Analog summing emulation idea

[Kiwburger]

Yeah, I've met hundreds, possibly thousands of people who could tell the difference between cables. Except in a double blind test. Then I've met, let me see now... none. Not one. What's more there's never been a double blind test where anybody could tell the difference.

Sheer ignorance. If you run cheap unbalanced line cables any distance, you will hear unwanted hum and crap. Simply replacing them with higher quality cables can noticably reduce the hum, because they are better shielded. The same applies to balanced mic cables, especially at high gain (ribbon mics etc). EMI shielding specs are measurable, and if you have any EMI problems (what studio doesn't), you will definately hear the difference a better cable makes. If you can't, maybe you have bigger noisefloor issues somewhere else in the chain, or a very noisy acoustic environment. In my view, you spend the money on quality cables for what you don't want to hear. Silence is golden.

[nuffink]

Yeah, I've met hundreds, possibly thousands of people who could tell the difference between cables. Except in a double blind test. Then I've met, let me see now... none. Not one. What's more there's never been a double blind test where anybody could tell the difference.

Sheer ignorance. If you run cheap unbalanced line cables any distance, you will hear unwanted hum and crap. Simply replacing them with higher quality cables can noticably reduce the hum, because they are better shielded. The same applies to balanced mic cables, especially at high gain (ribbon mics etc). EMI shielding specs are measurable, and if you have any EMI problems (what studio doesn't), you will definately hear the difference a better cable makes. If you can't, maybe you have bigger noisefloor issues somewhere else in the chain, or a very noisy acoustic environment. In my view, you spend the mone yon quality cables for what you don't want to hear. Silence is golden.

Except he was talking about speaker cables. And if it's ignorance, find me one, just one double blind test on cables with a statistically significant result.

Clue:- Try the cable manufacturers first, 'cos if the had the evidence surely they'd use it.

[jens]

[loads of unrelevant stuff]

fine and all - just has got nada to do with the described differences...

[stefancrs]

Yeah, I've met hundreds, possibly thousands of people who could tell the difference between cables. Except in a double blind test. Then I've met, let me see now... none. Not one. What's more there's never been a double blind test where anybody could tell the difference.

Sheer ignorance. If you run cheap unbalanced line cables any distance, you will hear unwanted hum and crap. Simply replacing them with higher quality cables can noticably reduce the hum, because they are better shielded. The same applies to balanced mic cables, especially at high gain (ribbon mics etc). EMI shielding specs are measurable, and if you have any EMI problems (what studio doesn't), you will definately hear the difference a better cable makes. If you can't, maybe you have bigger noisefloor issues somewhere else in the chain, or a very noisy acoustic environment. In my view, you spend the money on quality cables for what you don't want to hear. Silence is golden.

I've never heard any cable related audible hum or noise in any speaker setup I've listened to. Sound source and amplifier related hum and noise however is a different issue.

[Kingston]

Here's a slightly more advanced than a mere high school level introduction to ear. Although still sticking to the basics, it sets straight most blatant errors of human anatomy knowledge apparent in this thread.

http://www.du.edu/~jcalvert/waves/hear.htm

Mandatory reading for all placebo idiots!

You might be the one.

[friteuse]

friteuse, no need to worry about that. Just asked if you have other alike information. It's interesting to know *how* we sense that phase (time) difference - is it perceived as skin chill or disorientation or something else?

It isn't that esoteric . Here's a brief summary of the article's paragraph about hearing:

---
At first, such small phase differences are needed for the perception of direction. Here, a human being can percept angle differences of 3-5 degrees, what corresponds to a time difference of 10 µs. It has been showed with tests on instrumental acoustics that we even can percept signal structures in sounds down to 5µs and that we need them for a natural sound impression. This also becomes important when talking about impulse behaviour. Some analog tape machines have a built in phase compensation to improve the impulse transmission. This was adjusted with a 10 kHz square signal - so affecting frequencies above 20 kHz (as the first partial of a 10 kHz square is already 30 kHz).

In addition, we also seem to "feel" frequencies above our powers of hearing, what has been showed in newer tests, too, but is not finally prooven itself. In which way this happens, wasn't mentioned.
---

All together, I think, high sample rates play a rule at the limits of our hearing. It isn't that obvious like differences between cassettes and CDs. But since more and more people ask why analog equipement sounded better (or different), it should be considered when querying and improving actual digital techniques, beside all distortion and component tolerance effects.

[stefancrs]

I can't hear from where a very high or a very low frequency sound is coming from, I don't know about you though (I can hear up to 18kHz, but I haven't got a chance to hear where it's coming from...)

[friteuse]

I can't hear from where a very high or a very low frequency sound is coming from, I don't know about you though (I can hear up to 18kHz, but I haven't got a chance to hear where it's coming from...)

It's not about high frequencies in a sense of high tones. Of course, we percept direction not so good in very high frequencies.

It's about phase differences, no matter, at which frequency. Say, you'll have a 1 kHz sinus on both mics, but the left one should be delayed 10 µs. You'll need a very high resolution to represent this digitally (or you have to interpolate, what is effectivly like oversampling...)

[bernhardtjeff]

Here's a slightly more advanced than a mere high school level introduction to ear. Although still sticking to the basics, it sets straight most blatant errors of human anatomy knowledge apparent in this thread.

ha, i had to read that book in college.

i love these old scientific books from the 60s. everything is presented like this:

Want to know the answers to all the most vexing questions?
Modern Science can help!

it's weird... whenever i read old stuff like this, i always end up reading something later that basically indicates everything they said was bullshit. then i read something else that says what that guy said was bullshit. then i read something else... and so on, so i really don't know whom to believe. it's the exact same thing with this forum.

"i can hear a difference."
"no you can't."
"yeah, i can. this guy said so."
"well this guy said that guy's full of shit."
"well that guy is known to be an idiot."
"where? do you have any proof of that?"
"no, but here's something unrelated that says i'm right anyway."
"oh, yeah, that's bullshit, too. you're stupid."
"don't resort to personal insults, jackass."
"you're a hypocrite for saying that."
"yeah, but at least i'm right."
"this guy says you're wrong."
"that guy's an idiot. everyone knows it."
"find me just one person who can disprove what he said."
"okay: me. i can disprove it."
"fine, then do it. see? you can't. you suck."
"there's no point arguing with you. get outta here and go make tarnce with your high school buddies."
"see? personal insults again. for your information, i'm not in high school. i run my own studio and have 70 years experience working with audio."
"well, maybe your hearings not what it used to be, jackass. anyway, here's another paper this guy wrote who says you're wrong."



well, i guess that's what you get when everyone's an expert...

[bernhardtjeff]

What's more there's never been a double blind test where anybody could tell the difference.

hey, if you've got the cables and the gear, let's do a test. i have particularly bad ears, so i'd be a perfect ginny pig. just yousendit some wavs or mp3s (would 320k suffice?) or something to rocklegendallstar@yahoo.co.jp.

[deaf dunderkwac]

I heard a big difference when I changed out 50ft of 22 gauge shielded cable for 12 gauge at one of the first studios I worked at.
But that was wrong on so many levels (they even used the shield for the neg connection. The amplifier was on the verge of oscillation because of the added capacitance. Mind you, this was 1970.)



maybe I'm the one...


When I first started in tv production, I was being given a tour of a facility where quite a few of their monitors where not being feed a sync signal. People working there were oblivious to the incredibly high levels of about 16kHz screaming thoughout the floor.
Iasked the tour guide to guide me out of there as soon as the elevators doors opened.

As we age we lose alot of hearing in the ultra-high range.
If you bother to look this up, you'll find that women have better hearing, but if we haven't abused our hearing too much, by the age o 30 or so we all suffer a certain amount of lose in the highest octave.



(more trivial information)


@friteuse

it seems like only yesturday (well, 2002) where our staff had all these interesting discussions about where engines should process at. Of course processing has to work at higher bits, otherwise you couldn't add (mix) delays and such.
To get a smoother reverb sound it seems you would need add more delays to get a more diffuse sound.


Really, we are saying the same thing to each other, I sometimes fall into the engineering side more then I should, and perhaps I simplify more then I should.
Over the years I've found that if I use the same buzzwords everyone else has been using, alot of the real meaning to those buzzwords has been lost. A great many people get the concepts turned around So I avoid using them...


(just thinking outside the box with a new pair-of-dimes)

... directly.

[Kingston]

It's about phase differences, no matter, at which frequency. Say, you'll have a 1 kHz sinus on both mics, but the left one should be delayed 10 µs. You'll need a very high resolution to represent this digitally.

But it just doesn't work like that, no matter how much you want to believe it.

Study nyquist theorem, and find out why. Pay attention to the bits (pun intended) about sine vs. sampled point representation, and how they relate to each other, and how AD/DA converters deal with them. It's quite useless fighting against physics really.

[deaf dunderkwac]

aply same nyquist theorum to 10k. This is where it gets a bitmore interesting.
Not that it matters much.
There's still that issue of not all ad/da converters being equal and all.
It's hard to quantify, even with highly sophisticated test equipment.
We assume we understand everything about how the hearing/brain mechanism really works, only to discover something additional to our group knowledge.

A few years ago, around 1998, I performed a double-blind test on 10k sinewaves vs. 10k squarewaves and sawtooth waves.
I was at the time, amazed at the results, as I could hear the difference between the analog recording (on a well-adjusted Sony (mci) reel to reel at 15ips (dismissing the analog artifacts) and the digital done at 48kHz 20 bits on a Sony Digibeta video tape recorder.

This is all very academic, mind you, but I spent a lifetime learning to listen for the slightest variations in the signal. It was apparent to me that there was indeed a difference, and it could be verifyed on an o-scope that there was a marked difference in the waveform shape (both the analog and digital recorded versions differed from the actual waveform)
None of the 'participents' knew it was a test, btw

The average person couldn't tell the difference of either recording. All of them could hear the slight shift in the phantom image (cue phantom image gif)(where the heck is voidiod)
Some of the audio engineers noticed a change (they also didn't know it was a test, btw) some didn't...

It's there, but the debate for me is more along the lines of 'does it really matter to anyone but me'

answer... not anymore. Probably not ever. Analog methods worked ok, digital works better, but it gets to a point where it doesn't matter and we reached that point in the late 90's.

go make some music...
I got to get to work.

[Kingston]

There's still that issue of not all ad/da converters being equal and all.
It's hard to quantify, even with highly sophisticated test equipment.

I thought about explaining the sine vs. digital (AD/DA) signal represantation, but wikipedia does that better than I ever could. Basically the theory states that it's impossible to build the perfect lowpass filter (sinc), but that oversampling (withing AD/DA) and high samplerates (more relaxed lowpass filter) somewhat overcome that limitation.

What it means in laymans terms, is that instead of wanking about with highsamplerates, one should be using as high quality AD/DA and clocks as possible, to overcome those limitations, and to blur the line between 96khz and 44.1khz. Luxury converters of today can do 44.1khz at the theoretical limits. But that has only been possible for the last 5 years or so and it's still pretty expensive.

There was a great article (that I can find anymore unfortunately), that with the knowledge and limitations of todays technology, the optimal samplerate for pretty much anything is around 50-56khz. With those rates, the lowpass filtering is relatively relaxed (slope-wise), and DSP requirements aren't completely awful. The article stated that it was a trade off with best of both worlds.

It's unfortunate we have the large gap between 48khz and 88.2khz nowadays.

[friteuse]

It's about phase differences, no matter, at which frequency. Say, you'll have a 1 kHz sinus on both mics, but the left one should be delayed 10 µs. You'll need a very high resolution to represent this digitally.

But it just doesn't work like that, no matter how much you want to believe it.

Study nyquist theorem, and find out why. Pay attention to the bits (pun intended) about sine vs. sampled point representation, and how they relate to each other, and how AD/DA converters deal with them. It's quite useless fighting against physics really.

Sorry, maybe not well formulated. I'm fighting more against my poor english than against physics...
I think we both mean the same thing but you did not understand what I inteded to say or I did not understand what you said.

I wanted to say the perception of direction has to do with frequencies in that way, that the higher the sample frequency, the better the "direction resolution" (what ever the correct expression is for that). So, higher sample rates improve direction perception generally for any frequency and not, like bernhardtjeff assumed, for only high frequencies (tones).

My example of an 1 kHz sinus was a bit silly as I shouldn't have related it to mics. Taking an computer generated 1 kHz @ 44.1 kHz sample rate will make it impossoble to delay 10 µs without interpolation (as the smallest possible delay @ 44.1 kHz is 1/44100 ~= 25 µs).

Maybe I missed the point completly, but I see no direct relation to nyquist here... of course, nyquist is extremly important when dealing with AD/DA, but wasn't at all the topic I was talking about

Doesen't matter, I'll do like Dunder, go back to work and doing music now. And I will do it as I like to, nobody should be troubled when I do it @ 96 kHz