• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

AUDIO INTERFACES with frequency response reaching 100kHz? (or close to)

GeoffAturax

New Member
Joined
Nov 18, 2021
Messages
4
Likes
2
Hi,

I'm looking to purchase either the Sanken CO-100K microphone, or something similar, to capture material for sound design. The main reason being that it will enable me to capture high frequency content and have this preserved/revealed when pitching sounds down inside of my DAW. Nothing super complex or innovative, but it's something I've never had the opportunity to do until now.

Another first for me is that I am having to carefully examine potential audio interfaces with regards to their ADC frequency response. I want an interface that will allow for the capture of a similar frequency range as the CO-100K will allow. Am I correct in thinking that there is a direct correlation here; that the upper limit of an interface's frequency response will effectively shave off, to varying degrees, any incoming frequency content above it?

If so, then I am looking for interfaces which will allow me to utilise the CO-100K to its full potential and I'd be grateful if you were able to recommend (or just point me in the direction) of any interfaces you think might be a good fit.

Thanks in advance :)
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,827

Hmm. You need an ADC that can handle sample rates of at least 192kHz, giving you a range of 96kHz for audio, and a microphone preamp to match. They both have to be really low noise given how little energy there will be >20kHz. However your mic has somewhat unencouraging noise specs itself: 22dBA. What does noise look like at 100kHz? Good question to ask the company.

So for example the RME ADI-2 Pro and Millenia HV-3C.

The RME Micstasy is an all-in-one: https://www.rme-audio.de/micstasy.html

I'm not sure what mics are out there that reach that high reliably. Lab companies like GRAS and B&K do have capable mics, but they aren't for pro use and are very expensive.

I'm also sure that you'd need to filter the inputs. You'll have to use a lot of gain for ultrasonics and if you can't apply a shelf for normal audio range sounds you might cause clipping. The RME products can do that, for example.
 

Grooved

Addicted to Fun and Learning
Joined
Feb 26, 2021
Messages
679
Likes
441
If 96kHZ audio is enough, like said a 192kHZ interface is OK
If you really want to push to 100kHz, you will need a 384 or 756kHz one like the RME ADI-2 PRO FS or the Antelope Amari (I would pick the RME, lower cost and a sure thing).
Or last option, not all-in-one, a good mic pre-amp and the Cosmos ADC
 
OP
G

GeoffAturax

New Member
Joined
Nov 18, 2021
Messages
4
Likes
2

Hmm. You need an ADC that can handle sample rates of at least 192kHz, giving you a range of 96kHz for audio, and a microphone preamp to match. They both have to be really low noise given how little energy there will be >20kHz. However your mic has somewhat unencouraging noise specs itself: 22dBA. What does noise look like at 100kHz? Good question to ask the company.

So for example the RME ADI-2 Pro and Millenia HV-3C.

The RME Micstasy is an all-in-one: https://www.rme-audio.de/micstasy.html

I'm not sure what mics are out there that reach that high reliably. Lab companies like GRAS and B&K do have capable mics, but they aren't for pro use and are very expensive.

I'm also sure that you'd need to filter the inputs. You'll have to use a lot of gain for ultrasonics and if you can't apply a shelf for normal audio range sounds you might cause clipping. The RME products can do that, for example.

Brilliant info, thank you.

Excellent point regarding the low noise requirement, given the reduced energy higher up the spectrum. I had not given that much thought as of yet.

Question: With regards to the sample rates of ADCs, I understand the theory behind needing 192kHz to capture 96kHz, but I was under the impression that the advertised max sample rate of an audio interface does not automatically mean that it can handle such frequencies. Am I wrong about this? Is it really that simple or are there other potential limitations (particularly with cheaper devices) that may reduce the actual capabilities of an interface in real-world usage?

The Sanken CO-100K is not a definite yet. It's really the first microphone with such range that I have come across, but I shall keep looking. I'll be sure to ask the company about the noise specs as well. A quick search around online has brought no results in the way of a mic test.

The RME ADI-2 Pro looks a fine choice - particularly given the shelving off of normal audio feature as you say.

Many many thanks for your insight, it will set me off on a much better path than otherwise would have been!
 
OP
G

GeoffAturax

New Member
Joined
Nov 18, 2021
Messages
4
Likes
2
If 96kHZ audio is enough, like said a 192kHZ interface is OK
If you really want to push to 100kHz, you will need a 384 or 756kHz one like the RME ADI-2 PRO FS or the Antelope Amari (I would pick the RME, lower cost and a sure thing).
Or last option, not all-in-one, a good mic pre-amp and the Cosmos ADC

Thanks!

You have experience with the RME? Up until now, I have been restricted to interfaces in the range of $400-$500. I'm sure the improved recording results are in diminishing returns as you ascend up the price ranges, but I am still very excited to hear what this new realm of equipment will sound like.
 
OP
G

GeoffAturax

New Member
Joined
Nov 18, 2021
Messages
4
Likes
2

Grooved

Addicted to Fun and Learning
Joined
Feb 26, 2021
Messages
679
Likes
441
As of the DACs based on AKM chips (e.g. the RME), I would be cautious about checking their performance at 384+kHz first. Experience in thread https://www.diyaudio.com/forums/equ...ampling-speeds-oversampling-filter-issue.html
If I'm not wrong, the latest RME devices are build with ESS chip, not AKM
But you will need to check with the seller which one they sell.

Thanks!

You have experience with the RME? Up until now, I have been restricted to interfaces in the range of $400-$500. I'm sure the improved recording results are in diminishing returns as you ascend up the price ranges, but I am still very excited to hear what this new realm of equipment will sound like.
Used a lot of RME and other brands, and the first thing that comes to mind with RME has always been rock solid drivers. Never had a problem.
 

Trell

Major Contributor
Joined
May 13, 2021
Messages
2,752
Likes
3,285
I f I'm not wrong, the latest RME devices are build with ESS chip, not AKM
But you will need to check with the seller which one they sell.

The RME ADI-2 PRO is still using the AKM chip. So far only the RME ADI-2 DAC has changed to an ESS chip.
 

phofman

Senior Member
Joined
Apr 13, 2021
Messages
489
Likes
319
The aforementioned thread compares AKM4490/3 at oct and hex modes to ES9038Q2M 128xFs. There were three independent measurements of AKM by three posters, with basically same results. I would be happy if the AKM issue was investigated more. Maybe there was some measurement flaw, but I do not know of any.
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,827
With regards to the sample rates of ADCs, I understand the theory behind needing 192kHz to capture 96kHz, but I was under the impression that the advertised max sample rate of an audio interface does not automatically mean that it can handle such frequencies. Am I wrong about this? Is it really that simple or are there other potential limitations (particularly with cheaper devices) that may reduce the actual capabilities of an interface in real-world usage?
You can, for example, check @Archimago's blog, who reviewed the RME: http://archimago.blogspot.com/2018/09/measurements-rme-adi-2-pro-fs-adc.html And then Amir's: https://www.audiosciencereview.com/...nd-measurements-of-lynx-hilo-adc-part-2.3596/
As of the DACs based on AKM chips (e.g. the RME), I would be cautious about checking their performance at 384+kHz first. Experience in thread https://www.diyaudio.com/forums/equ...ampling-speeds-oversampling-filter-issue.html
@MC_RME Any comment?

I don't see why this problem would concern the ADCs, unless I'm missing something.
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,690
Likes
6,013
Location
Berlin, Germany
You can, for example, check @Archimago's blog, who reviewed the RME: http://archimago.blogspot.com/2018/09/measurements-rme-adi-2-pro-fs-adc.html And then Amir's: https://www.audiosciencereview.com/...nd-measurements-of-lynx-hilo-adc-part-2.3596/

@MC_RME Any comment?

I don't see why this problem would concern the ADCs, unless I'm missing something.
As of the DACs based on AKM chips (e.g. the RME), I would be cautious about checking their performance at 384+kHz first. Experience in thread https://www.diyaudio.com/forums/equ...ampling-speeds-oversampling-filter-issue.html
The main observation was that the AK4490/AK4493 always switch to a (low-pass filtered) NOS-like setting above 192kHz regardless of what digital filter has been set up. Those only are operational up to 4x speeds (176.4kHz and 192kHz).

For music playback this is totally irrelevant, though, because i) there will be -- or better said, there should be -- only little, if any, content above 20kHz anyway in and ii) we still have the analog lowpass filter which in case of the ADI-2 Pro/DAC is set to about 120kHz corner frequency. So only little imaging products will be generated and it's all so high above 20kHz that it simply doesn't matter.

It does matter for some specialist measurements tasks, for example you simply cannot generate a nice pure sine with low distortion/artifacts at, say, 150kHz.

The ADC is not affected as far as I can see but I did not test this thoroughly to make any solid statement.
 

phofman

Senior Member
Joined
Apr 13, 2021
Messages
489
Likes
319
For music playback this is totally irrelevant, though, because ...

It does matter for some specialist measurements tasks, for example you simply cannot generate a nice pure sine with low distortion/artifacts at, say, 150kHz.
Exactly, that's why I pulled it out it only after 384kHz samplerate and 100kHz frequencies were mentioned :)

The ADC is not affected as far as I can see but I did not test this thoroughly to make any solid statement.
DAC runs at the same samplerate, so high-bandwidth sampling will "unfortunately" switch the DAC to NOS mode.
 
Last edited:

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
Completely bypassing the oversampling digital filter past 192 kHz in DACs seems to be standard practice in the industry. If ESS don't do that, they're the exception.

Now, as for the ADC side, not only is analog bandwidth a concern, but antialias filters commonly get rather sloppy at quad speed as well, leaving an aliasing-free bandwidth that might only be 0.25 to 0.3fs (= fs - filter stopband). Parts that get to about 0.45fs even at 192 kHz (~86.4 kHz) include AK5394, AK5385, PCM4220/4222 (all high-end or at least upper midrange in their day), and that's about it. AK55xx at 384 kHz would also get the job done (~107 kHz), and slightly better at 768 kHz (~128 kHz). No idea about ESS filter performance at quad speed, their datasheet briefs do not care to mention such but I assume running at 384k/768k would work.

I am quite convinced that you wouldn't have to splurge on an HV-3C or an Earthworks ZDT amp by all means... but plain super linear, wide bandwidth mic preamps with no adornments are sort of rare. The RME QuadMic II might do the trick, it's specified to 200 kHz at -0.5 dB. Heaven knows whether it'll do that over the whole gain range, but given the mic's high sensitivity you shouldn't need to crank it much either.

So QuadMic II + Cosmos ADC is a combo I might look into.
 
Top Bottom