24bit vs 16bit and Hz

  • Thread starter Thread starter adam79
  • Start date Start date
Farview said:
Some converters sound better at one sample rate than the other. They shouldn't, but some do. Since some sound better at lower sample rates and some sound better at higher ones, it really must be something with the specific design of the converters in question and not really the sample rate that sounds better.
Click to expand...

I've seen discussions about this type of thing...it's almost like the thing with LCD monitors, where they have one resolution where the screen "mathematically" looks the best....and even with that, some people prefer something different. :D

Which brings up the question why there is sometimes this effort to persuade people that they really should like something different just because the math says so.....? ;)
 
miroslav said:
Which brings up the question why there is sometimes this effort to persuade people that they really should like something different just because the math says so.....? ;)
Click to expand...
Because normally the types of things that fall into that category are the ones where the difference is so slight (or non-existent) that math is the only consistent way of telling them apart.
 
Though math really shouldn't dictate preferences...even if shows a difference or NO difference.

We deal with a subjective art form, and like Mo Facta was suggesting, whatever "mojo" people need/use to achieve their end result, is all valid IMHO.
There's a million things in the audio world where decisions are made purely on a subjective basis...and I don't see a bunch of guys in lab coats running over and pulling out their calculators to try and convince people what they should prefer based on what the calculator is telling them. :)
 
miroslav said:
There's a million things in the audio world where decisions are made purely on a subjective basis...and I don't see a bunch of guys in lab coats running over and pulling out their calculators to try and convince people what they should prefer based on what the calculator is telling them. :)
Click to expand...

The only time I put on my lab coat is when people make claims for differences in quality that I know are incorrect. Yes, mixing is all about making subjective opinions. That's not always wrong when buying gear either, unless you're paying ten times more for something that you yourself would agree is no better were you to do a proper comparison. (I'm not referring to you personally Miroslav.)

--Ethan
 
miroslav said:
Though math really shouldn't dictate preferences...even if shows a difference or NO difference.

We deal with a subjective art form, and like Mo Facta was suggesting, whatever "mojo" people need/use to achieve their end result, is all valid IMHO.
There's a million things in the audio world where decisions are made purely on a subjective basis...and I don't see a bunch of guys in lab coats running over and pulling out their calculators to try and convince people what they should prefer based on what the calculator is telling them. :)
Click to expand...
The only time that tends to happen is when someone confuses their personal preference/opinion with actual fact.

Facts can be proven, opinions can not. That's the nifty thing about science, it's true whether you believe in it or not.
 
Yes...science is OK up to a point, but when you get to where "beyond here there be dragons"...then a little faith is needed too, 'cuz the science is not yet 100% complete, or it is only as complete as we currently understand it or have the ability to measure it.
If science just took for granted what is known at any present moment as the absolute final end-all...science would have stopped a long time ago, since each generation has had their moments, until the next generation went deeper.

So....some folks in the music/audio world still like to maintain a little faith along with the science... ;)
Maybe it's not all provable today...but if feels right, there's nothing wrong in going with it on a little faith.

I think we can all agree that science can't prove everything...and in the subjective world of audio, I just don't think it's as important as some folks like to make it out to be.
 
Farview said:
Also, the difference is subtle enough, even when there is a difference, that it really could just be in your head.
Click to expand...

OK...so tonight I messed around a bit with different sample rates...just to see how subtle things might be.

I wanted to take one source and convert it at 4 different rates using the exact same signal path/converter...with the only difference being the sample rate.

Since I couldn't find a way to play something identically 4 times....other than maybe tracking it to tape and then playing the tape back 4 times, which I decided against as it introduced a potential variable where the mechanical movement of the tape might be ever so slightly different with each pass...the only other option was to take a CD and just play the same track 4 times.
I kinda' didn't like the idea of taking a digital sound file, playing it back out to analog, and then converting it again back through my converter, but since I was just looking for differences, and not making a better/worse judgment, I decided this was OK.

So...I grabbed a CD that I knew was done VERY well, and had a lot of musical subtleties and air in-between, and did four passes, converting with each pass at 44.1, 48, 88.2 and 96...all at 24 bit.

Once in the DAW, I listened to each pass...and I have to say that I actually could hear a v-e-r-y subtle difference between the lowest (44.1) and highest (96) files.
The difference I heard was in the depth of the bass and also width of the image...but yes, it was V-E-R-Y subtle, and almost negligible.
I then cut identical 20 second pieces out of each, and converted all of them to 44.1/16...then I lined them up sample accurate, and listened to them over and over...and I could not hear any subtleties any more.
Of course, after about 20-30 times of the same short passage, and my ears lost what little objectivity there was.

Now I ain't no golden ear, so I can't speak for those pros, but I can understand why some of them might hear those things...and like Jay said, it may just be that a converter sounds a little different at 96 VS 44.1 because of the filtering or whatnot, but I don't have 10 converters to test against, so it may just be how MY converter does it. :)

So...do I think it's absolutely dumb recording at higher rates...well, I don't think it hurts anything, unless your system can't handle it when mixing huge track counts at higher rates...or you don't have a lot of storage. Otherwise, if you sleep better recording at higher rates...I say go for it.
That said...I don't think that when it comes to "Rock & Roll" those subtleties will be all that critical, and it does seem that once converted down to 44.1/16...even the subtleties vanish...so I wouldn't lose any sleep over doing the lower rates either.

I think folks should just go with what they feel most comfortable with. I will probably still track some things at 88.2, just 'cuz I feel like it :D...but I won't be concerned too much if I do some stuff at 44.1...and whatever I heard or couldn't hear is the limitation of my own ears. Someone else may come out with completely different conclusions.
 
miroslav said:
So....some folks in the music/audio world still like to maintain a little faith along with the science... ;)
Maybe it's not all provable today...but if feels right, there's nothing wrong in going with it on a little faith.
Click to expand...

There's nothing wrong in going with it on a little faith.

Just so long as that 'little faith' is not asserted to be an absolute. If "I can definitely hear a difference" is a faith-based pronouncement (i.e. a subjective assessment), then it can't be followed with "therefore there is definitely a difference."

On the other hand, if you repeatedly and consistently say "I can definitely hear a difference" during a series of double-blind tests, you have reasonable grounds for following it with "therefore there is a difference".
 
I think discussing audio sound quality is a little like talking about the existence of god and trying to decide which religion depicts god more accurately! :D
 
Maybe I can put a rather different slant on the "wing it or measure it" debate?

When "our guys" are developing a new product, amp, pedal whatever, they spend hundreds of hours playing and listening. Back to bench sloder, swear, tea, sloder, play, swear..........Eventually a concensus emerges that the sound is "right" and except for the ball park calculations to get the initial component values, virtually all the fine work is done subjectively...BUT!

They now have to make 1,000s of the fekkers SO! A specification, XdBs at Y frequency for Zinput, is painstakingly produced against which all products are checked. Can't make consistant products any other way....Two other things,

If you can't make decent recordings at 16bits 44.1kHz, 24bits and other sampling rates won't help you (except for "legroom!").

CD burning? We (son mostly) do everything (mostly) in Samplitude 8 se, 24 bits 44.1, export as 16bit .wav to desktop and burn. Never been an issue?

Dave.
 
miroslav said:
I think discussing audio sound quality is a little like talking about the existence of god and trying to decide which religion depicts god more accurately! :D
Click to expand...

Well, that sounds good on paper, but not really. Nobody questions the existence of sound to begin with. :)
 
ecc83 said:
If you can't make decent recordings at 16bits 44.1kHz, 24bits and other sampling rates won't help you (except for "legroom!").
Click to expand...

That's not really the point. People say that kind of stuff all the time, but it doesn't mean much when it comes to the actual process of recording.

People make decent recordings when they find *a comfort zone that works for them*...and that's why all the "mojo" stuff is valid IMO. If someone needs a $3k Les Paul to find his "mojo"....it's valid. Saying to him that he can play the same stuff on a $300 knock-off doesn't mean anything...because that's not HIS comfort zone.


RAMI said:
Well, that sounds good on paper, but not really. Nobody questions the existence of sound to begin with. :)
Click to expand...

But there's that old question that's never been answered....if a tree falls in the woods........ :)

Recording of sound isn't just a laboratory experiment based on analytical measurments....I think we can all agree that there is a LOT of faith that goes into each recording we do.
If it was all done and provable with math, we could just plug in the correct "formula" for every session... ;)
 
miroslav said:
But there's that old question that's never been answered....if a tree falls in the woods......
Click to expand...
You mean "If a man says something in the forest and no woman can hear him, is he still always wrong?" :eek:
 
I always liked the variation from the old "Murphy Brown" sitcom when Murphy says to dumb blonde Corky.....

"If a tree falls in the woods and hits you in the head....would you feel it?"

:D
 
Bit depth 24bit is no higher quality than 16bit. The only difference is headroom before you hit the digital distortion threshold at the 0db. Ableton for example runs everything at 32bit internally, thus you dont need to worry about ever going into the red.
Mastering Engineers want files in 24 bit so they have more headroom to play around with.

20-22,000 Hz is the extremities of the human ear (No human ear has every been able todetect higher frequencies in any test). The Nyquist law of sampling states that you need to record double that rate to get the up and down of each waveform to be able to reproduce that waveform exactly, hence 44,100 covers that range(the extra 120 was used for technical reasons). Some people record samples a litle higher than that at 48,000. Anything higher than that are prone to ultrasonic artifacts which can damage your hearing. Lookup Nyquist Shannone Law of Sampling in Google if you want the full details, and have a mathematical brain.
 
You must log in or register to reply here.
Back
Top