Thread: Solti's Ring on SACD -- the ULTIMATE?!

That was during the pre-digital era! I used analog gear--Studer, Revox, and Tandberg reel-to-reel decks. Around 1988-90 I bought a Sony PCM converter that used a video recorder as the storage device. It sounded a bit cold and clinical to my ears.

Originally Posted by Kontrapunctus

I used analog gear--Studer reel-to-reel decks.

When I started out, I worked with Nagras and magstripe in the film business. Digital is quite different. There aren't subtle shades of difference between decks and there's no generation loss. When you lay something down to digital, it is what it is.

It's good that you have experience because I can give you some specs that will make it clearer to you what is going on.

The big point to understand with digital audio is that high bitrates don't increase resolution. CD sound is already 20hZ to 20kHz stone flat with no distortion. Sound can't get better than perfect. High bitrates lower the noise floor, so instead of having 90dB of available range, you get 120dB or more. The thing is, most recorded music doesn't exceed 40dB, and even the most wide ranging music has about 60 dB. The average living room has an ambient noise floor that doesn't allow more than 40dB, and in order to clearly hear the entire range the CD format is capable of with headphones, you would need to turn the volume up to a level that would cause hearing damage.

What use is high bitrate sound then? Well, it's invaluable in a mix to be able to boost the level of a small sound like a flute cleanly without bringing up the noise floor along with it. I'm sure back in the analogue era, you ran into situations where boosting something made the tape hiss rise and fall along with your signal. Recording in high bitrate extends the range you can boost, making mixing easier.

Also, CD quality sound has a curve where resolution falls off. This falloff occurs at the very bottom of the dynamic range in the quietest of sounds- things that are not audible without a huge volume boost. In normal listening because this falloff occurs far below the level where human hearing loses its sensitivity. But in a mix, you might want to bring up a sound that far down, and you want it to be clear, not distorted.

High bitrate audio is useful for mixing where a super wide dynamic range is necessary. But once the mix is finalized and it is bounced down to redbook, there is no audible difference. At listening volumes, even loud ones, you don't get close to exceeding the specifications of CD quality sound.

So if there is no difference, why do SACDs sound different than the same album on CD? The reason is that they don't use the same master to make them. Sometimes it's as simple as just pulling a first generation master tape instead of a submaster. But SACDs are almost always remastered, and often they are even remixed. They replace analogue reverbs with digital ones. They apply noise reduction to eliminate tape hiss. They reequalize the sound to make it brighter sounding, or add a bass harmonizer to extend the bass response an octave lower. All of this monkeying around with the sound either makes it sound better, or it just makes it sound different. Then they lay a totally different old master onto the hybrid redbook layer and lower the overall volume so a casual comparison between the redbook and SACD layers seems to indicate that the SACD layer sounds "better".

All of this is at its core deceptive, because you aren't hearing an improvement because of the format. You're hearing it because of the remastering. Wait a few years and they'll pull the SACD master off the shelf and use it to produce a newly remastered CD. And it will sound *exactly* like the SACD. Paying more to get a non-standard disk with proper mastering doesn't encourage record labels to improve the mastering of CDs. It just encourages them to create new "whiz bang" non-standard, non-compatible formats so they can convince you to buy Pink Floyd's Dark Side of te Moon for the umptenth time.

As for your Kreisler SACD... When that was released on SACD, it was a huge improvement over the mastering on previous releases. Until the SACD series, the Living Stereo recordings had been given the budget line short shrift when it came to mastering. But RCA recently released a box set of Living Stereo CDs and used the SACD masters to produce it. With the exception of the center channel on some of the Living Stereo SACDs (because some were recorded in three track stereo) the $2 a CD box set sounds *exactly* like the SACDs. i know this for a fact because I have a pile of the RCA SACDs and the Living Stereo box an I've compared them.

Now that I've gone through my longwinded explanation, go back and read my previous comments. I'm not being an a$$hole. I'm simply stating the facts.

I agree that better mastering can make a huge difference on how old analogue recordings sound on cd, but I question the statement that redbook digital is perfect from 20 to 20k. By its nature it can only be an approximation. Perhaps you meant to the limit of the ear to tell the difference, but I've yet to hear any cd match live sound (or even quality RTR on a high end system).

Originally Posted by bigshot

Now that I've gone through my longwinded explanation, go back and read my previous comments. I'm not being an a$$hole. I'm simply stating the facts.

Calling us "audiophools" is what prompted my remark! (Admittedly, it is a clever pun...) You do sound like you know your stuff. In general, most SACDs sound warmer and less "freeze-dried" than most RBCDs to my ears. Be it different masters or whatever, I definitely prefer their sound, especially since I switched to a multi-channel system.

My understanding from what Bigshot has posted (and forgive my lack of technical knowledge) is that if the music sounds better that's because it's been RE-MASTERED and has nothing to do with the format (SACD).

So I think we can all at least agree that music does sound BETTER on SACD. Just not for reasons people can agree on. (Though I wouldn't be surprised if the music industry was indeed duping people into buying a format they don't need. That kind of stuff has been going on forever. From the VHS/Beta wars to constant computer upgrades that only wreak havoc.)

Further, I'm sure we can all agree that Solti's Ring could use a new and improved RE-MASTER. Whether Decca decides to call this SACD-hybrid or what have you; the point being it WILL sound BETTER.

So why don't they just RE-MASTER it and make some money?

I haven't found any correlation between remastering and better sound. Some sound a little better, some sound worse. A good case in point is the recent remasters of the Beatles catalog. The primary difference between the new CDs and the original CD issue is that the new ones are a little compressed. I much prefer the originals.

Generally, I look skeptically on remasters, because the original engineers and artists signed off on a specific sound. Having an engineer go in thirty years later with completely different equipment and rejigger everything may result in a different sound, but it's not as likely to result in something more faithful to the masters as a straight transfer off the masters themselves. The vast majority of remasters don't sound better or worse, they just sound different. An example of that is Let It Be (Naked). Phil Spector's Let It Be sounds nothing like what the Beatles intended their Get Back album to sound, but neither does Let It Be (Naked). They are both completely different than the Peter Sellers acetate of the way the Beatles left Get Back. Which one is better sounding? The new one of course, but I'll take the Sellers acetate warts and all over the modern digital mush version any day of the week.

When it comes to Solti's Ring, John Culshaw put everything he had into that recording and the CD releases have been faithful to the masters. I don't know why anyone would want anything different than that. Decca had excellent engineering at that time and the masters are four tracks with stereo on the orchestra and stereo on the stage with the singers. Culshaw carefully positioned the singers in front of the mikes to get a clearly defined soundstage. Why would you want to go in and monkey with that? Leave it alone. It is what it is.

Originally Posted by Rangstrom

I question the statement that redbook digital is perfect from 20 to 20k. By its nature it can only be an approximation.

Here is the longer, more technical explanation...
http://en.wikipedia.org/wiki/Nyquist...mpling_theorem

The short answer is that 16/44.1 is enough to completely and accurately reproduce from 20hZ to 20kHz at any human listening level. All of digital audio is based on this theory. Higher bitrates just give you more resolution at lower volume levels.

You can't believe what you read in stereo magazines any more. It's all designed to sell you something you don't need. I've been a hifi nut for a long time, and I remember the scientific and technical efforts that were taken to squeeze the last bit of sound out of analogue components. I was suspicious of the "perfect sound" motto too at first. But I read up on the science behind it and did my own A/B tests and came to the conclusions I've reached.

Since redbook audio is perfect, and a WalMart CD player puts out 20-20 clean at with more dynamic range than you would ever need, equipment salesmen have had to resort to outright deception to get people to spend more money. They do this by juggling numbers and creating false analogies. For instance, I'm sure you've heard about "jitter". It sounds logical. That little silver disk is spinning around really fast, maybe the sound could be jittering...

Well, the Audio Engineering Society published a study on jitter based on hundreds of listening tests conducted by audio engineers, musicians, golden ear audiophiles and just plain folk. They listened to a huge variety of equipment, from the highest high end systems to normal ones... even their own systems. The study concluded that the threshold of audibility for digital jitter was 100 times the range jitter is found in even the cheapest CD player at Costco. They couldn't hear jitter until it was magnified 100 times. Jitter is complete hoodoo.

Another argument is that even though redbook perfectly reproduces the audible range of sounds, frequencies beyond what humans can hear are important. (I've had stereo salesmen try to use this one on me to sell me a SACD player.) But the AES also conducted a test on that, and they found that although many people can perceive sound pressure of ultra high frequencies, these super sonic frequencies do not have any impact on sound quality at all. In fact, they induce headaches and listening fatigue. Another study found that you can filter off all frequencies above 10kHz (that's the top octave in the audible range) and the vast majority of listeners will say that it sounds just as good as a recording that goes all the way up to 20kHz.

The reason I call them Audiophools is because they don't do their homework. They trust equipment manufacturer's advertising, magazine advertorials and the hot air of audio equipment salesmen... and they pay through the nose for snake oil.

I agree with you that open reel tapes sound the best. The reason for that is that the manufacturing process on open reel tapes doesn't require remastering. You just plug a deck into a player playing the master and you hit the red button. Open reel tapes are more likely to be straight transfers off the master with no "sweetening" or compensation for technical issues, like the RIAA curve on LPs. But if you take the best sounding open reel tape in the world and capture it to 16/44.1, it will sound exactly the same. The difference between the master and the CD quality copy will be imperceptible. I know because I've done this test with both open reel tapes and half speed mastered Sheffield Lab LPs. I've also done direct comparisons between high bitrate music recorded and played back on a high end ProTools workstation and a redbook bouncedown of the same recording. No difference.

CD sound is all you need.

Originally Posted by Kontrapunctus

Around 1988-90 I bought a Sony PCM converter that used a video recorder as the storage device. It sounded a bit cold and clinical to my ears.

I noticed you added this... If that is the one with the separate Beta portapack (PCM 10?), I used that too. I worked for a Hollywood sound man who recorded the first TV program to be recorded digitally... Barry Manilow's Copacabana special. I did the transfers and dailies on that show using that deck. It took a second for the sound to resolve when you hit play, but the Nagras took a second for the crystal sync to pop in too.

Before my boss retired his Nagras, he did a battery of comparison tests involving test tones and real world sound between the Nagras and that Sony. I watched him. Learned a lot. That Sony deck outperformed his high end Nagras handily. The only drawback to the Sony was how it distorted when it became overdriven. The Nagras would take the red peaks cleanly in comparison. But the dynamic range of digital was so much wider, and the noise floor so much lower than tape, he just dialed back his levels 20% to allow headroom and it worked fine. The coldness you heard wasn't because of that deck.

I connected mine to a Sony Betamax that I had at home--I didn't have the nifty matching portapack unit.

The reason you call them "audiophools" is because you have no respect.

People like you are banned from audio forums around the world simply because your preconceptions about something being theoretically perfect inhibit you from actually hearing any differences, and you go around calling people names when they don't agree with you. That is why I think you are a crank.

Originally Posted by Amfibius

The reason you call them "audiophools" is because you have no respect. People like you are banned from audio forums around the world simply because your preconceptions about something being theoretically perfect inhibit you from actually hearing any differences, and you go around calling people names when they don't agree with you. That is why I think you are a crank.

I'm the one offering explanations and supporting evidence to back up my opinions, while you are resorting to ad hominem attacks. I'm sorry if your feelings have been hurt, but I'm not the one acting like a crank here.

My favorite Ring cycle. Solti's version is supurb.

Originally Posted by Kontrapunctus

I connected mine to a Sony Betamax that I had at home--I didn't have the nifty matching portapack unit.

I didn't have the small PCM unit--mine was more of a rack-style--don't recall the model number. (It would be nice to be able to edit posts a day or two later!)

Bigshot, interesting article, although it does not support your contention of perfect sound. The formula is based on sampling for an infinite time. For time limited sampling (i.e., the real world) it is a good approximation. It has been a long time since I worked in the area so my math skills are rusty, but it is a question of discrete versus continuous. As the number of sample points (discrete) increase the closer you get to continuous, but you need an infinite number (c) to actually reach continuity.

As I mentioned before, somewhere short of an infinite number will sound the same as the real waves to the human ear. Based on past experience with live music and various analogue sources, I don't believe RBCD is at that level. While it doesn't address the issue directly, Daniel Levitin's book--This is Your Brain on Music--deals in part with the brain's tendency to fill in gaps. I suspect this process lies to some extent behind digital's less than perfect sound (of course bad mastering doesn't help/those early cds using the standard LP rolloffs were unlistenable) and aural fatigue. Anyway the book is worth reading.

I do believe a better recording medium will be discovered eventually (I thought the experiment of adding barely audible white noise to playback was promising, at least for the aural fatigue issue). As to SACD, I'm on the fence. First at my age my hearing acuity has taken the expected hit and I haven't purchased either a dedicated SACD player or a DAC. I use my DVD player for SACD and the output from my Rotel CD player is excellent. I never did quad LPS and I don't do 5 channel. Still most of my serious listening is through BeyerDynamic T1s and I find most of the SACDs (of the 50 or so that I've heard) do sound better than the cd layer. But not all. That may be due to more careful production rather than a superior process but hey I'll take the improvement.

Thanks for the tips on the RCAs. I ordered a few of the Heifitz today.

That's like asking how many angels can dance on the head of a pin. Audio equipment salesmen would love for everyone to think that way... Just keep splitting the difference over and over forever. Keep moving that decimal point further to the left. Good is never good enough.

You can believe that you're hearing numbers on a chart, but the truth is, if you do a blind A/B level matched comparison between a 16/44.1 recording and a DSD SACD, you won't be able to tell the difference. I know because I did it myself and so have engineers at the AES. That doesn't mean that there's something wrong with SACDs. It means CD sound is that good. More samples and bigger file sizes doesn't mean better sound.

Resolution isn't the area that can stand improvement in sound reproduction. We've mastered that. Frequncy response and directionality are the big roadblocks to great sound today. Speaker design, EQ and room treatment along with well mixed multichannel sound are the keys to solving those problems. A new format or a magic electronic box aren't going to work. It's going to take money for high quality speakers, calibration and adjusting one's living space to suit the sound.