Is the CD good enough?

[Mallette]

>I still think digital has a much harder time with
the subtleties of either solo or massed strings, than with harder,
clearer, sounds of solo wind, brass or percussion, or pop voices.

That simply defies science, IMHO. If anything, the opposite would seem more likely if we were just dealing with intuitive hunches rather than science. Having used magnetic colloidal suspension fluid many times to view the tracks on tape (mainly to judge alignment) and loupes to look at the optical tracks on film I can say they "look" more like what we expect from the smoothing of tubes compared to the absolutely accuracy of high res digital.

I continue to maintain it's the engineering, not the medium. One of the high res recordings I did (not posted) is of pianist Denise Lippman and the head of the New England Conservatory of Music (can't think of his name at the moment) on cello. The detail of the bowing and fingering of the cello is the most up close and personal I've ever heard in a recording. One of the pieces I did put up was of Paul Garner on clarinet with Michele McDonald on piano. Turned up high I can ear his finger moving to half hole...that is detail I am not sure ANY analog tape could get as even with DBX and high speed large format tape it would be near the noise floor.

Again, please remember I have pretty neutral as to left/right brain. 90 percent of my collection of recordings are analog, and it was analog that led me to question digital and my favorite mike is 72 years old and I use ONLY vacuum tube mike preamps. However, the evidence as revealed to me simply does not support analog being, by nature, superior to digital in any way.

Dave

[garyrc]

To be more specific, the problems I have with recordings of massed violins are the following, described in a very subjective way -- the only way I can think of that might capture my impressions:

1) Recorded: The worst problem is that they tend to go "steely" rather than "gutty" -- they tend to sound a little more like their strings are made of metal than cat gut or a gut-like synthetic. They sometimes sound harsh, when none of the other instruments playing sound harsh. Strangely, when amateurs recorded our orchestra on a pro Ampex, no board, with old Telefunken microphones (probably the ancestors of the U47) suspended about 10 - 15 feet above the orchestra, there was no such problem. Likewise no problem with 2 ribbon mics (RCA 77?? probably) in just about the same position. In both cases, the orchestra sat on a beat-up wooden floor, with carpeting only in the audience area. The steely sound stands out more when using some speakers than others but I've heard it on speakers famous for how well they treat solo strings, including a variety of vintage Bozaks.

Live: A warmer, richer sound, no harsh sheen, but still with plenty of authority.

2) Recorded: Yes, as Marvel said, they sometimes sound mushed together, but they are often spread out over the stereo field pretty well, and are heard with some front to back depth. It is a different kind of lack of detail that I object to. The following kind seems to be missing in most recordings:

Live: Very exciting differences in attack, bowing, etc, between violins playing the same part at the same time; they can't all play exactly the same, and the slight differences sound good! Also, the scraping, some would say "nonmusical" friction sound one can hear when sitting very close to the orchestra is quite musical to me, and rarely comes through on a recording of massed strings but sometimes does when a solo string instrument is recorded.

[Mallette]

Yessir. IMHO everything you observe here is engineering. Gimme the 77's and 24/176.4 or a vintage Ampex half-track and I'll deliver you a warm, detailed image of however many strings you want to mass. However, I'll prefer the digital so I do not have to deal with noise, limited dynamic range, and some detail dithering due to nature of the magnetic process.

Dave

[Chris A]

I was researching bit width / sample rates and found the following:

http://mixonline.com/recording/mixing/audio_emperors_new_sampling/

This article describes tests run by Brad Mayer and David Moran that were reported in the September AES Journal titled "Audibility of a CD Standard A/D/A Loop Inserted Into High Resolution Audio Playback".

Don't throw your CDs away yet. They are better than some think.

Don

Honk if you love Horns

Don - I bet you know that this article you reference is really speaking the truth. However, there are a lot of folks that don't want it to be right. Anyone that tells you that CDs are no good (without qualification of which recordings) have really destroyed their credibility. CDs, as a format and a technology, were invented to replace records. The folks that invented CDs did a good job at doing that, regardless of the exhortations to the contrary that you might run into here. What has happened, in my opinion, is that many people listen to poor CD recordings and mixes (not to mention poor music). But many don't want to hear that (and especially from someone that they haven't been sparring with here for a long time).

I now just basically ignore arguments about recording formats and pay closer attention to recording engineers and the techniques they use. That is where I believe the differences are. For instance, I might recommend Chesky Records, Reference Recordings, and DMP CDs. They seem to employ really good recording engineers. That one CD exception to the rule that others may try to lay down here proves your point.

I'm afraid that much free advice on this subject offered here appears to feed egos, not the ears IMHO. Many audiophiles aren't buying better sound - instead I think they're buying something else. For instance, many people here think that electronics and turntables are the key to give you better sound. I believe it is the complement of those, namely high quality speakers, good room acoustics, good recordings by good recording engineers, and good musicians.

Just because you may get drowned out here doesn't mean that your notions aren't correct. I recommend talking to Roy Delgado (Klipsch lead design engineer) about his views on this subject. I think that you'll find a kindred spirit there.

Regards,

Chris

[PrestonTom]

Chris, you are a voice of wisdom!

There are certainly a number of things involved and that is why all sides of this issue end up being "correct". However there are the big issues and there are the small issues, and you have hit the nail on the head.

Good Luck,

-Tom

Incidentally, over the last several weeks I find that I just keep on playing Dave Mallette's recordings. They really do capture some difficult-to-capture details. That was a very generous of him to share those recordings (and much appreciated).

[Mallette]

Chris, Tom, and all:

Almost all the posts in this thread have some good points, and certainly Chris's is one of those. The question with Redbook is "Good enough for WHAT?" The designers of Redbook must be applauded for reaching FAR ahead of their time. In a time when a 20mb HDD was really big they pushed the envelope in every way to deliver a trully remarkable piece of technology that has stood the test of time. To do so they had to make some hard decisions. It's obvious they made the right ones within the context of the technology they had to work with.

That said, it is equally a fact that their are nuances 16/44.1 loses. I have ABC'd the same recording of applause at 24/176.4, 24/88.2, and 16/44.1. Even my 58 year old ears can easily detect the difference. However, those ears are pretty well trained and I suspect that 90% of younger ears might not hear much, and if they did they wouldn't care.

How many of the general population can tell the subtle but profound difference between a Stradivarious and an Amati? Not many, but that does not make the difference any less important. It is this level of difference we are discussing (or at least I am). The die hard LP lovers in this group are not, at least for the most part, luddites holding to the horse and carriage. What they hear is real and significant. However, it is NOT analog that is making the difference. It is resolution.

My answer would be that CD is good enough for the vast majority of material, but not all and not for the die hard audiophile who loves acoustic instruments and grand acoustic spaces. If the equipment were there, I'd record at 24/352.8. Why? Why not!! Is there such a thing as too much accuracy?

And, of course, a poorly engineered recording sux at any resolution.

Dave

[PrestonTom]

Dave, we are probably in agreement on these issues (we just don't realize it).

Unfortunately many of the arguments (LPs vs whatever, or 16 bit etc vs whatever) become reduced to a level where someone heard something on a LP (or a 16 bit whatever) and then listened to a 16bit (or a "special" high bit or sampling etc) and heard some difference.

What gets lost in all of this is that the mastering for the two may not have been the same, or even similar. This is when folks start "averaging". They tend to average their experience with LPs or 16 bit CDs or whatever and come up with an "average" experience with the format. Now, we really are talking about apples vs oranges.

The point Chris is making and that I agree with (I suspect you do also) is that there are all sorts of variations across these recordings (that went into our "average"). Some these variations are whether it is LP, CD or whatever. But there are other, and perhaps more important variations, that arise from the recording process itself (how it was done) and the mixing process (what "other stuff" was also done). I believe these variations are far more important.

Lets look at a simple one. Sheffield labs used to produce a "direct to disc" LP. No mixing, compression or other trickery was performed. These LPs could have a very strong sense of realism. Now compare to this to a conventionally recorded, and mixed, and compressed, etc, pop recording on a CD. The CD will not have that same sense of realism. This really is a case of apples vs oranges. Some would emphasize the medium (LP vs CD). While others, myself included, would point to all the other differences in how it was recorded. I suspect the real evils are compression and adding artificial reverberation etc.

I believe that CDs offer better resolution (even 16 bit ones) etc. But I also believe that many LPs can sound better. The difference is the engineering (now I am quoting you).

Personally, I have heard very few LPs that have had the same sense of realism, resolution, and detail that a well-recorded CD can have (not a typical CD). I would point to your recordings as being a nice example of such detail and resolution.

Good Luck,

-Tom

[kev313]

This discussion reminds me of an interview that Fremer did with Rick Rubin. Apparently Rubin feels vinyl is superior, although this was certainly a fluff piece and has nothing by way of a technical discussion. It is also enlightening in that Rubin clearly gives the audiophile crowd a big finger on the issue of compression. Bottom line is that compression is good for business and audiophiles ain't. That said, Rubin, who can clearly afford to indulge himself in such things, uses fancy pants power cords to the great dismay of snake oil theorists. http://www.musicangle.com/feat.php?id=38

[LarryC]

Whatever the potential is, or was, of CD, it is only rarely achieved, as is true for vinyl and tape too. If you want a 100 pcs of media to listen to, get your Chesky CDs or Sheffield LPs and listen to them endlessly. Mostly, the specialized labels have a lot of "who cares?" artists. If your listening world is "demonstrating" your exotic stereo to others, those items are fine. How many times are you going to want to hear Billy Burnette or Dave Grusin? If you like good sound and you like good music (whatever that is to you) then what counts really is what does the "average" CD, LP, 8-track or master tape sound like? Not the few primo examples, but the ones with music on it that you want to hear.

There is a fundamental underlying reality that engineers in the "LP era" didn't use as many "deadly" means as engineers do today in the CD era. By which I mean compression and all the rest that we now understand make the modern CD.

Great comment! From my standpoint, the leading edge of digital and CD recording technology is somewhat of passing interest, since if I want to listen to Mendelssohn's Scottish or Italian symphony by a fine orchestra and conductor that I know, it doesn't matter to me that I might be able to get more string detail and presence with a cutting-edge digital recording technology, if that work and that kind of performance isn't available with better technology. It takes years and decades to build that kind of library as you know.

My reactions are to what I hear from average CDs and average LPs in my collection and what I buy new. After Those are the only ones I can get, and I'm not going to toss out that collection. For me, an LP still has an edge in reality, and being closer to music and the way instruments and voices actually sound. Yes, in SMALL group instrumental and vocals, CDs are sometimes superior. However, I had to have a LOT of modding done to my CD player to get it to be comfortably close and sometimes superior to my LP system.

[Don Richard]

There is a fundamental underlying reality that engineers in the "LP era" didn't use as many "deadly" means as engineers do today in the CD era. By which I mean compression and all the rest that we now understand make the modern CD.

I really think that the "deadliest" thing that happened was the use of 8, 16, 24, track tape. Heck, you could slave as many tape machines together as you needed for as many retakes as you wanted. And you could do "flying switches" between several takes to assemble one "good" track. Multitrack production exploded because of the "perfection" that was attainable. This occured before CDs came on the scene and got worse after digital recorders were made for studio use. 48 and 96 track became common, and of course, they could also be slaved. The talent realized that no matter how "off" the performance was "it can be fixed in the mix". So subconsciously, perhaps, the performers stopped trying to nail a good performance in the studio.

Then came the boxes. There is now a box out there to fix any problem. Vocalist singing flat? Put er through a pitch shifter, voila!, magically in tune. Kinda dry from close micing? Digital reverb to the rescue. Timing off? Line it up in ProTools. Etc, etc.

Looks like a skill issue. Or several simultaneous skill issues. Performance skills, recording skills, mixing skills, mastering skills. And the chain always breaks at the weakest link.

[garyrc]

1) Lp v CD:

I found myself doing the same thing in the Lp and CD eras ----- creating two storage areas, one for acceptably good recordings, and one for recordings that were so bad that I would very rarely listen to them.

The most painful situation was/is when a great performance was imprisoned by a lousy recording. Those usually ended up in "good" area, but stirred some sadness and anger when played.

2) Compression:

Many Lps -- especially classical/orchestral -- were either compressed or recorded with limiters in the 60s and 70s, which drove people to buy dynamic range expanders (which weren't very successful).

To avoid compression, but still deal with big crescendos, the recordists sometimes set the level so that the very loudest passage would not be overrecorded (as Mercury often did, especially in the Fine era around 1962). The cost would be hiss coming through during soft passages (sometimes minimized by using 35 mm magnetic film), and any pops or clicks on the record being more noticeable during the soft parts. I preferred this to compression or limiting.

Some of the best sounding phono cartridges of the early 60s (e.g., Ortofon SPE, elliptical moving coil) could not track -- without a clear increase in distortion -- very loud, "heavily cut" passages, so some audiophiles had a "cartridge caddy" which contained some slightly worse sounding cartridges (Shure V-15) that could track almost anything out there.


Nothing is simple.

[Mallette]

Having said more that once in these pages that I'd rather listen to a first class recording on a boom box than a poorly engineered one on the world's finest system it should come as no shock that engineering is much more important to me than medium.

However, to return to the original question poseed by the originator of this thread: Is the CD good enough?

I must assume he refers to the the format, not the contents, so:

NO, it is not. All things equal (engineering, that is), vinyl will sound better and any higher resolution digital format will as well. Further, the highest level digital resolutions will sound better than the analog due to dynamic range, S/N, and the vagueries of the analog medium.

Point is, I like them all, and I HATE all badly engineered recordings no matter how they are presented.

Dave

[Rivendell61]

If the equipment were there, I'd record at 24/352.8. Why? Why not!! Is there such a thing as too much accuracy?

Hmmm, as I recollect the AES article does not add much constructive to the CD vs DSD discussion.
Not a very well designed ABX test if it was trying to claim inaudibility....but maybe it was not.

The prime article Don links has some misinformation:
High sampling rates do NOT provide more accurate timing resolution.
Redbook timing is in the nanosecond range--timing resolution is NOT constrained by sample rate.
Maybe I'm nit picking but that incorrect contention breeds constant false implications.....

Mallette,
Am I misinterpreting you?...or are you falling into the 'higher sample rates capture more detail' fallacy?
Rates have nothing to do with what is usually implied by 'resolution'.
A good converter should NOT sound better at very high rates. Quite the opposite in fact.
If your CardDeluxe seems to sound better--it may be it's way of suggesting that you one day invest in a better converter :-)

Mark

[Mallette]

If the equipment were there, I'd record at 24/352.8. Why? Why not!! Is there such a thing as too much accuracy?

Mallette,
Am I misinterpreting you?...or are you falling into the 'higher sample rates capture more detail' fallacy?
Rates have nothing to do with what is usually implied by 'resolution'.
A good converter should NOT sound better at very high rates. Quite the opposite in fact.
If your CardDeluxe seems to sound better--it may be it's way of suggesting that you one day invest in a better converter :-)

Mark

I was pushing it with the 352.8. However, there is no question that increasing the high frequencies captured has an audible impact on the sound...or prehaps it is something else we can hear but not yet quantify that we (or I) assume to be related to wider ultrasonic bandwidth.

I am only in a position to stand that in tests of the same material recorded through the same mikes at the same time that 24/196 sounded clearly, if subtlely, better than 24/88.2 and that 24/88.2 sounded definitely better than 16/44.1.

In LP recording, 16/44.1 sounds very good, like a first rate CD. 24/88.2 sounds just like vinyl.

I've been dubious of measurement as a gauge of "quality" since the absolute scientific proof that SS was superior to tubes in the 1970's. Only the brain can sort this out until our science becomes less about measurement and more about understanding of the functioning of the human brain.

Dave

[Don Richard]

I think that the AES paper shows that few people can hear a difference in digital storage sample rates. When people don't hear differences they don't buy. Low sales is the reason that the guy at the store thinks the hi-res media will fail at the consumer level. It is my theory that the music industry wants these formats mostly for better copy protection, not increased performance.

[Mallette]

Don:
I have absolutely no doubt that the AES paper is correct and also your other conclusions.
However, I do not record either to make money or to cater to a consumer audience. Since my 58 year old ears clearly hear these differences (though quite deaf to magic cables and such) and I have no doubt that many who frequent this Forum can as well, I elect to record at higher resolutions. Sheesh, I can hear the difference in a 78 recorded at 16/44.1 and at higher res.

I don't do this to make money, and it's a darn good thing since I'd have been broke a long time ago. I don't think those who spend a small fortune to purchase a hand made preamp that is flat from DC to light are going to want to feed it source material that isn't as wide range as technology can provide.

Whether one hears it or not, the CD has limits that are unnecessarily restricted at this point in history. The nice thing is that audiophiles are no longer tied to whatever the music industry decides is a format. It is my desire to both provide material at a higher standard and do what I can to make it easier for audiophiles to enjoy it.

Dave

[Don Richard]

Don: I have absolutely no doubt that the AES paper is correct and also your other conclusions. However, I do not record either to make money or to cater to a consumer audience. Since my 58 year old ears clearly hear these differences (though quite deaf to magic cables and such) and I have no doubt that many who frequent this Forum can as well, I elect to record at higher resolutions. Sheesh, I can hear the difference in a 78 recorded at 16/44.1 and at higher res. I don't do this to make money, and it's a darn good thing since I'd have been broke a long time ago. I don't think those who spend a small fortune to purchase a hand made preamp that is flat from DC to light are going to want to feed it source material that isn't as wide range as technology can provide. Whether one hears it or not, the CD has limits that are unnecessarily restricted at this point in history. The nice thing is that audiophiles are no longer tied to whatever the music industry decides is a format. It is my desire to both provide material at a higher standard and do what I can to make it easier for audiophiles to enjoy it. Dave

The reason I went to the high end audio shop was to demo hi-res audio to see if there was any difference that I could hear. Despite being engouraged to stay away from hi-res digital, I think I am going to get a universal player anyway, to see for myself.

Oppo, here I come.

[Mallette]

You will be happy with an OPPO, IMHO. However, remember that just having the ability to play high res digital and having material that does it justice are two different things. I have several DVD-A's with which I am not overly impressed. In fact, I think some of my own CD's sound better. However, I heartily recommend the Linn high res downloads. I REALLY like the Dunedin Consort "Messiah."

Dave

[Don Richard]

You will be happy with an OPPO, IMHO. However, remember that just having the ability to play high res digital and having material that does it justice are two different things. I have several DVD-A's with which I am not overly impressed. In fact, I think some of my own CD's sound better. However, I heartily recommend the Linn high res downloads. I REALLY like the Dunedin Consort "Messiah."

Dave

Thanks for the info, Dave. OPPO seems to have a good reputation and good reviews. And I'll definitely be downloading some "good stuff" to try.

[Mallette]

Thanks for starting this thread. Some of the "auld school" here groan and run away everytime this comes up. However, over the past years every time it does we get a bit closer to learning something useful.

While I have my opinions, what I come here for is to learn.

Dave