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 24th April 2014
  #3751
Quote:
Originally Posted by theblue1 View Post
Quoting without comment?
Yes, quoting without comment...this is the post:
Quote:
Quote:
...The rhythmic tap of a tambourine. The subtle resonance of an acoustic guitar string between chords. It's these finer nuances that pop out, unencumbered by the digital compression. Even if they're audible on the MP3 version of the song, they're suddenly more noticeable. The details are crisp.

Picking these things out is also made easier by the fact that the mix of the song itself feels--at least in my brief testing--more spacious. Almost three-dimensional. Each instrument has more room to breathe. Some have commented that it sounds like you're standing in the recording studio with the musicians. That's a fair way to describe it...

...But Pono does sound different. It surfaces new things to the listener. As many have pointed out, the sound is "warm," not unlike the analog sound of high-quality vinyl. The results will undoubtedly vary from album to album and speaker to speaker, but on the whole it does sound fuller and more pure than the audio files we're used to.
Inside Neil Young's Crowdfunded Quest To Fix Your Disastrous MP3 Situation ? Co.Labs ? code + community
See, no comment.

Quote:
Originally Posted by theblue1 View Post
...You said flat out that it sounded 'great' and when it was pointed out you hadn't heard it, you cited the sole person in this thread who had had a chance to hear one of the prototypes.

That's not exactly quoting without comment. That's accepting second hand, informal observation in a marketing context as veritable fact.

Seems to me.
[bold added] link to post
I think you should have linked to the thread to read my post in context of the discussion: https://www.gearslutz.com/board/so-mu...-pono-109.html

The discussion involved ellisr63, Don Hills and Karloff: I've never claimed I've heard Pono or that I think it is outstanding...ellisr63 said that Pono sounded outstanding and I referred to that as the main point of interest (to me) in a discussion about his post. Go and check.

I've already responded to you once about this in an earlier post; perhaps you missed my reply?
Old 24th April 2014
  #3752
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by theblue1 View Post
Wow...

It is both hilarious and tremendously sad that Harvest has been released twice in versions with some tracks compressed down to an average dynamic range of 9 dB.

Weirder still? The recent HDCD version has 2 dB LESS dynamic range on its most squashed track than earlier releases... oops!

Clean up your own kitchen, first, Neil.
In fairness, Neil might not be to blame. It could just be a record label decision.

Alistair
Old 24th April 2014
  #3753
Quote:
Originally Posted by 1070 View Post
its a really cool thread... people have their opinions! the usuall consumer dont care at all... they never did! great that pono is targeting that way and its defently a good thing for people who actually care.
if you listen to 40+ man classical recording, with the right speakers there is good market as well for serveral other high end audio people!
regarding the masses, nobody will ever care... they listen to youtube on their smartphone, spotify or whatever will come up... and that will defently never change in any way!
there is a difference between 192 and 44.1 but a 0815 pop record or urban record, made ITB or on a mpc you wont hear any diffrence... using a lot of mics in a big hall, sure there will be a difference!
i enjoy reading this thread a lot! go ahead and argue it out but at the end of the day it does not matter at all in my opinion.
I've heard over 160 live, unamplified orchestral concerts. I first began considering myself an 'audiophile' about 50 years ago after I got seriously interested in sound putting together my first stereo rig.

I'm afraid I don't really buy your reasoning. As a classical music consumer -- or as a reasonably well-informed recordist.

But, anyway, there's nothing new in the Pono push -- these formats, needed or not, have been in place and in the market for some time now. (Double rate and DSD discs have already failed to find a significant share of market after well over a decade of attempts to establish them as the next big things.)

Quad rate playback devices have been around for a number of years, and there are already quad rate portable devices on the market at price points above the Pono and even a little less (depending on features/retail/discount).
Old 24th April 2014
  #3754
Lives for gear
 
Arksun's Avatar
Quote:
Originally Posted by UnderTow View Post
You can also remove previous filter ringing by filtering lower and less steeply. That is the principle behind the so called apodizing filter: They set the filter lower to remove stuff but they do that by filtering out part of the audible band! Luckily most audiophile product purchasers (due to their age) can't hear high enough to be bothered by this.

Alistair
This is also the reason why when using a sample rate converter I prefer ones which allow a less steep filter at a lower freq point, because I can hear the difference of the impulse ringing and its impact on the audible range. Voxengos R8Brain Pro is a good example, where I prefer the sound of its conversion with the ultra steep mode set to OFF when its set to linear phase mode. The tradeoff of losing a little more of the high frequency content (which I can't hear anyway) is more than worth it.
Old 24th April 2014
  #3755
Quote:
Originally Posted by UnderTow View Post
In fairness, Neil might not be to blame. It could just be a record label decision.

Alistair
Is this not one of the albums they've been using to demo the Pono device?
Old 24th April 2014
  #3756
Lives for gear
 
Dpro's Avatar
Of course they are. They are trying to get back into the portable audio player market. It did use to be their ballpark and playing field till Apple all but ran it through with a sword and took it away.
Sony is not in good shape as a consumer electronics company. They are technically in critical condition in intensive care. This is them grasping at whatever they can.
Sorry but you just have to look beyond what people say in marketing.

Quote:
Originally Posted by Sounds Great View Post


Well there it is, a (scientific?) study that goes in the face of all the nay-Sayers here that say the public doesn't care about audio sound quality.



Well there you have it, as many (from both sides) have said, even if Pono is not all it's cracked up to be is got the talk about audio quality going for a lot of people, this is a good thing any way you look at it.
I don't think most here are saying the public possibly does not care about higher quality audio. They are saying in a lot of situations, cars, cheap ear buds,subways etc... People will not be able to hear the possible better playback of hi res audio given mitigating circumstances. In that sense they might wind up not caring.

Again you also confuse the fact of the argument here. Yes a lot of us feel hi res audio beyond 24/60k is not needed for great sound reproduction. We base that on proven science and tests. Then the Pono side says you can't have quality audio with out super hi res. That has not been proven with any science or actual proper resolution comparison tests. Notice I said proper? Its is rather important to make valid claims.

Neil and Pono drew a line in the sand with their comments and silly marketing. Had they not drawn that line? There would be no real argument going on here. That line is what is going on here.
While you may think outlandish claims that are false is good because it has people talking? I see a great big problem rearing it's head.

If they public decides they have been sold a bill of goods by Neil, the rock stars and the labels yet again? Because they discover they really don't need 192k and Pono to have great audio playback.
Well guess who gets the bum rap? The whole audio industry from the artists all the way to the labels.

I will say this again I am not against higher than MP3 resolution audio for the consumer. I am not against Pono per se. I am against BS marketing that makes unsubstantiated claims. Claims that in the long run confuse the issue more than correct it. Claims that have not shown the science to back themselves up with.

Wow I actually allowed myself to get sucked back into this.
Old 24th April 2014
  #3757
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by UnderTow View Post
This is about ADCs. It is the typical oversampling approach used by modern ADCs. You sample at very high SR with a few bits and then decimate down to lower sample rates with more bits. No magic going on there. The Pono on the other hand is a player. You can't add any resolution that wasn't already there.
Again, you can't extract more resolution than has already been captured. Whether that is done with more bits to start with or more bandwidth doesn't really matter. (That is an engineering choice more than anything else). But once the signal is captured, you can't add resolution that wasn't there to start with.
I don't know where to go from here. It is exactly the same technique as used by the ADCs, but on the playback end. You could continue onward until you had very, very high resolution over the range of 0 to 20 hz, if you wished, and you could calculate exactly what the effective output word length would have to be to represent it. It is EXACTLY the same technique used in the ADCs, when they as you put it sample at high sample rate with a few bits and then decimate down to lower sample rates with more bits.

This is exactly, exactly the same technique. It's just being used without the final decimation stage that would give you very good 44.1K or 48K sampling with an extended noise floor and more word length. Instead, they're giving you the bandlimited output delivered at 96K or 192K, which will incur a noise floor cost but give you the added wordlength. It is trading sampling rate for word length just like modern ADC designs. The high frequencies beyond 22K or so are gone, averaged out, period. They went into the resolution of the audible band, exactly as the 1 Mhz frequencies of modern sampling and decimation are gone because they went into the audible band's resolution.

This is what happens. It's very simple, so simple that I'm completely at a loss for how to spell it out more clearly. You can do that decimation at ANY stage. You could give consumers the original data stream from the first stage of capture and their players could decimate it down to 44.1 if they wished.

How is this difficult? They're even TELLING us this is what they are doing.
Old 24th April 2014
  #3758
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by chrisj View Post
You could continue onward until you had very, very high resolution over the range of 0 to 20 hz, if you wished, and you could calculate exactly what the effective output word length would have to be to represent it. It is EXACTLY the same technique used in the ADCs, when they as you put it sample at high sample rate with a few bits and then decimate down to lower sample rates with more bits.

This is exactly, exactly the same technique. It's just being used without the final decimation stage that would give you very good 44.1K or 48K sampling with an extended noise floor and more word length. Instead, they're giving you the bandlimited output delivered at 96K or 192K, which will incur a noise floor cost but give you the added wordlength. It is trading sampling rate for word length just like modern ADC designs. The high frequencies beyond 22K or so are gone, averaged out, period. They went into the resolution of the audible band, exactly as the 1 Mhz frequencies of modern sampling and decimation are gone because they went into the audible band's resolution.

This is what happens. It's very simple, so simple that I'm completely at a loss for how to spell it out more clearly. You can do that decimation at ANY stage. You could give consumers the original data stream from the first stage of capture and their players could decimate it down to 44.1 if they wished.
Clearly the topic is beyond your ability to comprehend. I give up trying to educate you.

Quote:
How is this difficult? They're even TELLING us this is what they are doing.
If you send me $1000 every month, I will make all your recordings sound better at a distance. Try it. Really. I am TELLING you that is what I will do. Trust me.

Alistair
Old 24th April 2014
  #3759
Lives for gear
 
Dpro's Avatar
Quote:
Originally Posted by chrisj View Post
I don't know where to go from here.... Snip happens

How is this difficult? They're even TELLING us this is what they are doing.
No Chris, Ayre is telling us what they are doing on their web page. Pono as far as I have seen is just using terrible metaphors and anecdotal evidence to try and make their point. Unless they have decided to at least show some transparency in their claims and post the information on the Ayre web page on Pono's.


Where do we go from here?

The rub here Chris is mainly BS marketing with no proof to substantiate. Like JJ said apodizing filters are not new and according to his own research they are poor. Do current his res audio converters sound worse than ones with apodizing filters utilizing 192k? Well in short I don't know because Ayre sure hasn't put up any audio tests to show it.
Years of Pro Audio conversion technology has not rushed to utilize it though.

In fact with that said, I rather like the sound my Apogee converters put out. Everyone has an opinion on this as well as some prefer Lavry,Burl, Prism etc... But the strange thing is none use an apodizing filter. Why Chris, is there something the whole converter industry has been missing?

I mean granted I am somewhat skeptical yet intrigued only for sake that because of personal family history, I know their could be an outside possibility. Yet that just makes scream louder for scientific analysis and data with examples.

Having a scientist for a father has perhaps breed that into me. He would expect nothing less.
Till it's proven beyond a shadow of a doubt? I see no reason for it.


People can chastise me for that but I the end it's the only thing that will settle this.

Last edited by Dpro; 24th April 2014 at 08:22 PM.. Reason: Grammatical corrections.
Old 24th April 2014
  #3760
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
I've explained this already. Real data at 192K means you can run a moving-average filter (for good or ill, that's what they say they're doing). It's much like if you upsampled to 192K to do that, except it's real data. Bits are cheap, why not? Then, you're filtering it back to around 20K again, but getting better resolution in the audio band, and suppressing noise in the audio band through this averaging.
You're being (intentionally?) disingenuous when you say that a 4-sample moving average filter has "better resolution" and suppresses noise in the audio band. The averaging indeed reduces noise, but that's it: you gain 3 dB of SNR; this is the extent of the "better resolution". But this slightly increased noise performance comes at the expense of a terrible frequency domain response.

JJ has already posted the Bode plot for a moving average filter. At 192 kHz, a 4-sample moving average is down -3 dB at 24 kHz, but the roll-off is so gradual that it's attenuating almost all the way to DC. In other words, a 4-sample moving average at 192 kHz actually reduces high frequencies in the audio band!

I don't think that's what you meant by "better resolution".
Old 24th April 2014
  #3761
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
"Combining 256 consecutive 20-bit samples can increase the signal-to-noise ratio at the voltage level by a factor of 16 (the square root of the number of samples averaged), adding 4 bits to the resolution and producing a single sample with 24-bit resolution."
And averaging 4 samples gives you 3 dB, or the equivalent of half a bit.

Quote:
Still: Pono gets as much as 6 db increased resolution over regular 24-bit playback through averaging the oversampled information.
SNR measures power spectra, so 3 dB not 6 dB.

Quote:
This is on purpose, it's in the audible band, and it's noticable from all accounts of hearing the actual device.
I somehow doubt that the people in Neil's car noticed the extra half-bit of noise performance.
Old 24th April 2014
  #3762
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by bogosort View Post
You're being (intentionally?) disingenuous when you say that a 4-sample moving average filter has "better resolution" and suppresses noise in the audio band. The averaging indeed reduces noise, but that's it: you gain 3 dB of SNR; this is the extent of the "better resolution". But this slightly increased noise performance comes at the expense of a terrible frequency domain response.

JJ has already posted the Bode plot for a moving average filter. At 192 kHz, a 4-sample moving average is down -3 dB at 24 kHz, but the roll-off is so gradual that it's attenuating almost all the way to DC. In other words, a 4-sample moving average at 192 kHz actually reduces high frequencies in the audio band!
That is exactly what I'm saying. The slightly increased noise performance wouldn't be there if not for the averaging, which is the same technique used with oversampling. And the response in the frequency domain is only 'terrible' if you're literally comparing its performance at radically supersonic frequencies–which Monty (xiphmont) swears will ruin amplifiers anyway and cause them to generate intermodulation distortion–with the same sampling rate converted normally, at no gain to the noise performance.

Guys, make up your minds. Do we care about listening to frequencies higher than 30K or not? I suggest not. And incredibly gradual roll-offs in the audio band as they approach cutoff are what we want, when the alternative is a ringy mess! Again we are hopping mad over tiny and gradual behaviors of the filters in areas where they're not going to injure the sound or even be noticeable as such. The audio band roll-off of a 4-sample (and they may be using 3) moving average at 192K is not going to be ruining the sound, full stop. I think it fairly likely that it would be preferred by a lot of listeners to a perfectly flat passband presented at 44.1K… that rings at 22.050K.
Old 24th April 2014
  #3763
Lives for gear
 

3db of noise isn't going to ruin the sound either, and surely any alteration of the frequency response within the audio band is a bad thing?

Chris
Old 24th April 2014
  #3764
Lives for gear
 

Double-bubble
Old 24th April 2014
  #3765
Lives for gear
 
paul brown's Avatar
Quote:
Originally Posted by Sounds Great View Post

Well there it is, a (scientific?) study that goes in the face of all the nay-Sayers here that say the public doesn't care about audio sound quality.

i'm trying to track down the study. i've got here. i'll cherry pick a bit. it seems to be the in-thing with surveys:

The CEA study, Notions of Quality: Audio Expectations of Consumers, also found consumers have varying opinions on which content formats provide the best sound quality. For example, 43 percent of consumers consider FM radio to provide superior sound quality while a similar number (46 percent) consider the audio from HDTVs to be high quality.

“Sound quality depends on both the audio source and listening device, and. manufacturers and retailers have the unique ability to educate consumers on what constitutes high-quality audio,” said Ben Arnold, CEA’s senior research analyst.


it seems, according to the CEA (and let us not forget who the CEA serves. hint: it'll be who pays them), that the listening public needs to be educated. we need to be educated as to what our audio expectations are. warning siren...where have we heard the term expectation before? e x p e c t a t i o n bias!

attention, need, satisfaction, visualisation, action. sales and marketing 101 a la dale carnegie. add more recent psychological and neurological scientific understanding. simmer...

sell the back catalogue again and a funky player. call it pono, call it HDtracks, call it digital vinyl!

i'm passionate about music. maybe obsessive. i'm asking for dynamic range back. i know that it is in the original recordings from the desk. in the audible range. no need to invent 'science' to support it.
Old 24th April 2014
  #3766
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
And the response in the frequency domain is only 'terrible' if you're literally comparing its performance at radically supersonic frequencies. . .
Did you not read what I said? The frequency response is rolling off in the audio band. You want a 4-sample moving average and a flat passband? Can't be done.

Quote:
And incredibly gradual roll-offs in the audio band as they approach cutoff are what we want, when the alternative is a ringy mess!
You're very confused. Our ears are much more sensitive to a filter's behavior in the frequency domain than to its behavior in the time domain.

Quote:
Again we are hopping mad over tiny and gradual behaviors of the filters in areas where they're not going to injure the sound or even be noticeable as such.
It's very strange to me that you're touting a 3 dB improvement of SNR (on a 24-bit signal) as being more important than the in-band frequency characteristics of a DAC. Crazy times!

Quote:
The audio band roll-off of a 4-sample (and they may be using 3) moving average at 192K is not going to be ruining the sound, full stop. I think it fairly likely that it would be preferred by a lot of listeners to a perfectly flat passband presented at 44.1K… that rings at 22.050K.
Just so I'm clear: you believe that a filter with 0.1 dB ripple at 22.05 kHz is more damaging to audio than one that's -1 dB at 15 kHz?
Old 24th April 2014
  #3767
Lives for gear
 

Quote:
Originally Posted by chrisj View Post
I think it fairly likely that it would be preferred by a lot of listeners to a perfectly flat passband presented at 44.1K… that rings at 22.050K.
If a HF rolloff is preferred by the listeners then the mastering engineer should have (would have?) implemented one. It is not the job of the playback device to distort the audio signal, quite the opposite.

Chris
Old 24th April 2014
  #3768
Gear Guru
 

Quote:
Originally Posted by Sounds Great View Post
Well there it is, a (scientific?) study that goes in the face of all the nay-Sayers here that say the public doesn't care about audio sound quality.
I don't think even one "nay-sayer" ever said the public "doesn't care" about 'better sound quality'. SOME of them at least probably do care. What people are saying is that increasing the sample rate will not really GIVE them the better sound quality they expect. Certainly not dramatic 'coming up for air after being underwater' sound quality!

Quote:
Well there you have it, as many (from both sides) have said, even if Pono is not all it's cracked up to be is got the talk about audio quality going for a lot of people, this is a good thing any way you look at it.
this rationale is like a guy selling worthless magic "organic" health beans, beans that are the same beans you can already buy in the supermarket, and half of them are grown with chemical fertilizers besides, and charging a PREMIUM price for them, but its "good" because it got a discussion going about healthy food!

Pono is flogging the already twice-failed "high sample rate" product when we all know better amps, speakers, headphones, quieter listening environments, and better, more dynamic MASTERS are what is going to make "sound quality" better. All things we can do without Pono. We are only using 7 dB of the 115 dB range of a CD as it is. All these compression-happy Analogistas crying about the FORMAT as the limiting factor make me laugh.

The "underwater" sample-rate hype from the Pono people is at worst, flat-out baloney, and at best a distraction from the real steps that should be taken long before doubling the sample rate and bit depth can be noticed.

The whole point of Stone Soup, is that in the end, you never really needed the Stone.
Old 24th April 2014
  #3769
Lives for gear
 
paul brown's Avatar
To underscore the high-resolution push on the part of the Big Three recorded media organizations—Sony, Warner, and Universal Music Group (UMG)—Hiatt introduced Jim Belcher, UMG's VP of Global Digital Technology and Production. In a presentation that showed the relative strength of download sales vs. physical media, Belcher clarified that while the U.S. leads the world in both total downloads and preference for downloads over physical media—60% of U.S. sales involve digital downloads—people in Japan, Germany, and France continue to prefer physical product over files.

There may be three main HRA retailers in the US, but more are gearing up to join HDTracks, Super Hi-Rez, and ITrax. Even while acknowledging that currently, "vinyl is selling very well for us," Belcher prognosticated a continued decline in physical products, greater reliance on HRA downloads, and the emergence of uncompressed and high-resolution digital streaming.


could they make it any clearer? download sales, up. cd sales, down. vinyl sales, up. create the 'need' for high-res downloads. say cd quality isn't good enough. liken the high-res to vinyl. just join the dots. give it to the marketing team to run with. pono is a pawn in a business plan. i don't deny the player could well be exceptional at its price-point. that is the gimmick. the visualisation. all the 'action' they'll want to come from the sale of content. they are the real players. who controls the content.
Old 24th April 2014
  #3770
j_j
Lives for gear
Quote:
Originally Posted by chrisj View Post
That is exactly what I'm saying. The slightly increased noise performance wouldn't be there if not for the averaging, which is the same technique used with oversampling. And the response in the frequency domain is only 'terrible' if you're literally comparing its performance at radically supersonic frequencies–which Monty (xiphmont) swears will ruin amplifiers anyway and cause them to generate intermodulation distortion–with the same sampling rate converted normally, at no gain to the noise performance.
There are two choices here:

1) You didn't look at the frequency response plots of a moving average filter that I posted a while ago, or you didn't understand them. Your moving average filter has MORE supersonic content than a proper filter, AND it has terrible rolloff inside the 20-20K band. Clearly audible rolloff.
2) You are continuing to repeat your original assertions, despite the evidence handed to you.

So, first, go back, please, and look at where in frequency the -3dB point is for a moving average filter, and where it is for an apodizing filter. What do we see?

1) The moving average filter has a lot more ultrasonic content.
2) The moving average filter has LESS in-band energy, because it rolls off faster

This makes several things clear:
1) You get a better SNR with the apodizing filter than you do with the moving-average filter.
2) You get better frequency response with the apodizing filter.
3) You get no pre-ringing with either one.

You do, of course, get a small amount of in-band phase shift with the apodizing filter, interestingly not more than 15 degrees per ERB, which is usually inaudible.

So, once again, what's the advantage of this moving average filter you seem stuck on? Nothing, that's what.

Now, as to your claim "same as oversampling". No, that's not true, oversamping devices use, for the most part, constant delay (i.e. symmetric) FILTERS, carefully designed filters with good frequency response. They do not use "moving average" filters, although in fact a well-designed filter is a weighted moving average, which is where you picked up the term, probably. That weighting is the key to this argument, by the way.

Attached see "4ma.jpg" which is, again, for your perusal, a 4-tap moving average filter. Not that in converting 96 to 48, it's 3dB point is 10.6kHz. If you use it for 192 to 48, that means you're down 3dB at 21.2, and down 1dB at 15.4kHz. Note that the ultimate rejection of this filter is a whopping 9dB. It's barely even a filter. For kicks, I calculated an "optimum" filter 4 taps long, it gets an also-whopping 11dB rejection. You get a tiny bit of improvement just from doing weighting even at 4 taps.

But let's consider reality. Under .25 millisecond there is little likelihood of pre-echo, yes? (I'm being too conservative there, but that's ok). Now, we're talking about 192kHz sampling rate, and .25 milliseconds before the center of a constant delay filter, so it's .5 milliseconds in total length.

***SO WE CAN AFFORD A 96 TAP FILTER*** Hello, now you are you starting to catch on? This isn't even an apodizing filter. It's a 96 tap constant-delay filter. What will this filter do, now? Let's see. With .1dB passband ripple (that's potentially much), it's down 95dB at 28kHz (vs. .1dB at 20). And it's still under the ear's established sensitivity. I'd use something a bit longer, and apodize part of it, which is easy, and the problem, if it exists, is done. See the response for the 96 tap filter, attached.

Note, also, the 96 tap filter allows LESS noise than your moving average filter, and so it provides BETTER SNR. Get it? Better.
Attached Thumbnails
Launch of Pono-4ma.jpg   Launch of Pono-96cd.jpg  
Old 24th April 2014
  #3771
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by j_j View Post
There are two choices here:

1) You didn't look at the frequency response plots of a moving average filter that I posted a while ago, or you didn't understand them. Your moving average filter has MORE supersonic content than a proper filter, AND it has terrible rolloff inside the 20-20K band. Clearly audible rolloff.
j_j, I am not the one who put 'moving average filter' in Ayre's promotional materials. I would be perfectly content if they implemented a 96 tap filter as you suggest. I am NOT responsible for the choices Pono makes—and I've suggested that their claims of 'water surface' vs 'lovely fresh air' may mean they're doing two taps at 96K and _three_ at 192K. Seems sneaky, but I wouldn't rule it out, and it would get people believing that 192K sounded 'better' than 96K, as the 96K would sound different. I don't know what they're going to do.

I will continue to suggest that as specified, flawed filter and all, Pono is going to sound good. And I'm really skeptical that anyone is going to hear a db rolloff at 20K or even 18K… and even more skeptical that people can tell the difference between exceptional filtering at actual frequencies of 50 kilohertz, versus bad filtering at 50 kilohertz. Especially when the content being filtered is sourced at 192K in the first place, so the 50K content is (a) legitimately sampled actual audio belonging to the music in question, and (b) typically very low in level anyway BECAUSE legitimate audio content above 40K is hard to come by. We are arguing about the handling of nearly nothing, inaudible nothing that's already at a very low level to start with.

If they do 2 taps at 96K and 3 at 192K, so that the 192K case hasn't got gentle (nobody is suggesting this is a steep slope at these frequencies!) roll-off at 20K, will you guys be more outraged or less? It will be in line with their chart, and you could argue that if they're gonna average in-band samples the 96K case gives them no options but 2 taps and the 192K gives them the choice of 3 or 4…
Old 24th April 2014
  #3772
Lives for gear
 
paul brown's Avatar
@j_j

i'd like to thank you for your continued patience in explaining the science. especially in simple terms underneath your pay-grade! i entered this discussion with all sorts of unfounded notions about filters and frequency rates mixed with confused science. you, and a couple of others, have prodded me to educate myself. you are a teacher. the sort who prompts a student to investigate for himself but also shares. Thank you.

i am quite open about my disdain of the record 'business'. they were getting their hooks into me with this high-res malarkey. distracting me from my dynamic range cries of foul! i'm back on track! if they want music to be more emotional, more appealing, more valued, then give us back dynamic range. the ear is very good at processing it. it seems so obvious.
Old 24th April 2014
  #3773
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by paul brown View Post
it seems, according to the CEA (and let us not forget who the CEA serves. hint: it'll be who pays them), that the listening public needs to be educated. we need to be educated as to what our audio expectations are. warning siren...where have we heard the term expectation before? e x p e c t a t i o n bias!
OMG. The anti-science types will never quit.
Old 24th April 2014
  #3774
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by nuthinupmysleeve View Post
OMG. The anti-science types will never quit.
Wait, did you just say anti-science? I thought the term was anti-scientific?
Old 24th April 2014
  #3775
j_j
Lives for gear
Quote:
Originally Posted by chrisj View Post
j_j, I am not the one who put 'moving average filter' in Ayre's promotional materials.
Ok, if you say so. But you're the one who keeps defending it against obvious problems, and making magical claims for it HERE.

Do you want me to just ask Charles what he does? It may be proprietary, it probably should be proprietary, and I wouldn't necessary encourage him to spill his secret sauce, of course, but whatever he's talking about, it's not likely simple filter of 'n' taps of value 1/n, standing alone, I'm pretty sure. Of course, it could be, because many people might interpret the rolloff positively. The only way we will find out about that is to stuck an allpass signal into something and just measure it.


And, of course, the SNR gain of 3dB for 96 to 48 or 6dB for 192 to 48 is probably not what's happening either, but that's what you're PRESENTING.

That's what you're PRESENTING, and that's what there is here to reply to.
Old 25th April 2014
  #3776
Lives for gear
 
paul brown's Avatar
Quote:
Originally Posted by nuthinupmysleeve View Post
OMG. The anti-science types will never quit.
too oblique or irony?
Old 25th April 2014
  #3777
Quote:
Originally Posted by paul brown View Post

i am quite open about my disdain of the record 'business'.
If you wanna be a stickler for accuracy, stop tarring the entire recording industry with the same brush. It's extremely clear Pono is a marginal project supported by a minority of artists and record companies.
It's like having a bad meal and saying you have a general disdain for catering.
Old 25th April 2014
  #3778
Quote:
Originally Posted by nuthinupmysleeve View Post
Record companies (at least the big ones) are ALSO just big megacorporations. They haven't cared about music for a LONG TIME.
Talking about respecting science - that's just an opinion, and cannot be proved.
Many, many record companies care about music of course. It is their product, their only way of making money. How can you be a dairy and not care about milk? You can't.
Some record companies believe they can make more money by pressuring artists to stick to a winning formula. It's a subjective viewpoint.
Record companies respond to the things people spend money on. If a lot of people watch The Voice, and the winner of The Voice sells a lot of records, then the bigger record companies will want to pursue that strategy.
It's no different from television, or McDonalds, or Apple and Google.
Record companies care about music you feel is less interesting, or artistic, or creatively pure - that's what you really meant to say.
Old 25th April 2014
  #3779
Lives for gear
 

Quote:
Originally Posted by paul brown View Post
@j_j

i'd like to thank you for your continued patience in explaining the science. ...

... if they want music to be more emotional, more appealing, more valued, then give us back dynamic range. ...
@j_j,
Yes, thank you.

@paul,
it's ironic that Chris J used to have a section on his website analysing a large number of classic hits and showing that dynamic range was a key factor in the "long tail" - how well a hit record continued to sell after the initial release.
Old 25th April 2014
  #3780
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by chrisso View Post
Talking about respecting science - that's just an opinion, and cannot be proved.
Yes, I didn't quote a source so you can consider that an opinion, albeit one shared by many.

Quote:
It is their product, their only way of making money. How can you be a dairy and not care about milk?
It's a widget. They care about sales of the widget. The qualities of music that sell are the qualities they care about, not what many would consider quality music.

Quote:
It's no different from television, or McDonalds, or Apple and Google.
Um... why would you assume differently? Big companies don't stay in business building stuff that doesn't sell.
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