Which DAC measurements do you like to see next?

[amirm]

AUNE S16 / S6, especially S16 because S6 is after some cuts off to save money, no FIFO reclocking, weak headphone amp etc, to my ears S6 is also much worse DAC than previous S16 (though hard to get these days, but comparison of these 2 units would be great !)

Welcome to the forum. I purchased the S6 from massdrop a while back. It says it ships on June 1. The S16 is quite pricey so I don't know that I will be buying it soon. If you are located in US and are willing to loan it to me, I am happy to measure and compare the two.

 

[maxxevv]

The listing looks wrong, that seems to be the Non-DSD version. It doesn't have a DAC onboard.

The NX4 DSD has a big DSD logo on it. (the original NX4 was a pure amplifier only)

 

[dc655321]

I have been waiting for a sale to buy the NX4. Just saw this ebay listing which is $40 cheaper than other sources. Anyone know if this is trustworthy? https://www.ebay.com/i/173121695267?chn=ps

I agree with @maxxevv that this is not the DAC you are looking for

If you want to measure the nx4 dsd, I would be happy to send mine to you. If you want to buy one, we can discuss that too ;-)

 

[amirm]

Thanks guys. I am glad I asked first.

If you want to measure the nx4 dsd, I would be happy to send mine to you. If you want to buy one, we can discuss that too ;-)

That would be great as I have no need to own one. Please start a conversation with me and I will give you the shipping info. Much appreciated.

 

[duo8]

The listing looks wrong, that seems to be the Non-DSD version. It doesn't have a DAC onboard.

The NX4 DSD has a big DSD logo on it. (the original NX4 was a pure amplifier only)

NX4 is a DAC, but obviously not the same as the NX4 DSD.
For some reason if a topping portable amp ends with a even number it will have DAC feature.

 

[garbulky]

I couldn't figure out what thread to post this on. I see that DAC technology is currently at very impressive levels in terms of specs.
I would be interested on seeing some articles about the current state of recording hardware namely microphones - both for price tranges within our reach and price ranges that are very expensive.
I would also be interested in seeing technical measurements of microphones. I am hoping that we can use this to get some idea of the best microphones out there in affordable price ranges.

 

[amirm]

I have not studied microphone measurements but I would think it would need some kind of isolated/anechoic chamber. And calibrated (ANSI traceable?) source. I would say it is outside of the scope of what we can do here for a good while.

 

[DonH56]

Measuring mics is a PITA. Most good microphone manufacturers have pretty complete measurements on their websites.

 

[dheepak10]

The NX4 DSD is a single chip ES9038Q2M, the D50 is a dual chip implementation. Not the same.

http://en.tpdz.net/products_detail/productId=25.html

http://en.tpdz.net/products_detail/productId=37.html

Thanks! This helps! When the review comes out, if NX4 performance is not as great as D50, then I might opt for D50 along with a decent headphone amp.
Noob Question - BTW why 2 ES9038Q2M chips? Is each channel processed separately in the D50?

 

[amirm]

Noob Question - BTW why 2 ES9038Q2M chips? Is each channel processed separately in the D50?

By paralleling channels, you get improved signal to noise ratio.

 

[tecnogadget]

What about the iBasso DX90 ?? I use it as a DAC coupled thought Line Out with an SMSL SA60 50w amplifier. It has ESS9018K2M in dual mono fashion.

 

[dc655321]

By paralleling channels, you get improved signal to noise ratio.

But, in this case (2 units) isn't it only a 3 dB improvement?
Not sure I understand where the improvement comes from...

 

[amirm]

Not sure I understand where the improvement comes from...

The signal adds up constructively but the noise being random, does not.

 

[Blumlein 88]

But, in this case (2 units) isn't it only a 3 dB improvement?
Not sure I understand where the improvement comes from...

To elaborate on Amir's reply. Imagine that noise is random at any given frequency (which of course it is). So to pick a frequency let us say 1000 hz. Noise at one thousand hertz in one device might be out of phase or a slightly different level than in another device. So noise at 1000 hz will usually increase, but not to double the amount. Because the phase and level are variable. On average over a wider band the noise level of two equally noisy devices will go up by 3 db.

If however we are feeding the same 1000 hz tone to each device those will be in phase and same intensity and will add to fully double the level or 6 db. So noise increases 3 db because it is random in level and phase. Our desired signal doubles to 6 db because it is not random. We gain a total of 3 db in signal to noise ratio.

 

[Blumlein 88]

I couldn't figure out what thread to post this on. I see that DAC technology is currently at very impressive levels in terms of specs.
I would be interested on seeing some articles about the current state of recording hardware namely microphones - both for price tranges within our reach and price ranges that are very expensive.
I would also be interested in seeing technical measurements of microphones. I am hoping that we can use this to get some idea of the best microphones out there in affordable price ranges.

Well this might sort of give you a little faith in mics though not answering your questions.

Here are two different $150 each Avantone CK-1 microphones. I measured a speaker sitting in a room so the response variation is from the speaker. I put each microphone as close to the same exact position as possible. They look well matched to me.

Same position for three Shure KSM 32 microphones which were made different years spanning a decade.

Some relatively inexpensive CAD M179 microphones made a few months apart.

Here I compare several different mikes to a UMIK1 measuring microphone. I generated calibration curves for the microphones from the published frequency response charts. While not completely identical they are generally close. Much of the difference is probably from trying to position a small pencil mic diaphragm in the same spot as a large condenser microphone diaphragm. Ignore differences below 200 hz as some of these are omni's, some cardioids, and some figure-8 patterns. Each of those patterns have variable proximity effect below 200 or 300 hz.

 

[dc655321]

To elaborate on Amir's reply. Imagine that noise is random at any given frequency (which of course it is). So to pick a frequency let us say 1000 hz. Noise at one thousand hertz in one device might be out of phase or a slightly different level than in another device. So noise at 1000 hz will usually increase, but not to double the amount. Because the phase and level are variable. On average over a wider band the noise level of two equally noisy devices will go up by 3 db.

If however we are feeding the same 1000 hz tone to each device those will be in phase and same intensity and will add to fully double the level or 6 db. So noise increases 3 db because it is random in level and phase. Our desired signal doubles to 6 db because it is not random. We gain a total of 3 db in signal to noise ratio.

Thanks for the extra clarity, @Blumlein 88. Much appreciated.
I was imagining the topology as leading to something like one would get with so-called "balanced" headphone outputs:
(Vx + Noise) - (-Vx + Noise) = 2Vx

Sounds roughly similar...

OTOH, given it's not uncommon to have SNR significantly greater than 100 dB, does an extra 3 dB really matter?
I mean, other than to marketing folks...

 

[SIY]

Measuring mics is a PITA. Most good microphone manufacturers have pretty complete measurements on their websites.

I've done a lot of mike measurement and, in fact, just finished my fifth mike measurement review for AudioXpress. It's actually not that hard if you understand the processes and keep your wits about you. Of course, it helps a lot to have an APx1701 on hand...

By and large, you don't need anything fancy for 1kHz and up, and can do pretty well below that as well. Noise floor is only slightly trickier.

 

[DonH56]

I've done a lot of mike measurement and, in fact, just finished my fifth mike measurement review for AudioXpress. It's actually not that hard if you understand the processes and keep your wits about you. Of course, it helps a lot to have an APx1701 on hand...

By and large, you don't need anything fancy for 1kHz and up, and can do pretty well below that as well. Noise floor is only slightly trickier.

In the primordial past (college, early 1980's) I used a fairly small box (about an 18"~24" cube IIRC) lined inside with absorption to measure near-field performance over most of the frequency range. Bass response was tough without a real anechoic chamber (which I had access to at the time; these days, the local college has a nice one, but I'd have to pay to use it or rope one of my friends on staff to help). The box when sealed had a pretty low noise floor after I wised up and suspended the measurement box within a larger box with thick walls and so forth to better isolate the measurement box from the outside world. And tended to test later at night when the building was quieter.

The bigger problem for me was getting a source that was known good over a wide range of frequency and dynamics. The good measurement mics often outdid the drivers used to send test signals to the mics. I had some piezo and compression drivers that were OK from maybe a few hundred Hz but sub-100 Hz I am not sure I was ever satisfied I knew what was limiting the measurements. My Earthworks was good to 5 Hz and at that time I had no good low-distortion driver flat to 5 Hz around.

But again, not my day job, so I was probably just being ignorant, or stupid, or both...

 

[DonH56]

Thanks for the extra clarity, @Blumlein 88. Much appreciated.
I was imagining the topology as leading to something like one would get with so-called "balanced" headphone outputs:
(Vx + Noise) - (-Vx + Noise) = 2Vx

Sounds roughly similar...

OTOH, given it's not uncommon to have SNR significantly greater than 100 dB, does an extra 3 dB really matter?
I mean, other than to marketing folks...

The signal goes up 2x (6 dB) but the noise is (hopefully) uncorrelated (random) so you have to RSS it instead of doing a linear subtraction: Nout = sqrt(Np^2 + Nn^2). The net result is signal goes up by 6 dB, noise by 3 dB, so you get a 3 dB noise advantage for each doubling in signal (including when you use a differential connection).

You do need to take some care as not all noise is uncorrelated, e.g. power supply noise, ground loop noise, etc. may scale with signal (especially in a single-ended environment).

Whether >100 dB SNR matters depends upon your system and ears.

 

[garbulky]

Well this might sort of give you a little faith in mics though not answering your questions.

Here are two different $150 each Avantone CK-1 microphones. I measured a speaker sitting in a room so the response variation is from the speaker. I put each microphone as close to the same exact position as possible. They look well matched to me.
View attachment 12810

Same position for three Shure KSM 32 microphones which were made different years spanning a decade.

View attachment 12811

Some relatively inexpensive CAD M179 microphones made a few months apart.

View attachment 12812

Here I compare several different mikes to a UMIK1 measuring microphone. I generated calibration curves for the microphones from the published frequency response charts. While not completely identical they are generally close. Much of the difference is probably from trying to position a small pencil mic diaphragm in the same spot as a large condenser microphone diaphragm. Ignore differences below 200 hz as some of these are omni's, some cardioids, and some figure-8 patterns. Each of those patterns have variable proximity effect below 200 or 300 hz.

View attachment 12813

Very nice. I do some very amateur recordings. I try to record in stereo as much as I can.
I use two MXL V67 g

An M-Audio Luna