Post by Petrus  December 4, 2009 (41 of 234)
|
|
audioholik said:
one bit is used also for the sign and the lowest significant bit is used for quantization. Petrus, give it a rest, even Bissie, your Guru, admitted that the difference between "16bit" audio CD and 24bit audio is audible.
Using one bit for sign does not take it away from contributing to the accuracy (basic math*, pal). What is this "using one bit for quantization"??? All the bits are used for quantization!!!
Do I need to give a rest to erroneus facts? This is not trolling for suckers, but for getting facts straight.
*) after 15 bit if we add one more, the scale is doubled. If we use the next bit for sign, the scale is also doubled, but the added double is on the negative side. Difference between the largest and smallest value is the same 2^16 in both cases.
|
|
|
|
|
|
Petrus said:
>Using one bit for sign does not take it away from contributing to the accuracy (basic math*, pal).
it may still contribute but not full time
>What is this "using one bit for quantization"??? All the bits are used for quantization!!!
but the LSB contains as much quantization error as signal so it's basically lost in quantization process
PS if you don't believe me try asking any professional mastering engineer - is the 96dB dynamic range really possible?
|
|
|
Post by Petrus December 4, 2009 (43 of 234)
|
|
audioholik said:
>Using one bit for sign does not take it away from contributing to the accuracy (basic math*, pal).
it may still contribute but not full time
>What is this "using one bit for quantization"??? All the bits are used for quantization!!!
but the LSB contains as much quantization error as signal so it's basically lost in quantization process
PS if you don't believe me try asking any professional mastering engineer - is the 96dB dynamic range really possible?
A recording engineer I know made +100 dB test CDs already in the nineties using a Sony DAT recorder. DR measured from the CD.
"it may still contribute BUT NOT FULL TIME"
You are not form the Nordic countries I take it?
Addendum: Now I start to understand your repeated postings of those "impulse responses" of PCM and DSD done with simulated 3 microsecond pulses. You really do not understand what microsecond is, or what frequency (measured in Hz) stands for and how time and frequency are connected. Amazing! "One bit used for sign, one for QUANTIZATION"! This would be hilarious if it was not so sad!
|
|
|
Post by canonical December 4, 2009 (44 of 234)
|
|
urbo73 said:
This is a bad analogy I'm afraid.
I think the analogy to resolution in digital images is quite good actually ... and I am *not* referring to having to zoom in to see this. I am referring to the way that older / early digital cameras produce images that look grainy, rather like the way that the CD format itself can sound quite grainy. By contrast, newer high-resolution digital formats don't tend to sound/appear grainy when done well.
> SACDs have 120dBs in the spec. Sounds good. But assuming the softest note is again at 30dB,
The softest note at 30dB? My Daikin airconditioner (internal unit) is rated at 23dB in night mode, and it is plainly and obviously audible when it is on.
|
|
|
Post by hiredfox December 4, 2009 (45 of 234)
|
|
jakeroux said:
Thanks for taking the time and sharing the thoughts - I’ve been looking forward to reading your findings. I would guess that you'll get some feedback in the coming days. While I'm not a musician and I use some pretty pedestrian gear, I had come to basically the same conclusion myself regarding stereo SACDs - no perceptible difference for me when compared to a well mastered RBCD, but they do often serve as a type of assurance that you are getting good production and mastering values (which as we all know, can often be a crap shoot with standard RBCDs). Of course, love ‘em or hate ‘em, MCH recordings are a different matter entirely, but for me they are an additional reason to purchase SACD.
And by the way, thanks too for the recommendation on the AKG K702s – I have a pair on order myself now.
Comments like this are one reason why I no longer get involved in technical discussions about resolution on here. There are far better sites with more qualified exponents of the digital arts to discuss such matters.
I repeat ad-nauseum:
I have high-end kit in stereo configuration whereby the superiority of SACD over any form of PCM playback is so evident that without prompting every one of my many friends - and we are all musicians of one skill or another - have never failed to make the distinction.
For this to happen, the differences cannot be subtle can they?
If your kit is not good enough to highlight such differences then up-grade it but stop being delusional.
|
|
|
Post by urbo73 December 4, 2009 (46 of 234)
|
|
canonical said:
I think the analogy to resolution in digital images is quite good actually ... and I am *not* referring to having to zoom in to see this. I am referring to the way that older / early digital cameras produce images that look grainy, rather like the way that the CD format itself can sound quite grainy. By contrast, newer high-resolution digital formats don't tend to sound/appear grainy when done well.
> SACDs have 120dBs in the spec. Sounds good. But assuming the softest note is again at 30dB,
The softest note at 30dB? My Daikin airconditioner (internal unit) is rated at 23dB in night mode, and it is plainly and obviously audible when it is on.
I thought I stated why it's a bad example quite clearly. The biggest mistake people make is exactly in this comparison. I'm not sure what more I can add, except that if people continue to make comparisons to imaging/video and audio r.e. bit depth and resolution, they will continue to not understand and draw exactly the wrong conclusions.
Yes, softest note at 30dB. That may never even happen w/o you turning up the volume like mad. Why? Because the noise floor in your room is at least that. See this chart if you wish for reference.
http://www.makeitlouder.com/Decibel%20Level%20Chart.txt
|
|
|
Post by urbo73 December 4, 2009 (47 of 234)
|
|
hiredfox said:
Comments like this are one reason why I no longer get involved in technical discussions about resolution on here. There are far better sites with more qualified exponents of the digital arts to discuss such matters.
I repeat ad-nauseum:
I have high-end kit in stereo configuration whereby the superiority of SACD over any form of PCM playback is so evident that without prompting every one of my many friends - and we are all musicians of one skill or another - have never failed to make the distinction.
For this to happen, the differences cannot be subtle can they?
If your kit is not good enough to highlight such differences then up-grade it but stop being delusional.
People get really worked up here. The guy was just offering his impressions, what's wrong with "comments like these"?? I don't get some of these posts to be quite frank.
You may repeat as you wish, but again, statements w/o an explanation, don't hold much water. Why can't people discuss w/o getting so worked up? Explain to me please how the differences are not subtle and why.
|
|
|
Post by canonical December 4, 2009 (48 of 234)
|
|
I don't think it's his kit. I think some people are just sensitive to some things, and others are sensitive to other things. Some people can detect the type of car from listening to the exhaust note, some people can pick not only the wine type but the area it was grown in (purely from tasting: not me), some people (myself included) care about the percentage cocoa in their chocolate ... To other people ... it's just choccies. In a famous statistical paper, the father of modern statistics, Ronald Fisher, could not believe that a particular lady could tell how her tea had been made ... just by tasting it ... i.e. whether the milk was added first, or later ... because it all tasted the same to him. But then he tested her ... and she got it right every time. Of course, if he had been a clumsy academic (like some others), he could have just randomly sampled 50 people, averaged out the people who can tell the difference against those that can't, and then incorrectly conclude that human beings cannot tell how their tea is made. But Ronald Fisher just focused on those that can ... and he got statistically significant results to prove it.
We don't all have to be the same ... and if it doesn't work for this chap ... there is probably no point in him going off buying more expensive kit. Maybe Cohiba cigars - but not more expensive DACs.
|
|
|
Post by urbo73 December 4, 2009 (49 of 234)
|
|
canonical said:
I don't think it's his kit. I think some people are just sensitive to some things, and others are sensitive to other things. Some people can detect the type of car from listening to the exhaust note, some people can pick not only the wine type but the area it was grown in (purely from tasting: not me), some people (myself included) care about the percentage cocoa in their chocolate ... To other people ... it's just choccies. In a famous statistical paper, the father of modern statistics, Ronald Fisher, could not believe that a particular lady could tell how her tea had been made ... just by tasting it ... i.e. whether the milk was added first, or later ... because it all tasted the same to him. But then he tested her ... and she got it right every time. Of course, if he had been a clumsy academic (like some others), he could have just randomly sampled 50 people, averaged out the people who can tell the difference against those that can't, and then incorrectly conclude that human beings cannot tell how their tea is made. But Ronald Fisher just focused on those that can ... and he got statistically significant results to prove it.
We don't all have to be the same ... and if it doesn't work for this chap ... there is probably no point in him going off buying more expensive kit. Maybe Cohiba cigars - but not more expensive DACs.
You talk about people's sensitivities (and I do agree some people are more sensitive than others to many, many things), but then give examples of associative memory. These are different things when you stop to think about it.
And even sensitivities can be measured. Associative memory cannot. And that is the difference. And I believe in the examples you give. I myself admit to having associative memories that can't be replicated, understood, or believed by others. But hearing sensitivities are something else. There are no "unknowns" with the output of a SACD or CD disc. There are either frequencies transmitted or there are not. Some can pick them out, some can't. There is no "hidden stuff". A lot has to do with knowing WHAT to listen to, not hearing problems. And then there are frequencies no human can pick out. That's just science. I've been tested several times for hearing, and I'm one who can hear the very high frequencies. So I know my hearing is dandy, after years of playing - acoustic music at that. But I was/am a bit OCD, so wanted to ensure my hearing was top notch.
|
|
|
Post by canonical December 4, 2009 (50 of 234)
|
|
urbo73 said:
... in the examples you give. I myself admit to having associative memories that can't be replicated, understood, or believed by others. But hearing sensitivities are something else. There are no "unknowns" with the output of a SACD or CD disc. There are either frequencies transmitted or there are not.
Your construction re associative memory seems (i) somewhat artificial, and (ii) I'm not sure that it has any relevance.
You argue that there are *no unknowns* in the output of a SACD and a CD.
Clearly, there are no unknowns if you are holding the control buttons deciding which layer (SACD or CD) to play, but as soon as you do a blind test (as you were doing), there are plenty of unknowns. In fact, the problem is very much like being given two glasses of wine, and being asked: which is the Mouton Rothschild? Or which is the 1986 vs 1991? You can do the test with you pouring the two glasses from the two bottles (no unknowns), or with someone else pouring the glasses. Some people will get it right almost every time ... others won't have a clue. Buying myself fancy Riedel wine glasses probably won't improve my score.
Another example: I can't reliably pick a Canadian accent from an American accent. To me, they sound basically the same. But many people clearly can tell a difference, and they find it very obvious indeed. It would be rather folly of me to suggest that there is no difference, simply because it is not apparent to me.
|
|
