When is Hi-Res not really Hi-Res?

[Bruce B]

But if you zoom in, you can clearly see the slope ends at 20.5k, well below the Nyquist, which is at 22.05

So... this is actually a 24/96 file that has been filtered from a tape machine.

 

[Julf]

But if you zoom in, you can clearly see the slope ends at 20.5k, well below the Nyquist, which is at 22.05

So... this is actually a 24/96 file that has been filtered from a tape machine.

Fair enough - so no real content above 20K. But what (in your view) is the benefit of a 96k sample rate?

 

[Julf]

So... this is actually a 24/96 file that has been filtered from a tape machine.

It is also very different from the Holst example I was discussing - where the energy drops to almost nothing just under 24k, and then goes up again, in a pattern looking very much like DSD noise shaping.

 

[Bruce B]

It is also very different from the Holst example I was discussing - where the energy drops to almost nothing just under 24k, and then goes up again, in a pattern looking very much like DSD noise shaping.

Correct....

 

[rbbert]

Actually it's not. I have a day job (Anesthesia). I'm in the studio from when I get home 'til midnight, where I have almost a 2 month backlog of projects. I will get to this when I have more breathing room.... sorry for the delay.

Bruce, I think you work too hard

 

[Bruce B]

Bruce, I think you work too hard

No... I just spend too much money and try to keep the wife in the lifestyle she is accustomed too!!

 

[garylkoh]

Bruce, the other thing that you might want to mention is the FR of the mics (and mic preamps) being used to do a lot of recordings. Many of the large capsule mics only go up to 16kHz. Especially when a mix of mics are being used, and there are very few small capsule extended range mics, the spectrum analysis end up looking like Redbook anyway.

 

[Julf]

Many of the large capsule mics only go up to 16kHz. Especially when a mix of mics are being used, and there are very few small capsule extended range mics, the spectrum analysis end up looking like Redbook anyway.

Absolutely. But again, the question is then what the benefit of a 96k sample rate is supposed to be for material that really has no content over 20kHz?

 

[Bruce B]

Absolutely. But again, the question is then what the benefit of a 96k sample rate is supposed to be for material that really has no content over 20kHz?

High sampling rates have little to do with freq. domain but more with time domain. But there are exceptions when you use dpa or Earthworks mics into a Millennia mic pre recording muted trumpet or piano at DSD128fs!

 

[Julf]

High sampling rates have little to do with freq. domain but more with time domain.

I am aware of the arguments having to do with the benefits of having to do less steep filtering, but I am not entirely sure I buy that argument. Modern DACs all do oversampling, and thus avoid steep filters on playback. On the ADC side, you also don't need a steep filter if you *know* there is no HF content, so you know there is little risk of aliasing.

 

[Bruce B]

The new Bob Marley - Legend

 

[LL21]

Hi Bruce,

If you are referring to the Remastered Edition of Legend produced by the Estate of Bob Marley, i have it. It is very different than the original CD. I like it a lot. What are your thoughts on it?

 

[Bruce B]

Hi Bruce,

If you are referring to the Remastered Edition of Legend produced by the Estate of Bob Marley, i have it. It is very different than the original CD. I like it a lot. What are your thoughts on it?

These tapes had a whole lot more bass content and much more clarity than the original CD

 

[asiufy]

Bruce,

Did you do the 96/192 version on HDTracks.com?


alexandre

 

[LL21]

These tapes had a whole lot more bass content and much more clarity than the original CD

I observed the same...in particular the clarity was startling. In general, i also felt the entire album felt more relaxed than the original CD.

 

[Bruce B]

The new Otis Redding - Dock of the Bay

 

[Julf]

I am here to learn - so, just for my education, how would the spectrum look like if you took some CD (44.1/16) or 48/16 material, and upsampled it in DSD for SACD, and then took that material and converted it to 96/24 PCM? My naive assumption would be that there would be a very steep drop in content at 22 or 24 kHz, and then a softly increasing noise spectrum (from the DSD noise-shaping) above that, again falling when the low-pass filtering for the PCM kicks in.

So, Bruce, what do you think of this spectrum:

I would say it is a DSD file resampled to 96/24, with low pass filtering applied above 32 kHz. Would you agree?

 

[Bruce B]

So, Bruce, what do you think of this spectrum:
I would say it is a DSD file resampled to 96/24, with low pass filtering applied above 32 kHz. Would you agree?

It could be either:

A pure DSD file with steep filtering above 32k

or

A PCM 24/96 file with steep filtering above 32k

 

[Julf]

It could be either:

A pure DSD file with steep filtering above 32k

or

A PCM 24/96 file with steep filtering above 32k

But if it is a PCM file, it would still have been resampled from something that was in DSD format, judging by that noise shaping?

 

[Bruce B]

But if it is a PCM file, it would still have been resampled from something that was in DSD format, judging by that noise shaping?

Yes, forgot to add that:

A PCM 24/96 file upsampled to DSD with steep filtering above 32k