocks (external or in their high end AD or AD/DA) it's the audio stages and how well those are designed which is not a function of price, but in some cases there are some really well regarded mixing engineers who are hearing something that is not something that is just altering the sound from neutral. Neutral or truth of what was intended originally recorded and hearing that (if you can - if you are in a room, with monitoring that allows such clarity) is the goal. It's commonly words like "opening" and getting ever closer to the holy grail of finally truly feeling like you are in the room with an artist or singer, they are with you right there. And in the production process not letting all the little bits along the way that can stamp that out. That's the difference between the numbers and testing and the process of creation. And why some guys are buying an Antelope 10m and after decades of thinking they were at the holy grail, they plug that into their rig, and it blows their minds. Again, in that neutral has truly begun to raise it's head lifting that last bit of darkness from the monitors you wouldn't have even known was there.

But, jitter I think once you are down to this spec:

"The reference clock of the Cranesong Solaris has less than 1ps (picosecond) jitter and employs a proprietary re-construction filter for accurate time domain response."

"The jitter measured for the clock, between our auditory spectrum of 10Hz - 20kHz, is typically around 0.045ps (44fs, femtoseconds)!"

Those quotes per Dave Hill at Cranesong with his evolution of his work with clocking is I think the most trustable from my standpoint around clocking and converters etc. And, the discussion isn't about anything with the clocking, it's what is surrounding it - design going in and coming out that truly is what people are hearing when they think they hear a difference.

I just got a Grace m900 for $499 and it uses a similar ADK chip as what Dave is using, I think those chips maybe are like $40-60 a piece - and in a $3000 converter or straight clock - that is hardly what makes the difference. These folks are incredible designers - the m900 to me I love because I know I can trust those Grace guys. And I know in both cases of these companies, they aren't mega corporations pushing next batch of new products to buy every year. Dave worked for decade(s) to get and refine his designs. And in pro audio I think that is what we want, to trust there isn't more to be gotten.

I am bias with Cranesong as I have Dave's amazing Hedd converter ((a $2000-3000 box) and the prior generation, just updated after years and years) which he would say, use it's clock, use it's converters, you don't need anything else. Yet, I still think no you need more gear, more-better expensive clocking etc. You need a Rubidium clock and an Antelope 10mx, but it gets to be basically a crazy pursuit. Again, I trust what some mixing guy is doing if he is putting out huge records like Havana and saying the secret to the warmth and mids is his 10mx. He is lucid, he is not on drugs. It's a tough call, but there is no way anyone or even the SOS article can just generalize.

I keep A/B'ing like the Grace m900 with it's ADK chip and Grace design standards to my 2005 Pioneer DV-563A for the DAC side of this equation, which has a Burr Brown DAC chip literally now well over a decade old. And on very basic average active monitors I can hear differences, but that Brown Burr chip and whatever Pioneer over spec'ed in these units that got discontinued because they had too much bang for the price of like $250, completely stands up and compares. Tiny bits of less mid-range smearing and just more of a FEELING of more power coming through and separation in the low end is what I hear with a modest but more modern DAC design, and not breaking the bank at all. And no clocking external wise at all in play.

The stuff really amazes me. Sorry, to dump like this I was away for a bit and missed the tail end of this discussion.

Really glad you cited that SOS article though.

best,

Dayz

 

[Dayz]

All well and good, in theory. Results vary in practice. So much so, that I don't want to take either position and that was my real point: There are variables to account for before unilaterally declaring either as better or worse.

My experience is primarily studio-based where cumulative jitter can become audible (and measurable). At home I prefer USB as an interface with the DAC as the master (RME ADI-2 Pro).

Exactly, you summed it up in two sentences.

 

[Wombat]

T

I generally will always go with SOS and I agree here we are talking about the smallest of differences in possible sound quality. There are so many variables, it's not just external or not external clocking. And, it's not just the clock stage that influences what people and most engineers hear with higher end clocks (external or in their high end AD or AD/DA) it's the audio stages and how well those are designed which is not a function of price, but in some cases there are some really well regarded mixing engineers who are hearing something that is not something that is just altering the sound from neutral. Neutral or truth of what was intended originally recorded and hearing that (if you can - if you are in a room, with monitoring that allows such clarity) is the goal. It's commonly words like "opening" and getting ever closer to the holy grail of finally truly feeling like you are in the room with an artist or singer, they are with you right there. And in the production process not letting all the little bits along the way that can stamp that out. That's the difference between the numbers and testing and the process of creation. And why some guys are buying an Antelope 10m and after decades of thinking they were at the holy grail, they plug that into their rig, and it blows their minds. Again, in that neutral has truly begun to raise it's head lifting that last bit of darkness from the monitors you wouldn't have even known was there.

But, jitter I think once you are down to this spec:

"The reference clock of the Cranesong Solaris has less than 1ps (picosecond) jitter and employs a proprietary re-construction filter for accurate time domain response."

"The jitter measured for the clock, between our auditory spectrum of 10Hz - 20kHz, is typically around 0.045ps (44fs, femtoseconds)!"

Those quotes per Dave Hill at Cranesong with his evolution of his work with clocking is I think the most trustable from my standpoint around clocking and converters etc. And, the discussion isn't about anything with the clocking, it's what is surrounding it - design going in and coming out that truly is what people are hearing when they think they hear a difference.

I just got a Grace m900 for $499 and it uses a similar ADK chip as what Dave is using, I think those chips maybe are like $40-60 a piece - and in a $3000 converter or straight clock - that is hardly what makes the difference. These folks are incredible designers - the m900 to me I love because I know I can trust those Grace guys. And I know in both cases of these companies, they aren't mega corporations pushing next batch of new products to buy every year. Dave worked for decade(s) to get and refine his designs. And in pro audio I think that is what we want, to trust there isn't more to be gotten.

I am bias with Cranesong as I have Dave's amazing Hedd converter ((a $2000-3000 box) and the prior generation, just updated after years and years) which he would say, use it's clock, use it's converters, you don't need anything else. Yet, I still think no you need more gear, more-better expensive clocking etc. You need a Rubidium clock and an Antelope 10mx, but it gets to be basically a crazy pursuit. Again, I trust what some mixing guy is doing if he is putting out huge records like Havana and saying the secret to the warmth and mids is his 10mx. He is lucid, he is not on drugs. It's a tough call, but there is no way anyone or even the SOS article can just generalize.

I keep A/B'ing like the Grace m900 with it's ADK chip and Grace design standards to my 2005 Pioneer DV-563A for the DAC side of this equation, which has a Burr Brown DAC chip literally now well over a decade old. And on very basic average active monitors I can hear differences, but that Brown Burr chip and whatever Pioneer over spec'ed in these units that got discontinued because they had too much bang for the price of like $250, completely stands up and compares. Tiny bits of less mid-range smearing and just more of a FEELING of more power coming through and separation in the low end is what I hear with a modest but more modern DAC design, and not breaking the bank at all. And no clocking external wise at all in play.

The stuff really amazes me. Sorry, to dump like this I was away for a bit and missed the tail end of this discussion.

Really glad you cited that SOS article though.

best,

Dayz

There is the subjective and undefined stuff creeping into the forum, yet again.

"mid-range smearing", "a FEELING of more power coming through", "secret to warmth and mids is his 10mx", "he is lucid, he is not on drugs", "most engineers hear".

 

[Dayz]

T


There is the subjective and undefined stuff creeping into the forum, yet again.

"mid-range smearing", "a FEELING of more power coming through", "secret to warmth and mids is his 10mx", "he is lucid, he is not on drugs"

Not for measurement sake. Just to explain perspective. It's subjective to generalize and have articles cited that aren't any more fact than anything else that is just subjective opinion. Next time I won't go off on that ramble. Apologies. Got caught up in being addicted to fun and learning ;-) Sorry. However, it is fact that highly regarded mixing engineer is not on drugs and is lucid, so... e.g. THIS portion was not "SUBJECTIVE STUFF" ;-)

 

[Wombat]

Not for measurement sake. Just to explain perspective. It's subjective to generalize and have articles cited that aren't any more fact than anything else that is just subjective opinion. Next time I won't go off on that ramble. Apologies. Got caught up in being addicted to fun and learning ;-) Sorry. However, it is fact that highly regarded mixing engineer is not on drugs and is lucid, so...

so.....so what?

 

[Blumlein 88]

@Dayz Hmmmm, I bet a good many mixing and mastering guys are on drugs, yes even when they work.

BTW, I have an Antelope Audio Zen Tour interface with one of their special oven clocks in it. I do my own recordings too.

I know how things feel the way you describe them. In my long experience however if you test those feelings, they mostly are a result of expectation bias from a multitude of sources. Our brain is so easily influenced. Our ability to measure things is well beyond our ability to hear. Yet feelings are another thing resulting from many things beyond just the sound entering our ears.

So if the Antelope 10MX is altering the sound via better clocking them why does it measure as worse timing than the on board crystal clock? And what is happening to create that sound that somehow isn't measured? The most likely explanation is the expectation of the users that this highly super accurate 10MX clock must be good. It would be really simple. Do some tracks with and without and see if these fellows can nail down that warm feeling when the 10 MX is in use and accurately tell us it is missing when there is no external clock.

The Crane Song stuff is good gear. Would have it myself were I wealthier. The news to me is much less expensive gear can perform well enough to sound just as good.

I don't have the files up currently, but I offered 8th generation copies and digital originals. The copies were sent thru a lowly Focusrite 18i20. People couldn't hear the difference when they weren't labeled. Tests like that can lead you to truth in audio better than feelings. Nevertheless, the feelings are powerful and difficult to over-rule for humans (myself included).