Neil Young speaks out against excessive compression

[scottfeldstein]

So you are proposing that the standard be 'good enough for people who don't really care, listening on an iPod.'

No, I think the standard should be "all the quality that people can distinguish."

 

[RBrownBass]

I won't even get into the alarming trend for musicians who use a million dollars worth of studio equipment just so they can sound as if they went back in time to 1971 and recorded their album in a garage.

Yeah, I love Fenders myself

Not alarming at all, at least not on Talkbass. Aaaaalllll the great recordings were done a certain way, so anything that comes afterward is unquestionably a waste of time and money. Our ears are already conditioned.

 

[chuck norriss]

Bruce Swedien: Recording Michael Jackson

I posted this article elsewhere on TB but Bruce Swedien of Quincy Jones/Michael Jackson fame also avoided compression for the most part:

“I’m not a big fan of compression or limiting at all — I can’t emphasise that enough. On many of the recordings that you hear today, all the excitement and all the colour is gone because they’re so over-compressed. I never did that. I would never have a compressor or limiter on the [master] bus, for instance. I want all that transient information there. And no compression or limiting on any drums or percussion. That’s one of the biggest mistakes that I hear, I think, in modern pop recording. The stuff is so compressed they’ve limited the living doo-doo out of the sound.”
That’s not to say that he leaves the dynamics of the performance completely untouched, but is much more inclined to achieve the required dynamic control through to-tape fader rides while overdubbing and automation while mixing. “I’m a nutcase about details in the mix, so I’ll use automation to a degree, but only very subtle compression. I have a pair of the new variety of [Universal Audio] LA2As that I just love, so I will use those, but it’ll only just be tickling the meter, at the most one or two decibels. I don’t like what happens to the sound when you compress any further, and that’s very important to me.”

 

[swspiers]

Get a random sample of 100 music listeners. Tell them you'll pay them $50 if they can identify the higher quality audio. Give each of them the files to listen to, one after the other, in random order.

My hypothesis is that the number of people who can pick correctly is going to prove statistically no better than chance. They'll be guessing.

And that test will fail with a statistical corelation below the threshold. Because there is operator bias all over that test.

The most interesting facet about the reaction of a regular person to compression is listener fatigue, which is not readily apparent or even identifiable until time has passed.

The typical mp3 or AAC file is not only compressed audibly, but dynamically as well. When the volume level is evened out, sure you can 'hear' everything better, but you also get tired of listening quicker.

Another thing to consider is how well does a person know the music that is being listened to in both formats. If we play 100 people a song they've never heard, there is virtually no chance they will hear a difference.

But if it's an old favorite that you've had on vinyl, 8-track, CD, Mp3 and SACD, pretty sure you'll hear a difference.

Another thing, unless they played in a band or orchestra, the statistical chance that those same 100 people will be able to identify the difference between a clarinet and an oboe is also negigible. How many people do you know who can sonically tell the difference between a trombone and a baritone? Or a Jazz Bass from a P-Bass?

 

[bongomania]

Even though it's only a handful of people on here, it's kind of surprising to me that on Talkbass, where the specifics of often inaudible frequencies put out by bass amps are endlessly argued, that Young is being knocked for wanting better sound quality more readily available.

But what is limiting the availability? It's not the technology--both the software and the hardware for distributing and listening to high-data-content audio files has been readily available for many years. It's not the CD format--the physical discs themselves are capable of storing data at much finer resolution than the commercial standard that's used. It's not the iTunes/piracy format, as you can download a high-res file just as easily as a low-res one.

What's limiting the availability is the consumers themselves. They do not want to pay for high quality, and they are impatient--they want their download right away. And they want 5000 songs on their iPhone, not 500.

Young's problem is that he's looking at the completely wrong problem. The real problem is he needs to convince the consumers that they want higher resolution. Everything else is already in place, and has been for a long time.

 

[swspiers]

I'd bet if the listening were done through high end monitors there would be an audible difference, but in practical terms I doubt the average user would ever have the equipment necessary to discern the difference.... so it's kind of moot.

Actually, there's a lot of snake-oil in high end audio. A complete system capable of resolving the detail of the music to the degree needed to discern the difference can probably be put together for less than $1000, certainly for less than $2000.

 

[scottfeldstein]

And that test will fail with a statistical corelation below the threshold. Because there is operator bias all over that test.

No idea what you're getting at. I'm sure we can come up with a methodology that would preclude the "operator" from influencing the results. Double-blind it, maybe.

 

[prd004]

Yeah he's saying he needs "some rich guy" or Steve Jobs to invent a digital format that retains more data, but that will take longer to download???????????

FLAC, WAV, AIFF, how many other examples already exist, and have existed for years?

I respect ol' Neil for many, many things, but this is not one of his finer moments. This, and his claim that he has already designed and built a car that runs on water instead of gas...

I've got an HD CD player in my home studio, and it's damn close to ten years old! Before iPods and mp3's became the rage, HD CD was the direction the industry was heading, and I am so sad it went the other way

 

[David Jayne]

I no longer own a high-quality music system. My best listening experience is an ipod with very good set of Sennheiser headphones. I've ripped certain songs into itunes in the lowest, and highest possible quality formats. I've never been able to tell the difference. Maybe this equipment isn't good enough to tell, I don't know. I'd love to have a nice 'system' again someday but that will be awhile.

 

[ghostsontheroad]

too much Ad hominem in this arguement...Ad hominem - Wikipedia, the free encyclopedia

 

[5StringFool]

Actually, there's a lot of snake-oil in high end audio. A complete system capable of resolving the detail of the music to the degree needed to discern the difference can probably be put together for less than $1000, certainly for less than $2000.

I agree, but most people don't even want to spend anywhere close to a grand on a stereo system. Most of the folks I know buy theirs at wal mart for less than $300, and those are far from true hi-fidelity.

 

[scottfeldstein]

I am so sad it went the other way

I'm not. I think it got pretty weird there for a while when everyone was downloading crappy home mp3 encodings off Napster. Even when we did the job ourselves off of our own CDs we somehow thought that 128k mp3 was pretty good. Weird.

But I have to say that I think the quality is pretty darned good now. And, not to belabor the point, I think most people can no longer distinguish between a 256k AAC and the master files from whence it came. Certainly not using the playback equipment they actually own.

i'm all for getting those last few wrinkles out to make the audiophiles grin. Don't get me wrong. But the genie is out of the bottle, folks. Digital downloads are here and they are not only here to stay, they're here to sit on your couch, eat your food, leer at your girlfriend and hog your remote control.

And I'm glad about it. Audio quality aside for a moment, isn't everyone? It's happening slowly, but internet distribution is beginning to lower the barriers to entry for musicians. Being published on vinyl by a record company used to mean having to fit through a pretty narrow bottleneck, get by a lot of gatekeepers. There's only so much vinyl, only so much distribution, only so much promotion, only so much store shelf space, etc. Not so anymore.

One of the most obvious vestiges of the Old Way is "regions." Why is it that I can't buy a recording but someone in Finland can? Does it cost more for Apple (or anyone) to let me buy it? Of course not. Who gains by restricting it? Not sure. Lawyers? Distributors? Shifty middle men of some sort? The musicians surely don't. And neither do I. That's just one example of what I hope digital downloads are going to soon destroy.

 

[GoesThump]

Ok, there might be something to it if the original recording was done on 1/2" tape at 48 kHz rather than digitally recorded at 44.1.

Perhaps I'm misunderstanding you here, but as written, the above doesn't make much sense. If you're recording at 48 kHz, that implies that you are recording digitally. Did you mean the old DASH recorders?

I know of no analog recorder that plays back to 48kHz. That's almost in bias land.

GT

 

[swspiers]

No idea what you're getting at. I'm sure we can come up with a methodology that would preclude the "operator" from influencing the results. Double-blind it, maybe.

First off, I have massive respect for you. It's kinda weird for me to be debating someone who I honestly believe overall has superior knowledge to me in most things.

But this is an area I'm passionate about.

Operator error simply means that the person devising the test has a bias (which we both do, since we're on opposite sides of the debate).

The problem with most of the methodology in this this area is the assumption that "100 random people can or cannot tell the difference between audio formats if given x amount of music in random order" Which on the surface, seems ok.

But apply my clarinet and oboe analogy, and things get a bit muddier. There are too many independent variables at play, which would all have to be eliminated to account for the + or - of any result.

Age, education, ability to hear frequencies, male/female, familiarity with the music, quality of the playback source, emotional reaction to the genre of the music, experience listening to hi-rez playback: all of these and perhaps dozens of others would skew any reult.

I pose that most people cannot tell the difference between a clarinet and an oboe, a trombone and a baritone, or a jazz and a P. But that would in no way prove that there is no difference.

 

[GoesThump]

Then, going from 16bit/44.1 to 24bit/96k is amazing. You're right that it depends on the source material, but in most cases, the difference can be pretty clear.

Yes source material matters. The quality of the converters matters too. I'll take old Apogees running at 44.1/16 over some random consumer soundcard running at 24/96 anytime.

GT

 

[swspiers]

Yes source material matters. The quality of the converters matters too. I'll take old Apogees running at 44.1/16 over some random consumer soundcard running at 24/96 anytime.

GT

+1 and no contest on the Apogees.

 

[onosson]

I pose that most people cannot tell the difference between a clarinet and an oboe, a trombone and a baritone, or a jazz and a P. But that would in no way prove that there is no difference.

I think a lot would depend upon the (explicit or implicit) goal of the test. Is it to determine whether there is a difference, or is it to determine whether or not the difference matters?

Of course, then we need to decide what we mean by "matters"!

 

[scottfeldstein]

First off, I have massive respect for you. It's kinda weird for me to be debating someone who I honestly believe overall has superior knowledge to me in most things.

It's the gray beard, isn't it. Be honest.

There are too many independent variables at play, which would all have to be eliminated to account for the + or - of any result.

Age, education, ability to hear frequencies, male/female, familiarity with the music, quality of the playback source, emotional reaction to the genre of the music, experience listening to hi-rez playback: all of these and perhaps dozens of others would skew any reult.

No, it wouldn't. Randomization sort of takes care of that issue for us. Or, if you like, we could employ statistical analyses after the fact to control for factors like age, sex and experience. It's been a while since I've had to pass statistics and research methodology courses, but I think we could drop an ANOVA test on that data and get some answers.

I pose that most people cannot tell the difference between a clarinet and an oboe, a trombone and a baritone, or a jazz and a P. But that would in no way prove that there is no difference.

Now you're making sense to me. Yes, just because most people can't discern a difference doesn't mean there isn't one. Or, more importantly: just because most people can't hear it doesn't mean nobody can.

 

[scottfeldstein]

Perhaps I'm misunderstanding you here, but as written, the above doesn't make much sense. If you're recording at 48 kHz, that implies that you are recording digitally. Did you mean the old DASH recorders?

I know of no analog recorder that plays back to 48kHz. That's almost in bias land.

GT

You're right, I'm wrong.

But in any case I agree with those who are saying that the source matters--as does the equipment used to process it into it's destination format.

 

[swspiers]

You may be assuming that the consumers aren't there. I'm not so sure of that. HDTracks is doing fairly well, SACD is alive and well in the Jazz/Classical and Blues world, home theater is thriving with HD video and audio soundtracks. There is a built-in market that will always be looking to by the next format of 'Wish You Were Here', and there is profit to be made by doing so.

High end retail stores may be fading a bit, but mid to high end is actually doing well sales-wise. There's no shortage of mid to high end companies and products, but the means of acquiring them have changed. This is the era of internet-direct sales. Same with music. Heck, my last 2 basses were bought online- one from Ebay, the other direct from Carvin.

There is also an assumption that people want low quality over high quality all of a sudden. I think it's actually a continuation of a trend that is at least as old as the 8-track tape. Even the 33 rpm LP was a technical compromise: 45 rpm LP's were just too big!

True quality was NEVER the overiding motivation of consumer behavior: it's always been faster, smaller, cheaper. It's not a trend that just popped up over the past decade.