What High-Resolution Audio Is Not!
In the quest to properly, simply and accurately define what high-resolution audio is, I thought I would use a standard scientific technique and talk about what it is not. This comes as I communicated with several members of the CES Audio Board working group about the issue. I pulled up an old email from Dr. Sean Olive, who is the Director of Acoustic Research/Corporate R&D at Harmon and the recently elected head of the Audio Engineering Society (AES). I quoted a sentence that he wrote to me that addresses the issue of analog tape and vinyl LPs being considered “high-resolution”. Here’s the quote:
“It’s good to know there are still some labels left dedicated to making high quality surround recordings. I applaud you for educating people about what is and isn’t HD (analog tape/vinyl being an example that isn’t HD).”
So if you don’t believe me…then take it from Dr. Sean Olive (and plenty of others), a very prominent authority on acoustics, audio engineering and recording. Analog tape and vinyl LPs are NOT and never will be considered high-definition…except possibly by the committee given the task of defining HRA!
The world of source audio fidelity can be broken down into several levels of fidelity based on the era of the recording AND the equipment/techniques used to make a particular track. It’s hard for advocates of a certain “warm sound” to accept and recognize that their format of choice may not be included in the highest strata of fidelity from a technical point of view…but really who cares? If a particular format works for you then go with it. I’m just hoping that high-resolution audio can actually mean something.
So here’s my list of the top five formats that are not and never will be “high-resolution audio”:
1. Analog tape recordings – As we’ve seen over the last few days and in some past posts, analog tape has limited dynamic range (the equivalent of 12 to 12-bits of PCM digital and that’s the first generation master, will you will never hear.), has all sort of noise, is subject to distortion and speed variations and loses fidelity with each pass of the tape over the playback heads. It can sound wonderful and many, if not most, of the tracks/albums you purchase from HDtracks and soon Ponomusic will be from analog tape masters…not HD.
2. Vinyl LPs – Except in the case where a Direct Metal Master is cut while the ensemble is performing an entire side of an album, all vinyl LPs are cut from analog masters. And since analog tape is never going to be high-resolution, then vinyl LP copies aren’t either. In fact, analog tape does a lot better job of delivering fidelity because of the limitations of mastering for vinyl but we already know that. And by the way DMM aren’t high-resolution either.
3. Compact Discs – are the benchmark or par value in the fidelity arena. I consider a 44.1 kHz 16-bit PCM CD that has been recorded and mastered with care to be standard resolution. After more than 30 years and billions of replicated copies as well as burned CD-Rs, this is the measure of good quality sound…or at least it has the potential to provide excellent fidelity. It’s really a challenge to do better than the Redbook specification.
4. MP3 – and ALL the other “lossy” compressed formats that drove down fidelity in the search for portability, transferability (ripping CDs to give to your world wide web friends…think Napster and Bit Torrent) and file size (what would you rather have “perfect sound forever” or “10,000 tunes on a single iPod”?). It doesn’t matter if you encode at 64 kbps (the lowest HD-Radio flavor) or 320 kbps, a compressed audio file will still never be a high-resolution file. It might aspire to “CD-Quality” because consumers can’t tell the difference but to be accurate, a lossy file means something has been removed. I want everything that the artists and engineers captures…even if I can’t “technically” hear it.
5. Cassettes and 8-Track Tapes – in the interest of being inclusive and because I suspect there’s still of lot of dashboards with cassette players in them, audio cassettes, 8-tracks, miniDiscs and DCCs (digital compact cassettes) do not qualify as HRA. Just say no.
So there you have it, my list of formats that will never make in the world of high-resolution audio. But then you have to ask what’s left?
I’ll talk about that in an upcoming post.
First of all, thank you very much for this informative and well-reasoned post, and all of the others. I have an EE degree. I am old enough to read about CDs in _Spectrum_ before they came out commercially, while I was in engineering school. I also have a terminal CS degree. However I am *not* an expert in audio technology and engineering. My point is a “devil’s advocate” one. There is research that shows that people, in controlled experiments, cannot distinguish a CD from a 320Kbps MP3 (from the same source). Should that not be more important evidence of high resolution than the fact that MP3s encode less information than CDs? What would you answer to a that?
You make a very valid point. If no one or at least the unwashed masses can’t distinguish between the original and a lossy compression version of the same file, then we should move on and be satisfied. Right? That’s what we’ve been submitted to for at least 20-25 years. But there is the intellectual side of the issue as well..perhaps its a philosophical point…but I need the real thing. I want to travel to the Louvre and stand in front of the Mona Lisa (which I have done), I want to sit on real leather instead of the fake stuff and I want everything that was played into the air during a performance to be put back into the air where you experience it…not matter how subtle or whether it can be “heard”. It just matters to me.
There have to be somethings and some places where we don’t have to sacrifice the integrity of the music…or anything else.
This is probably an unsatisfactory answer but I feel there is meaning in listening to a real HD-Audio recording at 96 kHz/24-bits over an MP3 at 320 kbps. You can’t know until you’ve sat and listened for days…there’s just something that’s right about it.
Mark,
Of course your 100% right on your definition of HDA BUT, as you know there is a big gorilla in the room
Your definition of HDA isn’t going to allow HDTracks, Pono, en all to resell that catalog of baby boomer rock to us at least one more time before we all die off. LOL
What will the world be if we don’t get a HDA branded version of Dark Side of the Moon, even if it can’t sound any better than the 41 year old analog master it is made from. The Audiophile market has been selling me snake oil for almost 50 years now, why would you expect them to stop now.. 😉
Best Regards,
Sal
I know…that’s the unfortunate part AND one of the big reasons why I believe that the whole “high-resolution audio” initiative is bound to fail UNLESS consumers are informed and told the truth. There are so many vested interests, powerful organizations and just plain faith in the audio world that it will never make into the mass marketplace. I’m in favor of giving consumers the very best “reference” standard copy of every album and track ever recorded…but just include the provenance!
Seems to me that folks are confusing resolution with accuracy. Yes, you can create a high resolution copy of an analog recording but it will not be accurate to the original performance. Isn’t this similar to taking a 12 mega pixel scan of a blurry 2 mega pixel photograph? Perhaps if you used that visual analogy, they might get it. Then again, I suspect those that are advocating for analog sources might have a financial stake. That said, I do enjoy high resolution copies of classic analog recordings, but I also recognize that the improvement I’m hearing is a result of the mastering process.
I like that comparison. It’s simple and makes sense. There’s nothing wrong with analog recording and playback…it’s a viable format. But it has limitations that should be recognized.
I believe that all Mark is trying to do here is define HRA. Not everyone is an Electrical Engineer and can understand the technical process of producing an analog sound wave with a microphone into an analogous electrical signal, storing and recording this electrical signal, digitizing the signal, analyzing the artifacts than contaminate the original signal including noise, distortion, the effects of resistance, capacitance and inductance on the signal. That is OK. This is complex stuff and most music lovers probably do no understand or care about all of this. They just want to put an LP on the turntable and hear the music. If it sounds good enough to them, then whether it is HRA doesn’t matter.
But the people who produce this music have to understand all of this in order to provide the best possible sound that can be provided to the music lovers. And maybe some can not hear a difference between an analog record vs. a blue ray recording done at a higher sampling rate and bit depth. This is fine, but please don’t be in denial and hold on to the old beliefs that analog tape recording or mass produced LP records were high resolution. When you look at an old VHS tape on a CRT type TV set, do you think that is high resolution video compared to a blu-ray video displayed on a 1080P monitor? Lets move on with the technology and all Mark is doing is telling us what-is and what-isn’t HRA. Perhaps the old VHS tape is good enough for you and you don’t have any desire to improve it, but that does not make it high resolution. Let go of the past technology and welcome the new technology. As an audiophile, this new technology is what I have been waiting for……..for decades.
Good summary. Thanks.
Dr. Aix’ postings are something to look forward to each day on the train home. I’ve listened to every medium, from a crystal set (to use earphones when parents thought I was asleep) to 78’s, 45’s, 33’s (with their RIAA curves) and so on. I remember Avery Fisher “fooling” entire auditoriums by having a live orchestra simulate playing while he played his “hi-fi”, and so on. But one wonders whether that is the point.
An engineer can easily make true hi-rez sound just like that crystal set- -that’s one end of the spectrum (and I own CD’s that sound just terrible compared to the same recording on an LP). The other might be termed, “Maximizing the recording process from live performance to whatever emanates from one’s speakers/phones to minimize distortion of any kind.” Critically comparing 16/44.1 to 24/192 is a fascinating exercise, but avoiding the purchase of music advertized as the latter when the source material is so badly flawed as to set one’s teeth on edge may be where the needle used to hit the vinyl, so to speak.
Accordingly, could it possibly be more productive to chart a course from the current status to a place where the industry should be. Pono is getting lots of criticism, but the proof of the pudding won’t be how good his candy bar processes signal, it’ll be his success in obtaining distortionless (I know there’s no such thing) source material.
Enough. Thanks again for the daily missives.
Thanks John…you sound like you’ve lived somewhat of a parallel life. I remember my first crystal set and the earpiece in only one ear. Those were the days.
I think your point about Pono, the device, and Ponomusic, the service, is exactly right. And I can tell you that the major labels are supplying the same masters to ALL of the “so-called” high-resolution download services. I’ve been in discussions with all of them. Right now there are only 1500 or so transfers that have been done. Just how many places will be able to successfully sell the same standard definition material dressed up in high-resolution “clothes”?
Hi!
First: thank you very much for your blog. I really enjoy reading this!
One question. you wrote: “Except in the case where a Direct Metal Master is cut while the ensemble is performing an entire side of an album, all vinyl LPs are cut from analog masters.”
Did you really mean DMM, or direct cut?
It wouldn’t matter either way…both cutting methods would avoid the analog tape stage but would still no quality as HRA.
I think we can cut away the confusion by focusing on the recording. Thus I would propose the following definition: “High resolution audio is a recording at minimum 88.2 kHz, 24 bit PCM or 2.8 MHz DSD.” The recording can then be reproduced at whatever playback resolution, even compressed to MP3 or cut into vinyl, but it would still remain a high resolution audio recording. We do not worry about the playback resolution, just the recording resolution.
The analogy in the video world would be Avatar which was recorded in true 3D HD and can then be reproduced and played back in 2D HD or 2D SD. The recording remains true 3D HD. A studio may convert a 2D SD movie to 3D HD and sell it with whatever marketing hype, it remains a 2D SD recording!
Sorry, this was meant to be posted in your latest blog – on Mar 23.
I’m pretty much with you on the definition of minimums. The problem is that every single mention of High-Resolution Audio focuses on the delivery side of the picture and ignores the “provenance” or source quality. If you get a track as a 192 kHz/24-bit file then they want us to believe that the recording format doesn’t matter.