[FYFL]

Perhaps you should try to at least watch that video to better understand the point I was trying to make. And “blind test” was not my idea but self proclaimed “guru” who claims to be “science” driven as well.

 

[bigshot]

I generally don't click on links until it is explained to me why I should. I have other things to do with my time. I choose how I spend it.

 

[FYFL]

I generally don't click on links until it is explained to me why I should. I have other things to do with my time. I choose how I spend it.


So do I.... While trying to walk you step by step what was the topic of that video.

 

[bigshot]

"He touches on high resolution and high frequencies in this video. Claims that it all matters to a certain degree."

ok. that doesn't tell me very much.

 

[gregorio]

He touches on high resolution and high frequencies in this video. Claims that it all matters to a certain degree.

Sure, it can all matter, to a certain degree WHEN MASTERING!

That's not what he said. He was talking about using reference recordings to compare your own work, style with. And mentioned lower fidelity files as less than optimal option due to slight tonality differences.

There are not "slight tonality differences" with lossy codecs, the differences are pretty huge. He demonstrates that fairly early on by presenting the "difference file" results of a null test.

Low res compressed samples shown noticeable roll off in high frequency range.

Yes, lossy codecs remove more high freqs because more masking occurs higher in the freq range. HOWEVER, he specifically stated that Everything in the difference file is "largely masked". You seem to have missed that part and/or not understood what "masking" is! Masking is part of the human auditory system, where frequencies become inaudible in the presence of certain other frequencies. The main way that lossy codecs manage to save so much space is by analyzing the freq content and not encoding those freqs which will be masked/inaudible. So the result, at high bit rates is audibly identical, even though it's missing a significant amount of frequency content. At lower bit rates you have to start throwing away frequencies that are *maybe* not fully masked/inaudible and at lower rates still, you have to throw away more and more freqs, that are more and more likely to be audible.

He was emphasizing caution to use lossy file as your reference in determining your own “style” or approach that you identify with while mastering your own work.
He didn't try to encode or manipulate file as a work project per se but a typical playback solution via DAC or similar.
His point as well as my point is, that lossy files exhibit audible degradation in “sound quality”. In tonality as well as in resolution.

Oh dear, you seem to have completely and utterly missed his point and not by a small margin!! That video was made by the Education Director of iZotope, it's a demonstration of "Ozone" (a complex software package for mastering engineers) and is specifically about it's "Track Referencing" feature, with the new "EQ Match" and other modules.

His point is that you should NOT use a lossy codec for the Track Referencing feature, because even though the missing freqs would be completely inaudible (at higher bit rates) it will affect the Track Referencing analysis and hence the EQ Match and other modules. For example, the EQ Match module could remove freqs that are audible in the master you are working on, that have been removed by a lossy codec in the reference track (because they were masked/inaudible in the reference track).

Clearly, your 2nd sentence is completely wrong, the whole video is about manipulating a mastering project (with Ozone) and is nothing about "a typical playback solution via DAC". It starts, logically enough, by explaining how to choose a reference track in the first place that won't screw with Ozone's analysis, such as lossy codecs and loudness normalized files for example.

While trying to walk you step by step what was the topic of that video.

Yep and there's the problem: You're trying to walk bigshot "step by step what was the topic of that video" but you have completely missed the topic of the video yourself!

G

 

[Paolo Pinkas]

CDs were superior to LPs in just about all specs from the very beginning. CD sound was superior to LPs from the very beginning.

In theory CDs are superior to LPs (for example: accuracy of reproduction, dynamic range and stereo imaging).
Vinyl is not a perfect format and it has got all kinds of problems. Nevertheless lots of people think it sounds the best.
And many have regretted giving away their LPs when CDs arrived on the market.

Michael Fremer once wrote:

"Most of the time I don't hear Vinyl's shortcomings. Most of the time records sound dead-silent and out of that comes a glorious sound closer to live and far more three-dimensional than any CD or really most digital of any kind. Whatever is creating that illusion is not important to me only that it is created.

CD sound is observed by many to sound like unlistenable crap. That is why people no longer sit and listen to music (just an observation but true). Over time science will be used to figure out why the brain does not like CD sound even if the measurements "prove" otherwise. That is because measurements are coarse indicators of reality the brain so well processes."

It is pretty simple: Listen to what sounds best to you.

 

[bigshot]

I have tens of thousands of records in my collection. None of them match the fidelity of CDs. Some of them are well mastered and minimize surface noise and distortion, but that doesn’t mean that the format itself is better. Some collectors fetishize obsolete media. While they say that they love obsolete technology because of the sound, that is only justification of their fetish. It isn’t factual. It’s easy to test to show that LPs aren’t superior, and the limitations of LPs are self evident. But as you say, people can “like” anything they want to. Everyone is free to prefer inferior things for personal reasons.

 

[old tech]

In theory CDs are superior to LPs (for example: accuracy of reproduction, dynamic range and stereo imaging).
Vinyl is not a perfect format and it has got all kinds of problems. Nevertheless lots of people think it sounds the best.
And many have regretted giving away their LPs when CDs arrived on the market.

Michael Fremer once wrote:

"Most of the time I don't hear Vinyl's shortcomings. Most of the time records sound dead-silent and out of that comes a glorious sound closer to live and far more three-dimensional than any CD or really most digital of any kind. Whatever is creating that illusion is not important to me only that it is created.

CD sound is observed by many to sound like unlistenable crap. That is why people no longer sit and listen to music (just an observation but true). Over time science will be used to figure out why the brain does not like CD sound even if the measurements "prove" otherwise. That is because measurements are coarse indicators of reality the brain so well processes."

It is pretty simple: Listen to what sounds best to you.

Quoting Fremer as your source is a bit like quoting Ajit Kulkarni to make a case that homeopathy is superior to modern medicine.

Fremer has always been an irrational digiphobe from way back and attracts likewise irrational vinylphiles (outside this world, he's is seen more as a crank). If you want a good laugh, have a read of his diatribe to the editor of the audio critic and the editor's rational response.

https://www.biline.ca/audio_critic/mags/The_Audio_Critic_21_r.pdf

 

[sonitus mirus]

Quoting Fremer as your source is a bit like quoting Ajit Kulkarni to make a case that homeopathy is superior to modern medicine.

Fremer has always been an irrational digiphobe from way back and attracts likewise irrational vinylphiles (outside this world, he's is seen more as a crank). If you want a good laugh, have a read of his diatribe to the editor of the audio critic and the editor's rational response.

https://www.biline.ca/audio_critic/mags/The_Audio_Critic_21_r.pdf

Wow, and that was back in 1994. There was so much packed into that reply about early digital mastering for CDs. I think that single editor's reply did more to validate all of my other research on the matter of digital audio than than anything else I have read over the last decade.

 

[Davesrose]

Wow, and that was back in 1994. There was so much packed into that reply about early digital mastering for CDs. I think that single editor's reply did more to validate all of my other research on the matter of digital audio than than anything else I have read over the last decade.

I do think it's funny that for the 80s, there was debate about the optimal mastering from the analog tape recording. Since vinyl was an established medium, perhaps your early PCM could have had issues. Now the tables have turned in which everything is based off digital recordings, and mastering for vinyl is the outlier. I could understand collecting records for tangible reasons: IE large cover art, the tradition of wiping the surface of the record with the anti-static brush and flipping the record, and now also even shining a light to view a hologram. I've never understood the argument that one medium sounds more natural then another....the same record on a TT can sound different with a given cartridge and tracking force. Less variables with digital since the source is linear, and only main colorization is amp stage.

 

[bigshot]

That elicited a chuckle from me. At least if you are going to pick cherries, do it on a cherry tree not one full of nuts!

 

[jeremymayf]

Interesting thread, I enjoyed reading the different points of view throughout the 16 pages. It's difficult to deny the science behind the method of digital sound reproduction. I would like to raise one question that I believe has not yet been discussed.

I recall reading about how SACD frequency range extends far beyond that of CD- indeed far outside the range of human hearing. At the time of reading circa 2002, it was argued that those frequencies, though inaudible, interact with the frequencies that humans can hear. And so, the argument continues, this interaction of inaudible/audible frequencies (or lack thereof in CDs) is one reason why we can differentiate between someone talking in a room vs a recording of someone talking. SACD freq response 20Hz to 50 KHz, vinyl 7Hz to 50kHz, wheras CDs can only go up to 22.05 kHz. Including these frequencies in SACDs then, will make the sound they render more lifelike in the same way vinyl did without the degradation of the physical media i.e., warping vinyl, turntable needle cartridges etc.

Now I can debunk part of this myself. The speakers reproducing the sound rendered by the SACD/vinyl media are in all likelihood incapable of reproducing sounds outside the 20 Hz to 20,000 Hz range--negating the argued benefit of including those frequencies. However, if we had speakers that could render from 7Hz up to 50kHz, would including those frequencies in recordings add (dare I say) a sense of realism to their reproduction? Curious what the more scientific responders think? Is it live or memorex?

 

[danadam]

it was argued that those frequencies, though inaudible, interact with the frequencies that humans can hear

If they affect the frequencies that humans can hear, then a microphone will capture that change.

if we had speakers that could render from 7Hz up to 50kHz, would including those frequencies in recordings add (dare I say) a sense of realism to their reproduction?

I suspect it could increase chances of high frequency IMD in your playback affecting reproduction of audible frequencies. It would not be faithful to the source, but it could, for some and by accident, add a sense of realism. After all "a sense of realism" is a subjective thing, I think.

 

[gregorio]

And so, the argument continues, this interaction of inaudible/audible frequencies (or lack thereof in CDs) is one reason why we can differentiate between someone talking in a room vs a recording of someone talking.

Firstly, there is no audible interaction of inaudible/audible freqs. This has been known and demonstrated for many decades and is actually pretty easy to test for yourself. If audible and inaudible freqs interact, all a human can hear is the audible freqs. For example, beyond a certain frequency a square wave can’t be discriminated from a sine wave. That point is where only the fundamental freq is within the audible range and all the higher harmonics that interact and make it a square wave are ultrasonic.

Secondly, those of us who often work with film/TV sound know that there is nothing in human speech beyond about 12kHz. Breath sounds, lip smacks and some other mouth sounds (teeth hits for example) can produce freq content into the ultrasonic range but dialogue doesn’t.

SACD freq response 20Hz to 50 KHz, vinyl 7Hz to 50kHz, wheras CDs can only go up to 22.05 kHz.

SACD has a range well in excess of 100kHz, however it works by adding a great deal of noise in the ultrasonic range. Sony therefore put a filter in all it’s SACD DACs at 50kHz as all that noise could easily cause IMD. And they required 3rd party SACD DAC makers to implement a 30kHz filter.

Vinyl theoretically allows freqs up to about 50kHz but only really under lab conditions. However, the practical limitation for commercial music production was considered to be around 14kHz - 16kHz, as noise, distortion and other non-linearities start to exceed the actual musical signal.

Including these frequencies in SACDs then, will make the sound they render more lifelike in the same way vinyl did without the degradation of the physical media i.e., warping vinyl, turntable needle cartridges etc.

In addition to all the above, controlled testing has demonstrated that at reasonable listening levels no one could tell the difference between SACDs and the same SACDs converted down to 16/44.

The argument you’re quoting isn’t really an argument, it’s marketing nonsense long since discredited and only wheeled out these days by those suffering from marketing induced placebo who haven’t actually done a controlled DBT for themselves.

G

 

[bigshot]

There has never been any test that shows that super audible frequencies make any audible difference in the playback of commercially recorded music. A/B testing between recordings with natural ultrasonics and without show that people can’t discern any difference.