i used a DSD recorded Pentatone (Jaarvi's Stravinsky chamber music) for my comparison test. How's that?Originally Posted by bachman
By the way, what do you think about a set of operas from the fifties recorded on four track being released on SACD? Do you think they have more hires content than multitrack stuff from the 80s?
Last edited by bigshot; Jul-28-2012 at 20:50.
advertorial much? Nyquist who?
No big deal? He obviously hasn't tried it! How long does your SACD player take to switch layers? How big a difference in levels are there between the two layers?
Last edited by bigshot; Jul-28-2012 at 20:52.
You, by your own admission, cannot tell the difference, so it doesn't matter what you use. Which is fine ... not everyone can. Some people can tell whether milk is added BEFORE or AFTER to tea ... I probably wouldn't have a clue. So what. I don't spend my life telling the tea drinkers they are wasting their time, and how to make a cuppa.
I haven't heard them ... and I don't like the methods Esoteric use [ Apparently, Esoteric do NOT generally get access to the master tapes, but get sent 24bit 96kHz masters, and then they use some cumbersome messy process where they convert back to analogue, and then Analogue to DSD, so there is an extra unnecessary step in there, and an absence of the master tapes. But I digress. ]
I would say this though ... I have a recording from Sony Columbia of the Tchaikovsky violin concerto with Isaac Stern ... I think it is 1958 ... Sony direct converted it from analog master tapes to DSD ... and it is simply stunningly gobsmackingly outstandingly fantastic. And I have another Sony Columbia of Copland conducting Copland (both on CD and SACD --- both from the same master). And the SACD version is natural and warm and lush and just wonderful ... like vinyl without the hiss and the pops ... whereas the CD version is dry and harsh and digital and lacking in the same warmth and colour.
Last edited by bachman; Jul-28-2012 at 21:09.
It isn't very easy to prove a negative, but it's relatively simple to verify a difference if tere is one. Withall audio equipment, all that takes is a couple of preamps to balance line levels and a switcher. It's surprising how many people buy into expensive equipment and formats and never take the time to check for themselves.
It's pretty easy to get a clue just by reading the specs though. The main advantage of 24 bit and other "hi res" formats isn't high resolution at all. It's dynamic range. In digital, dynamic range and noise floors extend downward. Better specs mean better sound at lower volume levels. At the upper range, they're identical.
This is all well and good, but redbook already has a dynamic range that is double what people need for listening to music. In order to hear the difference between a CD and a "hi res" format, you would have to turn up your stereo so loud, you would incur hearing damage.
24 bit is great for recording and mixing. At work, I've recorded and edited using a 24 bit ProTools workstation. I've also supervised sound mixes. 24 bit is a huge improvement over the old 24 track 2 inch tape format (which I have also worked with back in the day). You can pull up quiet stuff in the mix and it's clean as a whistle. But do I need it in my home? Nope.
What makes you think that is due to the format of the disk and not the mastering? In order to make a fair comparison, the mastering needs to be identical. The only way I found to be sure of that was to use a DSD recording that was only released on SACD hybrid. That way the label has no motivation to hobble the CD layer (something I found LOTS of examples of on SACD hybrids from the major labels.)
The Pentatone SACD I tested sounded absolutely fantastic. Some ofthe most natural and lifelike sound I've ever heard. But it sounded just as good on the redbook layer as the SACD.
Here is the SACD. You can get it through a third party on Amazon for about a ten spot shipped. Check it out and see what you think. It's a good performance too, so you can't lose.
http://www.amazon.com/Stravinsky-His...dp/B0000XKB9W/
Last edited by bigshot; Jul-28-2012 at 21:22.
Wow...and people accused me of being contentious...!
I was reading a story regarding an incident about Neil Young who, after being well-known for his harsh criticism of digital, decided that HDCD encoding made enough of an improvement for him to request his back catalog be re-released in HDCD versions. During that time, he was working with the people at Pacific Microsonics to review test releases of the new versions. The system they were using to evaluate them had several different HDCD-compatible DACs so that they could check the results on all of them. At one point, while Young was in the adjoining room, the tech staff switched DACs and played back a bit of the tape, only to be interrupted by Young, still in the next room, asking if they'd just switched to so-and-so DAC. As it turned out, they had. They then had him stay in the other room, and went through the DACs on a random basis, asking him which unit was being used at the time. They were amazed to see that he got it right 100% of the time. Needless to say, I wouldn't be able to tell such a thing from an adjoining room; in fact, unless we were talking about DACs with drastically different levels of quality, I probably wouldn't be able to tell such a thing while sitting in the "sweet spot" in the room. It appears that some people simply have better hearing than others -- and I'm not sure whether it would be considered a blessing or a curse.
For what it's worth, here's the précis (the only part available for free) from the AES study:
What I find most interesting about this study is that you had three different audio sources:It is currently common practice for sound engineers to record digital music using high-resolution formats, and then down sample the files to 44.1kHz for commercial release. This study aims at investigating whether listeners can perceive differences between musical files recorded at 44.1kHz and 88.2kHz with the same analog chain and type of AD-converter. Sixteen expert listeners were asked to compare 3 versions (44.1kHz, 88.2kHz and the 88.2kHz version down-sampled to 44.1kHz) of 5 musical excerpts in a blind ABX task. Overall, participants were able to discriminate between files recorded at 88.2kHz and their 44.1kHz down-sampled version. Furthermore, for the orchestral excerpt, they were able to discriminate between files recorded at 88.2kHz and files recorded at 44.1kHz.
1) 88.2 kHz native
2) 44.1 kHz native
3) 88.2 kHz downsampled to 44.1 kHz
What really leaps out at me is that the participants had no problem distinguishing between sources 1 and 3, but only were able to distinguish between 1 and 2 on one particular excerpt. It seems odd that 88.2 downsampled to 44.1 would be so easily distinguishable from 88.2 native, but 44.1 much less easily detected. Frankly, if I were reviewing the results, my initial thought might be that there was something wonky in the downsampling process that imposed its own sonic signature on the resulting 44.1 file that wasn't there with 44.1 native. Nonetheless, it seems clear that, at least on the orchestral excerpt, participants could distinguish between the higher and lower sampling rate. This might suggest that the difference between Redbook and "high-res" might be audible, but only on music of a certain level of complexity.
Last edited by regnaDkciN; Jul-30-2012 at 08:30.
I actually find the difference most noticeable with just a plain solo piano recording (especially in the bass), or a solo cello, or a small chamber ensemble like a piano trio.
Also, it's nice that they proved this with statistically significant results using ABX testing ... but that's really a a very difficult and onerous task to achieve. You have to remember not 2 .. but 3 ... sources ... and then say whether the X sounds like the A or the B. That's a very cumbersome and difficult test to prove. I don't know what else has been proven with ABX testing? Does anyone know? For example, has the difference between mp3 and CD been proven with ABX testing? And if so, which mp3 rates?
A much simpler test is to simply ask people: do you prefer A or B? Or which one do you think is hi-rez: A or B? Much easier.
http://mil.mcgill.ca/wp-content/pape...pling_2010.pdf
Papers of this sort do not mean much until corroborated. Still a nice read, but the study gives rise to more questions than it answers. For example: Just how good are black box "professional" software mastering tools? Why are sound engineers so often recording in 1 fs (44.1kHz, 48kHz) only? etc.
Dithers are very important.
Complexity of sound is different than we often think it is. When I was testing mp3 and AAC, they had no trouble encoding digital recordings of Mahler symphonies, but I stumbled across one recording that stubbornly artifacted at a much higher rate than all other kinds of music. It was a Sammy Davis Jr song from the late 40s on the Decca label. Every single track recorded at that session artifacted. I had to bump up the rate to get it to render properly.
Also, the main difference between "hires" rates and CD sound is the depth of the noise floor, not resolution at normal listening volumes.
I am a bit confused. By your own admission, you cannot tell the difference between hi-rez and CD sound, so I am surprised to see you describe the "main difference" as being XX or YY. This makes no sense.
In any event, the main difference to my ears has nothing to do with dynamic range or the noise floor. For my listening needs, CD has puhlenty of dynamic range and I don't really need any more, for home listening. To my ears, the main advantage of SACD ... given a proper hi-rez recording that has been kept clean and untainted through the editing stage ... is that the sound is natural and warm, that bass has impact and clarity that you do not get on CD ... and the problems of CD ... digital and harsh ... are mostly absent. I ascribe these benefits to the fact that at least 6 times more data is collected, with dramatically less need for interpolation i.e. that hi-rez is just more accurate than lower rez. But there are a whole variety of possible explanatory variables where hi-rez exceeds CD format standards ... and I don't really care. Ultimately, what matters is that it sounds better.
It makes perfect sense. If you check up on how digital sound works, you'll find that 24 bit is used for recording and mixing, because the lowered noise floor allows quiet elements to be boosted in volume without bringing up a lot of noise along with them.
However, For the purposes of playing back music in your home, even the most dynamic music rarely even uses half of the dynamic range of regular CDs. For normal playback of music, in order to hear down to the noise floor of 24 bit, you would have to turn your volume up to the level of a jet engine. You would suffer hearing damage. Total overkill.
Due to the Nyquist Theorum, the fundamental principle digital sound is based upon, 16/44.1 (CD sound) is capable of perfectly reproducing sound within the range of 20-20,000 Hz with a dynamic range of well over 70 dB. So the sound quality of CD sound in the range of normal listening volumes for music is identical to 24 bit. The added resolution of 24 bit is down in the depths of the dynamic range where you'll never hear it. If you're hearing a difference, it isn't because of the format itself.
SACDs aren't quite 24 bit, but the same applies to them.
Do you understand the difference better now? I'm trying to explain it clearly.
Last edited by bigshot; Jul-31-2012 at 17:35.
Posting Permissions
7Likes
LinkBack URL
About LinkBacks
Reply With Quote