This guy set up a way to test if you can hear them http://www.phy.davidson.edu/fachome/dmb/wmpviz/beats/beats.htm

There are detailed books on acoustics, like Benade, they are pretty dry. Benade's page at Stanford: https://ccrma.stanford.edu/marl/Benade/BenadeHome.html

Here is an interesting article, it is simplified a bit that focuses on difference tones: http://hjertmann.com/Combination_Tones_as_Harmonic_Material.pdf

Of course, I could be hearing things. I don't hear as well with the tin foil.

@Vitaliy frequency differentials have been studied since Tartini mentioned them in the 18th century. I personally can hear them, and most trained musicians can hear them. In addition, there is an extensive body of literature in physics on the subject.

However, since the effects are produced as part of hearing, the idea takes some getting used to. There is a similar branch of science concerning "filling in" of details by the visual cortex, but combination tones are considered "cochlear" rather than "perceptual", and as such represent physical phenomena. There are some who believe that the tones are primarily neurological.

Combination tones, like harmonics, comprise part of the texture of music. If you take them away, the music will sound different--maybe better in some cases.

Church organs have pipes that generate combination tones. So for the sine wave like character of the organ pipes, the combination tones add a dimension to the sound. Since the tones are used for low pitches, an accurate recording can be made that, when played on a very high quality speaker will generate the tones. They can be reproduced on very accurate headphones.

However, resultant tones derived from upper harmonics cannot be recorded or generated without increasing the recorded range by at least an octave.

I remember the first day in physics class when these tones were described as being "in ear" as opposed to being "in air". Takes some getting used to.

Very basic info: http://en.wikipedia.org/wiki/Combination_tone

Fuck spending 1000s on speakers et al - your room and your ears bugger it up far more than buying a 10k set of monitors can ever do. I spend more on a room than gear when I build a new space. Buy some speakers you like and know them with stuff you made or know intimately - cost a tenner don't matter! Back to Led Zep - their greatest album was mixed on a broken Yamaha amp with NS10's (also battered)

Huh. Spending on speakers is fun thing to do. At least it produce real difference usually :-)

Also, it is clear that good treatment is requiring for mixing, much less is ok if you just listen to things. Brain and ears are quite good dealing with reflections. if you have good amount of books, various stuff and soft furniture it is not bad already. Cheap way for mixing is monitors directed to you and being near field while back fall being far away and with some simple sound proofing.

Graphs - arse - I like Hendrix and Led Zep however "inadequately" they were recorded, hell a 58 strat would be decimated by graphs and tests these days, but they hugely sound better than a modern variation - and yes I have both - and same with 60's Les Pauls Teles Thunderbird and 335 etc and yes I have them too as well as their modern variations.

If it sounds good it is aint it?

Similar thing with soft synths and the largely pointless discussion about what's "better" I have a Jupiter 4 6 8 MKS50 80 101 3030 909 blah - Jupiters are a fuck to tune very session - same balls about amps - have a Jensen AC30 JCM50 and a room of other old glowing shit - are they better ? No Idea - what get used - what sounds best fastest.

Highest earning single last year (on ipad?) - made originally on garage band and a shit old epi acoustic - how do I know ? leave it you you to discuss. If you only knew !

Just use whatever sounds right - fuck trends and graphs.

Fuck spending 1000s on speakers et al - your room and your ears bugger it up far more than buying a 10k set of monitors can ever do. I spend more on a room than gear when I build a new space. Buy some speakers you like and know them with stuff you made or know intimately - cost a tenner don't matter! Back to Led Zep - their greatest album was mixed on a broken Yamaha amp with NS10's (also battered)

Hey but what do I know !

Concentrate on making shit not talking about it!

Happy New Year - Im back to me mix working like a jobber, in the box on PooTools, as 90% of everything you hear is submitted these days not a valve or 192k in sight.

@vitaliy

Here's my math. It doesn't matter if it is phase or time difference because the reproduction system has to recreate it.

The problem with my math is that it gives a sampling rate of 15.5x2 or 31KHz

The article http://en.wikipedia.org/wiki/Interaural_time_difference states: "The normal human threshold for detection of an ITD is up to a time difference of 10μs (microseconds)"

Which suggests that we can detect locational differences much smaller than I originally stated.

This is a debate that has been going on for a long time and there have been tens of thousands of pages written about it. So I will add only a few lines to this.

First, music contains combination tones which result from harmonics. These tones form in the audible range from inaudible harmonics above the range of hearing. If you strip off the top harmonics, you remove the combination tones sounding one octave or more lower, and these then disappear from the way hear the recorded sound. Note the phrase "the way we hear the recorded sound", because that isn't the same as the sound itself. Sound waves form in the ear as well.

Second, you cat (and you dog, but cats can hear higher) can hear up to 10.5 octaves, so if you pets are important to you, make sure to capture those higher frequencies.

60 is a very good sampling rate for you and you pets.

Remember that the analog audio chain is a weak link, even at 16/44. Indeed, one company (Wolfson) has built an excellent business by producing DACs which just sound better than the rest (even at 16 bit depth). Initially, Creative used the Wolfson chips in their early high-end MP3 systems (such as those I have used for a decade). Now I see that smartphones need higher quality audio too... and guess who is still able to make a better system...?

http://www.theinquirer.net/inquirer/news/2261101/samsung-galaxy-s4-set-to-ship-with-wolfson-audio-chip

I have used 24/192 in distortion measuring equipment. It has measurably lower distortion than 16/48, but a human would never hear the difference.

@sammy

I think we have few discussions on this. If you remove all except equipment it is mike and preamp quality (either in recorder or separate), ADC and recorder features are not very important unless it is utter cheap crap.

@sammy Order of Importance: Talent/Performance, Acoustics/mic placement, microphone, Preamp, Convertors.

Of course, sometimes talent, performance, and acoustics cannot be under complete control of engineer, so... yea, I guess I'd start out with good mic.

@Vitaliy

Where I wasn't clear is that it 2Mb/s DSD starts to make sense, 192/24 is about 9.2Mb/s (Stereo).

Your link in a step leads me to: http://en.wikipedia.org/wiki/Interaural_time_difference

Which suggests that it is either time or phase that is being detected, time for impulse, phase for tones, particularly above 1500Hz (ITD and ILD). A click is an impulse.

I'm of the belief that the high sample rates help to preserve the time and phase differences that we can detect. The previous assertions have been about volume and hearing frequency range.

The whole clicking experiment was in 'Audio World' several years ago and out of interest I tried it for myself.

Your link also leads to : https://www.meridian-audio.com/ara/coding2.pdf in which Bob Stuart says:

"Other listening tests witnessed by the author have made it quite clear that the sound quality of a chain is generally regarded as better when it runs at 96kHz than when it runs at 48kHz, and that the difference observed is ‘in the bass’. "

This ties up with a lecture from Richard Lord I attended a good few years ago. Bob Stuart's conclusions are on page 19, the rough synopsis is that 58K/20bit

I have vinyl, CDs, DSDs, SACDs and DVD-As at home. I agree that vinyl struggles against CD in many aspects. I have a 1960 recording of 'Take-5' which was transcripted from 30ips tape to DSD which shows me that great audio was available back then. I also have a SACD from a Police recording (Sting,Summers,Copeland) which I find very unpleasant to listen to.

From my diving medical I know that I have a hearing notch at 12KHz and Tinnitus which by rights should give me pretty poor hearing on a 'golden ear' basis. Therefore I can't argue that much!

@vitaliy The very high sampling rates make most sense to male hearing.

As a very simple experiment, if you have someone stand 2 metres behind you and click their fingers most men can detect when the clicking has moved about 30cm. If you do a little trig and math you find that the time difference between click arriving at the left and right ear is very small, double it for Nyquist and 2Mhz sampling rates start to make sense.

The time difference is : 0.0000657 seconds.

It is believed that one of the reasons that good vinyl can sound better than CD is that the timing differences between the left and right channels are better preserved.

I would say more compressed sound was pretty well popular before mobile devices like ipods and phones became commonplace. Upload a commercial track from 80s then upload a commercial track from the 90s... that's probably the most drastic difference you'll see, and it was a bit before the mobile took off too much.

I misunderstood when you said noise floor, I thought you meant the noise floor of the track/recording itself. I agree that you do need more compression for different environments – it's like, when you watch a movie on an airplane vs in a theatre, you definitely want a lot more dynamic range in one and not the other.

I have to disagree that loudness wars started as answer to background noise levels. It started out of continuing competition for outstanding sound compared to other companies, and the fact that human hearing percieves louder sounds as being "better" or atleast more sonically full (we hear more frequencies more evenly at louder volumes)

With digital, we have quieter noise floors than ever, and loudness war has been getting worse.