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 10th May 2014
  #4831
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by j_j View Post
I'm aware of several players (none sold now, but who knows about ones sold nowdays, I don't) that had the DAC reference coming off the V+ with NO FILTERING. Guess what AM you could see from the DAC output? Could it be the servo operating? Yep. It could be. :(

Just for an example. Different discs might indeed sound different, but not due to the data on the disc.
True, with a broken player all bets are off.
Old 10th May 2014
  #4832
j_j
Lives for gear
Quote:
Originally Posted by bogosort View Post
Pointing out that science doesn't know everything is not an argument; it is meaningless at best, and completely dismissive of everything we do know, at worst.
This is just the old creationist meme aimed at science in another field, basically, it's called the "argument from ignorance", that attempts to enforce a false dichotomy that goes "since we don't know everything, we don't know anything".

For instance, we know that "digital edges" on audio are the stuff of a desperate myth. We know that the human ear can't hear anything more than 5.5dB below a noise floor in an ERB. We know that the actual noise floor due to the fact air is made of discrete molecules is about 2-3 dB below the actual unimpaired threshold of hearing for a person with an average sized eardrum.

We also know that the 20-20K white noise due to atmospheric brownian motion is about 6dB SPL, give or take, at the ear drum, but that's not what matters, what matters is the level in a given band at the ear.

We know a lot of things.

We also can disprove things when they are shown to produce contradictions. So we also know that many, many things are wrong.

Basically the argumetum ad ignorantum is nothing more or less than a way to deny all knowledge, and keep mankind in a state of myth and ignorance. Personally, I find that argument not only fallacious (which it is) but also one of the worst false memes ever developed by mankind.
Old 10th May 2014
  #4833
Motown legend
 
Bob Olhsson's Avatar
 

Quote:
Originally Posted by bogosort View Post
True, with a broken player all bets are off.
It's amazing how often people assume that software and gear isn't faulty and then attack science over bad sound.
Old 10th May 2014
  #4834
j_j
Lives for gear
Quote:
Originally Posted by Bob Olhsson View Post
It's amazing how often people assume that software and gear isn't faulty and then attack science over bad sound.
It's also kind of amazing that people do not recognize bad equipment any longer. It used to be that funny stuff automatically caused people to check their equipment.

Now some of them complain about "digital audio" because people who don't understand it have taught them a whole load of mystical bull poop.
Old 10th May 2014
  #4835
Problem is that faulty design has been shown to yield outstanding results.
I think many musicians and studio personnel have taken that on board over the years, especially since the internet.
Since the Velvet Underground and later Punk, perfect musicianship is no longer as highly valued.
Since the digital explosion of the 80's and the rediscovering of bypassed analogue, a lot of people value imperfection over perfection.
Probably a majority of digital and virtual products are attempting to mimic flaws in older technology.
Noise, wow and flutter, tape hiss, badly scaled and out of tune oscillators, distortion, saturation, 'humanize' features on digital sequencers and drum machines.
Old 10th May 2014
  #4836
Motown legend
 
Bob Olhsson's Avatar
 

People are trying to sound like their heros and think it's about the old gear. In truth it's largely about the amazing chops and touch that were necessary to avoid trainwrecks in a world with lots fewer crutches.

They are also not trying to sound like Tascam consoles and tape machines. Too much digital gear has cheap analog stages that are not remotely comparable to the professional gear of 40 years ago.
Old 10th May 2014
  #4837
Quote:
Originally Posted by Bob Olhsson View Post
People are trying to sound like their heros and think it's about the old gear. In truth it's largely about the amazing chops and touch that were necessary to avoid trainwrecks in a world with lots fewer crutches.
I disagree.
I span both eras and have worked with plenty of superlative musicians.
I like the sound of flawed vintage outboard because it has character and there are happy accidents. Clinical perfection very often isn't pleasing to the ear.
The early EMI consoles, Helios and Neve had flawed aspects and some lower quality components to what we can access today, but they have more guts and character compared to the higher spec modern digital devices.
Of course it all depends what you are trying to do, and a modern DAW is better sounding and better specc'ed than a 1990's DAW.
A lot of superb, well known, successful musicians regularly use vintage products with flawed design not because they are trying to emulate their heroes, but because what they hear sounds good to them.
It isn't all about the old gear no, but all things being equal, 'old gear' can add to the already fantastic performance.

Old 10th May 2014
  #4838
Gear Addict
 

Quote:
Originally Posted by bogosort View Post
But that's the thing. It's easy to cite the limits of science and then follow with some arbitrary conclusion: "science doesn't know everything, therefore"

-- digital audio could be flawed, or
-- ultrasonic perception could be a factor, or
-- we don't know enough about human hearing to make definitive claims

etc.

But using that formula, any possible conclusion is equally valid: "Science doesn't know everything, therefore"

-- microscopic aliens could be orbiting Jupiter, or
-- elephants could be telekinetic, or
-- one out of every billion humans is immortal

etc.

Pointing out that science doesn't know everything is not an argument; it is meaningless at best, and completely dismissive of everything we do know, at worst.
This is beautifully succinct, I will abuse this comment in so many conversations. I will try to remember your delightful paraphrase of Russell

Concerning the idea of making hi-res audio cool for all the kids, well, how many people buy flac from bandcamp? The market that Pono is based on established music and specific niches. An independent artist typically can't afford to make a hi-res master as well as a CD master. They barely cover the cost of mastering for one format, after taking out loans to pay for their album recording. There is no advantage to them to market a hi-res format; the recording at 44.1/16 is already more hi-res than their demo.

Pono is initially for musicians who already have a market, so it can be considered just another way that record companies and artists evergreen their IP, by selling (in effect) a hi-res dongle. Do I really wish to buy a device to listen to albums I already have bought in maybe 4 formats already, just to see if there is some slight corner of experience I haven't yet squeezed out of a song? And are these artists really that insecure that they think their art is misrepresented? Just look at the quality of printing in an art book for misrepresentation! An artist's work is always distorted through the lens of the viewer, as it is translated to them.

The hi-res classical recordings aren't listened to critically on a portable anyway (a bus or train is no way to listen to classical music); the people who want hi-res for accuracy listen in a controlled environment.

High quality build costs more. We already see that the bigger a company grows, the more they try to increase the profit margin by reducing build costs. The current laptops have terrible speakers, because people value convenience, slim form factors and light weight over sound quality. For most people good sound these days is a question of if it is shrill enough to cut through the crowded sounds of a cafe. There is a reason that the audiophile market is a niche market. It would be so easy to have made a better analog section to the iPhone as technology has improved, but consumer convenience is more marketable, so we see slimmer form factors. The only way that the marketing can make a dent is to blow enough smoke that people assume there is a healthy fire, regardless of whether the format delivers any better experience or not.

I guess I am trying to see how this fits into industry and the marketplace, as I already have enough skepticism of the technical value of 192/24 as a delivery format. Maybe I have a cynical and skewed view, but the format doesn't stop people from mastering badly, or from choosing widely accessible formats for mass consumption. Ultimately the independent artists need to supply the marketplace with an accessible format, as they need to recoup their recording costs in a marketplace that devalues art. A new format won't mean the independents earn more, and the loudness war remains a concern regardless of format.
Old 10th May 2014
  #4839
Quote:
Originally Posted by DarkandKurious View Post
Concerning the idea of making hi-res audio cool for all the kids, well, how many people buy flac from bandcamp?
How many of the kids' favourite artists are on Bandcamp? Not many (if any) of mine are.

Quote:
The hi-res classical recordings aren't listened to critically on a portable anyway (a bus or train is no way to listen to classical music); the people who want hi-res for accuracy listen in a controlled environment.
This is another slight fallacy that keeps cropping up in these debates.
A portable player like Pono doesn't stop you from using it in a home environment.
If I load my HQ music library on to a device, I want to listen to it in a car and on a flight (yeah, noisy), but I also want to plumb it into my home hi-fi system, and if I'm in a very quiet hotel room in the middle of nowhere, why can't I plug in some good quality headphones and enjoy HQ music from an HQ player also?
Old 10th May 2014
  #4840
Motown legend
 
Bob Olhsson's Avatar
 

Quote:
Originally Posted by chrisso View Post
...The early EMI consoles, Helios and Neve had flawed aspects and some lower quality components to what we can access today, but they have more guts and character compared to the higher spec modern digital devices...
I don't buy that they are more flawed than modern consoles. Quite the contrary. Dynamic range specifications are pretty meaningless.
Old 10th May 2014
  #4841
Many bits of kit from the 50's, 60's and 70's have design flaws and components that have been improved on since then, but to replace that wiring, opamp or chip set removes some of the character.
Why has Michael Brauer got three or four huge racks of vintage outboard.
Certainly many synths and drums I've used over the years have much better designed and built modern counterparts - but have less character.
When it's easily affordable to achieve clinical perfection, many of us now realise it's the moments that fall between the cracks that are special.
The slightly out of tune synth, or the guitarist's mistake on an otherwise flawless take.
Old 10th May 2014
  #4842
Gear Guru
 
Karloff70's Avatar
 

Quote:
Originally Posted by chrisso View Post
Many bits of kit from the 50's, 60's and 70's have design flaws and components that have been improved on since then, but to replace that wiring, opamp or chip set removes some of the character.
Why has Michael Brauer got three or four huge racks of vintage outboard.
Certainly many synths and drums I've used over the years have much better designed and built modern counterparts - but have less character.
When it's easily affordable to achieve clinical perfection, many of us now realise it's the moments that fall between the cracks that are special.
The slightly out of tune synth, or the guitarist's mistake on an otherwise flawless take.
It's basically been reversed. Character is the new clarity. Clarity is a given starting point for all, as opposed to hard won as previously. Now everyone is gagging for saturation of any and every flavour.
Old 10th May 2014
  #4843
Lives for gear
 
Ephi82's Avatar
 

Quote:
Originally Posted by chrisso View Post
This whole Pono debate for me is about better quality consumer audio.
If Pono is no better than CD, but sells gazillions of players because it's cool to the kids, that isn't a victory for me.
If some of the people in this thread will be proven 'right' because Pono sells well, that will be a hollow victory. The whole marketing platform is about bringing better quality audio to the masses, not about being a cool, hipster gadget that makes the founders wealthy.
If it sounds better than many of todays portable players, and if it is marketed well, it will be successful.

It will have to be both, if not, big fail.
Old 10th May 2014
  #4844
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by walter88 View Post
I think nuthinupmysleeve was talking about the higher frequencies stored in the 192k files, before the ma filter. Chris it seems you've worked with the moving average filter alot recently. Would there be a difference if it was operating on a 192 file containing the higher frequencies vs. a 44.1 upsample? Are the higher frequencies a prerequisite for the quality you're hearing?
I think it would have quite the effect, BECAUSE the moving average filter is so eccentric and odd in its behavior.

Thing rolls off very gently and there's loads of naturalness, but it does NOT really filter properly. It's like a tone coloring filter rather than strictly a 'lowpass' filter. As such, there will be a difference between upsampling (and getting the odd characteristics combined with brick-wall lowpassing from the upsample) and using 192K data (and getting quite a lot of that legitimately sampled data bleeding through from the choice of filter).

That's just my estimate as I'm not working with 192K files over here, but one thing we know and I can confirm about the quad moving average filter is, it acts more like a comb than a proper lowpass. It'll do the things I've said about boosting resolution down low, but right up top it acts weird. The response is not so different from a simple acoustic cancellation: maybe that helps headphone listening? Making it sound more open by sounding like there's an acoustic reflective surface in there somewhere, but without actually overlaying a time-delayed copy that would blur the auditory information?
Old 10th May 2014
  #4845
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by chrisj View Post
Less. Moving average filter, remember? Rolls off a bit earlier than it should, for 'warmth'.
I guess it depends on how much there was above 20k. It's a silly pedantic argument.

Yes, it's a different filter, but yet... how is it that rolling off high end EARLIER results in audio that is no longer "underwater?!?!"





Gawd I love this stuff.
Old 11th May 2014
  #4846
Gear Addict
 

Quote:
Originally Posted by nuthinupmysleeve View Post
With Pono, there is no indication it will sound "better"... it just has more high frequency information. Snake oil.
Nuthinupmysleeve, I agree with you completely, and 192 does sound exactly like 44.1 (even a 44.1 downsample from the 192) to my ear in blind testing. But unfortunately I think it's almost too late to argue sampling rates (although I hope the arguments continue) because of the acceptance of 192 in hardware (even Lavry has a 192 DAC), and the way files of different sample rate are being created to compare and tout the superiority of "hi-res".

I would be willing to bet that nearly all of the files Pono is comparing to their "perfect" 192 (96, 44.1 CD) are downsamples from that 192, and they probably don't even know it, or consider it. I can't hear the difference with downsamplers I've tried, but blind testers at AES 2010 could with the one they were using, so that throws a HUGE wrench in this whole discussion. I was thinking anything they were hearing was expectation bias, but now I'm not so sure. Of course the 192 sounds better, because everything else was probably created from the 192 using a sampling rate converter.

Of course when even 192 is not enough and we're on to 384 KHz, THAT will sound better than the 192 that's created as a downsample from the 384.

It would be a never ending cycle based on an imperfect compromise (SRC).
Old 11th May 2014
  #4847
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by walter88 View Post
I would be willing to bet that nearly all of the files Pono is comparing to their "perfect" 192 (96, 44.1 CD) are downsamples from that 192, and they probably don't even know it, or consider it.
…but that's how lower sample rates are made, with any delta-sigma converter. It's always being broken down to lower sample rates and higher word lengths, and then dithered when the word length must be limited. That is literally the best way to do it, and directly sampling at the desired sample rate the worst.
Old 11th May 2014
  #4848
Gear Addict
 

Quote:
Originally Posted by chrisj View Post
…but that's how lower sample rates are made, with any delta-sigma converter. It's always being broken down to lower sample rates and higher word lengths, and then dithered when the word length must be limited. That is literally the best way to do it, and directly sampling at the desired sample rate the worst.
Chris, you know alot more than I do, but the listening results of the 2010 AES showed the SRC could be reliably picked, and the direct sampling could not.
Old 11th May 2014
  #4849
j_j
Lives for gear
Quote:
Originally Posted by walter88 View Post
Chris, you know alot more than I do, but the listening results of the 2010 AES showed the SRC could be reliably picked, and the direct sampling could not.
SRC can be done right, but one must be careful to have exact level match, etc, when trying to see if it's audible.

Even a slight gain difference will show up in a test. It's not easy to know exactly what you are hearing.

I do slightly disagree with Chris, though, direct conversion is not bad, but it is definitely harder.
Old 11th May 2014
  #4850
Gear Addict
 

Quote:
Originally Posted by chrisj View Post
…but that's how lower sample rates are made, with any delta-sigma converter. It's always being broken down to lower sample rates and higher word lengths, and then dithered when the word length must be limited. That is literally the best way to do it, and directly sampling at the desired sample rate the worst.
I'm probably showing my ignorance (I'm not sure) calling SRC "imperfect" and "compromise", and I probably shouldn't have said "huge" wrench (ok, medium sized wrench). I'm still trying to understand your statement about the delta-sigma converter (the ADC, correct?) but I guess you're saying the signal is not generally sampled at 44.1 in the ADC, but sampled at a higher rate and converted down to 44.1 for ADC output (sorry if I got this wrong). So I guess my comment about the "direct sampling" in the test was probably not what you were talking about.

edit: and I probably shouldn't have said "native" either because that's probably incorrect terminology, but I meant recorded to 44.1k and recorded to 96k.

edit: and I probably shouldn't have said "native" either because that's probably wrong terminology, but I mean recording to 44.1k and recording to 96k, for what that's worth.

Apparently what they were doing was recording at 44.1k somehow, recording at 96k somehow, and comparing the two, they couldn't tell the difference on 4 out of 5 musical pieces. They could tell the difference on 1 of those pieces, which I think is important because it's the only blind confirmation I know of between 96k and 44.1k in that respect.

They also made Pyramix SRC to 44.1k from the 96k, and those SRC could be picked out reliably against the 96k on all 5 pieces, and they said that this stage was the common way for mastering to make lower rate files for CD, etc. And that would include lower rate files for HDTracks (like 96k), if the original was 192k.

Regardless, it puts out the question of whether software SRC is always audible, and if it is and it's commonly used, that's one of the only good reasons to buy "hi-res". The authors were pushing for "hi-res" and I think they sort-of failed, but they declared victory because of the one piece victory in "native" and the home run with the SRC.

I'm just saying that Pono is probably just as affected by this SRC practice, even if they're doing the SRC themselves, and if it is audible, it gives them an argument FOR 192 that's a little sketchy at best. It adds a variable that's probably all over the place in different SRC implementations. And for whatever reasons, the blind testers generally couldn't tell the difference in the "native" 44.1 and 96, but could on the SRC. Maybe they didn't level match properly as JJ said, or maybe the SRC just really wasn't that good. But I think it's probably been used on a lot of records.

Edit: and I probably shouldn't have said "native" either, because that's probably incorrect terminology, but I meant recorded to 44.1k, and recorded to 96k.

Last edited by walter88; 11th May 2014 at 07:28 AM.. Reason: added "edit"
Old 11th May 2014
  #4851
Motown legend
 
Bob Olhsson's Avatar
 

I think it's important to understand that we mostly listen to digital audio that has been processed numerous times long after the initial conversion. This includes inside the player. The fact that one conversion is transparent doesn't mean that dozens of them in succession will always be. Each DSP calculation is still a destructive process.
Old 11th May 2014
  #4852
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by Bob Olhsson View Post
Each DSP calculation is still a destructive process.
Let's not single out digital here; any processing, digital or analog, at minimum adds noise. Done properly, a DSP calculation is no worse than measuring an analog signal across a resistor, and typically far better. The trade-off is that while resistors are much, much, much noisier than the rounding error in a double-precision floating point calculation, there are usually far more calculations performed in a digital process than there are resistors (and other components) in the signal path of an analog process.

Still, in the calculus of noise and "destructivity" (nonlinearities), DSP almost always wins by a landslide.
Old 11th May 2014
  #4853
Motown legend
 
Bob Olhsson's Avatar
 

Obviously analog is far more destructive however in practice a lot fewer processes are involved and a lot more care is taken than has become typical of a digital recording.

My point is that eliminating dozens of resamplings that wouldn't be necessary at a higher sample rate could make a higher sample rate end product sound better. Comparing just a single sample rate conversion out of context is comparing apples to oranges in the real world because distortion accumulates.
Old 11th May 2014
  #4854
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by Bob Olhsson View Post
I think it's important to understand that we mostly listen to digital audio that has been processed numerous times long after the initial conversion. This includes inside the player. The fact that one conversion is transparent doesn't mean that dozens of them in succession will always be. Each DSP calculation is still a destructive process.
I don't understand this. If a process is 100% transparent (in the range we hear) why would repeating the process twice be any different than repeating it 100 times? What would make it destructive?
Old 11th May 2014
  #4855
Gear Head
 
bandpass's Avatar
 

Quote:
Originally Posted by walter88 View Post
it puts out the question of whether software SRC is always audible
Software SRC can be made as accurate as you want it to be; by choosing a suitably long filter, aliasing/imaging distortion can be pushed below the system noise floor. So a good SRC implementation shouldn't be audible, with a linear playback chain, and with good level matching (and of course when both rates are >= 44.1k).
Old 11th May 2014
  #4856
Motown legend
 
Bob Olhsson's Avatar
 

Quote:
Originally Posted by Sounds Great View Post
I don't understand this. If a process is 100% transparent (in the range we hear) why would repeating the process twice be any different than repeating it 100 times? What would make it destructive?
"Transparent" only means one pass can't be heard because the distortion is so low. Every conversion expands the bit rate which then needs to be dithered and truncated to something that can be handed off or written to a file. The numbers have in fact been changed by every single signal processing task. If you repeat the process enough times, I can assure you the distortion and, hopefully, dither eventually accumulates to the point of audibility.
Old 11th May 2014
  #4857
Gear Addict
 

Quote:
Originally Posted by bandpass View Post
Pono demo in L.A. this coming Wednesday.
Did anybody go? Can't find any info about this at all, no announcements or write ups.
Old 12th May 2014
  #4858
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by Sounds Great View Post
I don't understand this. If a process is 100% transparent (in the range we hear) why would repeating the process twice be any different than repeating it 100 times? What would make it destructive?
I have tried this myself. Recorded through the same a/d and d/a over and over. I got bored after around 8 times or so and I couldn't notice a difference. I should have kept going. Maybe I will try again.
Old 12th May 2014
  #4859
Gear Head
 
bandpass's Avatar
 

Quote:
Originally Posted by walter88 View Post
Did anybody go? Can't find any info about this at all, no announcements or write ups.
Very little info about it: strangely low key. There are just a few comments here:
Computer Audiophile - Come Hear Pono And Talk Computer Audio At The Audio Salon
Old 12th May 2014
  #4860
Gear Addict
 

Quote:
Originally Posted by bandpass View Post
Very little info about it: strangely low key. There are just a few comments here:
Computer Audiophile - Come Hear Pono And Talk Computer Audio At The Audio Salon
Thanks Bandpass! Says next show in Newport Beach May 30. The last comment says they're still showing a Meridian-based prototype. That the Ayre-based unit is still to come.

Last edited by walter88; 12th May 2014 at 06:40 AM.. Reason: removed 96k. misread.
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