What is supposedly wrong with pcm DACs per se? And what exactly do people mean by pcm dac? Pcm encoding adapted for a truncated word length d to a core ? I'd have thought a full r2r converter would have the most straightforward characteristics if it works properly.

A mic outputs microvolts to millivolts; an ADC usually needs a volt or two to give optimum signal to noise ratio and to minimise quantisation errors. The ADC may be required to take inputs from various sources such as a mixing desk, and would usually work at around line level. A mic is usually...

There seem to me a few adc dac scenarios would be potentially capable of good results. I'm assuming the convertor core is good for linearity, noise and speed of course. 1) Use linear phase brickwall FIR at both ends, accurate as possible, and therefore precision long window sinc. With high n...

I think Rob is saying the time inter sample errors introduced by non-sinc interpolation reconstruction are complex and significant, whereas the errors in oversampled ADC anti-aliasing with decimation are likely to be more benign, and these benign errors will be what is reconstructed.

Maybe so. But you'd be optimising what was possible, not creating the impossible. With digital you can do much more in terms of filters. But that very first step is the tricky one, filtering the analogue. And you can't get too far, I don't think, with an analogue filter.

...and finally, for completeness (for the moment.....)..... A pure linear analogue time-continuous sinc brickwall phase linear filter would be just about impossible to make in practice. Even getting close to one would be extremely difficult.

A further point is that if you fed the analogue music signal into an ideal brickwall filter, analogue or digital, you'd get time domain pre-ring, unless the signal is already hard band limited to below the filter cutoff frequency. And that pre-ring would get reconstructed at the DAC as an...

Sorry to cause offence. You can't simplify beyond a certain point and still accurately convey the essentials here. If everything in the total ADC/DAC chain is done as close as possible to the Shannon Nyquist criteria, a filter with pre-ring is a necessity, both in the record end anti-aliasing...

No. Correct sound to hard-bandlimited input signal means both correct FIR choice, which includes extended pre-ringing, and rigorous implementation of same. There is little possibility of approaching this ideal in the real world for the total system; the best chance would be if the ADC and DAC...

Yes, normally there are no time isolated single sample wide impulses in music. And pre-ring should never manifest as such as an added artifact in a perfectly reconstructed music file, with exact filter and sample rate matching, true enough. However it does I believe represent energy which could...

I misused the term 'causal' I think, in referring to the published impulse response. Any FIR is causal in absolute real time of course. I was intending to refer to the post-peak part of the impulse response. The published impulse response I referred to from 'Stereophile' would not be linear...

Thanks Rob I guess I forgot the necessarily fairly severe FIR windowing, with the slow decay of coefficients on a sinc function. I found an article online from 'Stereophile', by John Atkinson, looking at the impulse response for the original Mojo. Did you use a time-asymmetric window over a...

Thanks all. I was a bit lazy and didn't realise Rob had gone into such depth here. I have started reading the first link which is very illuminating! Having compared the Mojo to a couple of other DACs I'm wondering whether a) The Mojo can be expected to be more faithful in all respects to an...

Questions re Rob Watts approach: 1) Does Bob seek to implement a pure brickwall FIR sinc function or something else? Why? 2) Is the subjective result distinguishing his designs seen by him as primarily due to tap length/computational accuracy? Are there other factors? 3) How does the recording...

Not seeing the questions I wanted to ask answered, either in the first posts or in a recent interview.

Hi Rob I have just bought a used Mojo version 1 and had a few questions about your design approach. I haven't followed the thread in depth I must admit. Can I ask them here? Apologies if I'm out of order.

People define thd, snr and dr in various ways, with and without weighting curves, etc. I think the ability to reproduce a sine wave passably well in the time domain is a decent indicator of real dynamic range and resolution. If you look at the -110dB sine wave on P4 of the PCM63 data sheet and...

The issue of what we mean by dynamic range arises. Crest factor, or peak-to-rms ratio, over a certain time frame or for a song, is one way it's measured for a particular track in a particular medium, and relates to DR compression used during recording/mastering. DR for a DAC usually means...

I agree DACs are getting better, with the newer higher grade delta sigma and hybrids much better than older ones. They probably sound pretty similar especially with the same filter. Some are still only really good for 18 bit or so while others like the PCM1792 are good for 20 bits plus. DCS who...

And I thought there was a coffee molecule. I think this might be my last remark other than perhaps for politeness and completeness for the moment. Said that before. I have no desire whatever to make anyone dissatisfied with their audio gear or music format. That probably sounds patronizing...

Yes you could all be right and everyone's happy, so fine, I'm happy too. Thanks for talking.

I listened to the Clapton thing again! I think about a lot of things unrelated to audio and this forum!

Well I might not have any others if I keep obsessing about audio all day every day and arguing with people:smile_phones: Dither in PCM is primarily to reduce quantization noise or randomise non-linearity in the DAC. It can also push low level signals over the 1st bit threshold (if the DAC chip...

Yes you can FT reconstruct any periodic waveform perfectly including a repeating recording if you have an infinite resolution available in the complex frequency domain. The imaginary component provides the phase information. Band limiting curtails the maximum frequency you need. The length of...

With the right dither and an original quanitzation comfortably sufficient for -120dB, i.e 24-bit, yes for frequencies well below sample rate and particularly periodic signals. In the general case, no. You need the bit depth.