There is no room for bias in good scientific evidence mate. Things are either facts or not facts and Monty doesn't express opinions, he merely tells you the facts. If you want to spend your life in denial - as countless thousands of hifi buffs do - then be my guest. But please don't try to infer that facts are opinions when they simply are not.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
High Definition iTunes Music Downloads May Be on the Horizon
- Thread starter MacRumors
- Start date
- Sort by reaction score
There is no room for bias in good scientific evidence mate. Things are either facts or not facts and Monty doesn't express opinions, he merely tells you the facts. If you want to spend your life in denial - as countless thousands of hifi buffs do - then be my guest. But please don't try to infer that facts are opinions when they simply are not.
I don't need to prove anything. It's quite the other way around. The need to prove it lies with those who claim that they can tell the difference between 24/96 and 16/44.1. Logic dictates that we should NOT be able to do so, so if people believe otherwise, where's the proof?
And "I trust my ears". "Don't tell me what I can and can't hear you cheeky sod" etc etc etc etc etc is NOT proof.
A CD has a dynamic range of around 100dB. So to hear the noise of the CD above ambient room noise, you'd need to be listening to the loud parts at around 130dB. With speakers with typical 90dB sensitivity, you'd need to drive them with a pair of 10kW amps, on full volume, to get 130dB. If you have less power than that, then you can't hear the noise floor of a 16bit recording, let alone a 24bit one.
And remember we are talking about hearing the noise during the quiet passages! You wouldn't have a hope in hell when it's blaring away at the full 130dB. Not that the speakers could stand it at that volume anyway. Most domestic speakers top out somewhere between 100dB and 115dB maximum. It's painful beyond about 125dB and a typical rock concert - a loud rock concert is around 115dB. So unless you listen at well beyond rock concert volumes, the noise floor of a normal CD is quiter than sitting perfectly still in a silent, empty room. i.e. CD background noise is in all domestic circumstances inaudible.
Regards the other parameter, sample rate, a CD with a 44.1KHz sampling frequency can perfectly reproduce signals up to 20KHz. This is not an opinion, it's mathematical fact as proven by Nyquist in 1933. It's a well established medical fact that the human ear is unable to hear anything above 20KHz at best.
So for people to claim they CAN hear a difference between 16/44.1 and 24/96 is a remarkable claim that needs to be substantiated with solid evidence. Whilst listening to normal hifi and normal volumes, they would need to be able to hear things that medical science says you can't hear, at volumes quieter than the quietest rooms in existance. It's a quite preposterous claim when you think about it.
Last edited:
Format Comparison
You ever gave it a try yourself?
Here's a link to Barry Diament's page, legendary mastering engineer for Atlantic and one of the very first to make CD masters.
http://www.soundkeeperrecordings.com/format.htm
Enjoy.
You ever gave it a try yourself?
Here's a link to Barry Diament's page, legendary mastering engineer for Atlantic and one of the very first to make CD masters.
http://www.soundkeeperrecordings.com/format.htm
Enjoy.
I was told by audio engineers a few years ago that mastering is more important, more influential to good music then 24bit/96KHz or (24/192) vs. 16/44.1 Is that what you are saying? Perhaps that's what people are hearing, better mastering on 24/96 then concluding it's better than 16/44.
I forgot to ask those audio engineers why are DVD audio, Blu-ray, SACD in in 24/96 if 16/44 is all you need. Maybe you know the answer? Marketing?
If 16/44 is the best we can do, then I am happy as I have my music ripped in iTunes from CDs in Apple Lossless. No need to spend more money. But also baffled that Apple won't sell CD quality music when there's clearly a market for it.
I forgot to ask those audio engineers why are DVD audio, Blu-ray, SACD in in 24/96 if 16/44 is all you need. Maybe you know the answer? Marketing?
If 16/44 is the best we can do, then I am happy as I have my music ripped in iTunes from CDs in Apple Lossless. No need to spend more money. But also baffled that Apple won't sell CD quality music when there's clearly a market for it.
... not a single reference to the fact that Apple still does not support FLAC in iTunes. LAME.
Get with it Tim Cook! No one cares about Apple Lossless ...
I care about Apple Lossless and re-ripped my entire CD library to the format. What is the incentive for Apple to support FLAC when their own lossless format works so well within the Mac and iOS ecosystem?
I've been a hifi buff for 35 years and have tried those and upteen others. (There's a link somewhere I will dig out that explains why those files are not a reliable test, but I can't find it at the moment.) Incidentally, I have a load of 24/96 flac recordings and a bunch of DSD's as well, many of which I prefer to the ripped CD equivalent.
For a long time, I was convinced I could hear a difference between the formats (as opposed to the mix), even though it made no sense. But I was skeptical, not least when in properly controlled tests, people consistently cannot tell the difference.
What finally swung it was when I decided to downsample all of my HD music to 16/44.1 so it would play on my old Apple TV V1 without it resampling. The quality of the conversion (downsampling) process is important with many converters not doing a great job and introducing audible artefacts. I used the sample rate converter included as part of Channel D's Pure Music. This is very high quality SRC.
What I discovered was that the 16/44.1 files sounded absolutely indistinguishable from their 24/96 counterparts. I could not and cannot tell the difference under any circumstances. (Playing them from a Mac Mini using Audirvana and a Benchmark DAC2-HDR into Bel Canto REF1000m's and Wilson Benesch Discovery loudspeakers).
The files do sound different from the CD rips however, because the mastering and mixing is different. Some of them are better than the CD, but interestingly not all. Phil Collins' Face Value is definitely better on the CD for example.
So forgive me but I honestly do not believe anyone who says they can tell the difference. I do not suggest they are lying (perhaps a few are), but mainly they are failing to understand that what they THINK they hear is not actually the truth. That what you think you hear is a function of your brain and your ears and since your brain also controls your emotions, your moods, your expectations and your perceptions, then we are not very good judges of things on any objective basis. We think we hear things that really are not there. This is why when proper controlled tests are done, listeners really cannot discern one thing from another. I don't think ANYONE has ever been able to reliably demonstrate being able to tell the difference in sound between one interconnect and another, for example. Despite the entire industry making them and making extravagent claims, no-one has ever shown they can hear a difference.
----------
Exactly.
16/44 lossless downloads would be a HUGE improvement over the current 256Kb lossy material on iTunes. 24/96 or 24/192 offers no improvement as a format, but sometimes the mastering has been done with more care because it's been done by audiophiles for audiophiles... and the results can then be very very good.
EDIT, Sorry I missed your question about why these HD audio formats (SACD for example) exist? I don't know, probably marketing I guess, yes. I suspect Sony thought for a while that the SACD format "Bitstream" aka DSD was technically and therefore audibly superior to PCM (the format used by CD and DVD-A), and it suited them to have something new to sell to the consumer. In fact DSD is a terrible format for anything other than live recordings of say classical concerts, because it's really hard to mix DSD and converting it to PCM without compromise is not straightforward. They would have been much better recording in 24/96 (24/192 is pointless under all circumstances) and then downmixing the final cut to 16/44.1 for distribution. Mastering in 24/96 is potentially useful because it's important to have the noise floor as low as you can get it for multiple overdubs and re-recordings. For end-user playback it is pointless.
Last edited:
Just to make sure I understand you correctly -
Those very very good 24/96-192 masters would sound just as fine with reduced word length and brick wall filtered to fit the 16/44.1 standard?
That is mathematically correct. The "high def" master carries no more audible information than a 16/44.1 CD quality data stream. Additional bit depth (24 vs 16) increases the signal to noise ratio. The higher sample rate increases the maximum frequency that can be reproduced, but 44.1 kHz covers the human audible range already.
While risking to sound esotheric here,
imagine a world - a forest, a seashore, a city - filtered down to 16/44.1
I'm sure this would be a rather dull place.
Great cymbals, a trumpet, violin or soprano sax produce harmonic content well above 20kHz, up to 100kHz to be precise. While they can not be heard by definition, they can be perceived by humans. While I have no proof for this at all, I had the pleasure to rehearse and perform with a wide range of exceptional instrumentalists during the last 25 years. I never heard a CD recording that came in the slightest close to the real thing, though.
To be fair, the majority of productions doesn't try to reflect the (the sometimes shocking ) reality of real instruments, but much rather a producer's or band's take on it, and what they want to offer for at-home listening. Instruments or vocals will often be highly disfigured, as an artistic choice. All too many times I've witnessed showcases of high end gear where at some point the seller jumped up saying Do you hear this sax?!, and it became obvious to me that this person has probably never heard a saxophone in real life.
Music at home will always be an image of music, and no matter which style (true acoustic recordings, or the most artificial electronic music, where probably no real-life counterpart even exists ), some productions paint a rather low res picture while other producers and engineers aim to put more on their canvas. In my own experience, these are the people who not only work in 24/96+, but also make their music (or their remasters) available at full resolution. They do it because they are proud of their products and because they know it can't be reproduced in the same way @16/44. These people are no mere believers, they are highly trained listeners, sometimes even with what they (or others!) call golden ears.
Interestingly, some of the first high res masters I've acquired where free additions to quality vinyl releases. To me, this prooves a/ the engineers are really proud of their job and b/ a SACD or DVD-A copy is really cheap to make and add as a freebie! (Bowie, Ziggy Stardust or Talk Talk, Spirit of Eden vinyl reissues are recent examples from the pop business).
There's no doubt that we will see a lot of smart people making a lot of money in years to come only based on stating higher numbers and claiming it has to be better just because of that. Also, I'm convinced that high res audio doesn't make sense for everyone, if only for the fact that we need a certain equipment to enjoy the difference, let alone the attitude or will. However, I'm convinced there is a difference. It takes a good production to make it audible, just as much as it takes a good way of re-production. I think Apples involement can have an impact only by giving a larger platform to producers who actually care. I know many of them.
Long rant, I hope there's a certain point in this. The plain scientific approach to me personally is a dead end.
Please, everybody wanting 24bit and 92KHz:
- I'd be fine with going that far, however
- as long as 95% of all music is mastered in such a bad way that even the maximum quality attainable with 44/16 is never even remotely reached, it just DOESN'T MAKE SENSE to store CRAP with higher bitrate!
Now if you want to know HOW GOOD 44/16 can sound, get something like the first track from the HDCD "Flux" by Elisabeth Karsten, or another similarly well made HDCD.
FIRST STEP: Get all music up to that level. SECOND STEP: Let's crank up the bitrate to get even more out of it. The other way around DOES NOT MAKE SENSE! It will very likely lead to typical crap marketing "look we've now got more bits" even though the sound will be the same old rubbish upsampled.
- I'd be fine with going that far, however
- as long as 95% of all music is mastered in such a bad way that even the maximum quality attainable with 44/16 is never even remotely reached, it just DOESN'T MAKE SENSE to store CRAP with higher bitrate!
Now if you want to know HOW GOOD 44/16 can sound, get something like the first track from the HDCD "Flux" by Elisabeth Karsten, or another similarly well made HDCD.
FIRST STEP: Get all music up to that level. SECOND STEP: Let's crank up the bitrate to get even more out of it. The other way around DOES NOT MAKE SENSE! It will very likely lead to typical crap marketing "look we've now got more bits" even though the sound will be the same old rubbish upsampled.
Sort of. If the bit depth is reduced and it's resampled using decent alogrithms, then yes absolutely. (Just truncating and poor resampling can produce distortions that are theoretically audible. But done properly, what you say is true.)
If you want to test this out for yourself, what I suggest you try is this: Take one of your prized 24/96 (or 24/192) recordings, downsample it to 16/44.1 and then upsample that file back up to 24/96 or 24/192. So now you have 2 files, the original and a copy which is at the same bit depth and sample rate as the original, but which has all of the information that cannot be stored within the 16/44.1 format stripped out of it.
If you don't have access to a quality SRC, by all means send me the file and I will do it for you. But there's a demo version of Audiofile Engineering's Triumph app, which includes iZotope (which is one of the best SRC's) if you have a Mac and want to have a play. Just import a file and choose Render to convert it. It's all pretty straightforward. Set the quality slider to max of course.
Now play the two files back to back a few times and see if you can tell which is which.
This is a good test because it eliminates your DAC from the equation. If you just compare a 16/44.1 file with a 24/96 file, it's quite possible your DAC does a better job converting one of the files, compared to the other. If you feed it with 2 files of the same format, this possibilty is eliminated.
If you can honestly, genuinely, sincerely tell which file is which, I will eat my shoes.
Last edited:
But that's exactly what you do get. Your eyes filter everything outside a 430 - 790 THz electromagnetic window, even though radiation outside those frequencies is all around us
And likewise our ears do just the same with respect to sound, and anything lower than 20Hz or above 20KHz, you can't hear.
It's not dull at all. It's the world we inhabit. You can no more hear (or in any other way detect) 30KHz sounds than you can see X-Rays. If someone claimed to be able to see X-Rays, you'd say they were deluded. An audio buff says he can hear 30KHz and that's deemed OK. Moreover, it's forbidden by the hifi fraternity to even question such claims. "I believe my ears" is the cry. It's quite quite bizarre.
Last edited:
But that's exactly what you do get. Your eyes filter everything outside a 430790 THz electromagnetic window, even though radiation outside those frequencies is all around us
Hah, maybe we'll meet some day for a drink (or shoe?!) and listen to some *real* birds ; )
Thanks for your conversion hint, I will certainly try this once I'll find some quality time (iZ user, nice to 'see' it's worth the money).
even if we grant this, why does it matter since most mics won't pick up that extra sonic information, some equipment in the chain won't pass it, and most playback systems won't reproduce it accurately(if at all)?
how do you distinguish between mere believers and people who *know*?
I would say with 98% certainty this is the issue. A few labels run their own download sites or allow their tracks to be released on HD download services, but most don't and probably don't want to. Unfortunate, and since the population caring about audio quality are too small to generate a significant pirating community, it is unlikely to change anytime soon.