I have a ueseless opinion, in terms of the broader argument. I loved Cd's, FLAC Hi Res, SACD. Brother was always trying to convert me to glow plugs and vinyl. Although I am still resisting the glow plugs( thanks Nelson), I am slowly turning to vinyl. I dont know why, but it has something no SACD or HiRes FLAC has ever shown me. I know SY and DF96 are correct in the theory of what they are saying, but not everything is known or can be measured. At the same time, I thk it will be, evntually.
@Scott
I think the thing you're missing is that the DAC reconstruction filter interpolates 'correctly' and actually 'resonates' i.e. it doesn't pass the stepped waveform, and it doesn't merely join the points together with straight lines. The peaks of the output from the filter will be higher and lower than the points that are fed into it, because the filter creates positive and negative sinusoidal peaks in between the incoming points and, however this is done, there is always a final bit of analogue filtering at the end. The output is changing smoothly and continuously, just like the waveform that was sampled.
There are arguments over the methods for bandwidth limiting at the input, how the reconstruction filtering is done, and how many bits of resolution are needed, but the basic sampling theory does seem to stand up to scrutiny.
I think the thing you're missing is that the DAC reconstruction filter interpolates 'correctly' and actually 'resonates' i.e. it doesn't pass the stepped waveform, and it doesn't merely join the points together with straight lines. The peaks of the output from the filter will be higher and lower than the points that are fed into it, because the filter creates positive and negative sinusoidal peaks in between the incoming points and, however this is done, there is always a final bit of analogue filtering at the end. The output is changing smoothly and continuously, just like the waveform that was sampled.
There are arguments over the methods for bandwidth limiting at the input, how the reconstruction filtering is done, and how many bits of resolution are needed, but the basic sampling theory does seem to stand up to scrutiny.
Nowt wrong with the theory, its sucky implementations of digital which drive some people to vinyl.
That looks rather like yet another statement along the lines of "We don't know everything, therefore we know nothing". Not true.buzzforb said:
LP may, sometimes, have a wider bandwidth than CD. I say 'sometimes' because I suspect that many cartridges can't recover the really high frequencies and are more likely to wipe it off the vinyl on the first playing. People mostly won't notice this because their ears and tweeters probably don't go up that high anyway.
LP definitely has higher noise and distortion than CD. It may have more hum too. Why do some people prefer vinyl? I suspect it is partly nostalgia, partly some HF rolloff making things sound smoother, partly low-order distortion making things sound richer. There could even be a slight preference for more noise. Why do I say this? Because there is a thriving trade in 'tube buffers', many of which add very similar faults to CD so that it sounds more 'analogue-like'.
I listen to LP, and tolerate its weaknesses for the sake of the music. I listen to CD, and sometimes have to tolerate poor recordings for the sake of the music. My perception is that as CD is a more transparent medium than LP it is easier to hear recording mistakes on CD. With the benefit of hindsight it might have been better if CD was 18-bits/48kHz, as that would enable smoother filters at each end and so reduce the digital problem which can easily be seen on a scope (but probably not heard): Gibb's phenomena from the bandlimiting, which has nothing whatsoever to do with sampling or digitisation.
Well, in studio work it should be better to use higher resolution and sample rates, but 8 times out of 10 it doesn't make a lick of difference because the monkey at the controls has everything cranked up to "11".
If the industry made a habit of properly mastering the music, I would be happy with any of the available formats.
I actually think the Loudness War is being intentionally inflicted on us in order to ruin the CD (44/16) format so they can usher in the "newer, better sound". I suspect that the plan is to eventually have two masters of the material; a butchered, compressed, LOUD version destined for CD and iTunes, and then a properly mastered, high dynamic range, zero clipped, more expensive (that's what it is all about, after all) version destined for ??? (HD tracks?, Linn?... a new media?)
Mark my words. It is coming. Just have to wait for all of us with 80's CDs to die off, or convince ourselves that the hi-res stuff is "better".
In the meantime, we have to suffer through the ironic times of downloading hi-res, high bitrate, high sample-rate stuff that was butchered in mastering and sounds like sh!t anyway.
Last edited:
For me, a really interesting question is whether a CD recording transcribed to vinyl sounds better to vinyl-o-philes than the original CD. If this could be established i.e. a vinyl-o-phile lets his guard down and recommends an LP that could later be shown to have been derived from a 16/44.1 recording, it would prove something. Not sure what, but it would be interesting.
Yes, this is a study I have some credence in:
CD vs SACD vs DVD-A - Long but interesting read
There are some hi-rez recordings that are better because greater care was taken in creating them, but could have been as good at CD resolution. My point that the translation isn't perfect is rather hollow if it is undetectable.
Still, why don't I think Stevie Wonder has come to play in the next room? Just not real. It's someplace else than the ADA loop, but it is there. The search continues.
There is no advantage in the studio using a higher sample rate than will be used for distribution. Band-limited samples do not improve with re-sampling to a higher rate. If they appear to improve, then either the listener is fooling himself or the original bandlimiting was flawed.cogitech said:
There may be an advantage in using more bits in a studio, as it allows for less precise level matching during mixing. The final result can then be scaled to fit the output 16-bit (?) dynamic range window.
The point is, the butchering of the music through dynamic range compression and cranking the levels to clipping makes the minutia above completely irrelevant!
Scott6113,
It is your speakers that are not able to convey the realism, not the electronic side of things if it is high enough resolution with low enough distortion. Electronics are down in the 0.001% distortion range and speakers are up to 10% or more in many cases. People are just concentrating on the wrong end of the reproduction chain, it is obvious to a system designer and not a single component concept that can get you closer to where you are trying to go.
It is your speakers that are not able to convey the realism, not the electronic side of things if it is high enough resolution with low enough distortion. Electronics are down in the 0.001% distortion range and speakers are up to 10% or more in many cases. People are just concentrating on the wrong end of the reproduction chain, it is obvious to a system designer and not a single component concept that can get you closer to where you are trying to go.
DF96. I am unfortunately lumped into the second paragraph, but realize it does not necessarily reperesent true HiFi reproduction at its most accurate. I do not intend to dismiss current knowledge, nor negate it because i think there are unknowns, but it cannot be denied that different people prefer different things, and not all of it can be so easily defined with numbers. I call it the human condition. I think it is what makes us unique, but that is just the egocentrism talking.
I amk fully wiling to accept this. I am building your chip amp simply because I like the way you have presented much of your experience on the forum. I get very similar vibe from Sy, but have to listen to his impass before passing judgement
Agreed. It was my wife's kitchen system which she liberated from the patio: Polk outdoor speakers which are weatherproof but not dynamic. Yes, I know modern amps do not audibly distort.
I wonder now if all this is expectation bias:
I moved from a 12 year old DVD Video/Audio player to Flac-Rega Dac source. It was a big deal: bass, dynamics punch detail. Was I fooling myself?
The DAC in the LG blu-ray player is doggy poo so I have it going through the Rega too.
I got a few 24 bit files for the new source gear. Big difference between CD and 24/44 Beatles. I have both. Remastered? I mean one time I started to listen and knew something was wrong. Then I realized the media server was in the old file directory. The old capitol CD was really muddy. I'd upload both if it were legal. I think anyone could hear that.
I got the Doors Strange Days, Paul McCartney's first solo album on 24/96. I had the vinyl on each. The Doors were as good (no scratches is nice), the McCartney much better. Remastered?
Still the best sound in my collection are CDs. Certain CDs from Blue Note, Deutche Grammophon, few others. So I can see the point about rez by itself can't improve the sound after a certain point.
I can hear the difference between filter 1 and 4-5 on the Rega. The treble has less grain. I move it around when I get to certain CDs but most of the time leave it at 1. Am I fooling myself? Reviewers say there's a difference. It's subtle but it's there. Are we all fooling ourselves?
My big speakers are pretty darn close. I'm getting parts for a Class D amp just for fun. After that, time permitting I'll experiment with a horn, kit first time, but I have a fresh design in mind and it won't go away.
Scott,
I understand because I have some CD's my wife bought years ago from something like Capital's CD club where you would get so many CD's per month or whatever and those just sound atrocious, I can barely listen to those. Then I put on a CD I bought and it is night and day the difference in the sound. I know some say remastered from the original source material but it depends on who did the remastering, they are not all the same. I have some older Chesky CD's that are just so clean and sound so real that on one that has a mix of demo sounds you would swear someone was really in the room with you, these were given to me at a CES show so they were made for demonstrations, they are super clean and clear. I design speakers so I have a leg up on most people, I don't have to buy a commercially produced speaker and this does make a difference, I am not stuck to all the same devices that everyone else has to select from. Good luck with the horns, I started out in this business as a waveguide designer so I understand how dynamic that can sound when done correctly with good devices. Some people seem to think they are all the same and that compression drivers are just commodity products but I beg to differ on that account.
I understand because I have some CD's my wife bought years ago from something like Capital's CD club where you would get so many CD's per month or whatever and those just sound atrocious, I can barely listen to those. Then I put on a CD I bought and it is night and day the difference in the sound. I know some say remastered from the original source material but it depends on who did the remastering, they are not all the same. I have some older Chesky CD's that are just so clean and sound so real that on one that has a mix of demo sounds you would swear someone was really in the room with you, these were given to me at a CES show so they were made for demonstrations, they are super clean and clear. I design speakers so I have a leg up on most people, I don't have to buy a commercially produced speaker and this does make a difference, I am not stuck to all the same devices that everyone else has to select from. Good luck with the horns, I started out in this business as a waveguide designer so I understand how dynamic that can sound when done correctly with good devices. Some people seem to think they are all the same and that compression drivers are just commodity products but I beg to differ on that account.
That McCartney album was so real I actually felt embarrassed. It's just him in a room with a state of the art, new at the time 4 track reel to reel. It's like barging in on someone practicing.
https://www.hdtracks.com/index.php?file=catalogdetail&valbum_code=HD00888072328105
It used to be on his site for $2 less but I can't find it now.
Here is the big disappointment: Amazon.com: The Capitol Albums Vol. 1: The Beatles: Music
They give you both the original mono and the stereo-ized remix and they both bite. Vinyl is definitely preferable.
Here is the day to that box set's night: Amazon.com: The Beatles [USB]: The Beatles: Music
BTW I got that paperweight one from ebay and since the stem was broken, the seller gave me $30 off. I paid $160 net. Really worth it. A day in the life, when the orchestra builds and builds...I never knew the horns were as prominent. And early stuff like Do you want to know a Secret? the mikes are practically down the lads' throats.
I guess if the better versions come 24 bit, does it really matter that the fact that they are 24 bit is moot? If offering a new version is the occasion for a revisit to the master tape and stripping away the compromises they had to make for 60s-70s equipment, then I'll keep buying high res. I'll just be less disappointed when only a CD is available.
The horns are a long way away, need to do research on the latest.
On another thread one of the engineers I know who was a recording engineer made a valid point in that many of the original tapes are very old and have now had problems of bleed through from layer to layer and also a drop in actual signal that is on the old tapes. If the tapes weren't stored correctly it makes perfect sense, it came down to how they were wound on the reels and other things like temperature and humidity control where the tapes were stored. Somewhat the analogous to old movie film and how the image fades over time in the emulsion. Some of this old music can never be retrieved at the original quality, the information is compromised and no matter the bit rate you can turn a turd into a diamond.
I have heard print-through, not to mention hiss in CDs and vinyl. It's coming from the tape. Imagine. Pre-dolby. On one of my Carpenter's CDs you can actually hear the hiss rise a moment before a new instrument joins the mix. The engineers try to keep it away from the quiet passages but their manipulations detract too. Some of my less popular artists (Lovin' Spoonful comes to mind), when they finally get to CD, must be from faded tapes. It had lost a lot. That's the strength of digital. The Sarah Brightman I hear today could have a revival in 500 years if the sun doesn't explode and sound as fresh if the bytes make it. Never going to happen with any analog format.
I hear they digitized the Abbey Road tapes of the Beatles in 2009 in 24/192. They're probably going to keep trying to release newer versions, keep the income stream going. The Beatles were #2 in album sales 2000-2009, 30 years after the breakup! The Apple Records President is no fool. Why bother with 24/192 if it makes no difference? Just marketing I guess.
I hear they digitized the Abbey Road tapes of the Beatles in 2009 in 24/192. They're probably going to keep trying to release newer versions, keep the income stream going. The Beatles were #2 in album sales 2000-2009, 30 years after the breakup! The Apple Records President is no fool. Why bother with 24/192 if it makes no difference? Just marketing I guess.
It's interesting that people can come from different perspectives: using plain, "rubbish" speakers has shown me that this does not compute. Feed a low-life driver with a clean enough, high drive capability signal, and people would be staggered by how good it sounds.
But, if you can't fix the electronics, then ultra-sensitive, high quality speakers are the next best thing ...
Frank
The missing link
It is hard to accept that CD resolution is sufficient for human ears when there are higher resolution recordings that make the typical CD sound terrible by comparison. It turns out that while the resolution itself isn't the difference, there is a difference. It isn't expectation bias, at least not always. The difference is the 24/96 product. Engineers tailor a CD for the typical customer. In the old days, a engineer's studio would often have a Crosley AM radio, a mid-fi equivalent, and actual studio monitors. They would have to make sure the recording sounded as good as possible out all three systems. Nowadays I bet they're checking with Apple ear buds and standard car stereos because that's how the mass market consumers listen.
This guy sends chills down a purist's spine: Everything About Audio Mastering | Delicious Audio
Wouldn't it be nice to find a product that corrected these problems in CDs we already have? If the compression were consistent it could be undone. Audacity? This article confirms what I suspected. I'm not imagining the superior sound of 24/96, I'm just mis-attributing it to the technology when it is the market giving me the benefit of sound closer to live. It follows that not every 24/96 is better than every CD and that the best CDs can sound every bit as good as 24/96, but on average, the recordings out there at higher res are better.
The discussion after this study said it best:
http://www.drewdaniels.com/audible.pdf
Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held upthroughout our testing: virtually all of the SACD andDVD-A recordings sounded better than most CDs, sometimes much better. Had we not degraded the soundto CD quality and blind-tested for audible differences, wewould have been tempted to ascribe this sonic superiorityto the recording processes used to make them.Plausible reasons for the remarkable sound quality of these recordings emerged in discussions with some of theengineers currently working on such projects. This portionof the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly,and listen carefully in a low-noise environment.Partly because these recordings have not captured alarge portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them,without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label.High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20(or even 10) dB of the available dynamic range, as so many CDs today do.
Our test results indicate that all of these recordingscould be released onconventional CDs with no audibledifference. They would not, however, find such a reliableconduit to the homes of those with the systems and listening habits to appreciate them. The secret, for two-channel recordings at least, seems to lie not in the high-bit record-ing but in the high-bit market. (emphasis mine)
Scott
It is hard to accept that CD resolution is sufficient for human ears when there are higher resolution recordings that make the typical CD sound terrible by comparison. It turns out that while the resolution itself isn't the difference, there is a difference. It isn't expectation bias, at least not always. The difference is the 24/96 product. Engineers tailor a CD for the typical customer. In the old days, a engineer's studio would often have a Crosley AM radio, a mid-fi equivalent, and actual studio monitors. They would have to make sure the recording sounded as good as possible out all three systems. Nowadays I bet they're checking with Apple ear buds and standard car stereos because that's how the mass market consumers listen.
This guy sends chills down a purist's spine: Everything About Audio Mastering | Delicious Audio
...there is a certain amount of improvement or “sweetening” that the Mastering Engineer may be able to do. Mastering Labs are generally equipped with a variety of high-end E.Q.’s and Compressors and other specialized gear that the Mastering Engineer can use to make your record sound better. Most well recorded albums will be sonically improved by mastering. Most of the time, records leave our mastering lab sounding bigger, more open, louder and more sort of “3D.”
The 24/96 or above product, in contrast, is made for, well, us. The engineers chuck the need to keep the volume up to cope with noisy environments through compression, stop upping the mid-bass to help mediocre speakers or buds, and optimize for the studio monitors.
Wouldn't it be nice to find a product that corrected these problems in CDs we already have? If the compression were consistent it could be undone. Audacity? This article confirms what I suspected. I'm not imagining the superior sound of 24/96, I'm just mis-attributing it to the technology when it is the market giving me the benefit of sound closer to live. It follows that not every 24/96 is better than every CD and that the best CDs can sound every bit as good as 24/96, but on average, the recordings out there at higher res are better.
The discussion after this study said it best:
http://www.drewdaniels.com/audible.pdf
Though our tests failed to substantiate the claimed advantages of high-resolution encoding for two-channel audio, one trend became obvious very quickly and held upthroughout our testing: virtually all of the SACD andDVD-A recordings sounded better than most CDs, sometimes much better. Had we not degraded the soundto CD quality and blind-tested for audible differences, wewould have been tempted to ascribe this sonic superiorityto the recording processes used to make them.Plausible reasons for the remarkable sound quality of these recordings emerged in discussions with some of theengineers currently working on such projects. This portionof the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly,and listen carefully in a low-noise environment.Partly because these recordings have not captured alarge portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them,without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label.High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20(or even 10) dB of the available dynamic range, as so many CDs today do.
Our test results indicate that all of these recordingscould be released onconventional CDs with no audibledifference. They would not, however, find such a reliableconduit to the homes of those with the systems and listening habits to appreciate them. The secret, for two-channel recordings at least, seems to lie not in the high-bit record-ing but in the high-bit market. (emphasis mine)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.