What should be and what is are not always the same thing. Experimentally, some modern dacs sound subjectively better in DSD mode than they do in PCM mode.
Edmund Meitner, founder of museatex and Meitner Audio, patented the attached similar concept in 1995. The general notion of both, seems to apply analog signal theory to digital signal implementation. Even though, they are both well regarded for their digital systems knowledge. Such notion being the view that SINC-function based FIR filter impulse-response ringing is undesired for reconstructing a sampled signal. That was probably subjectively true, except not for the reasons which they apparently suspected.
As we just found out in our investigation, high-performance sharply bandlimited, long-ringing FIR filter playback, essentially sounds like zero-ringing NOS playback. Suggesting that Denon's Alpha DAC and Meitner's patent were, unknowingly, intended to compensate for audible artifacts introduced by common OS interpolation-filters.
Last edited:
That doesn't mean anything to readers unless measurements can be presented for comparison because it may have been all in your imagination.
I do support @Markw4 statement. Some Delta-Sigma DAC's sound better in DSD mode indeed.
Either you are mixing up feedback loop orders and wordlengths, or I don't understand you, or both.
I think the reason why multibit sigma-delta modulators got popular is mainly the invention of mismatch-shaping algorithms that help to get a decent performance out of a multibit DAC without insane matching requirements.
Single bit can also work well when you use the old trick of embedding a pulse width modulator in the sigma-delta loop to make it a quasi-multibit system. The quantizer can then be multibit and can hence be dithered according to dither theory.
The only things that can generate distortion are clipping on intersample overshoots and roundings of intermediate results and rounding of the end result. My former colleague Frans Sessink suspects that the SAA7220 rounds intermediate results. The digital filters in the first Philips CD players certainly rounded the end result with first-order noise shaping and no dither.
As long as there is no clipping or folding and you don't round anything, an FIR filter that interpolates by an integer factor is linear and time-invariant and can't distort, no matter how short or long it is. It can have a crap frequency response and pre- and post-echoes, though.
What would you do or have you done to confirm that it wasn't all in your imagination? Any measurements or objective comparison data to share?
Thx for this link.
Reading the patent a solution was proposed to solve the following problem:
Pre ringing distortion is highly audible in music passages and is the major audio difference between analog processed playback and digital playback.
As you already mentioned, this is not the case.
On top of that one should be aware that pre ringing may already start much earlier where a high Fs digital recording is downsampled to 44.1 using the same sort of brick wall filters as used for OS.
Whatever smart type of OS filter will be used at a later stage in the DAC, this filter will never be able to remove eventual pre ringing in the available music content.
Feeding a Dac with a dirac pulse to visualise things may therefore be misleading.
Hans
If something sounds better than the other, it would first have to sound different. Confirming that can be done by comparing measurements and or level matched double blind listening test.
@ Even,
I read hundreds of posts (at least that is what it felt, did not really count) where you keep nagging on measuring etc and trolling about sound quality perception.
so I took a look at the threads you started. Do I need to say more? Specially the thread on the Ebay amplifier and the first reaction to that made me start the day with a smile.
You are really a fun boy, keep presenting everyone as the man who gets it all in other threads. Really funny - I should have done this earlier, I take your post not so seriously anymore
I read hundreds of posts (at least that is what it felt, did not really count) where you keep nagging on measuring etc and trolling about sound quality perception.
so I took a look at the threads you started. Do I need to say more? Specially the thread on the Ebay amplifier and the first reaction to that made me start the day with a smile.
You are really a fun boy, keep presenting everyone as the man who gets it all in other threads. Really funny - I should have done this earlier, I take your post not so seriously anymore
OK, if you want to play that silly game, let's presume a difference has been proved, now how do you answer the question? Do you see yet the meaninglessness of your question? Your questions in this vein are unanswerable as I'm guessing you already know, so, pray tell, why do you keep asking them?
Just because it's unanswerable to you doesn't mean it's the same for sajunky (whom I asked what he would do to confirm one's observation is all in his head or actual audible difference). Perhaps you should've consulted with sajunky first.
Since you are new here, some background info is what you need. Try this, Audio Myths Workshop - YouTube
As I suspected, you are new to this forum and the world of sound reproduction as well. The quality of sound reproduction equipment is judged on its level of fidelity, higher the better. If you aren't familiar, look up the term hi-fi (high fidelity). Something sounding better in reproduction would have higher fidelity to its reference source. I suspect that you thought it was something a listener likes better. That would be a valid criteria if you are judging a musical performance but this thread is about audio gear for reproducing such performance. Try to understand the difference between the two.