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 February 2015
  #5191
Lives for gear
Quote:
Originally Posted by Plush View Post
It's very clear that naysayers are in over their heads.

It is delusional to believe that the full capabilities of the SABRE chip in the PONO are not fully utilized.

After all, the hi-res playback was why the device was developed in the first place.

Monty is a charlatan. Who else will back him and join him on the clown cart?
This is very interesting--can explain the sense in which he is a charlatan? Because as far as I can see he's not a charlatan at all. Please explain
Old 10th February 2015
  #5192
Lives for gear
Quote:
Originally Posted by chrisj View Post
The hilarious part is, it's so trivial to demonstrate how superior something like a Pono 192/24 format is to CD or mp3 on technical grounds. It is SO EASY to measure this. You can even throw on some compression, alter the sound and make the ugly grunge jump right out so anybody can hear it. There is NO doubt that high resolution is objectively better on several fronts. Any archivist would go for the more high-bandwidth version. Any scientist measuring things about the music would grab the better data-set without hesitation, and get cleaner results.

But when it comes to love…
I really do not see what you mean by "superior" here. If by "superior" you mean "contains more information undetectable by the ear" then sure. By this logic, 384 is twice as superior as 192, right? I mean this seems like an inescapable conclusion, if this is what you mean by "superior."

The thing is audibility, and hear you are arguing that human ears are not machine like. Ok, sure. But you are advancing the machine like "superiority" of 192 as the reason we should prefer it: that is, we should prefer it even though we cannot reliably perceive it. Or, perhaps, cannot perceive it at all (which is what over 010 years of scientific testing and study of human hearing says, but...)

We should prefer it because why? I am 100% sure that in a decade people will be telling us that we need downloads at 768 because it's manifestly clear there is more info there and you can hear it in the reverb tails.

Last edited by PB+J; 10th February 2015 at 04:21 PM..
Old 10th February 2015
  #5193
Lives for gear
 
Santiago's Avatar
 

Can I just bring the conversation back to the point that, given that the Ipod Classic is no more, and that Pono has a fairly reasonable price, it sounds like a fairly good deal?

I'm a bit sceptical about the importance of above-CD sample rates in audio, but, even if we take that out of the equation and assume that it will be slightly better than an Ipod Classic, it's not like there's too much competition out there for portable digital audio, is there?

If it cost double what it does, all the discussion about its performance would be more warranted, but at its current price I don't understand why all the vitriol.
Old 10th February 2015
  #5194
Lives for gear
 
Santiago's Avatar
 

Can I just bring the conversation back to the point that, given that the Ipod Classic is no more, and that Pono has a fairly reasonable price, it sounds like a fairly good deal?

I'm a bit sceptical about the importance of above-CD sample rates in audio, but, even if we take that out of the equation and assume that it will be slightly better than an Ipod Classic, it's not like there's too much competition out there for portable digital audio, is there?

If it cost double what it does, all the discussion about its performance would be more warranted, but at its current price I don't understand why all the vitriol.
Old 10th February 2015
  #5195
Lives for gear
 
stratology's Avatar
 

Quote:
Originally Posted by Santiago View Post
Can I just bring the conversation back to the point that, given that the Ipod Classic is no more, and that Pono has a fairly reasonable price, it sounds like a fairly good deal?

I'm a bit sceptical about the importance of above-CD sample rates in audio, but, even if we take that out of the equation and assume that it will be slightly better than an Ipod Classic, it's not like there's too much competition out there for portable digital audio, is there?

If it cost double what it does, all the discussion about its performance would be more warranted, but at its current price I don't understand why all the vitriol.
If you don't care about audio above 16bit/44k, the competition is the iPod touch.
Costs less, endlessly more powerful (download music directly to the device, Internet access, apps, 5x the battery life, smaller, high res display, camera, video playback, etc.).


If you do care about audio above 16bit/44k, the competition are devices like the Fiio.
Old 10th February 2015
  #5196
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
He does not have my permission to make 128K lossy compression a mandate. He would like to have the power to mandate to the world that nobody may pursue audio resolutions he sees as wasteful, and is arguing forcefully against ever doing such 'wasteful' things.
Talk about a straw man! You don't have to grandstand here; no one, including the Evil Monty, is pushing a 128 kbps mandate. The issue is between CD quality and "high res" (or "snake oil", depending on your side).

Quote:
They DO have the power to control their broadcast mediums and mandate what people have to put up with, and they do. It's dangerous to sit around listening to these types and offering no pushback. You'll be on 96kpbs before you know it, and the same guys will be telling you it's fine, enjoy, don't be a hater.
You're lumping Monty in with the content providers like Youtube, which doesn't make any sense. Also, you do realize that as internet bandwidth has increased, so has the audio bitrates of content providers? Streaming quality has been going up, not down. But that's non sequiter. We're not debating how content providers should use their bandwidth; we're talking about distribution formats for purchased music, as in: should I buy the next Pixies album at CD quality or as 192/24 files?

A whole bunch of people are arguing that CD quality is not good enough, that we should be producing and buying "hi res" audio. Pono was designed to serve that crowd. Other people, like Monty, are arguing that CD quality is as good as it gets for humans, that increasing the sample rate and word length is just a waste of resources. This is fundamentally a technical debate, yet for some reason it has taken the form of a religious debate. I'm not sure why -- we don't typically see such passion in debates on, say, AES versus SPDIF -- but statements such as "It's dangerous to sit around listening to these types and offering no pushback" certainly don't help. Dangerous? Really?
Old 10th February 2015
  #5197
Lives for gear
 
Santiago's Avatar
 

Quote:
Originally Posted by stratology View Post
If you don't care about audio above 16bit/44k, the competition is the iPod touch.
Costs less, endlessly more powerful (download music directly to the device, Internet access, apps, 5x the battery life, smaller, high res display, camera, video playback, etc.).


If you do care about audio above 16bit/44k, the competition are devices like the Fiio.
The Ipod touch's memory is much smaller than the Ipod Classic (a maximum memory of 64GB), and it has a lot of unnecessary features, so I don't count it as competition. ( I say this as someone whose Ipod classic is chock-full of lossless files and needs to delete playlists when adding new ones).

I've heard bad things about the software, stability and playlist capacity of the Fiio, which is why the Pono sounds good to me. The additional improvement in sound quality is more of a bonus, really.

Last edited by Santiago; 10th February 2015 at 05:28 PM..
Old 10th February 2015
  #5198
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by chrisj View Post
Still not seeing a lot of comprehension of the ABX corollary: if you did infinite trials and got say a 55% confidence, that's enough (infinite) to statistically rule out chance, and you've proved a phenomenon is either only there part of the time, or that the human sensory apparatus regularly fails to be reliable and picks up the phenomenon (which exists 100% of the time) only part of the time—but predictably so.
I'd say you still understand neither binomial statistics nor significance tests.

Quote:
The superiority of say 192/24 is provable by measurement and instrumentation. It is literally impossible to argue that CD or mp3 is not heinously degraded by measurement standards and we know exactly, in great detail, what the damages are, where they come from, even the spectral properties of the distortions induced. We know all this with incredible thoroughness.
My high school English teacher used to admonish our papers with the saying Show, don't tell. So show us how 192/24 is provably superior for audio sources. Show us the spectral properties of CD's induced distortion.

Quote:
You can't take a human in their moment of relaxation and enjoyment, when they're associative and their senses are reaching out for magic and inspiration, and demand they function like a machine, 10/10 no error and hearing the same transient exactly the same each time. It is utter foolishness and contrary to the essence of what music (or any art) is.
The problem here is that the human brain is remarkably apt to invent its own reality. This has been documented in countless experiments. Therefore the brain cannot be trusted. If you don't understand and agree with that simple statement, then there is nothing more to say. A person may truly believe that they hear magic in one out of every hundred transients or whatever, but that does not make it so. ABX is one technique for short-circuiting a huge source of "brain lies", the visual cortex. The statistical interpretation of ABX results -- significance testing -- provides us with a reasonable way of assessing the likelihood of our conclusions. Otherwise we're just guessing.

Quote:
The hilarious part is, it's so trivial to demonstrate how superior something like a Pono 192/24 format is to CD or mp3 on technical grounds.
You'd instantly end countless hours of debate by providing such trivial to demonstrate proof! Do it!
Old 10th February 2015
  #5199
Lives for gear
 

Quote:
Originally Posted by bogosort View Post
You'd instantly end countless hours of debate by providing such trivial to demonstrate proof! Do it!
There really is no point.

All he could be referring to is
- greater bandwidth (even though 24/192 is higher than really necessary for distribution)
- the potential for lower noise floor (even though it's possible to achieve noise floor in the audible band equal to or greater than analogue systems with even 16/48, so yet again 24/192 is higher than necessary)
- lack of the distortion that's inherent to lossy compression (even though if the bit rate is high enough, these distortions can be demonstrably inaudible)

... there's no way of refuting that if your criteria for technical superiority is to change the input signal as little as possible, 24/192 is technically superior to CD and lossy compression given an input signal of at least 24/192 :p; we can reasonably presume that this is his criteria.

There is no point because that isn't what anyone criticising the claims of Pono is discussing.
Old 10th February 2015
  #5200
Lives for gear
 
stratology's Avatar
 

Quote:
Originally Posted by Santiago View Post
The Ipod touch's memory is much smaller than the Ipod Classic (a maximum memory of 64GB), and it has a lot of unnecessary features, so I don't count it as competition. ( I say this as someone whose Ipod classic is chock-full of lossless files and needs to delete playlists when adding new ones).
OK, I see your point. Limited storage capacity is less relevant with a device that has Internet access, because you can access your whole music library via iTunes Match.
You use lossless files, so iTunes Match is not an option, as it syncs 256k AAC.


I believe a device intended for use with lossless files needs storage capacity an order of magnitude larger, in the terabyte range, rather than gigabyte range, to avoid running into the storage issues you see with your iPod classic. Another missed opportunity for the Pono.
Old 10th February 2015
  #5201
Lives for gear
 
Santiago's Avatar
 

Quote:
Originally Posted by stratology View Post
OK, I see your point. Limited storage capacity is less relevant with a device that has Internet access, because you can access your whole music library via iTunes Match.
You use lossless files, so iTunes Match is not an option, as it syncs 256k AAC.


I believe a device intended for use with lossless files needs storage capacity an order of magnitude larger, in the terabyte range, rather than gigabyte range, to avoid running into the storage issues you see with your iPod classic. Another missed opportunity for the Pono.
Well, the Pono can be expanded, although I agree that the memory could be bigger, and something in the terabyte range would be amazing.

When my Ipod finally dies its replacement will probably be a Pono, perhaps a Fiio if by then they have sorted out the software and playlists...
Old 10th February 2015
  #5202
Lives for gear
 
bogosort's Avatar
Quote:
Originally Posted by -tc- View Post
... there's no way of refuting that if your criteria for technical superiority is to change the input signal as little as possible, 24/192 is technically superior to CD and lossy compression given an input signal of at least 24/192 :p; we can reasonably presume that this is his criteria.
But that would be an empty statement; clearly we are talking about capturing and playing back audio signals, not measuring, say, the switching noise of switch-mode power supplies. But you are correct, we should define our parameters. By audio I mean the electrical representation of musical sound. I suggest that a reasonable bound on audio signals is 20 kHz bandwidth (baseband) and 120 dB of dynamic range. Anything outside this cannot reasonably be considered audio.
Old 10th February 2015
  #5203
Lives for gear
 

Agreed.
Old 10th February 2015
  #5204
Airwindows
 
chrisj's Avatar
Then you already want more than 16 bits, perhaps something like 20 (which I could live with, it just doesn't divide cleanly into digital bytes and that's bad design).

I agree.

I can also measure very easily that you get substantial 30-40K content off plain old cymbals, though it's very attenuated, through even the most inappropriate 'measurement' mic ever, a SM57. Any condenser, ANY normal 'overhead or cymbal' mic will deliver this content still more accurately.

It's there, and it's trivially easy to represent it in the waveform with current technology (hell, with decade-old technology). Why on earth should we not do this? We are not making the argument of 'now we move to 48 bit and 3 MILLION k sampling rate!'. This whole thread is about a single $400 playback device which, I contest, exceeds any possible human's needs for audio quality.

How can we even be having the argument of 'NO! It has to be a lot worse than that! Snake oil! Heathens!' much less have 174 pages of it? You see, this is the best evidence of all that we MUST speak up politely and say no and demand a place for good-quality devices like the Pono. If we don't, there's literally 174 pages of ANGRY demands that we take our 16/44.1, or our 320kbps, or our 128kbps (hi, YouTube HD streams!) and like it.

It is sometimes possible to literally mandate that sort of thing. YouTube is. No matter what you upload, they will transcode it and that is your only outlet, barring stuff like Vimeo or paying your own bandwidth costs: and then who is going to hear about any of it, and where?
Old 10th February 2015
  #5205
Lives for gear
 

Meh.
I recently returned to vinyl on a decent deck with quality cartridge.
Game over.
Nothing is going to touch a DECENT turntable with a non abused record for me. It slaughters anything digital I have, with one or two exceptions and thats cause the particular record isnt that great. Good record, and you literally laugh when you compare it to its digital counterpart.
And records are fun.
Digital files are most certainly not.
Old 10th February 2015
  #5206
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by chrisj View Post

I can also measure very easily that you get substantial 30-40K content off plain old cymbals, though it's very attenuated, through even the most inappropriate 'measurement' mic ever, a SM57. Any condenser, ANY normal 'overhead or cymbal' mic will deliver this content still more accurately.

It's there, and it's trivially easy to represent it in the waveform with current technology (hell, with decade-old technology). Why on earth should we not do this?
Because that isn't audible. We have to draw the line somewhere. The upper frequency limit of our hearing (and a tiny bit extra) is a good place to do that.

Btw, I consider 192 KHz technically inferior as it is wasteful. Bad enengineering. Not surprising since it is driven by the marketing department rather than the engineering department. Marketeers are rarely any good at engineering.

Personally I wouldn't mind settling at 48 KHz to just have a single sample rate for all my projects. Imagine the cost saving if all interfaces and all software only did 48 KHz and nothing else! (Not to mention internet bandwidth wasted on pointless discussions ).

Alistair
Old 11th February 2015
  #5207
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by dodittydada View Post
Meh.
I recently returned to vinyl on a decent deck with quality cartridge.
Game over.
Nothing is going to touch a DECENT turntable with a non abused record for me. It slaughters anything digital I have, with one or two exceptions and thats cause the particular record isnt that great. Good record, and you literally laugh when you compare it to its digital counterpart.
And records are fun.
Digital files are most certainly not.

I listen to quality vinyl with my Ortofon X5 cartridge nearly daily.

I had a bunch of SACD disks but I sold most of them. I think it comes close to vinyl but still misses some of the more organic elements of non-converted analog.

The fact that you really can't listen to records in the car has me hoping that one day I can afford the equipment to record all my records to dsd and play them back on the road. This would be the next best thing.
Old 11th February 2015
  #5208
Lives for gear
 

Quote:
Originally Posted by chrisj View Post
Then you already want more than 16 bits, perhaps something like 20 (which I could live with, it just doesn't divide cleanly into digital bytes and that's bad design).

I agree.
You're right - 16 bits doesn't quite cover the full theoretical dynamic range of human hearing under ideal conditions if you stick to the common definition that 1 bit = 6 dBs. But that's not quite accurate. Of course you know about noise shaping. So in practice, well-prepared 16 bit audio can be enough to cover the 120 dB figure perceptually. Sure, go a few bits higher, sample a few kHz faster -- why not, it definitely makes some things easier (how about 60 kHz as proposed by Mr Lavry? Seems perfectly sensible to me. Or just stick to 48 kHz or everything -- also sensible).

But conflating CDs and the 'underwater sound' of low bit rate MP3s as a way of pushing 24/192 as the one true way to hear music, etc. is just silly. It remains unsupported by good evidence and is sensationalistic which is what annoys some people (and as Al brings up, there are technical arguments against 192 kHz sampling). The Pono system is geared towards making consumers think that bigger numbers are better, i.e. the inclusion of the little pink sample rate meter and pop-up info-graphic on each album page that will make them think that if they buy a 48 kHz album they're getting less than half the 'audio quality' than they should be expecting. Such presentation is either naive or dishonest, and neither is laudable. Aesthetics aside, the device itself seems good.

We don't need 'high res' audio, we need 'appropriate res' audio done properly with music that hasn't been smashed with a limiter (unless that's what the music needs, which is true sometimes). The implication by some that everyone should just silently put up with being mislead (even with if with good intentions) in order to receive the latter isn't good enough imho

Last edited by -tc-; 11th February 2015 at 12:24 AM..
Old 11th February 2015
  #5209
Lives for gear
 

Quote:
Originally Posted by chrisj View Post
our 16/44.1, or our 320kbps, or our 128kbps (hi, YouTube HD streams!) and like it.
Here you go again, 16/44.1 is not lossy data compression in the same way as mp3. If you want to attack it you have to explain what we can hear that it doesn't produce, not just lump it in with mp3.

Nobody is mandating anything, if you don't want lossy compression on your audio (I don't either) then don't upload it to Youtube.

Now in reality, uploading stuff to Youtube may be inevitable and may be done by others involved in the project, so of course it would be good if it were lossless. If you simply show the human-audio-band differences between the original 16/44.1 wav and the lossy encoded result you'd be showing actual evidence of the actual problem, which is more likely to be productive.


Chris
Old 11th February 2015
  #5210
Lives for gear
 
doom64's Avatar
A while ago when Bluray and HD DVD (remember that?) were released I recall reading about a little experiment someone at a Best Buy did.

At the time Blu-ray was receiving a huge marketing push. HD DVD? Not so much. Despite HD DVD having a superior format (VC-1 video with Dolby TrueHD audio as a standard) with initial movies released compared to Bluray (MPEG-2, TrueHD not standard) it was that marketing that clouded people's perceptions.

Sony had a deal with Best Buy. A side by side TV comparison. Bluray on one side and HD DVD on the other. The vast majority of customers, according to this Best Buy employee, said that Bluray was better than HD DVD. So, after a couple weeks the employee switched the HDMI cables between the players. The Bluray TV was actually showing HD DVD and vise-versa.

People continued to prefer Bluray, even after the switch.

Marketing is a VERY powerful thing. Conduct your own tests. I have. That's how I like to come to exacting conclusions and not waste money.
Old 15th February 2015
  #5211
Lives for gear
PonoPlayer Firmware Update Version 1.0.5
February 11, 2015

PonoPlayer Firmware Update Version 1.0.5

***ALERT: Because of a previous issue (that is fixed in this firmware version), please remove your microSD card prior to downloading and installing this latest firmware update. After downloading the firmware update, to install, please eject your PonoPlayer from your computer and unplug the Micro-USB cable from your PonoPlayer. ***

Watch the Firmware Update video for a walkthrough of the update process.
http://vimeo.com/111570190


Firmware Version 1.0.5 Release Notes:
Feature Enhancements & Bug Fixes

Added DSD playback capability

Playback of DSD file utilizes the DSD mode of the player’s DAC. The data in the DSD file is presented directly to the DAC.
Support for 2 channel DSD64 and DSD128 files
Support for DFF file container (containing uncompressed data)
Support for DSF file container including ID3 tags

Improvements to Audio file support
Corrected “Scanning Music Library” issue for MP3 files that have malformed UTF-8 tags
Added handling for truncated/malformed WAV files
Added handling for AIFF files with invalid chunks (some HDTracks files had this issue)
Added support for 32 bit WAV files

Improvements to USB mode and device management
Fixed issue where the PonoPlayer drives weren't always detected if the player was asleep when plugged into a computer
The screen rotation locks automatically while Music Transfer Mode is enabled
Detect corrupted files that cause the “Scanning Music Library” issue and display a dialog with the filepath of the file to allow users to connect via USB and remove the file.
Detect if the internal storage is formatted to an unsupported filesystem or has filesystem errors that the player cannot repair. Display a dialog allowing the user to reformat the internal storage if this occurs.

General Improvements
Improved support for microSD cards with certain types of invalid partition tables
Exit Balanced Mode if a cable is removed while the player is powered off
Prevent overwriting playlist number if microSD card is removed and firmware is updated
Fixes related to playback with Volume Leveling enabled
Old 16th February 2015
  #5212
Lives for gear
 

Quote:
Originally Posted by CarmenC View Post
PonoPlayer Firmware Update Version 1.0.5
February 11, 2015

PonoPlayer Firmware Update Version 1.0.5

***ALERT: Because of a previous issue (that is fixed in this firmware version), please remove your microSD card prior to downloading and installing this latest firmware update. After downloading the firmware update, to install, please eject your PonoPlayer from your computer and unplug the Micro-USB cable from your PonoPlayer. ***

Watch the Firmware Update video for a walkthrough of the update process.
http://vimeo.com/111570190


Firmware Version 1.0.5 Release Notes:
Feature Enhancements & Bug Fixes

Added DSD playback capability

Playback of DSD file utilizes the DSD mode of the player’s DAC. The data in the DSD file is presented directly to the DAC.
Support for 2 channel DSD64 and DSD128 files
Support for DFF file container (containing uncompressed data)
Support for DSF file container including ID3 tags

Improvements to Audio file support
Corrected “Scanning Music Library” issue for MP3 files that have malformed UTF-8 tags
Added handling for truncated/malformed WAV files
Added handling for AIFF files with invalid chunks (some HDTracks files had this issue)
Added support for 32 bit WAV files

Improvements to USB mode and device management
Fixed issue where the PonoPlayer drives weren't always detected if the player was asleep when plugged into a computer
The screen rotation locks automatically while Music Transfer Mode is enabled
Detect corrupted files that cause the “Scanning Music Library” issue and display a dialog with the filepath of the file to allow users to connect via USB and remove the file.
Detect if the internal storage is formatted to an unsupported filesystem or has filesystem errors that the player cannot repair. Display a dialog allowing the user to reformat the internal storage if this occurs.

General Improvements
Improved support for microSD cards with certain types of invalid partition tables
Exit Balanced Mode if a cable is removed while the player is powered off
Prevent overwriting playlist number if microSD card is removed and firmware is updated
Fixes related to playback with Volume Leveling enabled
Is this a good thing, or a bad thing?
Old 18th February 2015
  #5213
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by CarmenC View Post
Added DSD playback capability

Playback of DSD file utilizes the DSD mode of the player’s DAC. The data in the DSD file is presented directly to the DAC.
Support for 2 channel DSD64 and DSD128 files
Support for DFF file container (containing uncompressed data)
Support for DSF file container including ID3 tags
This is very cool, because I don't think the future is in re-selling the White Album another 357 times. This is a path to the proliferation of CURRENT recordings done in a minimalist, two-mic manner, on a real tight budget.

The thing about DSD is it's a simpler, cruder format than PCM, with some striking variations from normal digital practice. It's got really weird noise behavior, where the error just keeps dropping to impossibly low levels as frequency drops, but in supersonic frequencies it goes nuts and loads all the error/artifacts way above 20K. This is actually a situation (unlike PCM and totally the opposite of what Pono does with 192K PCM) where Monty's concerns about intermodulation are well founded: if you have bad amplification it might well get harsh with DSD content as the DSD throws so much HF crap down the pipe.

But the trade-off gets you a simpler form of encoding/decoding, and people whose reactions I trust (to be a signpost of something out there to be sensed, not 'any single perception of theirs is automatically true') have responded very favorably to DSD. It's analogous to the analog designs where it's all based around one FET or whatever: minimalism, as a path to more conveying of emotions, textures and other such hard-to-quantify experiences.

Better still, you can get a DSD converter for less than $1000, and coupled with about $3000-$6000 of great mic pres and microphones, that means for less than $10K any person could simply go out and buy a recording chain equal to any in history (pretty much). Then they can put the stereo mics (minimalist!) in front of any acoustic performance, capture it to DSD, and sell it to Pono listeners and anybody else who's got that playback capability.

Compare to YouTube playing everything back at 128kbps lossy joint stereo or whatever they're doing to save bits this week (could get worse!) and you're looking at a whole, distinct, LABELLED market for exceptional quality audio and human performances on musical instruments, in a way that's possible for newcomers to get into. All you need to do is make your recordings and try to sell them as 'Pono-ready DSD recordings' and people will immediately know what that is and why it was done.

I'm not going to be making DSD recordings, I'm going to be making 24/96K recordings, getting my sound with that, and saying 'um these are still quite good you know!'

But I was immediately pleased to hear DSD was being added: and that now becomes my big curiosity factor. I'm going to want to buy some kind of probably symphonic DSD content so I can check that out. It ought to be a lot of fun!
Old 18th February 2015
  #5214
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by chrisj View Post
The thing about DSD is it's a simpler, cruder format than PCM, with some striking variations from normal digital practice. It's got really weird noise behavior, where the error just keeps dropping to impossibly low levels as frequency drops, but in supersonic frequencies it goes nuts and loads all the error/artifacts way above 20K.
DSD's noise behaviour is not weird. It is just that as it is only 1 bit, the dither noise floor is at the same level as the signal. The signal is literally awash in noise. To make it listenable to humans, that noise is very heavily filtered to push it up and beyond the human hearing. Even with this very heavy filtering, it is mathematically impossible to fully dither a DSD signal. (That is part of the reason why modern converters use more than 1 bit. DSD is outdated technology and the only reason it is still around is for marketing reasons). Hence there is always a certain amount of quantization distortion in DSD signals. I think the level is too low to hear but maybe this distortion is what some of the DSD proponents like. (I doubt it. I think it is simply the marketing that sold them on it).

Alistair
Old 18th February 2015
  #5215
Lives for gear
 
doom64's Avatar
Quote:
Originally Posted by chrisj View Post
This is very cool, because I don't think the future is in re-selling the White Album another 357 times. This is a path to the proliferation of CURRENT recordings done in a minimalist, two-mic manner, on a real tight budget.

The thing about DSD is it's a simpler, cruder format than PCM, with some striking variations from normal digital practice. It's got really weird noise behavior, where the error just keeps dropping to impossibly low levels as frequency drops, but in supersonic frequencies it goes nuts and loads all the error/artifacts way above 20K. This is actually a situation (unlike PCM and totally the opposite of what Pono does with 192K PCM) where Monty's concerns about intermodulation are well founded: if you have bad amplification it might well get harsh with DSD content as the DSD throws so much HF crap down the pipe.

But the trade-off gets you a simpler form of encoding/decoding, and people whose reactions I trust (to be a signpost of something out there to be sensed, not 'any single perception of theirs is automatically true') have responded very favorably to DSD. It's analogous to the analog designs where it's all based around one FET or whatever: minimalism, as a path to more conveying of emotions, textures and other such hard-to-quantify experiences.

Better still, you can get a DSD converter for less than $1000, and coupled with about $3000-$6000 of great mic pres and microphones, that means for less than $10K any person could simply go out and buy a recording chain equal to any in history (pretty much). Then they can put the stereo mics (minimalist!) in front of any acoustic performance, capture it to DSD, and sell it to Pono listeners and anybody else who's got that playback capability.

Compare to YouTube playing everything back at 128kbps lossy joint stereo or whatever they're doing to save bits this week (could get worse!) and you're looking at a whole, distinct, LABELLED market for exceptional quality audio and human performances on musical instruments, in a way that's possible for newcomers to get into. All you need to do is make your recordings and try to sell them as 'Pono-ready DSD recordings' and people will immediately know what that is and why it was done.

I'm not going to be making DSD recordings, I'm going to be making 24/96K recordings, getting my sound with that, and saying 'um these are still quite good you know!'

But I was immediately pleased to hear DSD was being added: and that now becomes my big curiosity factor. I'm going to want to buy some kind of probably symphonic DSD content so I can check that out. It ought to be a lot of fun!
The thing about DSD is that it is good for maybe one thing: straight captures. Typically from master tapes. Any modifications to the signal must be done prior to A/D conversion.

Apparently it is very hard to edit/mix with. So, unbeknownst to a lot of audiophiles, DSD files are actually linear PCM conversions. Bwahahahaha! Bwahahahaha!
Old 18th February 2015
  #5216
Lives for gear
 
stratology's Avatar
 

Quote:
Originally Posted by chrisj View Post
Compare to YouTube playing everything back at 128kbps lossy joint stereo or whatever they're doing to save bits this week (could get worse!)

Complaining about YouTube again?? The default format of YouTube content has recently changed from Flash to HTML5, so, if anything, there may be a quality improvement in the future.


All YouTube content is provided for free, a large portion is not authorised by the artists.
Complaining about poor quality of free content or pirated content is bizarre at best.
Old 18th February 2015
  #5217
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by doom64 View Post
So, unbeknownst to a lot of audiophiles, DSD files are actually linear PCM conversions.

You are posting false information. Please post links to something credible showing that DSD is actually linear PCM conversions.

Quote:
Originally Posted by UnderTow View Post
DSD is outdated technology and the only reason it is still around is for marketing reasons).
Sure, that and the fact that the sound is far superior to PCM.
Old 19th February 2015
  #5218
Lives for gear
 
doom64's Avatar
Quote:
Originally Posted by Sounds Great View Post
You are posting false information. Please post links to something credible showing that DSD is actually linear PCM conversions.
Before stating that I'm posting false information try a Google search.

Sterling Sound's PCM to DSD mastering | Steve Hoffman Music Forums

Sites like https://www.nativedsd.com/ wouldn't exist if this weren't the case. My words were confusing though. I should have said that MANY DSD files are PCM conversions. Certainly not all, which was implied by my previous sentence that they must be sourced from tape for DSD to make any sense.

If mastering engineers/record labels/musicians/etc. would just go back to not hypercompressing CD releases we wouldn't need to be having this debate. Uncrushed 16/44.1 sounds fantastic!

If DSD were superior to PCM then it would be easier to manipulate. The fact that DSD --> PCM sounds fine but PCM --> DSD doesn't says a lot.
Old 19th February 2015
  #5219
Airwindows
 
chrisj's Avatar
Quote:
Originally Posted by doom64 View Post
If DSD were superior to PCM then it would be easier to manipulate. The fact that DSD --> PCM sounds fine but PCM --> DSD doesn't says a lot.
Not at all: the intractability of DSD is part of what makes it interesting. Makes it a 'capture medium' where it's decidedly nontrivial to do crazy things with it. I personally feel that any combination of DSD and super-high-res digital ought to suffice, but I'm open to somebody feeling that only a pure DSD stream (see what I did there? Doesn't that sound like a fresh mountain spring? ) will 'feel' totally right at all times to them.

To rephrase: if someone came and said 'every now and then I get the sense with the PCM->DSD that there's a digitalness to it I don't like, it feels more flat than it should', I'd have to say maybe they had a point (even though maybe they're AWFUL picky, and I'd like to have their playback system)

If someone came and said 'I can tell that every time, it's obvious to anybody', I would say bollocks!

But I'm not asking for that standard, that is an absurdly crude standard. This is the Pono thread, and we're talking about supporting a listening experience where people can settle into extended, familiar getting-used-to listening with treasured audio and music, stuff that casts a spell and puts out a sonic picture you can bathe in. And we're talking about soaking up all the impressions from this experiencing, including the tiny hints that say either 'hey, is that a mahogany back wall?' or 'DJSBFIESGIBHVDSYIFHSDBFHKDGFIUH' deep in the background that you'll never hear, but maybe every now and then you get a glimpse of it, to be thereby enthralled or confounded.

Raw DSD has a place. Again, the fascinating thing about it technically is how the error tracks with frequency. I made a dither that approximates this behavior with PCM, but nobody else has and the interesting thing about the behavior is, we know everybody hears 1K and 100 hz etc. real good. This is no longer 'nobody can hear past 20K', this is now 'fine, have super linearity beyond other digital formats down in the meat of the audio band, and the digital crap can all sit above 20K being not perfectly dithered'.

Whether it can be autotuned is not an interesting part of that conversation. We already have perfectly good PCM for that
Old 19th February 2015
  #5220
Gear Guru
 
Sounds Great's Avatar
 

Quote:
Originally Posted by doom64 View Post
If DSD were superior to PCM then it would be easier to manipulate.
I'm talking about superior sound, ease of use is a different question.


Quote:
Originally Posted by doom64 View Post
The fact that DSD --> PCM sounds fine but PCM --> DSD doesn't says a lot.
Sounds fine? Fine is not good enough to me. DSD sounds worlds better than PCM in my experience. At least as far as the finished 2 track.
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