Some commonsense

[esldude]

11 hours ago, mansr said:

I understand you want to compare something between a DSD256 file and a downsampled version of it. You also mentioned spectrograms.

 

Starting with a DSD256 file from a NativeDSD sampler (track 5 from this album) I happened to have handy, here's what I've done:

1. Downsample to 44.1 kHz.
2. Upsample to DSD256 rate, keeping 24-bit sample precision.
3. Subtract the original DSD samples (as ±1 values).
4. Plot spectrogram of difference up to 44.1 kHz.

 

Above 21 kHz, we see a little high-frequency content from the recording that was lost in the 44.1 kHz conversion. Below that, nothing. The level of the difference in the 0-20 kHz range is -170 dB RMS, -155 dB peak. This is too small to be representable in 24-bit precision.

 

Is this anything like what you were looking for?

So peak levels in that are maybe -80 db FS in the 20-25 khz range.  So if your 0 dbFS level plays back at 120 db SPL, these ultrasonics would be 40 dbSPL.  Considering hearing around 20 khz if you have it has a threshold of about 100 dbSPL, I'm going out on a limb here and saying that isn't going to be heard.  This is before we consider masking by louder lower frequencies too.  

[esldude]

3 hours ago, Jud said:

 

Is my impression correct that the differences there were less than with the 2L samples?

It has been awhile since I looked at those.  And depends upon which recordings.  I can look perhaps and get back to it later. 

 

Okay happened to have this one on file.  2L38.  192/24 minus 44.1/16.  The background here is set to go to light gray at -80 dbFS.  Light blue is the first color just above that.  Very little there.  I'm showing the left channel as it had slightly more content left than the right channel. 

 

 

Here is one where I took one channel and implemented a brickwall filter at 22 khz in the upper channel, and left the lower channel alone.  Same settings goes to light gray at -80 dbFS.  A little more, but not much there above -80 db FS.   The RMS value for what is left is - 69 db.  

 

[esldude]

This whatever it is that makes high sample rate better sure is well hidden.

[esldude]

7 minutes ago, jabbr said:

With all due respect to those who insist that 16/44.1 captures all that we could possibly hear, none have you have presented to me an audio reproduction system which sounds entirely realistic, to me, and in my eternal hope that future audio reproduction systems will improve on the current state of affairs, common sense tells me to preserve every bit of a recording --  holding out the real probability that we will need new types of recording, yet nonetheless.

 

Said more technically: y'all are entirely forgetting nonlinear mechanisms.

Hint: entirely realistic audio is not going to result from going to higher sample rate.

[esldude]

Just now, pkane2001 said:

 

You're on the right track. If only there was a way to add zero samples to a signal without affecting it. If only it could be done.... 🤔

 

https://dspguru.com/dsp/howtos/how-to-interpolate-in-time-domain-by-zero-padding-in-frequency-domain/

[esldude]

1 minute ago, pkane2001 said:

 

Also works by zero-padding in the time domain:

 

https://www.dsprelated.com/freebooks/mdft/Zero_Padding_Applications.html

That must be why you see people say the time and frequency domains are opposite sides of the same coin.  Hmmmmm. 

[esldude]

1 hour ago, Paul R said:

 

Seriously? There is at least 11 years of this argument all over this system. 

 

Notice I did not try to say hi-res was better, merely different. 

 

I certainly am not going to record at 16/44.1 or even at 24/48. I do have to trust what me ears tell me. That’s true even when they scream at me that something us wrong and I can not figure out what, exactly. 

 

You can if you wish, and I am sure your work will be very good indeed. (Said in serious way, not joking, sarcastic, or ironic in tone. )

 

 

 

I've done it at 44 and 48 with 24 bits.  And at 88, 96, and 192 rates.  Neither I nor the musicians had a preference.  

 

The two musicians with the best ears, a young girl and a middle age lady who also has a masters in music when asked to compare listened.  Listened some more.  And one said, "so what is supposed to be the difference?"  I'd asked them to listen to some different versions and tell me which was better.  When I told them the sample rates were different, the lady said, "Oh, I thought you changed something else. I've wondered what difference that would make.  I've been told it would sound better the more you sample it." Shrug. 

 

So I've mostly stuck with 48/24.  

 

I'm sure my gear was no good, my mike was wrong,  none of our ears were adequate, the moon was out of phase, it was recorded below ground and played back above ground,  playing by musicians when facing north and listening was facing southwest, a cold front was coming thru, global warming is causing many issues etc. etc. etc. ................

 

 

[esldude]

1 hour ago, sandyk said:

 But can he record classical material as good as that as George Graves has done commercially ?

So do George and I need to have a high resolution "record off" now?  Hahahahaha.  

 

I haven't claimed to be a great recordist.  In fact what surprised me, is once you learn the basics and do get a little experience, just how easy it is to get nice recordings.  Key being, decent mikes in the right place, a good room or space and good musicians.  You get those and there isn't much to it other than having enough sense to not process it to death.  

 

Now when you get to multi-tracking, difficult rooms, bunches of retakes and processing well I'm able to do okay(maybe), but experienced people do much better than me.  OTOH, experienced people also turn out work much worse than mine much too often.  The how to do it isn't really the main part of what makes commercial recordings sound like they do.  There are all these other reasons.   And many of the things done means you have no chance of getting the recording to reproduce like the real thing.  It isn't on the recording at least 99% of the time. 

[esldude]

1 hour ago, Paul R said:

 

I think you may be on ta sumthin... you recorded where three ley lines crossed didn’t ya? 🤪

How did you ever know? One of the musicians had written some short stories about let lines.

[esldude]

9 minutes ago, Le Concombre Masqué said:

Any event... how about "time-smearing" and minimum perceptible interaural time difference ?

Check and check.  You are all good with a touch over 40 khz sampling.  

 

Was it here that someone put up the link to a recent paper on lowest interaural time difference perception.  It indicated with the right conditions it might be a little lower than 10 microseconds.  Might be around 7 or 8 microseconds.  Excellent methodology.  Nothing like the ill conceived stuff by Kunchur.  The kicker is they used 48/24 to present signals and make this determination.  It has the ability to portray interaural time differences well below microseconds and was fine for their work.  

[esldude]

5 hours ago, Le Concombre Masqué said:

placebo effect reading 24/96, ie,  on that Springsteen's track I mentioned is a helluva drug then. Buy the 16/44 with less presence and and slightly blurred layering, IMO, as much as you like as long as greedy marketers keep selling 24/96 to happy fools placebo sensitive persons like me

Do you know if it is the same master?

[esldude]

1 hour ago, Le Concombre Masqué said:

how would I know for every commercial release?

I doubt it's trumped : subtly better : more presence, better delineation, consistent with the idea of better transient reconstruction and no time smearing. Worth it with my system and personal mania but not saying in a MF's manner "digital was shit and I was right to promote vinyl till 24/192 came and its day and night compared to 16/44"

Well that was my point.  Quite often the higher res is mastered a little differently.  So sure it could sound better, but you don't know the sample rate has a thing to do with it.