Saturnus,
You appear to be asserting that, for example, events that occur "between" samples, or events which occur closer together than 22.7µs, will not be accurately represented in the eventual analog output. This is a commonly held belief. It is also wrong.
Every signal or event is composed of sine waves of different frequencies, amplitudes and phase relationships. This can be proven with Fourier transformation. So provided that all of the frequencies concerned are less than half that of the sampling rate, their frequencies, amplitudes and phase relationships will be accurately captured and reproduced. Julf is right, 6µs difference is accurately conveyed.
Last edited:
To be fair, it may not mean that the track was sourced from MP3. It may mean, for example, that many of the samples used to construct the track were sourced from MP3. Many synths and drum machines etc also output at quite low sample rates and bit depths.
True. It still implies that "hi res" doesn't buy you much extra mileage on those tracks.
I'm trying to wrap my head around this whole Pono 192k/24bit FLAC format that will be available at the Pono music store. How come they didn't just use say straight up 192khz/24bit pcm wav format? Isn't the whole point of having hirez digital audio to have source material that was recorded and captured at 192k/24bit wav? Or is that incorrect? I don't get that?
IS the FLAC just for the condensed storage space and transfer speed aspect?
But then when I think about it....not many studio recordings out there in the world currently have ever been captured that way either. Most of the current recordings out there now are being done at 88.2/24bit or 96khz/24bit wav format AFAIK. Now I'm sure there are some people and studios recording at 192k...but still?
And a lot of the record company catalogs of past recordings were all cut on 15-30ips analog tape.
Then there's CD's which were made at 44.1k/16bit and those have been stamped and created from all sorts of analog and digital masters.
So what exactly is the process for creating 192k/24bit FLAC?
Is it just a SR conversion from something of lower resolution or are these going to be analog remasters, digitized using 192khz/24bit A/D converters or what?
Can anyone enlighten me?
That Beck article about the MP3 conversion to HiRez just feels like a scam and an easy one to pull off. And how would people know that unless they had Jurgen golden ears or actual test equipment to detect it if that's even feasible?
I just don't get it?....what is it were really listening to? Hirez what?
And buying a $400 hirez format "player" or even a $40,000 hirez format "player" doesn't really mean anything if the source material you're playing through it isn't....
This is the kind of thing happening in the HD video world for awhile now too.
People just start calling something 4k, HD 1080p, 1080i but the source material is 720p and it will always be 720p.
Man....seems like this same slippery slope with all this digital goo....
🙂
IS the FLAC just for the condensed storage space and transfer speed aspect?
But then when I think about it....not many studio recordings out there in the world currently have ever been captured that way either. Most of the current recordings out there now are being done at 88.2/24bit or 96khz/24bit wav format AFAIK. Now I'm sure there are some people and studios recording at 192k...but still?
And a lot of the record company catalogs of past recordings were all cut on 15-30ips analog tape.
Then there's CD's which were made at 44.1k/16bit and those have been stamped and created from all sorts of analog and digital masters.
So what exactly is the process for creating 192k/24bit FLAC?
Is it just a SR conversion from something of lower resolution or are these going to be analog remasters, digitized using 192khz/24bit A/D converters or what?
Can anyone enlighten me?
That Beck article about the MP3 conversion to HiRez just feels like a scam and an easy one to pull off. And how would people know that unless they had Jurgen golden ears or actual test equipment to detect it if that's even feasible?
I just don't get it?....what is it were really listening to? Hirez what?
And buying a $400 hirez format "player" or even a $40,000 hirez format "player" doesn't really mean anything if the source material you're playing through it isn't....
This is the kind of thing happening in the HD video world for awhile now too.
People just start calling something 4k, HD 1080p, 1080i but the source material is 720p and it will always be 720p.
Man....seems like this same slippery slope with all this digital goo....
🙂
Because it is exactly the same digital waveform data. Does your excel spreadsheet values change if you zip and unzip the spreadsheet file? No.
Quite a lot is still done at 48 kHz.
The assumption is that it will be the same stuff as you already get on HDTracks and other "hi-res" stores. Some actual 96/24 recordings, some upsampled from SACD, some upsampled from CD, and some re-recorded from (often Nth generation) tapes.
Fortunately the only test equipment you need is some (free) spectrum analysis software (such as audacity).
Exactly!
Not sure you can blame it on digital - digital just makes it easier to fake. But that is true of the actual musical performance too...
It will only be "hi-res" (in quotes). It will be a 192 kHz/24 bit file, but the highest frequency content will only be 22 kHz, and the dynamic range will the that of 16 bits.
Will it matter? Probably not. 44/16 is quite enough to represent any musical signal accurately, and a lot of people are of the opinion that the only benefit of hi-res (once the material has been processed into the final product) is the softer anti-alias filtering allowed by the higher sample rate - but then, why would you pay extra in terms of the premium price of the "hi-res" recording, storage space and bandwidth when you can just as well do the upsampling in the DAC?
I don't understand the hype about the player. Similar players have been out there for long. Even iPods play lossless music.
However, I have high hopes for the music store. If they manage to offer a good selection of highres music for download, that's a big break through. HDtracks and others do it already, but their music selection is lame compared to what's available with lossy music (just look at the iTunes store!). But if the hype about a yellow Toblerone playing FLACs helps to make a high-resolution music store with a decent selection become reality, then I am a pono player fan, too!
Cheers
mbrennwa
---
audioroot | Custom HiFi gear – custom audio for you
However, I have high hopes for the music store. If they manage to offer a good selection of highres music for download, that's a big break through. HDtracks and others do it already, but their music selection is lame compared to what's available with lossy music (just look at the iTunes store!). But if the hype about a yellow Toblerone playing FLACs helps to make a high-resolution music store with a decent selection become reality, then I am a pono player fan, too!
Cheers
mbrennwa
---
audioroot | Custom HiFi gear – custom audio for you
I maybe wrong be I was under the impression that it was not possible during recording in that each sample couldn't predict the value of next sample. In other words I was under the impression that it required a known set of data. But if you say that the next sample value can be predicted accurately by the preceding one then I have learned something new which is always a good thing. I can't help but wonder though why we need to record anything in the first place when everything can be predicted mathematically.
Even HDTV doesn't reproduce ultraviolet or infrared any better than "normal-resolution" TV - and the analogy really breaks down anyway because the eye works rather differently than the ear.
Instead of throwing up inept analogies, let's go to the core of the matter - do you actually believe that the Nyquist-Shannon sampling theorem doesn't apply, or is false?
No. I am not. I am talking about data capture and letting people buy the original recorded material. Every representation of the same information with less data may be accurate but is still just a data compression and there are several ways to do that.
I was referring to this:
So yes, you are confusing recording and reproduction.
If you represent exactly the same data, and totally accurately, why would you use a non-optimal method?
"Compression" is also a dangerous word - it could mean dynamic compression, or data compression - and data compression can be lossy or lossless. A lossless data compression method uses less space and bandwidth, but recreates the original data with absolute accuracy.
Still no. I was (and still am) under the impression that during data capture, ie. recording, that a single sample or set of samples cannot predict accurately the value of the next sample.
Please note that I specifically said data compression, not compression as a general term, and by data compression in this context we're naturally talking lossless compression. Otherwise the discussion is irrelevant.
Last edited:
You can not totally predict it, but if you know the input signal is bandwidth-limited, you can constrain the possible values of the next sample.
That has nothing to do with reproducing the digital data accurately once you have recorded the data. So yes, you are confusing recording and reproduction.
Behringer SRC2496 performs sample rate/bit depth up/down conversion on the fly.
I recently got one (last week), and initial impressions is that integer upsampling sounds subtley nicer with 44k/16bit source files on cheap headphones.
Next week I will connect to good amp and speakers, and Stax electrostatic headphones and take a closer listen.
If the Pono hi-res download service is to achieve serious audiophile market penetration, the source files will need to be disclosed before purchasing.
If the consumer is expected to pay premium prices for MP3/suspect source files, the world is going to tell them to get stuffed.
I like the concept/promises, but the ecosystem needs to be supported by honest disclosure.
A pocket portable seriously high quality capable playback device will be welcomed by music lovers, and the price is not out of reach for most.
The iPod range of devices offer great functionality, but IME the sound quality has always been sub-optimal.
My Toshiba 18" laptop has spdif optical out on the 3.5mm headphone socket.
This is a feature that would further extend the interest/market of this player if incorporated.
Dan.
Last edited:
So you are agreeing that a minimum sample rate of 192khz is needed to express a resolution of 6µs or better during recording?
No. What makes you think so? I suggest you re-read what a bunch of us have been telling you (and read up on Nyquist-Shannon as well).
I agree. I just don't think Neil Young will be any more successful than HDTracks & Co in getting the labels to honestly disclose the information.
- Status
- Not open for further replies.