Review and Measurements of Holo Audio May --- Probably the best discrete R2R DAC

[Esotechnik]

Correct me if my math is off, but isn't that like 40MHz for a 8Vpp signal?

Sorry, mistyping.
http://www.msbtech.com/support/DAC3_Manual.pdf
"Balanced output: 7.5 V RMS (20Vpp)
Slew rate: >1000 V/microsecond
Settling time: <90 nanoseconds
Dynamic range: 136 dBA measured
Noise floor: < -145 dB"

 

[Killingbeans]

But still, at 20Vpp the 1000V/us is what you'd need for 8MHz bandwith? Quite a bit overkill for audio, or am I wrong?

EDIT: 22KHz at 20Vpp only needs 3V/us to be unharmed, and for something crazy like 100KHz, 13V/us is enough... just saying.

I'd really like to see an MSB DAC get tested. I'm far from an expert, but from what I've gathered so far they seem be a little "creative" when getting numbers from FFT measurements. I could be wrong though

 

[Tks]

It's impressive in the same way as someone paddling a kayak across the Atlantic instead of buying a plane ticket. There's no benefit to the end user from them choosing to do things the hard way.

I actually disagree here in the same way a mechanical watch holds value for some folks.

Likewise the actual internals of this thing are a sight to behold.

Now is it $4,000 good? That, I can't say, seeing as it's the best R2R I've seen in all respects by a mile.

I agree with what you say in principle, but the primary goal of someone buying this isn't outright sound quality, but this brings it(and competes with Dac chips), and then the luxury furnishing expected of a device in this price bracket.

 

[maxxevv]

I actually disagree here in the same way a mechanical watch holds value for some folks.

Likewise the actual internals of this thing are a sight to behold.

Now is it $4,000 good? That, I can't say, seeing as it's the best R2R I've seen in all respects by a mile.

I agree with what you say in principle, but the primary goal of someone buying this isn't outright sound quality, but this brings it(and competes with Dac chips), and then the luxury furnishing expected of a device in this price bracket.

I have to agree in the parallels of it with watches and the technologies used.

The best non-network corrected watches are from Seiko and Citizen. Top rate quartz movements with accuracy that measures in at under 10 second deviation per year. Those are about US$2k ~3k a piece.
Cheap but properly made quartz movements typically keep within 2~5 seconds/ day

The very best of serially produced, mechanically actuated watches that have by far the best time keeping accuracy are probably the Omega co-Axial movement and Seiko's SpringDrive. Both do about 20 ~ 50 seconds in a month depending on model specifications. Those are US$4.5k and up.
Most run of the mill affordable mechanicals will do good to be within 5~10s / day or about 150s~300s / month.

They can't really compare to the best quartz in terms of accuracy. But there is an innate romanticism to them for people who choose to buy them.

Its a similar thing with DS chips versus R2R step ladder.

 

[slingshot]

I only wear a watch when I am on an airplane and my cell phone is turned off.

 

[scott wurcer]

I have to agree in the parallels of it with watches and the technologies used.

The best non-network corrected watches are from Seiko and Citizen. Top rate quartz movements with accuracy that measures in at under 10 second deviation per year. Those are about US$2k ~3k a piece.
Cheap but properly made quartz movements typically keep within 2~5 seconds/ day

I don't get the watch thing, I don't care what time it is plus minus 10 min. or so ever. I have not owned a watch for 50yr.

 

[maxxevv]

I don't get the watch thing, I don't care what time it is plus minus 10 min. or so ever. I have not owned a watch for 50yr.

Haha ... the vast majority of people out there don't care either. They just want to know the time when they need to, a wall clock, the handphone or computer screen are just fine.

Its a lot like the vast majority of people out there don't care how their radio sounds as long as they can enjoy their favourite radio programs or when they favourite song comes along in the broadcast.

 

[mt196]

I have to agree in the parallels of it with watches and the technologies used.

The best non-network corrected watches are from Seiko and Citizen. Top rate quartz movements with accuracy that measures in at under 10 second deviation per year. Those are about US$2k ~3k a piece.
Cheap but properly made quartz movements typically keep within 2~5 seconds/ day

The very best of serially produced, mechanically actuated watches that have by far the best time keeping accuracy are probably the Omega co-Axial movement and Seiko's SpringDrive. Both do about 20 ~ 50 seconds in a month depending on model specifications. Those are US$4.5k and up.
Most run of the mill affordable mechanicals will do good to be within 5~10s / day or about 150s~300s / month.

They can't really compare to the best quartz in terms of accuracy. But there is an innate romanticism to them for people who choose to buy them.

Its a similar thing with DS chips versus R2R step ladder.

Yeah, I buy automatic watches because I like the technology that is behind them, it sorta feel a throwback in the past other than they look good. But I won't go around saying they keep time better than my phone as r2r fans usually do for that type of dacs

 

[JohnYang1997]

Yeah, I buy automatic watches because I like the technology that is behind them, it sorta feel a throwback in the past other than they look good. But I won't go around saying they keep time better than my phone as r2r fans usually do for that type of dacs

They claim better sound, better detail, better resolution, more organic, more analogue.

 

[Esotechnik]

Samples from R2R DACs (and SAR ADCs) is not or less auto-correlated. Old type DACs has less jitter sensitivity than D-S.

But still, at 20Vpp the 1000V/us is what you'd need for 8MHz bandwith? Quite a bit overkill for audio, or am I wrong?

Passive output (without any active components), directly from R2R matrices can drive headphones >300 Ohm. At a maximum sampling frequency of 5MHz, audio amplifiers will give high IMD. Parasitic cable parameters provide some LC filtering. For example teflon + copper = hi-end

 

[Milt]

Sorry, Wolf, if this has been asked previuosly, but did you listen to May?

 

[WolfX-700]

Sorry, Wolf, if this has been asked previuosly, but did you listen to May?

of course

 

[Killingbeans]

Passive output (without any active components), directly from R2R matrices can drive headphones >300 Ohm.

But why would you do that? Won't the output impedance be several hundreds of ohms, giving massive anomalies in the frequency response?

At a maximum sampling frequency of 5MHz, audio amplifiers will give high IMD. Parasitic cable parameters provide some LC filtering. For example teflon + copper = hi-end

I'm a little confused about this. Are you saying that the MSB DACs has no output buffer?

Can you point to some measurements showing R2R DACs with no output buffer having less IMD than those with a buffer?

 

[Milt]

of course

And? Did you like it?

 

[WolfX-700]

And? Did you like it?

“After making a measurement that excites me. I think letting me evaluate its sound will only lead to unbelievable results. ”
This is my previous reply.

But just do I like its sound? I think I like it!

 

[Milt]

“After making a measurement that excites me. I think letting me evaluate its sound will only lead to unbelievable results. ”
This is my previous reply.

But just do I like its sound? I think I like it!

Hey I read Art Dudley, Herb Reichert, Michael Fremer & TAS, I'm well-versed in unbelievable results

 

[Esotechnik]

Can you point to some measurements showing R2R DACs with no output buffer having less IMD than those with a buffer?

https://www.audiosciencereview.com/forum/index.php?threads/msb-dacs-measurements.9112/

 

[maxxevv]

https://www.audiosciencereview.com/forum/index.php?threads/msb-dacs-measurements.9112/

Did you actually digest what was discussed there or are you quoting verbatim blindly ?

Those are tall claims unsubstantiated by anyone other than the manufacturer. And besides, is there a measuring device that actually measures down to those levels claimed ?

 

[Esotechnik]

And besides, is there a measuring device that actually measures down to those levels claimed ?

16-bit picoscope after gain and/or rejecting test sine.

I think there should be bit accuracy of at least 14 bits in the 24-bit DAC->SAR ADC signal path after magnitude scaling, time offset and DC removal.
ADC and DACs for measurement have "no missing codes" feature.
Delta-sigma have idle tones, limit cycles and "all missing codes" after multi-stage noise shaping.

"To give one example, you could take a DAC and do something very classical, like sweep the level of a sinusoidal signal from full scale to nothing, and then look to see how distortion changes with signal level. You might find some minuscule squiggles at lower levels and shrug them off as measurement errors, like, “OK, that is just the machine not correctly measuring noise.” But I got suspicious at some point and said, “Hang on, let me try to find explicitly whether something happens in the noise floor with the signal modulation, but then I have to do so without a signal present. How do you do that?” Well, you sweep a DC input to a DAC. You feed it a constant code, some small value, and measure the noise. Increase that code and repeat. Suddenly you’ll find that some of these D-to-A converters will do these frightening things, like the noise floor suddenly shooting up or an audible whistle actually just walking through the audioband as you sweep, going from supersonic down to zero and then back up."
https://www.soundstageultra.com/ind...s-of-mola-mola-hypex-and-grimm-audio-part-one

 

[maxxevv]

So what's the measurement device that came up with the above measurements then ?

You don't get a free ride by quoting verbatim blindly in these forums.