Yep, i did. And, a 12 foot satellite dish.
The market is not a good excuse for accepting inaccurate sound... MP3 is fine then, also. So, that cant be used to argue to keep using something - because it is cheap or popular - when better, more accurate exists. Well, not if we want accuracy in Hi-End music systems.
I have limited my thoughts only to what is more accurate to what real sounds, sound like.
I had to buy BW from Comcast to get a really good video download picture in USA. I think it was 11MBS? Visually, the video on this up and coming country, Thailand, is WAY better - more detail and more accurate in every way... same TV/Monitor 65 incher. I just ran an internet test and I am down loading at 34MBS. Without paying more.
Why stay with CD when there is more accurate for same price?
THx-RNMarsh
Last edited:
What are these "real sounds" you keep talking about, where are they and where did they come from?
Please don't exaggerate. What I actually say is what I find sounds better to me. It doesn't come from a theory, just from listening very carefully. Others who have heard it seem to find the same as me. Therefore, I suggest to try it. Obviously, there is no guarantee everyone will like it, and or that everyone will care if sound quality does happen to be better.
Doesn't always seem like it from here.
Last edited:
Exactly, for example Richard might consider it to be more accurate but not necessarily better.
CD is 16-bits (and should be dithered). In practice some CD player dac chips were only good for about 14-bits.
You seem to be asking about the uncertainty of amplitude within the bounds of a single bit. In theory, the uncertainty should come out as noise and if at a low enough level a signal should be so buried in noise as to be inaudible. However, that is based on averages over a sufficient number of samples.
The way to avoid problems from that with short time duration signals near Nyquist is to brick-wall filter out anything (when making a CD) that the reconstruction filter can't handle to within the limits of audibility.
Of course, the whole process is never done perfectly and most will agree that it is possible for a dac or a whole digital audio system to be so bad as to seem audibly problematic to almost anyone. The arguments about what is good enough in practice depends on what individual people find to be true. There is research that can be cited by people who agree with it, and that can be criticized by people who think it is wrong. Ultimately, it is human nature to listen first, then look for arguments to support one's own perception as the standard. That's true of humans in general, not just the ones who don't agree with you 🙂
When people like yourself come along and start picking at possible shortcomings of practical implementation considerations, the response tends to be to defend the theory as though it is under attack. The theory is fine, no worries. Where we disagree around here is more in the area of what is to be considered 'good enough.'
Last edited:
Chris719,
What is the difference between a Dirac pulse train at 44khz, and the actual train we speak of, every other sample full negative w/r to spectra. Specifically, for non ideal pulses which are NRZ half, that is 2 x oversampling with zero every other data point?
Jn
A Dirac pulse train is a sampling of a DC signal, or an attempt to sample a 44 KHz signal, so I do not understand why they would do that.
What is the difference between a Dirac pulse train at 44khz, and the actual train we speak of, every other sample full negative w/r to spectra. Specifically, for non ideal pulses which are NRZ half, that is 2 x oversampling with zero every other data point?
Jn
A Dirac pulse train is a sampling of a DC signal, or an attempt to sample a 44 KHz signal, so I do not understand why they would do that.
Last edited:
Wut, are you saying you failed spotting any difference between any sample in the listening test due to it not being live? Hey it's time to send your M2's over to me as your next obvious upgrade would be to install a live orchestra into your living room! 😀
I’m almost surprised that MQA was not yet mentioned as being the next step in “unlocking every detail of the original master recording”.
Ha,ha, what a fantastic commercial joke.
Hans
Ha,ha, what a fantastic commercial joke.
Hans
No, not at all. the amplitude of every bit is accurate to the level of the sampling system.
As per my example, the use of NRZ at any level of zero stuffing, cannot get back the flat 22khz sine sampled at 44.1k as the stream of data does not carry the correct amplitude throughout the data, but is sine modulated at the beat rate. Use of heavy duty DSP to calculate the proper trajectory of the signal is possible, especially in an ideal system, but that is not is being done in the NRZ style system, even with oversampling.
That of course, is not my argument. I have been happy with vinyl, CD, and MP3, based on my needs.
many here tend to think I have skin in this CD bitrate game, I do not. What I discuss are the fundamentals, what I learned back in '76 give or take.
I spend a very significant time at work teaching rather high level people the fundamentals that are biting them, and find that in general all of engineering seems to suffer this to some extent. Reliance on simulation math to cover the real world can be problematic when you try to fix real world problems with software, not understanding the real error cause.
True that.
Correct application of the theory is the issue. I point out real time window width/nyquist limitations, and some come back with "infinite train math shows your wrong.." Which is of course, a strawman. or shall we say, an inverted strawman..... a namwarts.
jn
Last edited:
Goody good then, but I fail to see your vote so far in the poll thread, just telling ya bro!
😎🙂
I typed out a rant how I hate journal drops where it's clear the individual didn't even bother to read the abstract, much less the meat of the paper but was hoping that the title and the link provenance was enough to bamboozle the audience. Fortunately I went to bed instead and spared myself joining the mob of poop slingers in the fullest extent.
Jneutron (and really most all of us) -- having wasted my time reading the last week's trainwreck, it would serve you/me/us well to start with where you stand rather than going into ostensibly neutral technical nitpicking. Just as you're not privy to my day to do job and what I have to obsess over, we lack the context of where you're coming from. I give you, among a few, a lot of benefit of the doubt but it took me a while to understand what you were trying to drive at and I too would have gotten stuck in the loop of reacting to reactions and not actually resolving the original point. It certainly would have saved you a ton of grief.
Happy holidays all, I'm back to lurking if at all band hopefully staying in the shop rather than online.
I just moved and do not have a system yet to listen, now.
Would love to, though.
-RM
You are absolutely incorrect on one point. Ton of grief.
Despite the few who tend to attack rather than understand what was said, I did not consider any of this dialogue to be a train wreck. Nor, do I consider any of the participants to be anything other than smart. Some would be better with a tad more emotional intellect, but the discussion is better with them than without.
Many points and counterpoints were discussed, wheat within the chaff so to speak.
I learned quite a bit about how digital audio has progressed from my first dsp stuff in the late 70's. When George posted the three Lavry links, several questions I had were easily answered, several questions came to mind, and several assumptions that were built into the system may or not be acccurate.
To me, discussions of this nature bring out technical things that would otherwise not raise to the surface.
As I've stated, I have no skin in this CD rate stuff, but wrestle with the extreme details I see questionable. Very early on in my posting, I stated the concern that close to nyquist requires larger windows. The Lavry papers helped me express it in terms used in digital audio.
Glad you are still here lurking..
John
Despite the few who tend to attack rather than understand what was said, I did not consider any of this dialogue to be a train wreck. Nor, do I consider any of the participants to be anything other than smart. Some would be better with a tad more emotional intellect, but the discussion is better with them than without.
Many points and counterpoints were discussed, wheat within the chaff so to speak.
I learned quite a bit about how digital audio has progressed from my first dsp stuff in the late 70's. When George posted the three Lavry links, several questions I had were easily answered, several questions came to mind, and several assumptions that were built into the system may or not be acccurate.
To me, discussions of this nature bring out technical things that would otherwise not raise to the surface.
As I've stated, I have no skin in this CD rate stuff, but wrestle with the extreme details I see questionable. Very early on in my posting, I stated the concern that close to nyquist requires larger windows. The Lavry papers helped me express it in terms used in digital audio.
Glad you are still here lurking..
John
JN, the amplitude of every sample is correct only within the AD converter error limit + noise. The error is amplitude dependent, i.e. reflects non-linearity of transfer function. The older converter even had skip codes. Recently, ESS DAC has been known for "THD hump", another expression of non-linearity.
And the ESS Sabre hump looks like the green line in the attached picture
Credits for the picture goes to: Article: Understanding Digital Audio Measurements | Audio Science Review (ASR) Forum
Credits for the picture goes to: Article: Understanding Digital Audio Measurements | Audio Science Review (ASR) Forum
Attachments
Last edited:
- Status
- Not open for further replies.