It can be measured comparing the source bits.
I am sorry for you, because the REAL details are lost with sigma-delta conversion, replaced by FAKE details uncorrect in time/phase. Those "artificial colorations" are harmonics that were present in the original recording.
I also would
But I still wouldnt use a reconstruction filter. 
This is utter nonsense. Just because a particular sinewave doesn't use all the levels in the picture it doesn't mean they don't all get used. If the frequency is a (phase locked) sub-multiple of the sample rate then they won't - but that's almost infinitely unlikely in practice. In the general case there's no hope of phase locking the input frequency to the sample rate clock.
No you don't in practice and no, its not only 3bit resolution. In any case if you fed it through a reconstruction filter how could that possibly change which DAC levels were used?
which is already oversampled, dont be sorry for me, i'm completely happy having an excellent performing dac that also sounds excellent. see this is that psuedo science i'm talking about, interested to test your discerning hearing to tell the fake details (your words, I wouldnt use them) from the real ones. phase distortion can quite easily be measured in the actual output of dacs using an analyser, you are saying these fake details passed through these tests? fake ninja stealth details perhaps? is DS perfect? nope, but I would and did take it over NOS dacs I tried. I think its a little bit funny to taolk of fake or real details, any sampling method is just the result of snapshots taken in time and no ADC or dac captures actual real details.
i'm not stuck on DS, just the dacs I prefer from what ive tried are DS, i'll get around to building up a multibit dac or 2 one of these days, but more for curiosity than anything else as i'm completely happy with the dac section of my rig and have moved onto other areas.
I also would
good for you!!
probably the next thing I play with will do away with the dac altogether using DSD over USB with a high performance analogue filter directly into an amp
Last edited:
When you are so completely happy, why invest your time here
, as you have moved onto other areas as well??Maybe there is still some curiosity, but at the same time fear of loosing some of your "certainties"??
what? since when does being completely happy with your setup mean you stop thinking and investigating? I dont know any diyers that get in deep; that sit back and rest....
maybe you, not me
why is it you feel the need to interupt threads like the wire? truly are you in this much denial?
i've just drawn a line under the dac section and moved on to other areas or I would never get the whole system finished, ive spent enough time and money on exploration of the conversion section already and the scope doesnt include anything but curiosity in this unit and how it gets the reviews when there are clearly a number of shortcomings to go along with some benefits (in the objective sense). I said very clearly on the first pages of this thread that I wasnt interested in buying this dac and why, why; other than being a troll do you feel the need to act confused when I say this now?
also if you truly want to keep the thread on topic, why would you provoke more discourse?
I simply addressed a post directly quoting me, you, you are just arguing because thats what you do
maybe you, not me
why is it you feel the need to interupt threads like the wire? truly are you in this much denial?
i've just drawn a line under the dac section and moved on to other areas or I would never get the whole system finished, ive spent enough time and money on exploration of the conversion section already and the scope doesnt include anything but curiosity in this unit and how it gets the reviews when there are clearly a number of shortcomings to go along with some benefits (in the objective sense). I said very clearly on the first pages of this thread that I wasnt interested in buying this dac and why, why; other than being a troll do you feel the need to act confused when I say this now?
also if you truly want to keep the thread on topic, why would you provoke more discourse?
I simply addressed a post directly quoting me, you, you are just arguing because thats what you do
Last edited:
I was replying to Sonic, and I quoted you only in the last sentence. anyway it doesnt seem that you are much open, just checking to feel sure that your choice was the best one.
I have heard a few ESS9018 dacs and it is not the best sound that you can get from digital (and I'm not saying either that the Metrum is).
Anyway, as long as you are happy with your sound, far from me pushing you in another direction.
Being happy with some sort of setup does not mean to stop thinking and investigating, I agree.
In the case of the Metrum dac, for the moment, I am completely satisfied; so much that I don't consider building / buying some dac kit and tweak it with I/V stages, power supplies, analog output stages and what more for quite some time to come. I've done this extensively, and I realize that there is nothing to gain in sound quality compared with the Metrum (and after all for me sound quality is the decisive factor, albeit I know and respect that DIY has much appeal and can be a goal in itself).
I did not interrupt the Wire thread. As mentioned before I was there because I am (still) looking for a solid state amplifier to drive my open baffle woofers; I even initially ordered boards and parts. My last posts there were on my doubts concerning SMPS's being superior to linear power supplies (not much activity on that SMPS-WireAmp front by the way....).
Who actually is arguing?
I am surprised that the Metrum, being a NOS dac, sounds that good, and I am interested to know what measurements are relevant to support this, what measurements are not relevant, in other words to get a bit more insight in the discrepancy of traditional measurements and perceived sound quality.
Members like you and SoNic keep on arguing and arguing; actually in the meantime I am more interested in the psycho-social factors which are behind this weird behavior than in the content of your posts (but I said that before..).
And oh, when you think I am a troll, you better report my post(s) to the moderators, huh? Trolling is against forum rules, so they will decide right?
Last edited:
I'm still interested to read your answer to the question I asked you earlier. Did I miss your reponse? I'll repeat it here:
"Let me ask you a question, if the information encoded on a CD is time-domain sensitive (i.e., music), does the digital sinc filter's time-domain response ("ringing") distort that information? Be careful in your answer, recall that you've already said that the system having the higher distortion is definitionally, the more colored."
To be clear, I'm asking specifically about DAC impulse response. This question is distinct from that of DAC phase response. All linear phase FIR interpolation filters feature, guess what, linear phase, even though their time-domain (impulse) responses can vary dramatically. The now common practice of including a linear phase digital filter with fast and slow transistion band options, such in the T.I. PCM179x series DACs, exemplify that fact.
Last edited:
If something was present in the original recording, how can it be artificial? In English this is a non sequitur. In Latin as well.
They have "slow" filter options to reduce the group delay through the filter for low-latency pro audio applications, not to satisfy a group of people who believe they prefer a worse-performing digital filter.
You appear to have totally missed the point of my question. I didn't ask why T.I. included these two filter options. I merely cited those T.I. DAC+DF chips as examples which demonstrate the distinction between FIR filter phase response and impulse response. My question was about the implications of sinc filter impulse response, with respect to time-domain sensitive information content. Now that you should be clear what the question was, perhaps you also would care to take a shot at addressing it?
Last edited:
You appear to have totally missed the point of my question. I didn't ask why T.I. included these two filter options. I merely cited those T.I. DAC+DF chips as examples which demonstrate the distinction between FIR filter phase response and impulse response. My question was about the implications of sinc filter impulse response, with respect to time-domain sensitive information content. Now that you should be clear what the question was, perhaps you also would care to take a shot at addressing it?
I understand the pre-ringing bothers you, but what kind of band-limited signals do you play through your DAC on a daily basis?
I don't have any studies to reference but I'd expect the ringing to be inaudible myself.
Many albums you are listening to through your NOS DAC have probably been passed through linear phase SRC in the studio.
Whether I feel pre-ringing bothers me, or not, you do agree that there is distortion of the time-domain waveform via typical CD standard recording/playback, and that the only question is whether that distortion is audible?
I think the only answer to the first question can be "yes". The impulse is not recreated exactly the same as the original, so there must be distortion.
The second question is much harder to answer. The "bad" measurements of a NOS DAC don't seem to be all that correlated to the sound quality, but no doubt the same thing goes for "bad" measurements of OS on impulse response... If they were, we'd have gone one way or the other and never looked back. It may seem we have done so since OS is the (almost) universally agreed upon method, but that may just be the result of gathering enough followers.
The people that have heard a good implementation of NOS are beginning to question the validity of (some of the) assumptions OS is based upon.
The second question is much harder to answer. The "bad" measurements of a NOS DAC don't seem to be all that correlated to the sound quality, but no doubt the same thing goes for "bad" measurements of OS on impulse response... If they were, we'd have gone one way or the other and never looked back. It may seem we have done so since OS is the (almost) universally agreed upon method, but that may just be the result of gathering enough followers.
The people that have heard a good implementation of NOS are beginning to question the validity of (some of the) assumptions OS is based upon.
Last edited:
Yes. The fundamental technical "fault" of the CD medium for music, as I see it, is that while CD can accurately capture the full frequency-domain components of music, it does so at the expense of time-domain distortion. This seems the only plausible explanation for why NOS sounds more natural than OS (a generalization, I know). NOS provides a slightly curtailed high frequency repsonse (-3.16dB@20kHz) in exchange for a perfect impulse (time-domain) response.
There isn't any inherently greater in-band distortion via NOS. The greater ultrasonic image replication content of NOS is above the audio band, so, unless the post D/A amplification chain is prone to significant IMD, there won't be any greater in-bamd distortion for NOS compared to OS.
Last edited:
Yet the fly in the ointment for this hypothesis (ISTM anyway) is the one chris719 has already hinted at - the perfect time domain performance of NOS is just going to faithfully reproduce the linear phase, pre-ringy time domain performance of the ADC (and potentially ASRC or whatever else got put in the signal path). So why would that last filter (present or absent in the DAC) make all the difference between NOS and OS when there's bound to be at least one non-optimal filter in line prior to it?
This is VERY good question.
ASRC does horrors to the signal and I try to avoid it like pest. A lot of stuff is registered at 88.2k which is probably better than 96k downsampled to 44.1.
I guess filters applied digitally when mastering are not that damaging compared to the aliasing filter done at the DAC?
That seemingly contradictory fact helped lead me to the hypothesis I had suggested in the final several paragraphs of post #120 here: http://www.diyaudio.com/forums/digi...ctave-dac-what-chips-used-12.html#post2968716
I suspect that much of the subjectively annoying aspects of CD audio stems from the dynamic interaction of the MULTIPLE (at least two) sinc filter responses across the recording/playback chain. Remove either the ADC sinc response via apodising, or remove the DAC sinc repsonse via NOS and, viola', much more natural sounding reproduction results. I would add that while eliminating one of those two (at the least) sinc responses in the chain substantially improves the sound, it may also be that removing BOTH would improve the sound to it's ultimate point. Something which could be done using high sample rate audio.
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.