The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Launch of Pono Studio Headphones
Old 8th May 2014
  #4771
Gear Guru
 
Karloff70's Avatar
 

Quote:
Originally Posted by j_j View Post
Hm. Perhaps it's actually fertilized the roses and is now encouraging flowers, then. :p
Old 8th May 2014
  #4772
You were there were you?
Old 8th May 2014
  #4773
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by I.R.Baboon View Post
Get a star to approach some other stars while they're on drugs, put them in a car, possibly give them some more drugs, play them a fraudulent, rigged sound comparison, result: "whoa, dude, that's incredible, i never realized our music didn't sound like **** before etc etc", that's Pono, i wish it all the best.
Why are you trying to make this about drugs? If that is your main concern, there are better forums for such discussion.
Old 8th May 2014
  #4774
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by DarkandKurious View Post
To my understanding, Pono is playing back PCM. An Antelope Eclipse (for example) is also playing back PCM. The Eclipse has a dynamic range of 123dB, and 24bit dynamic range is 144dB. Can someone really produce a consumer grade appliance that outperforms a mastering convertor? Simple economic considerations suggest to me that a battery powered device of small capacity will not have the power to outperform a high end converter running on mains.
Actually, battery power is what you'd want there. Many high end phono preamplifiers are based on types of battery power these days, and it's hard to line-condition mains power to be quite as clean and noiseless as battery power. That's an advantage to Pono, not a disadvantage.

Also, the bottleneck is not the existence of batteries, but the duration of playback you get. You can't start a car from wall current, but charge a 12V battery for long enough and it can put out that kind of energy that quickly. It's the same thing with a portable player: with the right parts, you could make it blow up any headphones, very briefly.

These are fair points, but you've got them backwards. It's going to be easier to do these things on batteries, and battery life becomes the issue. Though the form factor of the thing is big enough to hold a very nice battery that'll keep up with the requirements: there, the constraint becomes size, and we already know the Pono is a clunky ungainly Toblerone. You get things from that decision, like the ability to choose a better and larger battery designed to put out serious current if needed.
Old 8th May 2014
  #4775
Gear Addict
 

Quote:
It's going to be easier to do these things on batteries, and battery life becomes the issue. Though the form factor of the thing is big enough to hold a very nice battery that'll keep up with the requirements: there, the constraint becomes size, and we already know the Pono is a clunky ungainly Toblerone. You get things from that decision, like the ability to choose a better and larger battery designed to put out serious current if needed.
I don't argue that DC will provide cleaner power, and that some of the best audio gear (especially preamps) is powered by batteries. I would doubt the battery in a portable player is capable to supply the analog components with enough juice to do more than a conventional headphone amp. This is speculation on my part, admittedly, but it is a portable player, and who would buy a portable player that gives only one album in audiophile quality before needing a recharge?

I am still particularly interested in why a portable player (which will be used in environments which are typically 30dB or more noisier than studios) would need 24bit resolution, or why any of the high res formats discussed would offer any benefit when 24bit resolution exceeds the dynamic range of hearing. If the background noise was at say 30dB, to hear the lowest dither, the peaks would rise above 170dB. What gear has that kind of headroom, what ear could manage? Even with the best power supply, what kind of headphone amplifier has a S/N ratio to replay a 144db dynamic range? I would love an amplifier like this to record with - imagine a console with such clean preamps!

Considering putting any mix off the master tape on Pono (after all, that is what Neil Young loves in the analog world): the tape would still have a higher noise floor than 16 bit offers (even if it was unmastered) no matter how it was encoded to PCM (provided correct gain staging). So why the waste of storage, the extra number crunching, when the final mixes still have noise well above the dither, regardless of whether they were squashed in mastering or not?

I can understand the potential for better stereo imaging and an uncramped high frequency response offered by a higher sampling rate to an extent, it just seems to me that the form factor and intended use is at odds with an audiophile concept.

If they have discovered some nifty processing that makes things sound better, wouldn't that simply be software that doesn't really need a proprietary device?
Old 8th May 2014
  #4776
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
Actually, battery power is what you'd want there...
From a design standpoint, I'd much rather deal with power supply filtering than having my power budget constrained by runtime considerations. The PS is just one subcircuit and easily managed, while the power budget affects all your design decisions.

In any case, it won't only run on batteries, so they still have to filter the power supply. From a design perspective, the worst of both worlds. But if it somehow sounds different when running on batteries, then the designers did something wrong.
Old 8th May 2014
  #4777
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by DarkandKurious View Post
I would doubt the battery in a portable player is capable to supply the analog components with enough juice to do more than a conventional headphone amp. This is speculation on my part, admittedly, but it is a portable player, and who would buy a portable player that gives only one album in audiophile quality before needing a recharge?
You can push significant power with big (and expensive) Li-poly batteries. I don't know what kind of power typical headphone amps provide, but a 4800 mAh battery should be able to power 6W from a 5V supply for about 4 hours. I'd imagine that's more than enough to drive most low impedance headphones and all the electronics.

Quote:
I am still particularly interested in why a portable player (which will be used in environments which are typically 30dB or more noisier than studios) would need 24bit resolution, or why any of the high res formats discussed would offer any benefit when 24bit resolution exceeds the dynamic range of hearing.
A fair point. It's hard enough hearing the 16-bit noise floor of a CD without turning up the monitors to dangerous levels. Clearly, for most people in most listening environments, 16 bits provides all the dynamic range we need. Experts tend to agree that 20 bits takes care of all the outliers (which, I imagine, consists mostly of people listening to classical music in anechoic chambers). But computers (and programmers) like things that are aligned on 8-bit boundaries, so we went with 24-bit word lengths. It's overkill, sure, and it costs a byte per sample when compared with 16-bit, but it's the smallest multiple of 8 that outperforms the very best analog DRs.

For the vast majority of listening experiences, there is no audible difference between 16- and 24-bit playback. But 24 bits virtually guarantees enough DR for any audio application, thus it is justifiable as a delivery format standard.

Quote:
I can understand the potential for better stereo imaging and an uncramped high frequency response offered by a higher sampling rate to an extent...
Why would stereo imaging change with higher sample rates?

Quote:
If they have discovered some nifty processing that makes things sound better, wouldn't that simply be software that doesn't really need a proprietary device?
The de facto design goal is for a transparent DAC. Hopefully the entirety of Pono's sonic appeal doesn't rest on their applying the equivalent of a smiley EQ curve on the backend. The audiophiles, for one, will skewer them.
Old 8th May 2014
  #4778
Gear Addict
 

Quote:
Originally Posted by bogosort View Post
Why would stereo imaging change with higher sample rates?
My understanding is that ringing from the reconstruction filters at 192 has a perceptual effect on transients that can alter the perception of the stereo image, while the gentler slope of filters have a better phase response. It is not something I have experienced myself as I am personally quite happy with 88.2 for recording (because I'm miserly with hard drive space). Maybe this is something that was true in early designs and is outdated, or maybe it is just a false understanding I have. My brain starts to hurt when I think about the math on that level.
Old 8th May 2014
  #4779
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by DarkandKurious View Post
My understanding is that ringing from the reconstruction filters at 192 has a perceptual effect on transients that can alter the perception of the stereo image, while the gentler slope of filters have a better phase response. It is not something I have experienced myself as I am personally quite happy with 88.2 for recording (because I'm miserly with hard drive space). Maybe this is something that was true in early designs and is outdated, or maybe it is just a false understanding I have. My brain starts to hurt when I think about the math on that level.
Maybe I'm the confused one, but I don't see what the phase response of a filter has to do with stereo imaging. We can make a digital filter with any phase response we want, including zero phase. But even an IIR filter with nonlinear phase will surely present a coherent stereo image, as long as both channels have the same response (and since the filter is digital, this is guaranteed).

I think you're confusing the "phase" issue of multiple mics on a single source with the phase response of filters, which is an entirely different thing.
Old 8th May 2014
  #4780
j_j
Lives for gear
Quote:
Originally Posted by bogosort View Post
Maybe I'm the confused one, but I don't see what the phase response of a filter has to do with stereo imaging.
There are two issues. One is interchannel matching. In the modern age, there is no excuse for bad interchannel matching.

The other is only going to be relevant if the phase shift is big enough that the group delay of the filters pushes part of the spectrum (this is now in a per-channel setting) so far away from the rest of a transient that it creates a monophonic artifact that can reduce the ability of the ear to detect a transient onset. It can happen, it has to be severe phase shift and rapid change in phase shift with frequency to make it happen, though.
Old 8th May 2014
  #4781
j_j
Lives for gear
Quote:
Originally Posted by DarkandKurious View Post
My understanding is that ringing from the reconstruction filters at 192 has a perceptual effect on transients that can alter the perception of the stereo image, while the gentler slope of filters have a better phase response.
This depends on filter design. If a filter is designed as a symmetric FIR filter, there is no phase shift beyond the constant delay (which is sometimes referred to as linear phase, since phase = 2 pi frequency_in_Hzz delay_in_seconds). Such filters have been alleged to have pre-echo problems. Certainly some early, poor ones have demonstrated such. It's the same issue as pre-echo in a codec, although the problem in a codec is much worse, since the bandwidth of 1 filter output is much narrower than twice the transition band of the reconstruction filter.

Basically, "spread" of a filter's impulse response scales as C/bandwidth, where C is determined by the amount of rolloff at the band edge, and bandwidth is the bandwidth of the filter in Hz. C can be as small as 1/2 for very small rolloff of a perfectly optimum gaussian filter. In practice it's more like something under 10.

Now, IIR filters can not be symmetric (that would require infinite delay before output starts, and unstable filter coefficients that would require infinite resolution in calculating the output, which are both kind of fatal to the idea), and will have phase shift that is not pure delay. This can have a perceived effect monophonically, as well.

So, you take your choice, and test what you designed. Sorry. No real help there, perhaps, except to design and test.

This is why I advocate 64kHz as a sampling rate. It was a proposed standard at one point, and both analog and digital filters can be shown numerically to be well out of the way for a 20kHz bandwidth. I must also add that nobody has ever proven the need for this with a listening test.
Old 8th May 2014
  #4782
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by j_j View Post
I must also add that nobody has ever proven the need for this with a listening test.
As a scientist, please be accurate and add "that I am aware of" to that statement.

Thank you.
Old 9th May 2014
  #4783
Quote:
Originally Posted by Sounds Great View Post
As a scientist, please be accurate and add "that I am aware of" to that statement.

Thank you.
I would hazard to wager that j_j is more up to date on his scientific journal reading than you are.
Old 9th May 2014
  #4784
Lives for gear
 
doom64's Avatar
Not sure if this has been posted yet. It's a somewhat scientific analysis of why Pono/HD Tracks is a waste for the human ear.



The guy in the video even shows the difference between an HDTracks release and the CD version. The CD version is smashed and the HDTrack version isn't as smashed.
Old 9th May 2014
  #4785
Lives for gear
There are some interesting discussions in here, and I'm certainly learning a lot.

One thing I think is important is that sound is perceptual and listening is a behavior that can be learned.

For example..
We eventually hear the difference between 2 good preamps, one pricey/one cheap, even though they may measure similarly within the realm of "limits of hearing parameters".. we shouldn't theoretically be able to perceive a difference if they are both delivering clean, flat, undistorted signal. Yet we eventually do "hear" the undesirables in a cheaper preamp.

So why would bit rate and sample rate be any different?
Couldn't we eventually learn to hear the flaws in a lessor format?

24 bit to my ears is very much a cleaner experience. 16 bit sounds good to me, as the noise blends with sounds I usually prefer to sound distorted anyway: cymbals/guitar pick noise/sibalance. 16 bit deficiencies seem undetectable... Until you hear the 24 bit version.

16 bit noise is not exactly like tape hiss that sits at the noise floor. It's a noise that's only present when something else is making sound. It also is a "shhhh" kinda sound, so it's up there above 8kHz. I guess it's quantization noise???
Not aliasing noise from sample rate..

But, I'm pretty optimistic that there is a way we will all learn to hear the difference in sample rates once we get truly used to not having to sacrifice sound for the delivery medium.

My first good computer monitor could do 1280 x 900 (or something like that). I remember it feeling like virtual reality. I still have it, and after using higher resolution screens all the time, to look at that screen it just looks BAD

...still I wonder, and not trying to sound mean or anything.. how does someone with failing eyesight feel about blueray movies vs DVD movies ??? Maybe they can't tell the difference? Maybe they think Bluray is just a ploy to sell the same titles over and over???
Old 9th May 2014
  #4786
Lives for gear
 
doom64's Avatar
Quote:
Originally Posted by 21doors View Post
So why would bit rate and sample rate be any different?
Couldn't we eventually learn to hear the flaws in a lessor format?
The human ear hears a finite range. Go to an ear, nose and throat doctor and get a hearing test done to find out what your's is.

Quote:
Originally Posted by 21doors View Post
24 bit to my ears is very much a cleaner experience. 16 bit sounds good to me, as the noise blends with sounds I usually prefer to sound distorted anyway: cymbals/guitar pick noise/sibalance. 16 bit deficiencies seem undetectable... Until you hear the 24 bit version.

16 bit noise is not exactly like tape hiss that sits at the noise floor. It's a noise that's only present when something else is making sound. It also is a "shhhh" kinda sound, so it's up there above 8kHz. I guess it's quantization noise???
Not aliasing noise from sample rate.
16 bit noise resides around -96 dBFS. You probably can't hear it...you'll probably hear the noise from your speaker amp before you hear quantization noise.

Quote:
Originally Posted by 21doors View Post
My first good computer monitor could do 1280 x 900 (or something like that). I remember it feeling like virtual reality. I still have it, and after using higher resolution screens all the time, to look at that screen it just looks BAD

...still I wonder, and not trying to sound mean or anything.. how does someone with failing eyesight feel about blueray movies vs DVD movies ??? Maybe they can't tell the difference? Maybe they think Bluray is just a ploy to sell the same titles over and over???
I have bad eyes...but eyeglasses help LOL. I can see a clear difference between DVD and Blu-ray, assuming it's produced properly. There are diminishing returns, however. 720p high def is fine on smaller TVs. Past about 42" 1080p becomes necessary. The new kid in town is 4k (Ultra HD) and that only makes sense if you watch TV from 3 feet away on an 80" screen. Most consumers have 65" TVs or smaller so 4k makes zero sense for home viewing. More reading:

When does 4K Matter?
Old 9th May 2014
  #4787
j_j
Lives for gear
Quote:
Originally Posted by 21doors View Post
For example..
We eventually hear the difference between 2 good preamps, one pricey/one cheap, even though they may measure similarly within the realm of "limits of hearing parameters".. we shouldn't theoretically be able to perceive a difference if they are both delivering clean, flat, undistorted signal. Yet we eventually do "hear" the undesirables in a cheaper preamp.

What measurements would you propose to measure "clean, flat, undistorted"?

It is entirely possible to take the usual variety of "specs" and do all sorts of useless things that will sound different. So ...
Old 9th May 2014
  #4788
j_j
Lives for gear
Quote:
Originally Posted by 21doors View Post
16 bit noise is not exactly like tape hiss that sits at the noise floor. It's a noise that's only present when something else is making sound.
This is false. If the noise floor on a digital system is only there when something else is happening, the digital system is not dithered.

Dither is a constant, fixed level of noise.

It's not EXACTLY like tape hiss, because it's flat with frequency, and tape hiss isn't, but it is constant. If it's not, your system is broken, get it fixed, or get a new one.
Old 9th May 2014
  #4789
Lives for gear
Quote:
Originally Posted by j_j View Post
If you can't demonstrate something, you can't test it, so such claims are not within the realm of science, and must lie within the realm of personal faith. Personal faith can not be more than personal, no matter how many people share such personal belief.
Right, so before the microscope was invented you couldn't demonstrate that microbes actually existed and that was a matter of personal faith. But after the invention of the microscope, they became a matter of science. There are many examples like this throughout scientific history where there has been progression of something that could not initially be demonstrated (but was suspected), which then became part of the realm of science. Is audio a different type of science where something like this could never happen?
Old 9th May 2014
  #4790
Lives for gear
Quote:
Originally Posted by j_j View Post

It's not EXACTLY like tape hiss, because it's flat with frequency, and tape hiss isn't, but it is constant. If it's not, your system is broken, get it fixed, or get a new one.
It very well may be a problem outside the bit rate/sample rate realm, so I shouldn't jump to conclusions, and I should tell myself to look for reasonable explanations. I don't understand the deep working of conversion nd you would know this of course..

The noise I perceive is the same as that easily heard in 8 bit recordings, and is the same noise heard in 12bit things like old processors from the 80s.. The only difference is that the noise in 12 bit is quieter than 8 bit.. As for non-constant noise: Maybe a gating system of sometype is employed, I haven't the slightest, why the noise isn't constant like tape???
Move up to 16bit and it seems to be there, just like 8 and 12, only of course much lower..but it's there.
24 bit I can't detect any trace..

It would take NASA a few days to narrow down every variable in my setup for sure, and I don't want to get that deep, haha.

But I would not simply jump to blame converters and software as being broken and the reason for this sound. Simply for the reason that most music is going to be listened to on consumer devices, which will also display the same traits, of which so far seem to agree with what I'm hearing.

If it's due to my production techniques, well then maybe I can change my gain staging or something so that 24 and 16 sound more alike.

But when I listen 8 bit and 12 bit, it just seems like the same sound, so the conclusion is pointing in my mind at bit rate..
Old 9th May 2014
  #4791
j_j
Lives for gear
Quote:
Originally Posted by 21doors View Post
But I would not simply jump to blame converters and software as being broken and the reason for this sound. Simply for the reason that most music is going to be listened to on consumer devices, which will also display the same traits, of which so far seem to agree with what I'm hearing.
Unless you're doing heavy EQ and boosting stuff captured in 16 bit by many dB, you really should not hear the noise floor as anything more than a constant, very low level hiss.

But you should always process in at least 24 bit if not 32float. Dunno what your equipment does.

I have encountered quite some equipment where the inventors/designers/whatever did not quite understand digital signal processing, so it's hard to know offhand.
Old 9th May 2014
  #4792
Gear Head
 
bandpass's Avatar
 

Apparently, the Pono demo tracks were from HDtracks:

Quote:
Originally Posted by Lisa Hershfield, operations manager at HDtracks
Those musicians listening to Pono in Neil Young’s car were actually listening to recordings from us.
Old 9th May 2014
  #4793
Gear Addict
 

Quote:
Originally Posted by MarsBot View Post
Right, so before the microscope was invented you couldn't demonstrate that microbes actually existed and that was a matter of personal faith. But after the invention of the microscope, they became a matter of science. There are many examples like this throughout scientific history where there has been progression of something that could not initially be demonstrated (but was suspected), which then became part of the realm of science. Is audio a different type of science where something like this could never happen?
But....even after the microscope you can't see microbes with your eyes so nothing changed and putting microbes into TV screens or whatever makes no difference. Like someone said early in this thread...if we find a way to more accurately measure the speed of sound..or light for that matter, that does not change the speed of light.
Old 9th May 2014
  #4794
Gear Addict
 

Quote:
Originally Posted by doom64 View Post
The human ear hears a finite range. Go to an ear, nose and throat doctor and get a hearing test done to find out what your's is.



16 bit noise resides around -96 dBFS. You probably can't hear it...you'll probably hear the noise from your speaker amp before you hear quantization noise.



I have bad eyes...but eyeglasses help LOL. I can see a clear difference between DVD and Blu-ray, assuming it's produced properly. There are diminishing returns, however. 720p high def is fine on smaller TVs. Past about 42" 1080p becomes necessary. The new kid in town is 4k (Ultra HD) and that only makes sense if you watch TV from 3 feet away on an 80" screen. Most consumers have 65" TVs or smaller so 4k makes zero sense for home viewing. More reading:

When does 4K Matter?
4K matters to me because I got a hi def 39" computer monitor for $500!
Old 9th May 2014
  #4795
j_j
Lives for gear
Quote:
Originally Posted by blindjoni View Post
But....even after the microscope you can't see microbes with your eyes so nothing changed and putting microbes into TV screens or whatever makes no difference. Like someone said early in this thread...if we find a way to more accurately measure the speed of sound..or light for that matter, that does not change the speed of light.

I think the bot putting out comments in regard to using tools to find microbes was just trying to create controversy, really. The whole idea is just so laughably silly on the face of it that it can't be offered seriously. I mean, are we going to hear how the ear has a spatial sensor like the eye, now? Or what?

It's worse than turning up the level by 70dB and then pointing out you can hear a difference in the noise floor between 16 and 24 bit PCM.

Yes, you can. Does it apply to usage? No?

I've replied to people using completely inappropriate visual analogies in another thread in mastering. Visual analogies do not work in audio, the ear and eye are very, very different.
Old 9th May 2014
  #4796
Gear Addict
 

Quote:
Originally Posted by bandpass View Post
Apparently, the Pono demo tracks were from HDtracks:
Which begs the question again: If the 192k files are the ultimate, how were the 96k and 44.1k files they claim sound underwater made? If HDTracks has 192 and 96 and 44.1, it's very likely the album was mastered at 192, and then (if you believe the 2010 AES listening test paper) downsampled to create the 96 and 44.1. ("standard practice" in mastering, they said). So this is most likely what they're listening to in the car, and possibly the basis for their whole underwater scenario. According to the AES paper, the testers were able to reliably tell the downsample (Pyramix SRC) from the original in blind tests. If this is the case, their whole underwater story is based on the quality and capability of a sampling rate converter, not the capability of 192 vs 96 vs 44.1 per se. It's possible they found an album that was not mastered in this way, but it's more likely it's as I've just described it.
Old 9th May 2014
  #4797
Lives for gear
Quote:
Originally Posted by j_j View Post
I think the bot putting out comments in regard to using tools to find microbes was just trying to create controversy, really. The whole idea is just so laughably silly on the face of it that it can't be offered seriously. I mean, are we going to hear how the ear has a spatial sensor like the eye, now? Or what?

It's worse than turning up the level by 70dB and then pointing out you can hear a difference in the noise floor between 16 and 24 bit PCM.

Yes, you can. Does it apply to usage? No?
I've been thinking about this, and maybe it has something to do with this:

If I blend a reverb send with a vocal that is peaking around -1dBfs, I usually start to detect the reverb around -28dBfs or so.. So I might back it down to -30dB and still know it's gone if I turn it off. (The sound that's affected is sometimes in the lows, where the dry signal is loudest! --[I always low pass verbs like crazy] )
Ok.. That was just to point out that we can hear stuff that you would think should be totally masked.

So it takes an open mind to consider this... But here it goes:

When I look at many CD albums with a spectrum analyzer, the highs are rolled off above 12kHz.
On many albums, 12kHz peaks are never above -50dBfs.. And 15khz peaks are usually lower of a roll off.
Can we hear 15kHz? Many can, some will have trouble.
I'm going to guess that maybe the noise I'm hearing it really is just the noise floor messing with the sound up there. If the noise floor isn't already much lower than the signal, then maybe during bit reduction, something goes wrong...

Let's say the harmonics up there in the 24bit file are sitting +10 above the 16 bit noise floor during peaks (-86 dBfs). The attack and decay of a cymbals harmonics might sound ok in 24bit. Could the noisefloor maybe become one with the signal during down-conversion, totally messing up the original harmonics???
Then as those cymbal crashes peak up through the noise floor, they are now butchered versions of their 24bit former selves??? Could the attack and decay get messed up since it's so close to the noise floor?
Old 9th May 2014
  #4798
Gear Maniac
 
Decompress's Avatar
 

Quote:
Originally Posted by j_j View Post
There are many things we do not know. There are some we do (always with the possibility of refinement). There are many things we can disprove (ditto).

This does not explain either Pono or DSD, nor does it provide any kind of excuse for much of the mistaken claims here, which are in either the "can not be demonstrated" or "can not be tested' catagory. If you can't demonstrate something, you can't test it, so such claims are not within the realm of science, and must lie within the realm of personal faith. Personal faith can not be more than personal, no matter how many people share such personal belief.
That's funny, because the people who were exercising their "personal faith" and complaining about what turned out to be the sound of jitter, back before jitter could be demonstrated, tested, or was even known to exist, had their shared "personal belief" proven correct once science caught up with what their ears had been hearing all along.
Old 9th May 2014
  #4799
Lives for gear
 
Ephi82's Avatar
 

Just a little perspective on PONO's chances in the marketplace:

Consumers have bought so many Beats headphones by Dr. Dre that Apple announced that they will be buying the company for, hold on, $3.2 Billion

Interesting no?
Old 9th May 2014
  #4800
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by 21doors View Post
When I look at many CD albums with a spectrum analyzer, the highs are rolled off above 12kHz.
On many albums, 12kHz peaks are never above -50dBfs.. And 15khz peaks are usually lower of a roll off.
Presumably that's how the albums were recorded and has nothing to do with CD or 44.1 kHz sampling. A 192 kHz version of the same albums would show the same roll-off. Either that or your equipment is seriously broken.

Quote:
Let's say the harmonics up there in the 24bit file are sitting +10 above the 16 bit noise floor during peaks (-86 dBfs). The attack and decay of a cymbals harmonics might sound ok in 24bit. Could the noisefloor maybe become one with the signal during down-conversion, totally messing up the original harmonics???
Then as those cymbal crashes peak up through the noise floor, they are now butchered versions of their 24bit former selves??? Could the attack and decay get messed up since it's so close to the noise floor?
You can easily test your theory: set up a two-channel 24-bit mix, with a looped cymbal crash on one channel and white noise from a signal generator on the other. Set the noise to -90 dB and then listen to the mix, muting the noise noise on and off, listening for differences. If the cymbal volume is reasonable, you shouldn't hear a thing.

This represents a properly dithered 24-bit to 16-bit reduction. If you raise the noise to -60 dB, you're emulating a reduction to 10 bits. It's noisy as hell, but you still clearly hear the cymbal: its attack is clear and its decay should sound like a smooth fade into the noise floor.

Uncorrelated noise -- the result of a properly dithered quantization -- is the least offensive thing you can add to a signal. This is because we live in a universe full of uncorrelated noise and our brain-ear system is used to ignoring it. It doesn't "mess up harmonics" or the signal envelope, even at surprisingly high volumes. If it did, we wouldn't be able to talk to each other in crowded places!

Bottom line: if your 16-bit mixes sound worse than your 24-bit mixes, you're doing it wrong.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump