Announcement

**MylesBAstor** · 02-08-2016, 06:26 AM

Originally posted by JCOConnell View Post

Sometimes I go a while without listening to audio on my stereo systems ( watching hdtv for a couple weeks maybe) and then I decide to play
a few CDs, and WOW what an improvement over dolby digital tv sound! (even though I have a nice home theather playback setup). The CDs just
sound so more dynamic and detailed. The sound is very satisfying from both a sound and a musical standpoint. Then something happens, I put on
a vinyl LP and its WOW all over again! CDs by comparison didn't sound so good anymore, only once I had my ears exposed to the sound of a good LP.
This situation has happened to me at least half a dozen times and I am now absolutely convinced LP has higher fidelity than CD due to this experience,.

I think it was HP who said, "if you want to enjoy digital, stop listening to analog." Of course that was many years ago and digital has made tremendous strides since HP penned that line.

The more important question is whether there's still life in those old silver discs?

After all, pepple have been buying CDs now for the better part of 35 years and have amassed large collections. This summer when I was at Magicos, I heard RBCD or just 88/24 played back from the top Baetis music server through the Pacific Microsonics HDCD 2 DAC and the sound was unlike I've ever heard from digital. (of course, we can argue that just ripping the CDs to the hard drive improves the sound--or as I have done ripping them to DSD and more 0s and 1s!) In fact so good, had me questioning whether we needed high-rez. Would I sell my LPs or tapes? No. But there a world of music out there and why not experience it in the best sound possible. I never could understand the debate between would you rather have good sound and average performance or great performance and lousy sound? Why must the two be mutually exclusive????

MikeL here also wrote about his experiences with his Trinity DAC that plays back RBCD. Maybe there's still some life left in those silver discs after all.

**JackD201** · 02-08-2016, 07:56 AM

The EERA Tentation was the box that did the same for me. Robert too in his review. Turned out there was more to the old horse. I don't obsess about digital all that much anymore to be honest. I think 16/44 at least has gotten to a point in my set up that I can really just hunker down and enjoy. I just have to remember to not lead off with Analog. Going back takes a bit of mental adjustment but it's now doable in the span of a song or two.

**JCOConnell** · 02-08-2016, 12:09 PM

well, I've got about 1500 CDs in my collection ( about 1000 pop/rock the remainder classical and jazz) lying around here. I'm certainly not ready to just toss them in the trash bin, but I am not as enthusiastic about playing them as I am my much larger LP collection. I am glad a lot of stuff I have on CD is mutually exclusive to what I have on LP so theres no direct format conflict in most cases.

**Alrainbow** · 04-12-2016, 06:57 PM

Cd are not dead to me anyway. But one must consider that dsd done well is much closer to analog
It's not about impact as I a few Super audio CDs. Hat will blow wind at you and still seem fine meaning not blaring
But dsd is just easy on the brain. I have some greet people Music but one cannot play pcm then dsd of the same songs and not hear the people sound. My msb or my new lampi as good as they are playing is pcm. But it does seem to me that my brain gets adjusts pretty fast if I just play pcm. Also dsd can be played louder like analog

**DexDuophonic65** · 05-10-2019, 06:08 PM

I think a major reason this debate has raged-on ad infinitum (or ad nauseum!), has to do with: the constant comparisons between digital vs. analog seemingly soooo dependent on using reissued/vintage material (obviously mastered in an all-analog chain during the "golden age") with...how, basically, a copy now sounds transferred to an entirely different piece of software technology (and with 40+ years' age possibly affecting the conditional integrity of whatever remains as the source).

I mean, the max frequency response of the typical 2tr./15ips/1-4" stereo mixdown master was -give or take- 30khz with Dolby encoding...and, (probably?) a S/N of around 75-80db. Presuming that material wasn't recorded on Ampex 456 and hasn't completely disintegrated/stuck together by this time (which, if you're talking a '70s and '80s album: it most likely was)...the record label gets into the area of using safeties from much further down the dubbing chain or any surviving LP cutting master (more of these around; since the dominance of vinyl would've necessitated keeping the album in print --- but with all the wrong EQ only applicable to records) to use as the source for a reissue.

Try copying that to a domain, where: there are no mechanically-induced intolerances and, even on the most budget players, the S/N is at least a whopping +101db(!). Unfortunately, CD mastering has gone the way of how 8-tracks and cassettes were mass-marketed: meant to sound full ("loud") on portable and/or budget equipment on the consumer end. The volume discrepancy of trying to transfer hissy vintage material to a format with, like, 50db more headroom is not what the (average) consumer wants to hear. So: the record label winds-up no-noising the reissue to death and cranks up the signal to compensate (if it's sourced from a LP cutting master: they crank up the bass as well). Of course, this is not a problem when mastering records (because their S/N averages -87db and there's the whole host of limitations unique to that format).

The oldest cd I have is a 1990 MCA copy of Dave Mason's "Alone Together" album and: it is (still) one of the best I've ever heard. Even playing it on a similar-vintage, mid-grade Alpine-era Luxman unit does nothing to diminish the involving sonic qualities of it.

**EdAInWestOC** · 05-13-2019, 08:47 AM

The limitation of 16/44.1 has to do with aliased content between samples. If you want to read a technical explanation read the following:

http://www.edsstuff.org/sampling.pdf

It is easiest to understand when you think about a 22.05kHz sine wave. If this sine wave is sampled at 44.1kHz you will get two samples per cycle of the waveform. A modern DAC can compensate for the lack of data and reconstruct the sine wave but if there is anything other than a simple sine wave (as there is in music) the original waveform cannot be accurately reproduced.

The good news is there is little content at that frequency in music but there are harmonics and out of phase information that gets aliased. Some listeners complain about a Redbook's inability to reproduce a listening hall's space and that can be blamed on the lack of sufficient data.

The link above points out that a digital encoding and playback system needs to take into account the signal that is being encoded. Two times the highest frequency is sadly insufficient for a dynamic waveform. Its OK if we are encoding the output from a sine wave generator but that is not music.

The unpredictable nature of music makes for a tough waveform to successfully reproduce at 16/44.1. There is lots of unpredictable waveform content that all contributes to the sound of instruments and their placement in the listening space.

In order to encode the music accurately the sampling rate needs to be increased but there is a law of diminishing returns. We cannot keep on upping the sampling rate in order to capture all of the relevant content. Then the bottom line comes into focus and the unfortunate truth is that an accurate digital encoding and playback system for music requires a much higher than expected sampling rate.

That all assumes that you want as faithful as possible fidelity when we encode and playback PCM data. We have to keep into mind what the terminology is. We sample the original waveform and we cannot expect to capture all of the relevant data unless we raise the sampling rate so high that we should have just stuck with analog to begin with. The resulting size of the digital data causes storage and transfer issues due to the very large files size.

When a digital PCM system is designed we ask the question are we capturing enough data to do the job? When it comes to music "good enough" is different things to different people. Have we ever asked ourselves why Redbook digital sounds so good for bass and for small group recording but it sound less stellar when we try to encode/playback a large orchestra? Or record and playback a cymbal crash? The low sampling rate in Redbook is not sufficient to do the waveform justice.

As a lifetime audiophile I find digital PCM to be a trade off. We enter this hobby with ultimate fidelity as our goal and if we wander into PCM digital, we start by throwing away relevant data between samples. Yes, analog has its problems too but this subject is about Redbook audio and its issues.

I wanted the CD to be the thing we were all told. I wanted a better music source with great S/N and dynamic range content. The truth is that we have finally arrived at the best possible Redbook players but it is still not what we were promised. Analog is still king when it comes to high fidelity because there is nothing thrown away. It is real time and everything that can be picked up can be recorded and played back.

There are lots of things that can go wrong with analog recording and playback but when it is right there is nothing better. Digital recording ends the problem with multi-copy signal degradation that plagues analog. Thats a big plus but there are many gotchas in digital and especially in Redbook. It can sound remarkably good but I have many Redbook CDs that shows what it can do wrong.

It is not perfect sound, not even close.

**Musica** · 05-13-2019, 10:16 AM

"Analog is still king when it comes to high fidelity because there is nothing thrown away. It is real time and everything that can be picked up can be recorded and played back."

With all due respect there is no perfect playback medium. But if I had to pick one it would be tape.

WRT analog, IMO you downplayed the resulting sonic degradation associated with S/N, dynamic range not to mention wow and flutter, sensitivity to vibrations, ticks/pops, etc. You also exaggerated the limitations with higher sampling rates and made a leap suggesting that, "we should have just stuck with analog to begin with." Storage is ridiculously cheap and data transfer technology can easily manage Quad DSD or 24/192 PCM even via wireless.

I would also encourage you to investigate a digital system budgeted on par with your analog as you (and many others) are orders of magnitude off in sources' budget to truly hear what good digital can do. As a data point, your cartridge alone costs ~20x more than your DAC..

**Kingrex** · 05-13-2019, 10:54 AM

I have been private chatting with some fellow audiophile on Audiophilestyle. These members are developing software to take away the nasty compression and other mishandlings of digital file in years past. Many of the CD you have that don't sound good is not a result of any limitation in the medium. Its an injected noise, distortion that was intentionally/un-intentionally put there for purposes of radio play, auto playback or just not really knowing what they were doing.

I am for some reason an ABBA fan. Even liking the band, I sometimes have a hard time getting through a digital file as the sound is a little to thin and brittle. One of the gentlemen on Audiophile Style sent me some ABBA files where he removes the DolbyA formating plaging the files. My jaw dropped when I played it. That night I must have played the 6 or so songs he sent me 3 or 4 times. My wife and I have been playing the files all weekend. Its not a dramatinc change. It's not like OMG it's so much better. It just became so natural and correct.

Let me take this a step further and this will piss people off. Another member sent me some ABBA files he played with. These too were better than the rips of the same songs I have made. Even with the checksums saying its the same file, his sound different. How can that be. His files were ripped with less noise as he uses linear PS on his optical drives and all HD in his servers. It actually makes an audible difference. Now will come the no it does not. It's in your head.

There is a reason some people are really enjoying digital and others not. Do those not all that satisfied use purpose built switches and have linear PS on there routers and modems. I doublt it. But think, what are your doing to your whole digital chain when you have one device putting a bunch of nasty noise on your ethernet and now its in your server poluting everything in there. It's the subtle finishing touches that help a digital playback system rise to the occasion. Mistakes anywhere in the setup are akin to a poorly setup cartridge. It might be a great table, arm, cartride etc, but set up all wrong and it's not going to give magic.

**Alrainbow** · 05-13-2019, 12:51 PM

A few more thoughts on cd playback and the medium
first off a really good CD player can give you a very good playback of the medium. It was made to play this back alone so all inside of made well is optimized for it.
Took me 2 years or more to catch up to my msb UMT plus
on this even streaming to my own dac.
Next is this there is not a whole lot of cd s made that are bad yes some are far better but little are truly bad by percentage.
It's the whole playback chain that's makes it bad or good.
But the great ones do make one feel bad of the lesser ones lol.
As for home theater playback of cds
money must keep in mind most are not made to playback high quality but midfi so tough to judge. It's the same if we use a Hifi speaker to play back movies

**Steve Lefkowicz** · 05-13-2019, 12:53 PM

I understand there is renewed interest in playing from the CD itself these days, but at least for me, I view the disk merely as a storage medium. I consider my shelves of CDs as an archive so I can reload if my hard drive and its backups fail. On regular CDs my CD/SACD player, a very good Marantz SA-8001 (that's very appropriate for my budget oriented system) simply doesn't sound as good as my PC/iFi based digital rig. Even comparing SACD to the dsd files I'd take the PC/iFi system over playing the disks.

But I certainly agree with the original premise and would rather listen to LPs on my Linn!

**Kingrex** · 05-13-2019, 03:45 PM

I'm going to fire up my vinyl rig again. I haven't seen it since October when the preamp blew up for the third time. I get the whole holding the records, looking at the covers, just the ritual of playing a record. I also really enjoy the sound when it's working properly. Records are very pleasant to listen to.

**Tech7738** · 05-14-2019, 08:57 AM

Originally posted by DexDuophonic65 View Post

I think a major reason this debate has raged-on ad infinitum (or ad nauseum!), has to do with: the constant comparisons between digital vs. analog seemingly soooo dependent on using reissued/vintage material (obviously mastered in an all-analog chain during the "golden age") with...how, basically, a copy now sounds transferred to an entirely different piece of software technology (and with 40+ years' age possibly affecting the conditional integrity of whatever remains as the source).

I mean, the max frequency response of the typical 2tr./15ips/1-4" stereo mixdown master was -give or take- 30khz with Dolby encoding...and, (probably?) a S/N of around 75-80db. Presuming that material wasn't recorded on Ampex 456 and hasn't completely disintegrated/stuck together by this time (which, if you're talking a '70s and '80s album: it most likely was)...the record label gets into the area of using safeties from much further down the dubbing chain or any surviving LP cutting master (more of these around; since the dominance of vinyl would've necessitated keeping the album in print --- but with all the wrong EQ only applicable to records) to use as the source for a reissue.

Try copying that to a domain, where: there are no mechanically-induced intolerances and, even on the most budget players, the S/N is at least a whopping +101db(!). Unfortunately, CD mastering has gone the way of how 8-tracks and cassettes were mass-marketed: meant to sound full ("loud") on portable and/or budget equipment on the consumer end. The volume discrepancy of trying to transfer hissy vintage material to a format with, like, 50db more headroom is not what the (average) consumer wants to hear. So: the record label winds-up no-noising the reissue to death and cranks up the signal to compensate (if it's sourced from a LP cutting master: they crank up the bass as well). Of course, this is not a problem when mastering records (because their S/N averages -87db and there's the whole host of limitations unique to that format).

The oldest cd I have is a 1990 MCA copy of Dave Mason's "Alone Together" album and: it is (still) one of the best I've ever heard. Even playing it on a similar-vintage, mid-grade Alpine-era Luxman unit does nothing to diminish the involving sonic qualities of it.

You've kind of mashed up several different issues but ended up hitting the nail on the head anyway!

It's a real shame than the vinyl/CD comparison ends up being about analog vs digital. If anyone were actually familiar with the two entirely different paths audio has taken to get to each medium, anyone could realize there's a whole lot more being compared than the audibility of storage media. In fact, if you could isolate just the media and its distortions, and feed each with an identical input signal, any listener would be shocked at how similar they could sound. But that's not what's been done, not at all. So when a CD and the vinyl version (or tape version) are compared, what we actually are comparing is two different versions, neither actually being a clone of the master, each likely made at different times, definitely in different mastering facilities, by different people with different skill-sets, and possibly with entirely different end goals. Never, ever assume that the original goal was a faithful replica of the master in either case!

16/44 had a bad wrap in the beginning, and that's been more than adequately sustained. There's a lot of complicated psychology going on here. The digital hate, based on a smooth analog signal being chopped up into pieces (even if the concept is seriously flawed) didn't sit well from the beginning. Ever see a list of CDs that sounded better than the original vinyl? Of course not. Not because there couldn't be one, but what would be the point? I just might prove the CD could sound better, and who really want's that?

Yes, of course there are terrible sounding CDs. Some of the early ones had quite a few issues. But so did early vinyl, then it got better as it went along. There are terrible sounding records and tapes too. Is that the point? Do bad samples define a medium? Well, they have, for better or worse.

But also, we have to try to keep the discussion based in reality. "the max frequency response of the typical 2tr./15ips/1-4" stereo mixdown master was -give or take- 30khz with Dolby encoding...and, (probably?) a S/N of around 75-80db." Ok, hang on just a minute. No tape format, not any of them, is flat from 20- 30kHz +/- 1dB. See what I did there? I stated a real frequency response, not just a maximum possible frequency. I included a range and deviation. Can you get 30kHz off tape at 30ips? Well, sort of, if you're really careful and look really hard. But that's not what defines frequency response. You need several things, a set of frequencies to work between, a means to define variance (+/- XdB) and then really, we need a level at which to take that response measurement. Any linear PCM system running at 44.1kHz sampling or higher has flat response from 20Hz to 20kHz +/- .5dB at any level up to 0dBFS. Simply not true of tape. LF response and HF response are speed dependent. You want HF, you throw away some LF response. And tape response is level dependent. It changes depending on how hot the signal is. Dolby did not improve FR at all, in fact, it made it worse at low levels. Dolby A encode>decode does not track perfectly, but close enough, and since the tracking issues are worse where the signal is least, it hides its tracks quite well, an provides 10dB of s/n improvement. But Dolby A was not universally used. Dolby SR was batter in all aspects, but even less standardized outside the film industry. Outside of the dynamic range improvement, Dolby didn't help with tape's core issues.

16/44 is better at replicating its input than tape is. That's not open to debate, it's clear and measurable. And that doesn't address listener preference at all, which is another matter entirely.

The transfer to lacquer from tape or digits didn't improve anything, though. Lacquer, with all its limitations, and there are more than a few, is remarkably transparent. Below its maximum modulation levels (which vary with frequency), and allowing for separation losses, it's largely a garbage-in/garbage-out medium. Turns out, so is PCM, but if you're going to really compare the two, you need to start with the same garbage. And that, folks, you just don't have. Lacquer adds distortions, but that's not what the core problem is. It's the garbage going it. 16/44 doesn't add nearly so many distortions, it does replicate the input garbage pretty well. But if you don't use the same garbage, you can't make the comparison. And, simply, you can't by listening to the two different releases.

I have no issue with the "vinyl sounds better" idea with regard to specific recordings, often it can. I do have big problems with "analog sounds better" because that comparison cannot be done fairly by anyone outside of the industry. Some of us inside the industry have actually done that comparison, which is why I have big problems with listeners making that statement, when what they really mean is "the vinyl issue sounds better than the CD issue", which I won't argue with.

**Alrainbow** · 05-14-2019, 09:29 AM

The biggest misstep in red book was its bit depth. 16 is 24 times less sound volume steps less than 24 bit is. The result is 24 has a seemingly lower noise floor
now why I don't know but I'll bet I can pick it out on many systems. I do know 24 bit depth has many more loudness
levels only.
How do others feel and know why
i have 44.1/24 or 88.2/24,it's obvious even st this low bit resolution

**Tech7738** · 05-14-2019, 01:07 PM

Originally posted by Alrainbow View Post

The biggest misstep in red book was its bit depth. 16 is 24 times less sound volume steps less than 24 bit is. The result is 24 has a seemingly lower noise floor
now why I don't know but I'll bet I can pick it out on many systems. I do know 24 bit depth has many more loudness
levels only.
How do others feel and know why
i have 44.1/24 or 88.2/24,it's obvious even st this low bit resolution

This is a common misconception.

First problem, there are no actual 24 bit ADCs, with possibly one exotic exception. All of them generate 24 bit words, but with 20 bit performance. The limiting factor is thermal (Johnson)noise. You can't get around it easily. The exotic exception uses staged ADCs to "fake" a composite 24 bit ADC. But it's a rare exception not even found in studios.

Related, there are no sound systems capable of reproducing the theoretical 144dB dynamic range of 24 bits. If you had a very, very quiet room with ambient noise at 20dB (NC20), a signal peaking at 0dBFS would be theoretically reproduced at 164dB SPL, well above the threshold of pain and permanent hearing damage.

Looking at it another way, put 0dBFS at a practical maximum SPL of 105dB, the total dynamic range of your listening room is only about 85dB, easily within the capabilities of 16 bits.

And one more perspective...there are no recoding studios or performance venues with 144dB of dynamic range between the loudest sound and the ambient noise floor. None, not even close. Even microphone self noise is typically at 20dB (assuming preamp noise is lower, not always a safe assumption).

24bits is overkill, very seriously, for recording releases. It's somewhat helpful in recording because level setting is less critical, allowing engineers to be careless and not trash a take.

Second problem, there are no "volume steps". None. This is easily seen by looking at the output of any DAC on a scope. You will never see any steps, they don't exist in the output. If they did, they would translate to a ridiculous amount of audible distortion. All waveforms are reconstructed by the reconstruction filter. No sampling frequency or data ever exists at the output. 100% smooth audio.

Third, related to the above, there are no "volume levels" at all. There can't be, or the system couldn't record audio.

And lastly, the comparison. What, exactly, is being compared? 44.1/24 to 88.2/24? How, exactly? Same identical input? How did those versions come into being? Was one downsampled to create the other? Upsampled to create the other? Made using an entirely different recording chain? A completely different recording, not even the same music? See, none of those are valid comparisons.

You like 88.2/24 because you now it's 88.2/24, and have a whole bunch of biases that dictate the belief. Or, the 88.2/24 is just a better recording, which is not only possible, but likely. Downsample it to 44/16, and run a true ABX test, there will be no audible difference.

I know, I know...blasphemy. Sorry, guys. I've been around on this. In my world, I really needed to understand what the real differences are, if there's a point, and where to invest resources. Recordings are made at 48, up to 192kHz, and 24 bits because we can, and clients ask for it, and there's a slight benefit to originating at 24 bits. But that's it. The rest is psychological, which is both powerful and real, and not invalid at all. Expectation makes the world go 'round. No point in taking it away, even if the biases are based in misunderstanding of the basic concepts of digital recording.

Announcement

Audible limitations of 16/44

Audible limitations of 16/44

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment