Best No oversampling dac to buy??

[theREALdotnet]

No because I cannot recommend any reconstruction filterless DACs.
Hell... only the earliest of Sony CDP had no oversampling but at least they had a filter.

Thank you. I’ve been scratching my head what “no oversampling” has to do with “no filter”. Surely, every DAC must have a filter to reconstruct the audio waveform in the intended frequency band. Equally surely, a DAC does not necessarily need to oversample before D/A conversion. Why the conflation of terms?

 

[Trell]

Thank you. I’ve been scratching my head what “no oversampling” has to do with “no filter”. Surely, every DAC must have a filter to reconstruct the audio waveform in the intended frequency band. Equally surely, a DAC does not necessarily need to oversample before D/A conversion. Why the conflation of terms?

A NOS won't reconstruct the audio waveform to a flat 20 Hz - 20 kHz but have a dip/shelf in higher frequencies. At my age I won't be able to hear that in any case, though, and statistically you won't either.

 

[solderdude]

Surely, every DAC must have a filter to reconstruct the audio waveform in the intended frequency band.

Yep, and that's exactly what is missing in (reconstruction)filterless DAC designs often referred to as 'NOS' but this is factually misnaming.
Oversampling and reconstruction filter are 2 different things. A good upsampler, of course, should use a (steep) filter to calculate the intended sample values in-between the given ones. A filterless upsampler can also be made where the in-between values are simple 'sample and hold' from the last known value to the next known value.
This is what is simulated in the RME when the filter is set to 'NOS' (there is some small filtering to prevent overshoot). The same misnaming for the sake of the audiophools who do not know the right naming and adopted the abbreviation NOS but mean (reconstruction)filterless.

 

[tuga]

the best NOS DAC currently on the market is the Holo Audio May afaik:
View attachment 279446

Is that measured in NOS mode?

 

[Head_Unit]

a bunch of ultrasonic garbage that is not removed.

But is there REALLY? And not just for NOS, I've long wondering what would happen if you did not have an output filter at all since it causes problems. Yeah in theory all this aliasing of images blah blah blah. However that supposes there is significant ultrasonic content coming into the A/D in the first place, which I seriously question how much of that there is in the recording space, and further question how much of what there is actually makes it through the microphones. I've never seen any serious investigation of this. Maybe that's a good thread topic; some members might be set up to measure that?
(I exclude discussion of recordings of stressed plants...
https://www.sciencedirect.com/science/article/pii/S0092867423002623

audible high frequency roll off

Yes, I recall the Luxman "Fluency" car unit as -1 dB at 15 kHz, and -3 dB at 20 kHz.
- I would posit few people would hear that rolloff* with actual music. Hmm I'm not set up to do it but it almost begs for a thread of test files...
Nevertheless people conditioned to ±0.000000000000123 dB flatness were freaking out. Ironically, now that I think about it, many of them the same people shoving compression driver horns into cars with umpteen 18" pro woofers, who probably couldn't hear 12 kHz out of a flute next to their ear anyway!
*if implemented with a lowpass in a "regular" system; never mind the changing nature of at least the Wadia's response. I don't recall anyone ever discussing the Luxman changing frequency response on a sample-to-sample basis but maybe it just never came up.

NOS is a technique to extract hard earned money from audiophiles.

Like anything else in this audio world, surely sometimes. Not originally; the "Fluency" guys (who I "met" in a meeting with a lot of translation going on but no sake sad to say) seemed genuinely concerned about improving time domain behavior. I see it simply as a different approach, each with its own foibles. For instance, pre-ringing in more conventional DACs: supposedly inaubible, but is it? How would you know? And let us not forget, everyone has different hearing with differing sensitivities to various phenomena. Also let us not forget that 99.872% of audio comparison testing in unblind and hence unscientific.

I stumbled on this paywalled reference
https://www.sae.org/publications/technical-papers/content/940259/
and this interesting quasi-summary
https://samplerateconverter.com/educational/nos-dac

 

[solderdude]

But is there REALLY?

There is US crap when there is a signal. The higher the frequency and amplitude of the signal the higher the US crap.
That may well be filtered by transducers and or bandwidth limiting in most cases.
There could be cases where the US crap may bite the signal. Think the popular class-D amps for instance.
Use it with bandwidth limited tube amps and there will not be much of an issue.
Most 'NOS' DAC 'filterless DAC' would be more appropriate) aficionados are more likely to use such obviously flawed DACs than people seeking signal fidelity so there most likely is not an issue (well there is a bit with the treble but they seem to like that) with the US content and is just either filtered by the system or the tweeter does not care.

The biggest issue is the amplitude of higher frequencies which is not 'correct', low to mid frequencies is not an issue, they will be rendered fine, the US 'errors' are well above the frequency limit of the average audiophool as they probably can't hear a thing about 14kHz anyway... if they are lucky.
Some folks seem to like that roll-off that isn't really a roll-off anyway only on average there is a roll-off.
It highly depends on the frequency and point where the sample reaches the max amplitude versus the original amplitude before the sampling process.

 

[NTK]

a bunch of ultrasonic garbage that is not removed

But is there REALLY? And not just for NOS, I've long wondering what would happen if you did not have an output filter at all since it causes problems. Yeah in theory all this aliasing of images blah blah blah. However that supposes there is significant ultrasonic content coming into the A/D in the first place, which I seriously question how much of that there is in the recording space, and further question how much of what there is actually makes it through the microphones. I've never seen any serious investigation of this. Maybe that's a good thread topic; some members might be set up to measure that?

Here are some measurements of the Meitner MA3 by Stereophile that showed the ultrasonic garbage. The ultrasonic garbage are the "images" you get during the first order hold (FOH, aka stair-steps) D/A conversion when done NOS and sans filter.

Meitner MA3 Integrated D/A processor Measurements

Sidebar 3: Measurements

www.stereophile.com

[Edit] Compared to the Benchmark DAC3 HGC which is a properly implemented DAC.

Benchmark DAC3 HGC D/A preamplifier-headphone amplifier Measurements

Sidebar 3: Measurements

www.stereophile.com

 

[MaxwellsEq]

The filters in digital systems are essential to minimise aliasing and imaging. If you are playing music designed to be reconstructed with a DAC at a certain sampling frequency, failing to filter at half that frequency can introduce artefacts in the audio domain, such as IMD.

@Blumlein 88 wrote an informative post on this : https://www.audiosciencereview.com/...-imaging-and-upsampling-a-real-example.35687/

 

[voodooless]

However that supposes there is significant ultrasonic content coming into the A/D in the first place, which I seriously question how much of that there is in the recording space, and further question how much of what there is actually makes it through the microphones. I've never seen any serious investigation of this.

There is no need. You can find this information in any ADC datasheet. Nevermind that most microphones don’t do much ultrasonics either. Luckily ADCs are also oversampling, so the filters can be done mostly in the digital domain, and can be very steep, just like in their (proper) DAC counterparts.

 

[gvl]

Yes, I recall the Luxman "Fluency" car unit as -1 dB at 15 kHz, and -3 dB at 20 kHz.
- I would posit few people would hear that rolloff

The rolloff starts early, below 10khz, while small in absolute terms at any particular frequency point, it’s wide and causes enough HF energy loss overall to be audible.

 

[earlevel]

But is there REALLY? And not just for NOS, I've long wondering what would happen if you did not have an output filter at all since it causes problems. Yeah in theory all this aliasing of images blah blah blah. However that supposes there is significant ultrasonic content coming into the A/D in the first place, which I seriously question how much of that there is in the recording space, and further question how much of what there is actually makes it through the microphones. I've never seen any serious investigation of this. Maybe that's a good thread topic; some members might be set up to measure that?

No, it doesn't suppose there is significant ultrasonic content into the A/D to start with. I'm not sure what you're saying, because you acknowledge images. The images are result of PCM. Of course, the DAC is not putting out impulses, at least ideal ones, so there will always be some filtering (if implicit). But if you sampled a 1 kHz sine at 48 kHz, the first image is at 47 kHz. The point being that you don't need to record "significant ultrasonic content" to get significant ultrasonic content if you leave off the filter.

I probably don't fully understand what a NOS DAC might be—from what I've read it seems a big of a non-specific description. At face value, it would be one without an oversampling filter, and to me that implies an R-2R ladder DAC. But as @solderdude points out, people take it to mean no reconstruction filter. But since they aren't using (and can't use) ideal impulses, whatever they are doing results in convolution that is in effect, some type of filter. @solderdue mentions sample and hold (zero order hold), which is the typical shortcut, and in fact that's a weak filter. A problem with that is that if you're not running a higher sample rate, you have frequency droop in the audio band nearing the Nyquist frequency. That's usually compensated for by the filter. But any way you want to look at it, exactly what a "filterless" DAC outputs depends on...how it outputs its signal. That opposed to, for a DAC with a "reconstruction" filter, which would also make up for any design choice that precede it int he output path.

I don't know, but to me this sort of thing is an extension of "we were so happy with analog, then digital came along and ruined music" sickness. Oh, for the days of hiss, pops and hum. Now we (some) have to worry about what their filters looks like on paper. Before digital, I'd never heard people concede they couldn't hear something, yet were concerned it could be making them feel uneasy or something.

 

[Head_Unit]

Before digital, I'd never heard people concede they couldn't hear something, yet were concerned it could be making them feel uneasy or something.

There are times I can't hear my wife, yet I see her mouth moving and I feel uneasy. Does that count?

 

[Head_Unit]

But if you sampled a 1 kHz sine at 48 kHz, the first image is at 47 kHz. The point being that you don't need to record "significant ultrasonic content" to get significant ultrasonic content if you leave off the filter.

Ah, OK, correct. Whence some of your equipment might not be happy, unless we presume those inaudible ultrasonics do indeed affect us. Ignoring those possibilities, what I should have said is that if there is little ultrasonic content coming into the A/D, then there would be very little audio band content (images distortion). (I think???)

 

[earlevel]

Ah, OK, correct. Whence some of your equipment might not be happy, unless we presume those inaudible ultrasonics do indeed affect us. Ignoring those possibilities, what I should have said is that if there is little ultrasonic content coming into the A/D, then there would be very little audio band content (images distortion). (I think???)

I think I follow what you are saying, by also referring back to what you said earlier. I think you are talking about how much aliasing there might be baked into the signal from the ADC, with the implication that if the source signal had more content over half the sample rate, the more of it that would make it past the band-limiting filter and get baked in as aliasing. An interesting topic for discussion, but I think a little off topic for the thread. (?) I think we have to assume the ADC is doing its job, and now what about this "NOS" DAC business. Whatever may be baked in at the ADC stage (and presumably, it was digitized on proper gear by professionals who would notice aliasing if it did happen) is going to not go away with NOS or non-NOS (I had to say that ) DACs. No?

 

[Head_Unit]

Whatever may be baked in at the ADC stage (and presumably, it was digitized on proper gear by professionals who would notice aliasing if it did happen) is going to not go away with NOS or non-NOS

Perhaps I am mentally conflating aliasing in the ADC versus DAC...evinced further by my total mental inability to now at this point remember why the heck there is a lowpass filter...
(Somehow I was thinking that if there was not ultrasonics really getting into the recording in the first place then lack of a lowpass filter like in an NOS design doesn't really matter much).

 

[dc99]

What is it you prefer about non-oversampling DACs? These days they seem to be pretty rare.

IMO, they will become more and more popular, as more and more Hi-Res (96/192kHz) recording are being made. When you listen on 96/192 sampling rate, there is no need to do over-sampling and no need to apply the destructive reconstruction filter.

 

[Audiofire]

IMO, they will become more and more popular, as more and more Hi-Res (96/192) recording are being made, there is no need fore over-sampling and no need for the compromising reconstruction filter.

IMO, they will become more and more popular, as more and more Hi-Res (96/192kHz) recording are being made. When you listen on 96/192 sampling rate, there is no need to do over-sampling and no need to apply the destructive reconstruction filter.

 

[staticV3]

When you listen on 96/192 sampling rate, there is no need to do over-sampling and no need to apply the destructive reconstruction filter.

Tell me you've never learned about the Nyquist-Shannon sampling theorem without telling me that you've never learned about the Nyquist-Shannon sampling.

 

[gvl]

Given early DACs ran with 4x filters the statement about 192kHz not needing digital filtering isn’t that wrong.

 

[dc99]

Tell me you've never learned about the Nyquist-Shannon sampling theorem without telling me that you've never learned about the Nyquist-Shannon sampling.

Filtering in the digital domain is different to time domain filter.