Originally Posted by arnyk
Umm, too you. Here's reality - Stuart's paper started out in the late 1980's as an attempt to fix the mistakes that were made in the original RedBook standards. It appears that he gave it a retread and tried to get it to fly 10 years later in the late 1990s. Eventually his ideas got some traction and the major recording and electronics companies blew a few hundred million on SACD and DVD-A which may have overkilled his idea, but they were based on his basic thesis that significant audible improvements over RedBook CD format were not only possible, but likely to be commercial successes. Didn't happen in the mainstream marketplace and 100's of millions were lost and careers were seriously damaged. People in high places got fired for hitching their company's wagon(s) to the DVD-A and SACD stars.
That is remarkable progress Arny. You went from yesterday not knowing his paper was published at AES and seemingly not aware of it, to all of sudden knowing all about it and blame Bob for existence of DVD-A and SACD? Even after I explained his position was against SACD, you pin that format on him too? How about blaming him for the current recession we are having?
The reasons those formats failed are clear and have nothing to do with Bob. Consumers have shown incredible apathy for audio/video fidelity. They simply will not pay for them, or put quality ahead of convenience. Neither DVD-A or SACD allowed ripping of their content, coming at an era where you could do that. Nor would they allow consumers to buy individual tracks as they could by either stealing the bits or buying them from iTunes. And of course prices were higher. For *general* consumers, all of these issues soundly trump quality differences and hence the formats failed.
As another counterpoint, both of these formats allow for multi-channel playback. Surely you are not going to say consumers can't tell the difference between 2-channel and multi-channel music. Yet, even though that differentiation existed and clearly identifiable in blind testing, it didn't save either format. Clearly that was not enough to trump lower value and lack of convenience.
It is important to make another point. There is a temptation here to extrapolate the topics we are talking about to general public. That is wrong. This is an enthusiast forum. We don't care what the general public thinks of our hobby or desires for fidelity. Take video. My wife would not buy anything but DVDs. They are cheaper and she finds them on sale at the stores she shops. I am the only one buying blu-rays and at extra cost. To her, DVD is "good enough" as to not be worth spending anything extra for quality. Truth to be told, I watch those DVDs with her and if the story and acting is compelling, I enjoy them just the same. By your logic, this must mean that there is no fidelity difference that is observable in blind tests against DVD whereas we know the reality is otherwise. Mass consumer preference for a format is not the same as us voting for what is improved fidelity or not.
If I asked die-hard audiophiles if they would buy either one of these formats they would say yes. But sadly, they do not make a reasonable enough market to enable distribution of higher fidelity content. So the formats did not die because the fidelity wasn’t there necessarily but because there is not a mass market for it. With HD Tracks selling high resolution music, the bits that were in these formats may be coming back anyway.
So no, Bob is not the reason either one of these formats got created, or failed. The industry had a format war on its hand and Bob was doing its best to speak about his wisdom after nearly two decades of designing such gear at that time. He didn't work for Sony and Philips who wanted to advance their patent position which was lost on the CD (time period was expiring). Nor was he part of the Panasonic, Toshiba and crew who wanted to advance the same for DVD-A. You want to know why these formats existed, you need to look in that direction
As I noted earlier, despite lack of involvement from Bob and his company, his company's lossless compression became part of HD DVD and then Blu-ray. We should be thankful of his contributions rather than putting him down just because he dared to say we need to go a hair past CD to assure inaudibility of the distortion in our distribution format. He strives for excellence and we need more people like him, not less.
Sounds great on paper, but fails the sniff test when you do good listening tests. Stuart has had the resources for over 35 years (Meridian was founded in 1977) to come foreward with even just one proper listening test that showed that what he was talking about actually did what he kept claiming - provide superior sound quality.
His job isn't the same as yours Arny. Your goal is how crappy we could make audio until folks can hear distortion. His job is to provide a specification that has a design margin that allows us to establish inaudibility without much guesswork. That is a higher bar than you strive. By your standard, when building a bridge, we should go by the average weight of all the cars going on it, rather than increasing that by a safety margin. If we are talking general public, your standard is more than enough. Indeed it is too good as consumers are good with lossy compression even though we know that doesn’t achieve full transparency.
That is beside the fact that if Bob had documented the tests you ask about, you would find 1000 faults with it as to dismiss the results. Yet if Monty says there is some listening test that proves having higher bandwidth is bad for us, you give him high five without even asking for the test details. The mere mention of what you want to hear is good enough evidence. But 37 page paper written and presented at AES is not worth anything but to be used to bad mouth one of the industry leaders in digital audio.
Here is an example test you and I discussed:
"DVD-Audio versus SACD
Perceptual Discrimination of Digital Audio Coding Formats
Listening Comparison Test between DSD and High Resolution PCM (24-bit / 176.4 kHz)
Dominik Blech and Min-Chi Yang
Erich-Thienhaus-Institute (Tonmeisterinstitut), University of Music Detmold, Germany
The test was ABX and designed to see if anyone could tell the two sources apart. Almost no one did except for four people:
"The four highest scores fell into the region of “critical probability.” This amounted to only 2.76% of all the tests. These four tests were carried out by four separate listeners, all of whom chose stereo music examples, and in all four cases headphones were used—thus excluding the influence of the listening environment to the greatest possible extent."
So we have four people hearing the difference between SACD and DVD-A. Your counter was here: http://avsforum.com/avs-vb/showthrea...2#post20277042
saying the test was done in 2004 and M&M test was in 2007 so the latter result is more valid. Never mind that the M&M test had nothing to do with comparing SACD to DVD-A, and didn't follow the same protocol or anything like it. But hey, you like its conclusions better so you ran with it. My answer is a few posts later: http://avsforum.com/avs-vb/showthrea...9#post20277189
Additionally, as I just noted, in M&M test, one person got 8 out of 10 right. If I am building a format that I claim to be fully transparent to the source, I wouldn't want anyone getting such a high score. The fact that this statistic doesn't rise to 95% wouldn't be enough for me. I would want the format for high-end enthusiasts to do better than that. Again, the target here is not the general public so it is not important to me that 50 other people got lower scores. I would have taken that one person aside and tested him some more rather than summing everyone together and declare “50-50” results.
As I point out below, if the above is true and I think it is, Stuart wants people to use almost 2 bits for every bit that would be required in the minimal case.
You think it is? You didn't read the report as Swampfox did? Did you read Swapfox’s proper summary of how he is rounding up to the next sampling rate since we don't have 58 Khz as a valid option? How about the fact that Bob goes down from 16 to 14 bits? He doesn't get credit for that? You are forever non-objective when you look at evidence like this Arny
Of course not. I can actually construct a technical argument that you need 48 KHz sampling to do a great job of capturing the music on a 15 ips half track analog tape master. But, it all falls apart when you include the human ear.
Once more, enthusiasts want all that existed there. If that requires 48 Khz, then you don't get to chop down the content for them because you have this M&M test in your pocket. If you claim to have a copy of the master tape in digital, then you better deliver that as is. What is so wrong with doing that Arny?
As you correctly imply, Reel to Reel does not have a brick wall filter on its frequency response at 22 Khz. So capturing it at 44.1Khz may indeed leave some of the response on the table. Why not capture that and give it to the discriminating consumer? Why go on and on, advocating a lesser standard?
Do you need 24 bits or 20 bits to capture the music on that half-track master? Don't be silly! The noise floor on those great-sounding old tapes is in the 12-13 bit range depending on how you do the digital. With aggressive noise shaping, 11 bits might work.
For others who don’t know what we are talking about, when digital systems use fixed bit depth, you create distortion since analog waveforms do not have such steps. The way to solve this is to add some amount of random data in the form of noise to the signal prior to digitalizing it. That converts distortion into noise which is a good thing. What is not a good thing is that we have now raised the noise floor of our system relative to the source. Noise shaping says instead of adding noise that has equal power in the audible frequency band, we make it so that it has the least amount where the ear is most sensitive and more of it where it is not.
Given the above, the best use of noise shaping is when you have bandwidth above 20 Khz to stuff the noise there. With 44 Khz sampling, your response stops at 22Khz so you are now essentially forced to put the quantization noise in the audible band. This is one of the other reasons it is better to have some headroom in our sampling rate as to give us a place to “park” these dither noise. If you used 88 Khz for example, you now have 50% more bandwidth to push the noise into.
The above is the type of analysis that Bob does in his paper (and then some) to arrive at his recommendations. He puts thought in there combined with excellent grasp of technology and science. Whether you agree with this conclusions or not, it is worth a read. As I have said, I have always used his paper to prove that we don’t need astronomical sample rates and bit depths for transparency. Its use in reverse by Arny is very odd.