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 6th April 2014
  #2431
Gear Head
 
bandpass's Avatar
 

Quote:
Originally Posted by Kenny Gioia View Post
I'm betting that a test that I would put together (Record and mix a song at 192kHz and export it at 44.1 16 bit then up sample it back to 192kHz and run an ABX test) would have all the participants not being able to tell the difference.

But that is my opinion as I think I would also fail that test.
Here are four 24-96 samples (decent 192k samples are pretty thin on the ground) with down/upped versions for ABXing. (The 'c' samples are the difference files, just to check that all is okay in the conversion).
Old 6th April 2014
  #2432
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by Sounds Great View Post
Concentrate on what happens in the range we hear. Does raising the sample/bit rate effect the result of sampling in the audible range? I think it does.
I don't buy the '192K' meme at all. I think it's gonna cut down on depth of the audible sound through adding extra processing noise in most cases and that it's a trade-off.

Still don't mind as that design choice is going to make Pono better at playback at the sample rates I prefer. I find that if you don't like bright anyway, 44.1K is perfectly serviceable so long as you deal with the glare of pretty much anything loud and bright. For better treble handling, you go up a bit from that, probably no more than 96K. I guess real treble freaks could demand 192K and I won't call them liars, because I can hear what 44.1K does, it just doesn't _bother_ me particularly. Lack of depth bothers me but I've always liked music that wasn't super-hyped-and-sparkly and the music I like can fit within 44.1K handily.

That includes the old stuff, and the vinyl. The records I like, you'd have to use 24 bit to not collapse 'em into lame shallowness but you wouldn't need more than 44.1K for most of them. It's loud brightness and close-up sounds that get you in trouble with 44.1K.
Old 6th April 2014
  #2433
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by bandpass View Post
Here are four 24-96 samples (decent 192k samples are pretty thin on the ground) with down/upped versions for ABXing. (The 'c' samples are the difference files, just to check that all is okay in the conversion).
Dammit, man, you're shaking my faith that 44.1K can be reasonably acceptable if you don't push it.

Some of the 16/44.1-and-back versions are atrocious. In '4' the cymbal crash splats horribly. I thought I could say 44.1K was workable if you liked recessed, darker tonalities but now I'm not so sure :P
Old 6th April 2014
  #2434
Lives for gear
 
GJ999x's Avatar
@Chrisj - i've lost the will to follow this thread properly and probably never had the technical knowledge to do so in the first place but just wanted to say, thanks for Deckwrecka, that played a nice subtle part in the first track i got signed!! Great freebie!

Quote:
Originally Posted by chrisj View Post
Dammit, man, you're shaking my faith that 44.1K can be reasonably acceptable if you don't push it.

Some of the 16/44.1-and-back versions are atrocious. In '4' the cymbal crash splats horribly. I thought I could say 44.1K was workable if you liked recessed, darker tonalities but now I'm not so sure :P
Old 6th April 2014
  #2435
Lives for gear
 
paul brown's Avatar
Quote:
Originally Posted by chrisj View Post
Dammit, man, you're shaking my faith that 44.1K can be reasonably acceptable if you don't push it.

Some of the 16/44.1-and-back versions are atrocious. In '4' the cymbal crash splats horribly. I thought I could say 44.1K was workable if you liked recessed, darker tonalities but now I'm not so sure :P
i think it is important to know how the conversion was implemented. i have tried with various LPFs. if the difference was so obvious, it made me investigate the filter. they are complex as a very clever retired scientist has often said in this thread. perhaps the poster of the files can detail his process. it is not scientific until this is done.
Old 6th April 2014
  #2436
Gear Guru
 

Quote:
Originally Posted by chrisj View Post
In the mid-sixties?
well nobody had digital delays in the mid 60's
but as soon as they were available, they were employed in vinyl mastering. Starting in the mid 70's.

Quote:
When I'm talking about sonorous classical records I'm thinking Mercs and Shaded Dogs. It's news to me that MY records went through a 16/44.1 digital delay. On the other hand, this could explain a lot about why I got turned off to records past a certain point. Seemed like the mojo went away, but the problems (skips, rumble, noise etc) stayed.
my point is that Analog Purity ain't what it used to be, and apparently, it never was. It's pretty funny how nobody even noticed for all those decades! It is becoming increasingly clear that whatever it is about these analog formats that people like, 192kHz audio will not be providing it! I am frankly shocked to see the Analog Crowd lining up to support this. It's almost as if they want to FINALLY 'give up' their opposition to digital - and "Hi-Resolution Digital" gives them a face-saving way to do that.

The worst part about the DDL thing is you can now never be "sure" which of LPs are "contaminated" but my best guess is most of them. How will we ever know what to like and what to hate? This is of course strictly about cutter-head delays as applied to just about every LP. But for classical albums, after the mid 80's, what percentage of classical albums were even SOURCED analog in the first place? Classical to tape dropped off very fast.

Quote:
From sterling sound LINK they are promoting their NON-digital service and offer this contrast to just about "everyone else"
Quote:
when cutting from analog tape the preview and main signals can be derived in two ways
1. the first is to convert the signal to digital and using a digital delay line or workstation output two feeds one for the computer and one for the cutter head. The problem with this approach is that it requires an analog to digital and an digital to analog conversion of the signal...
from a Gearslutz post by a mastering engineer. Mr Berson is a former mastering engineer at Europadisk and now has his own place in NYC:

Quote:
Originally Posted by Cellotron
Even this wouldn't insure it by any means - because a lot of places (read: the vast majority) even if they receive analog tape send the direct feed to the pitch/depth computer and then use a digital delayto send the program signal to the cutter head. ... ...Ironically vinyl mastering studios were some of the very first adopters of digital technology - using 14-bit digital delay lines (later updated to 16-bit) by the likes of Studer and Ampex issued in the mid 70's.
From Opus 3 records

From Record Industry

from The Masters

Vinyl Envy

GS post from Pete Lyman
Old 6th April 2014
  #2437
Gear Head
 
bandpass's Avatar
 

Quote:
Originally Posted by paul brown View Post
i have tried with various LPFs. if the difference was so obvious, it made me investigate the filter. they are complex as a very clever retired scientist has often said in this thread.
That's why I wasn't too keen on wiring up a new filter for this—better to go for something tried and tested. Down and up done using sox with all default options except for a mild shaping on the dither to 16.

The difference files and spectrograms for all are in the zip. It looks pucker to me: the difference spectrograms show no energy apart from the dither below 20k (so I think JJ's in-band level matching requirement is satisfied) and, as expected, they look equal to the originals above 20k.

If folk can hear a difference between the 'a' & 'b' versions, also interesting to know if anything can be heard in the difference files ('c' versions), either as is, or cranked up.
Old 7th April 2014
  #2438
Lives for gear
 
nuthinupmysleeve's Avatar
 

BTW.. if you CAN hear a difference, hit this guy up with your results:

Think You Have Golden Ears? Take the Scientist Challenge!
Old 7th April 2014
  #2439
Lives for gear
 
nuthinupmysleeve's Avatar
 

duplicado!

Last edited by nuthinupmysleeve; 7th April 2014 at 03:35 AM.. Reason: duplicado!
Old 7th April 2014
  #2440
Gear Guru
 

Quote:
Originally Posted by nuthinupmysleeve View Post
BTW.. if you CAN hear a difference, hit this guy up with your results:

Think You Have Golden Ears? Take the Scientist Challenge!
Quote:
All you have to do is prove that you can reliably hear a difference between CD, 320kbps mp3, and any higher resolution format, and we will write a glowing, feature-length article about you. That’s a promise.
cool. They will even let you choose the material and practice with it.

well, we seem to have at least a half a dozen people right here in this thread who can EASILY do that. Right guys? I think it would be great for the Honor of Gearslutz if one of them wins the glowing feature-length article instead of one of the rock stars getting out of Neil Young's Cadillac. Those guys already have glowing feature-length articles written about them.

Quote:
...But the more I went looking for evidence, the more that I found there was none to report....In the end, I was unable to find any reliable evidence that anyone, trained listeners included, could hear or appreciate a difference between these formats when listening blind. I did however, find mountains of studies that pointed the other way. (We even did one of our own, just be sure.)...All this was a genuine surprise to me, especially when I thought back on all the anecdotes and impassioned testimonials I had heard to the contrary.
Old 7th April 2014
  #2441
I have to say I could see some real practical issues with such a challenge, because of the problem of multiple versions.

It seems to me that anyone sponsoring such a test would have to be almost annoyingly anal about making sure there wasn't a slip up in practice or method that let an inappropriate set of comparison files into the fray. Happily, I ain't the guy running the challenge. heh
Old 7th April 2014
  #2442
Lives for gear
 

disk mastering digital delays

Quote:
Originally Posted by joeq View Post
well nobody had digital delays in the mid 60's
but as soon as they were available, they were employed in vinyl mastering. Starting in the mid 70's.
Howard Holzer (Haeco) had a delay in the mid 70's and Ampex marketed one in conjuction with their introduction of 1/2" two track in 1978.
In Los Angeles, few if any mastering rooms were using digital delays in the 70's or 80's, and if so, name them.
Lacquer cutting was then a highly competitive business, you don't want to loose a shootout based on the sonics of some converter.
Digital delays were definitely used when the master was digital, i.e. PCM-1610/1630 or the JVC 900 system.
Preview machies were still the main signal source well into the mid 80's, at least on the left coast.
Old 7th April 2014
  #2443
Gear Guru
 
Sounds Great's Avatar
 

So who is "this guy"?

Some guy wants to test me and prove me wrong? So how, where, when does this happen?
Old 7th April 2014
  #2444
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by Sounds Great View Post
So who is "this guy"?

Some guy wants to test me and prove me wrong? So how, where, when does this happen?
I'm sure there are ways you can prove it. Shouldn't be that hard. For example, do a screen recording with the files, showing the actual test and your success rate? I'd imagine he'd want to see your files (I would too) to ensure they are the same volume and they are otherwise identical to ensure there are no "tells."
Old 7th April 2014
  #2445
Airwindows
 
chrisj's Avatar
It's pretty damn easy if you get to shoot out hihats and cymbal crashes at 24/96 against 16/44.1 (never mind mp3). Just sayin'. If it was just the sample rate, or just the bit depth, that would end up REALLY fatiguing doing an ABX. Even with very reasonable differences it's tiring doing repeated ABXes because it's an insensitive test involving continuous disruptions of your attention, but you just need key elements to be strikingly different.

I wouldn't want to do the classical bit with the strings and harpsichords under ABX conditions. It's not even about 'relaxing into the music' or being unable to do that, when you're ABXing. You've got to find really resilient blatant differences that will be unmistakable many times in a row. By the same token, if you ABXed the sample rate difference on something like a bass guitar's string being plucked, it'd be unreasonably hard.

Listen to the hats and cymbal crash on that '4' example, back and forth, and tell me it's not obnoxiously obvious. The whole character of the thing is different in a palpable way. Of course, I am using revealing speakers. NS10s.

Oh, you thought speakers had to do a GOOD job of putting out high treble energy to reveal this stuff? Not so. I do think I would find it impossible to distinguish on Avantones, though. But the NS10s are more than capable of throwing out a big difference in output given such a striking difference in signal energy.

Anyone thinking a 'scientist challenge' is relevant to THIS set of audio files is foolish, I think. You would have a hard time picking a type of sound more revealing of digital highs, than loudly mixed raw cymbals and crashes. That stuff puts half its energy up there, which is not the case for many other sounds.

I know this world, you see: back in the days of Classic Mac, I ported an ABX tester to MacOS for Arny Kreuger and did some ABX tests. Probably my most impressive result was that I told the difference between 320K mp3 and 16/44.1 uncompressed audio, on a castanet recording, something like 18 trials out of 20. That's well over 95% confidence right there, and I think it was only 16/44 though this was years ago. I listened for the personality of the 'clicks' rather than any frequency-related 'tell'. I was much poorer at picking out wow and flutter ABXes, which is probably why I like vinyl so much.
Old 7th April 2014
  #2446
Gear Guru
 

Quote:
Originally Posted by Sounds Great View Post
So who is "this guy"?
Justin Colletti

Quote:
Some guy wants to test me and prove me wrong? So how, where, when does this happen?
no, he wants to prove you right! And then write an article about the person who can do it.

Quote:
The unproven claim that comes up the most often is that there is a clear audible difference in sound quality between a super-high-resolution format like 24/192, and either standard full-resolution audio at 16/44.1, or a high-resolution mp3 at 320kbps.

That these claims are in fact, unproven, may surprise some people. If you’re one of them, don’t worry, you’re not alone.

When I first started to look into the claims about the sound quality of super-high resolution formats, I, like many others, instinctively believed that they must sound better. I mean, just look at the numbers! They’re so much bigger! Duh!! But the more I went looking for evidence, the more that I found there was none to report.
he is trying to find some to report. He is hoping for someone who can succeed. I don't think he is going to write an article about some guy who could not do it. "Can't do it" is the default at this point in time. He is looking for a man who will bite a dog. This offer has been up since September, BTW.

He previously did an online test with his readers, presumably audio-savvy individuals, of ACC vs WAV - even though this test was online and readers could at least potentially have downloaded the files and examined them in the computer, (i.e. cheat) STILL the percentage of correct answers was 49%.
Old 7th April 2014
  #2447
Gear Guru
 

Quote:
Originally Posted by cathode View Post
...
Preview machies were still the main signal source well into the mid 80's, at least on the left coast.
"Well into the mid 80's" is 30 years ago. That's a lot of vinyl. My point is that some rather large percentage of the vinyl people are oohing and aahing about IS 16-bit PCM, period, and the worst part of it is they don't know for sure which ones they are. This inability to "tell" casts doubt upon how much of a risk there really is of losing a shootout, for example.

Steve Berson quoted in my post above, and others in this thread have some differing points of view ("because a lot of places (read: the vast majority...")

From everything I have read, it seems pretty clear that digital look-ahead started being used almost as soon as it was available, quickly became routine, and soon enough, places that Did Not use digital delays were unusual enough as to merit special mention. If some analog purists wish to "believe" all of THEIR records are "pure", though, it would not surprise me in the least.
Old 7th April 2014
  #2448
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by chrisj View Post
It's pretty damn easy if you get to shoot out hihats and cymbal crashes at 24/96 against 16/44.1 (never mind mp3). Just sayin'.

(snip)

I know this world, you see: back in the days of Classic Mac, I ported an ABX tester to MacOS for Arny Kreuger and did some ABX tests. Probably my most impressive result was that I told the difference between 320K mp3 and 16/44.1 uncompressed audio, on a castanet recording, something like 18 trials out of 20. That's well over 95% confidence right there, and I think it was only 16/44 though this was years ago. I listened for the personality of the 'clicks' rather than any frequency-related 'tell'. I was much poorer at picking out wow and flutter ABXes, which is probably why I like vinyl so much.
I think you should take him on! His profile of you can also include info on your plugins/etc. Good free publicity. What the heck!

Seriously, if anyone can do it you might, especially with your plugin development expertise.
Old 7th April 2014
  #2449
Lives for gear
 

Quote:
Originally Posted by joeq View Post
"Well into the mid 80's" is 30 years ago.

From everything I have read, it seems pretty clear that digital look-ahead started being used almost as soon as it was available, quickly became routine, and soon enough, places that Did Not use digital delays were unusual enough as to merit special mention. If some analog purists wish to "believe" all of THEIR records are "pure", though, it would not surprise me in the least.
And well into the mid 90''s.

In 1984,
JVC Cutting Center was using a preview machine, MCI with custom JVC electronics.
KM Records was using a preview machine, an MCI,
Capitol Records was using a preview machine, Studer.
Mastering Lab was using a preview machine, MCI with custom electronics.
Bernie, Studer with custom electronics.
Precision Mastering, ATR with a preview head, and that was very trick.

No, your claims are totally unsupported.
Old 7th April 2014
  #2450
Gear Guru
 

Quote:
Originally Posted by cathode View Post
In 1984...No, your claims are totally unsupported.
What are you saying? Are you saying NOBODY used digital delay for vinyl in 1984?

In my link which you did not bother to read, Steve Berson from Europadisk says:
Quote:
Ironically vinyl mastering studios were some of the very first adopters of digital technology - using 14-bit digital delay lines (later updated to 16-bit) by the likes of Studer and Ampex issued in the mid 70's.
Are you saying that "well into the 1990s" all (or even the majority of) vinyl everywhere all over the world was cut using a fully analog signal path? Is all vinyl today still using these machines?

The sad fact is that little documentation of the analog "purity" of most LPs was ever made or kept. Probably because nobody was "bothered" by the sound. From the mid 80's on, more and more and more of them went through a digital stage, even when the masters were analog. The even sadder fact is that nobody can tell which ones are which by playing the records!
Old 7th April 2014
  #2451
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by joeq View Post

He previously did an online test with his readers, presumably audio-savvy individuals, of ACC vs WAV
The test needed for this is not going to be on line.
Old 7th April 2014
  #2452
Lives for gear
 
nuthinupmysleeve's Avatar
 

Quote:
Originally Posted by Sounds Great View Post
The test needed for this is not going to be on line.
A test of .wav vs. .aac can certainly be online.
Old 7th April 2014
  #2453
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by nuthinupmysleeve View Post
A test of .wav vs. .aac can certainly be online.
So I am going to listen to something on my computer's sound card?
Old 7th April 2014
  #2454
All these tests are so trivial! We need a real test, like, if you drown, that proves you weren't a witch!
Old 7th April 2014
  #2455
Lives for gear
 
Silver Sonya's Avatar
 

Worth a read.

- c
Old 7th April 2014
  #2456
Lives for gear
 
camus's Avatar
 

That 808.2 CD player costs more than the converters most mastering engineers are using. I had a tough time reading on after that. Yes, I am a cynical bastard sometimes.
Old 7th April 2014
  #2457
Gear Maniac
 

^^ That's bad enough. Check this out.

Ayre, the Pono converter tech consultants, got caught red handed a couple years ago buying up $500 Oppo players and putting them back in a box with their own name on. Then charging punters $10,000 for the privilege.

Ayre? or Oppo?
Old 7th April 2014
  #2458
j_j
Lives for gear
Quote:
Originally Posted by nuthinupmysleeve View Post
It's true nyquist wasn't specifically speaking to this fact.
Actually, Shannon proves that in-band is fine, once you meet the constraints of the Shannon Sampling theorem.

Nyquest made the conjecture, Shannon proved it.

But Shannon did in fact speak precisely to the question of what happens anywhere in-band.
Old 7th April 2014
  #2459
j_j
Lives for gear
Quote:
Originally Posted by nuthinupmysleeve View Post
If that is true, then why has everyone failed in double blind testing?
Just for pedantry, "everyone who's take the test".

Someone is sure to attack you for claiming a negative.

The answer to them is, then, "take the (*&(*& test".
Old 7th April 2014
  #2460
j_j
Lives for gear
Quote:
Originally Posted by paul brown View Post
i would have thought it was the product of design. probably using mathematics. likely using proven theoretical understanding at the time.
Well, partially, the other part being the estimation of human hearing frequency range.

Which seems pretty accurate for people over 20 or so, in this day and age, at least.
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