|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
4.241.217.241
In Reply to: Anybody read "The Death of Dynamic Range" webpage? posted by darkmoebius on November 22, 2006 at 21:20:31:
First of all, the claim that digital has "wide dynamic range" is not technically factual. Digital has only 3 dynamic reference points (sample points), representing the minimum, average and maximum.Because of this, it limit's its actual range and scale, and noise-shaping schemes do little to improve upon it.
Given the above, one can easily see how compression is detrimental in the digital domain.
Follow Ups:
In every audio format, there are good and bad recordings. I really don't think one formatt is superior over the other. Though, technically this may seem an ignorant statement from my side. I really can't favour one format over the other coz I failed to find any consistency in one particular format over another re recording quality etc..Ive heard and made some cassette recordings that were superior to their LP counterparts ( mainly done with TOL Nakamichi decks from the golden era..). Ive heard CD recordings that were superior to their SACD counterparts and vice versa etc..
But what i've found consistent is the playback hardware and how one can really enjoy one's music collection (in whatever format it's on)
thru a really well made CD Player, TT Deck or Cassette Deck.Technical discussions can be eye-openers but I feel that in the end what matters most is how we personally perceive the sound in context of our varying tastes and preferences.
JMHO...
On a side note, can you please explain to me why the Balanced Outputs of Accuphase players are so clearly superior to their Single-Ended counterparts ? After listening to my DP-57 in Balanced mode I'm seriously not able to listen thru the Single-Ended outputs at all !! Would this be different when listening thru an Accuphase amp?
I didn't find this difference to be so big in most other digital players Ive had..
Thanks and enjoy the music..:)
AP
# The path of the righteous man is beset on all sides by the inequities of the selfish and the tyranny of evil men # Samuel L. Jackson (Ezekiel 25:17)> Pulp Fiction <
Hi,I would like to say something about the matters George was talking about.
The points of references are the sample points being 44.100 times per second and each of them 16 bits in voltage level, being coded digitally. This results in over 65.000 voltage values that can be used for the initial point to be represented in the waveform.
All other than this, like 18 bits or 20 bits or 24 bits aren't on the cd, only 16 bits are on the CD.
They are corrections made possible through some digital mathematics and interpolations.This coding during mastering is done fairly correctly, so it should be a nearly perfect representation of the original analog information at the points of references.(one of the sample points in a second)
Why then does digital sounds DIGITAL, being flat, cold, heartless?
If the data was correct during encoding, then something must be wrong during decoding..!!!
Why do I state this, did anyone ever wondered why there was so much difference whenever you do something with or on the cd, like an extra weight.
There are extra thin discs to put on it, this creates a difference in sound.
Between several drives used in the players there are also differences in the resulting sound experienced.This must NOT be, since the error handling in the digital domain would have so much extra data for interpolating and the whole interleaving manner of building up of the digital information should be able to restore all the incorrect data whenever there are faults during the read out.
So WHY then is there so much differences after all.????
After some real experimenting with drives and mechanical resonance alterations I concluded that the read out is very faulthy.
The data that is burnt in the process of manufacturing the cd record is more accurate than expected.
Problem I think now is that during read-out the TIMING and SAMPLE value are mis-aligned in TIME as a result.Therefore the shape of analog information after read-out and D\A conversion is not near the original.
Most of the data used in the D\A conversion is data that is created by interpolation, so another data than actually available after all on the disc itself.The data on the cd is still correct, but the read-out is where the problem lies.
Therefore the sound sounds digital and cold and flat...So it is not that digital has a problem.
If the timing and samples are read out more correct than the resulting analog waveform will be very close to the original waveform that has been encoded into the digital data put on the CD.Therefore this is also why the dynamics within the CD also is dead.
There is to much interpolating and miscalculations after read-out has performed that the data has to many 1's in the data, which makes the total digital value per sample be to high, resulting in a more flat sound character and resulting analog output dynamics are less.BELIEVE ME, i could not believe my ears when I heard the differences after making the corrections on the spindle servo ONLY. (the way the CD turns during the laser read-out)
I still haven't done anything on the focussing and tracking part, this is very interesting for me, since I will then do the correction on the bits parts. (the amount of 1's in the data are incorrect, therefore the 0's also, but they don't cause the problem)I am trying to get someone from the audio business to listen to my modification, just to let some one who cares would hear the differences and let people be convinced by such person, since I haven't got the influence in that area.
Till now, no-one even gives me reply.. I find that so dissapointing, just cause I have really developed something that matters and is so much more than all the other modifications around and even exotic manufactured CD players.
I don't even need other clocks, only thing I added was a somewhat better OPAMP in the end stage...just for some nicer sound character.This is so frustrating, but on the other hand I and my friends will be the only persons who finally really enjoy CD sound like never before or ever will without this optimization.
It really comes very close to analog, hard to believe but it really is....I could not believe it myself, the resolution heard now was unheard before.
The music has so much more involvement end dynamic nuances in alle of the parts in IT, in voices, instruments, depth and width.
So why am I complaining...By the way, I live in the Netherlands...
Grtzzz..
Andre, (Natural Timbre)
HowdyThere are plenty of discussions about the existence of jitter as well as ways to minimize the effects of jitter here and other places. Certainly the things you mention are often useful and there are players out there that incorporate similar techniques as well as others.
George - 24-bit data has 2^24 voltage levels. That's 1,677,216 different voltage or current levels depending on the D/A chip.
Irregardless, its limited to only 3 points of reference.
What are those three points? Nearfield, Farfield and dead and buried?
. . . you're the first person who has ever claimed this on this board, despite all the digital engineers who show up to tell you you're wrong.And the word you're looking for is "regardless." "Irrespective" is a legitimate word, but not "irregardless."
HowdyCould you expound on what you mean by "Digital has only 3 dynamic reference points (sample points), representing the minimum, average and maximum." Even with a good background in DSP I don't have any idea of what you are trying to say.
All audio-related functions within the digital domain are dependant on sample's and points of reference.Although DAC's are capable of a wide range of voltage samples, they nonetheless have the same unchanging points of reference. A minimum threshold, an average threshold and a maximum peak threshold.
Analog on the otherhand has a continuous "fluid" threshold within its range of scale. Do you understand what I am saying here?
....that each signal level gradation in digitized recording has a minimum threshold, average level, and maximum threshold.... (There are 65,536 such gradations on a Redbook CD.)If so, you basically described "quantization error"....
Although dither (in the A/D encoding process) does make the behavior more "fluid" when played back with a DAC using classic (oversampled) digital filtering.... Because the random bit triggering on the media (induced by dither noise in the A/D process) translates to an interpolated signal (from the D/A digital filter) whose levels are between the gradations of the media. The extra bits in an oversampling DAC enable such interpolation to take place.
Don't let them drag you down George.
nt
HowdyI understand that, and I think others will agree that it's a little more clear than your first statement.
Don't get me wrong I'm not a digital apologist (I understand it too well for that) but it's still a leap to your conclusion about dynamic range from just that statement. There are plenty of recording systems that use higher precision than just 16 bits and hence can ameliorate the effects of any particular sampling reference.
And why would what George states be a problem, anyway?
bring bac k dynamic range
The issue here is that digital is only capable of representing one sample and point of reference at any given moment. No amount of differing has been able to correct it.When it shifts from one sample and reference point to another, it does so in a nonlinear, stair-stepping fashion.
These 2 problems result in a clearly skewed representation that is devoid of a true, fluid range that clearly lacks the ability to present 2 or more levels of threshold simultaneously.
This is one of many problems that needs to be addressed if digital is ever going to replace analog as a competitive hi-fidelity medium.
I'm still not with you. As far as I can see you failed to clarify your position to a bunch of experienced EEs.Please explain "point of reference".
Please explain "shifts from one ... reference point to another"
Please explain "clearly skewed representation"
Please explain "the ability to present 2 or more levels of threshold simultaneously"
And please do so in a way that makes it clear that you are not discussing quantisation error and a lack of reconstruction filtering.
Also please define the sort of digital system you are comparing to the sort of analogue system. Is this about the mathematics of sampling and/or quantisation? About ADC/DAC chip limitations? About ADC/DAC systems? Mathematical analogue? Magnetic tape? LP?
digital is perfect, ABSOLUTELY perfect!
HowdyIf you had read our posts you would know that. Werner, myself, and others here are well aware of the shortcomings (and strengths for that matter) of digital. Also, if you note, we are willing to give you the benefit of the doubt, but so far we haven't found common terminology to really understand what you are trying to say.
nt.
Not only have you people disrespected my knowledge on the subject, the complexity of the issue is apparently beyond your ability to understand.I will not be participating in this thread any further.
HowdyIf you know the standard terminology why don't you use it? If you don't know the standard terminology or even attempt to use it you'll find it very hard to communicate in the future.
Despite your protestations I and others here obviously know a lot more than you give us credit for and despite misunderstandings at times we still manage to discus technical matters.
nt
nt.
nt
nt.
In wading through this conversation, it appears you've been completely unable to clearly state your position with any clarity. You speak in vague, undefined generic terms. When asked specific questions or given the opportunity to clarify, you denigrate and bluster instead of addressing the issue. That leads the casual observer to this discussion to suspect that you are in over your head.I've heard some wonderful digital recordings and many poor ones. I've also heard many wonderful analog recordings and a vast number of poor ones. The reasons for poor recordings with either method would appear to be due to something other than the specific technology involved.
My knowledge on the subject, as well as the details of my world-changing breakthru designs will you never in your lifetime have a clue about!!!!
you spout gibberish and act like you're being persecuted for your superior knowledge. Next will come the obligatory references to Galileo and Edison.You either don't understand digital technology or you are incapable of coherently communicating in the English language.
In either case, I will bet my fortune that you never come up with a "breakthru" (sic) format.
What beneficial knowledge have you exhibited?
It is true I have not done that.You claim to have invented the CD format or at least been involved with its development.
Where is your evidence for this extravagant claim?
Based on your demonstrated misunderstanding of the basic tenets of digital sampling theory, I simply don't believe your claim.
And it doesn't matter anyway. This is the Internet, your resume is not relevant. You are judged by the quality of your information and the clarity of its expression.
You flunk on both criteria. Your posts contain essentially no information except that you love Accuphase very much.
nt.
I have about 4000 LPs. How many do you have?When you can barely write a coherent sentence and cannot spell simple English words, it is probably unwise to call other people morons and fools.
"My knowledge on the subject, as well as the details of my world-changing breakthru designs will you never in your lifetime have a clue about!!!!"What do these "breakthru designs" entail? Feel free to email me if you're not comfortable stating it here.
I have shared a few of them on this site in the past. One, which I have just opened a thread about on the general asylum, is my full spectrum, frequency-modulated optical analog playback technology.It has been locked in the development stage for the last 30 years do to an industry-wide lack of interest as digital cd has stolen the show!
It is essentially a 12" laserdisc format, and will remain as such until compact disc designs can store a full analog album.
If any of you have anything positive to say, then please join in the thread.
As noted, I haven't seen an explanation of anything from you in this thread. Just bluster and bombast and I don't find them very useful in technical discussions.
This post is made possible by the generous support of people like you and our sponsors: