Korg Forums

What is the exact difference between these 2 sampling rates?
Considering Korg's new 1-Bit recording system with the MR-1000, I'm getting more interested in recording at these rates then the standard 48kHz.
Are most of you using these rates?
Storage seems not to be a problem anymore as 1 TB starts to go under 100.-€.
I'm curious to hear from all of you about this issue, specially in regard to mastering and the new Korg technology
Anashwaran

Anashwaran wrote:What is the exact difference between these 2 sampling rates?
Considering Korg's new 1-Bit recording system with the MR-1000, I'm getting more interested in recording at these rates then the standard 48kHz.
Are most of you using these rates?
Storage seems not to be a problem anymore as 1 TB starts to go under 100.-€.
I'm curious to hear from all of you about this issue, specially in regard to mastering and the new Korg technology
Anashwaran

Note that this is completely different from 1-bit.
The Parameter Guide (and Help system) have descriptions of these different options. As the manual describes, the 96kHz options use sample-rate conversion on input and output; the OASYS itself always runs at 48kHz. The only reason to use the 96kHz option, in my opinion, is if you are running a DAW at 96kHz and wish to use a direct digital connection from the OASYS.

Best regards,

Dan

Hi Dan,
Yes, I studied the manual now to check for the difference. ( I should have done it first ) But as you said I use Cubase 4 and can record with the Optical connection this quality. So with the 1-bit system of Korg it would be possible to stick with that quality until the Master-Mix.
Then there would be only 1 down-process for the CD instead of 2.
And this one can be more controlled regarding the samples. Is that correct?
Anashwaran

Anashwaran wrote:Hi Dan,
Yes, I studied the manual now to check for the difference. ( I should have done it first ) But as you said I use Cubase 4 and can record with the Optical connection this quality.

It's not a question of quality, but rather of convenience. The OASYS always runs at 48kHz. At the S/PDIF output, there's an option for sample-rate conversion to 96kHz. From the perspective of the OASYS, this does not offer an advantage in quality; instead, it just makes it easier to hook up to a system which is already running at 96kHz.

Personally, I still record at 44.1kHz, and simply use the analog outputs from the OASYS.

Anashwaran wrote:So with the 1-bit system of Korg it would be possible to stick with that quality until the Master-Mix.

How would the 1-bit recorders affect your use of Cubase?

Anashwaran wrote:Then there would be only 1 down-process for the CD instead of 2.

What would the second "down-process" be?

- Dan

Dear Dan,
I was a little bit shocked when you say there is no difference in quality between 48 and 96kHz, but you must know it.
I was under the illusion that the sound changed on the Oasys, that it was more transparent, pristine, or more spacy. So that was just an autosuggestion. But it is logical.
Now the second point about the 1-bit recording:
If you go to the Korg Website (that you know probably better than me) :

http://www.korg.com/mr/Future_Proof_Rec ... lained.pdf
it is said that:

After recording and editing all your tracks in the DAW/hardware system of your choice, you can mix directly, or via an analog summing mixer, to the MR-1000. This gives you the superb fidelity of high-rate 1-bit technology, which preserves all the transients, nuances, and “air” from the original recording

and further:

The 1-bit advantage
In a 1-bit system, the audio is recorded at super-high sample rates, commonly 2.8224 MHz, and now with the MR-1000, 5.6448 MHz. At this high rate a 1-bit system is able to reproduce frequencies from DC up to 100 kHz, which exceeds all other digital systems and even magnetic tape, which can reproduce up to 50 kHz. At these high rates there is no longer a need for steep filters, which removes a possibly coloring element in the encoding chain.
Future
Even better, by remaining in the 1-bit format that the converter already used, there is no need for the decimation filter process during recording, and no need for the interpolation and oversampling filter processes during playback. So what comes in goes out, with no extra math in the process, thus eliminating any need for the data/audio to change. Remember, decimation filters and their design have a major influence on the sound with PCM recording, and 1-bit recording eliminates the need for them.

This means that if you record in a DAW directly with 96 kHz you stay at that level up to the Master Mix on the 1 Bit level. Then this Mix has to be brought down to 44kHz by a different interpolation- procedure as is done inside the Oasys ( although this is not bad) or inside normal programs and Daws.

Both MR recorders use state of the art converters: the Burr Brown PCM4202 for A/D conversion,and the Cirrus Logic CS4398 for D/A conversion.

So there will be only 1 process of conversion on the best possible level. Or a Master-Studio can do the Mix from the recorded 1-Bit version with much better algorithms than we can do ourselves.
I start making publicity here for Korg, but the idea attracts me. I might still be in a big illusion, who knows?
ANashwaran

Anashwaran wrote:Dear Dan,
I was a little bit shocked when you say there is no difference in quality between 48 and 96kHz, but you must know it. I was under the illusion that the sound changed on the Oasys, that it was more transparent, pristine, or more spacy.

Sample-rate conversion adds another processing step to the output. A purist might choose to avoid the extra step. However, as noted, it might be convenient if you are running your DAW at 96kHz, and wish to connect digitally.

Also, note that the "96k Enhanced" setting includes a high-frequency exciter effect. As noted in the manual and help, this also affects the main analog L/R and headphone outputs. So, this will alter the timbre of the output.

I understand the advantages of 1-bit recording. It's currently the best means available to capture a stereo analog signal. If you are mixing in the analog domain, using an analog mixing board, analog summer, analog outboard gear, etc., then it makes good sense to record the resulting stereo analog signal in ultra-high-quality, future-proof 1-bit. The same goes for live 2-track recordings.

In this case, it would probably be best to simply leave the OASYS in the analog domain (unless you are recording multiple passes of the OASYS), and record it as part of the analog mix - rather than first recording its sample-rate-converted digital output.

On the other hand, if you're mixing "in the box" - that is, by bouncing the final mix within Cubase - then sending the mix out to analog (which is necessary to record as 1-bit) requires an additional chain of A/D and then D/A.

Best regards,

Dan

Thanks Dan, for the clear explanations.
Will that extra step from a Daw at 24bit 96 kHz to the 1Bit not keep that quality ? And is that not possible on a digital level?
Another question:
I record the voice normally with a condensator-mic which goes into the Oasys ( + phantom ) and here I get a very good quality with all the effects of the Oasys. This now goes digitally (optical) into the DAW. Until now I used 48kHz. Will the quality here change with 96 kHz?
Thanks for more
Anashwaran

Anashwaran wrote:Thanks Dan, for the clear explanations.
Will that extra step from a Daw at 24bit 96 kHz to the 1Bit not keep that quality ?

If you already have a stereo mix in the DAW, recording it into a 1bit recorder would have no advantage, and would presumably slightly degrade the audio due to the extra D/A and A/D stages. If you're mixing in the analog domain, using an analog mixer or summer, analog processing, etc., then the 1-bit recorder would offer a quality advantage.

Anashwaran wrote:And is that not possible on a digital level?

Do you mean, is it possible to digitally convert from linear PCM to 1-bit? I guess that this would be possible in software, but I don't think that it would offer any practical advantage.

The whole point of 1-bit recording is that it's a better representation of the analog signal. Once the signal is already in digital form, there's no advantage in re-encoding it.

Anashwaran wrote:I record the voice normally with a condensator-mic which goes into the Oasys ( + phantom ) and here I get a very good quality with all the effects of the Oasys. This now goes digitally (optical) into the DAW. Until now I used 48kHz. Will the quality here change with 96 kHz?

The OASYS always runs at 48kHz; 96kHz is only done by sample-rate conversion. See my comments above; the same caveats apply whether you're using the synth features or the audio inputs.

Best regards,

Dan

Some more illusions gone.
Thanks for your time.
But it will save also my time, because I will not have to redo some of my recordings if things are as you discribe.
And I might even save some money!
Thanks again for the rectifications
ANashwaran

My apologies for the belated reply to this old thread, but this is an under explored topic.

What is the nature of the "enhancer" algorithm on the sample rate converter on the SPDIF port ? Is this a dithering process or, is it something else ?

I ask, because I noticed a fairly dramatic improvement when I last tried this. It is almost worth running at the higher sample rate just for this effect, so I am very curious.

Perhaps there is a way to emulate this effect within the Oasys w/o bumping up the SR ?

Sincerely,
Craig Hansen-Sturm

chansensturm wrote:My apologies for the belated reply to this old thread, but this is an under explored topic.

What is the nature of the "enhancer" algorithm on the sample rate converter on the SPDIF port ? Is this a dithering process or, is it something else ?

I ask, because I noticed a fairly dramatic improvement when I last tried this. It is almost worth running at the higher sample rate just for this effect, so I am very curious.

Perhaps there is a way to emulate this effect within the Oasys w/o bumping up the SR ?

Sincerely,
Craig Hansen-Sturm

It is a high-frequency exciter/enhancer, similar to the Stereo Exciter/Enhancer effect.

- Dan

Thanks Dan.

While remaining at 48kHz, would you care to hazard a guess as to the settings required in the Exciter/Enhancer effect to achieve a similar effect to the 96k setting?

jg::

Excellent question jg::,

I would be very interested in an estimation of that setting as well, if Dan didn't mind putting his thoughts to it. I happen to be at a point in my mastering for my fisrt CD that is pretty crucial. I am utilizing quite an array of other keys besides the Oasys but the "O" is at the very hub of my studio. There is also a modest amount of processing gear that I am putting to use. I have never really considered myself much of a mix man but have always managed to get my recordings a notch or two above the "basement tape" demo. In recent months I got my studio sort of "dialed in" to a point where the dynamics are good, EQ'ing is real, real close (though I tend to have a "hollowed out" midrange at times and boosting those frequencies doesn't always yield the expected results somehow), but in general it doesn't really sound like a rookie (like I am) did the recording/ mixdown.

All of my stuff is copyrighted so I'm gonna provide some samples of what is going on the CD (entitled U Can Dream). I am hoping to get some good criticism (other than just getting ragged on because someone didn't like my arrangement or particular melody line or style of music) that will provide some suggestions for improving my overall mix.

In any case, I'll keep an eye open for Dan's suggestion if he choses to add anything about your query.


Kind regards,
UCanDream