I tried to wade through this MQA: Questions and Answers Temporal Blur In Sampled Systems | Stereophile.com which purports to support the microsecond level of time acuity claim. The writing feeling like the "baffle with BS" type and the references are not all that solid. However I find the whole MQA thing somewhat suspect. Just maybe they have a point but I can't penetrate the text well enough to see it.
I agree with the feeling that it is baffle with BS. It really looks like the author doesn't really understand the science. "wading" is a most appropriate term.
I could see the issue if DAC's simply fed a S/H, and then into a brick. But when the reproduction math has access to samples before and samples after, the signal fidelity can be dropped into nano territory. As I recall, the width of temporal data doesn't need to be infinite to drop uncertainty very very low.
Perhaps someone should introduce the author to the acronyms DSP, FIR, and IIR..
Ah, I went a bit further in the reading. I believe "floobydust" just about covers it.
When a paragraph purporting to answer the question "what is it" is indecipherable, it's time to move on. Anybody with sufficient knowledge of the topic should be able to explain it clearly and concisely.
John
Last edited:
This reminds me of the guy who applied for a position and wasn't even granted an interview. He inquired how this could be, as he knew he was a great fit for the job, and the HR person reviewing his resume said they were looking for someone with DSP experience.
He had made the mistake of spelling out "Digital Signal Processing".
We are at the mercy of ignorant morons. But we all knew that.
As I mentioned on another thread. If you view MQA as 'Bob's retirement fund' it suddenly makes sense.
We have had this argument before. Interested parties can search for it. People may judge for themselves which alleged claim/assumption seems more "extraordinary" and so where the burden of "proof" lies. As most audio is bandlimited there will be little or no energy at the frequencies at which the composite cable approaches 7.5 ohms characteristic impedance, unless the amplifier has stability problems.jneutron said:
It is my belief that almost all people (apart from you) who aim at 8 ohm impedance for a speaker cable do so out of ignorance about low frequency transmission lines; they believe that an 8 ohm line at audio frequencies is easily achievable. We have certainly seen evidence in this thread that this is probably the case for Mr. RNMarsh.
Your interest in 8 ohm cables arises more out of a subtle form of Fourier denial (or is it Laplace denial?); too subtle for many people. Essentially you are claiming that the low frequency performance of a circuit cannot be derived from its impulse response, because you claim that the impulse response contains relevant facts which are missing from the LF response. If true this would undermine the whole of circuit theory; we would never know when we can trust our calculations without doing a full wave analysis.
MQA just removes the GD. That is why they cannot talk straight about what it does.
When they know the input characteristics and can compliment the output side as well, all GD can be removed/backed-out. Other recordings are guessimates based on what recording industry majority uses for gear and its GD.
Then layered on top of that is the compacting of file size etc. which helped then do their thing to the data more efficiently.
THx-RNMarsh
When they know the input characteristics and can compliment the output side as well, all GD can be removed/backed-out. Other recordings are guessimates based on what recording industry majority uses for gear and its GD.
Then layered on top of that is the compacting of file size etc. which helped then do their thing to the data more efficiently.
THx-RNMarsh
😉
(I am a transplanted one in the UK - originally from about 4000 miles left of where you are located)
Don't get me started on HR . . . 'effing flesh peddlers (yeah, infinitely worse than marketers)
😀
You did not ask me. You assume too much and jump to conclusions... a fad here, it seems. That was not the reason I did it. I did it to contain the fields and lower the Ls.
If you buy yourself a portable EMI sniffer you will see that the electromagnetic fields radiated from lamp cord speaker wire is very large and far reaching.
Lowering the Ls is a positive thing to do when using dynamic drivers/tweeters to prevent HF roll-off.
Killed 2 birds with one stone. Three actually.... it reduced any RF pick up.
THx-RNMarsh
Last edited:
I don't see that nor do I quite get your point. The impulse response contains all the information at all frequencies, the two halves of your sentence seem contradictory.
We've been over this before, the characteristic impedance graphs are fairly useless at tiny fractions of a wavelength but characteristic impedance and speed of propagation are intimately tied to L and C. This is not a coincidence so all forms of analysis must converge to the same answer unless incorrect values are used.
Last edited:
Actually, we have not.
Despite peer reviewed research extending back into the 1970's, you still bypass the issue and instead obfuscate.
You claim that audio circuits in essence do not require bandwidths in excess of 50 Khz (latest claim).
But we are sensitive to inter-aurals at the level of 1.2 to 1.5 uSec.
Until you understand this, perhaps you should remain out of the discussion. Your understanding of E/M is ok, you just don't see the forest for the trees.
Only one other statement in your post will be responded to, the rest of your post is rendered useless as a result of ignorance of what is really being discussed.
This is somewhat interesting, but requires you be taught (or shown) quite a bit before you understand.
Perhaps the most important concept you need to understand is settling time.
Until you do, you will never understand.
To wit: given an LC model of a wire to the load:
Drive the load using a steady state sine, and measure the lead or lag of V and I.
Drive the load using a different frequency steady state sine, and again measure the lead or lag.
Now, if one alters the stimulus from the first frequency to the second, how long will the system take until the appropriate lead (phase) has been arrived at for the second frequency??
The answer is of course, the settling time. This is a consequence of the energy storage of the elements being discussed. edit: to ignore this entity is to presume the speed of light has been violated and that the prop velocity is infinite. Is that what you really intend??
You should already know and understand this. I make the assumption you took circuit analysis and state space control theory. If not, I can recommend some books.
John
Last edited:
Not sure what "inter-aurals" means in this context? Are you meaning the difference in arrival time at the two ears?
Got any refs to test results on this?
Where does it say that? What they they won't say straight is that they have invented a new lossy compression scheme that they think is better. And BTW it will be tied to proprietary hardware ($).
... from the audible descriptions and the way they are doing their thing. If they just came out and said this... there would be little interest and competitors would be all over it fast. The compression scheme, IMO, is needed to do the number crunching for the GD reduction more effectively. ... where is the patent to read?
I talked to JC a couple months ago about this and he seemed to have more info than I did but thought it was likely GD also.
In my experience there are two issues which cause a lose of detail at upper-midrange and higher.... rising distortion products with rising freqs and GD.
BTW -- when DC servo was introduced and response to near DC was achieved, the comments were universally about how clear the bass became..... we are even more sensitive- according to research - to the low freq GD than to high.... more audible. Especially, with the proliferation of the DC-servo concept... the multitude of series RC networks at I/O in a system of several products in series made an even larger improvement to the sound.
THx-RNMarsh
I talked to JC a couple months ago about this and he seemed to have more info than I did but thought it was likely GD also.
In my experience there are two issues which cause a lose of detail at upper-midrange and higher.... rising distortion products with rising freqs and GD.
BTW -- when DC servo was introduced and response to near DC was achieved, the comments were universally about how clear the bass became..... we are even more sensitive- according to research - to the low freq GD than to high.... more audible. Especially, with the proliferation of the DC-servo concept... the multitude of series RC networks at I/O in a system of several products in series made an even larger improvement to the sound.
THx-RNMarsh
Last edited:
We'll see I guess, the above makes no sense to me at all. That time spread stuff is plain old sinc function nonsense and has nothing to do with group delay, I'll remind you sinc reconstruction (that which they want to avoid) is linear phase (pure delay).
EDIT - In this context what you added makes even less sense. There are few (any?) sound cards or players with DC response.
Last edited:
- Status
- Not open for further replies.