Hi Def... Analogue Master Quality?

Posted by: DaveBk on 11 November 2011

Question... Given we all are interested in Hi Def, and at present there are quite a few classic albums on HDTracks, what is the real definition of Analogue Masters from the '70s? To expand my question -

Were high range microphones used with reasonable sensitivity over 20kHz?
Was there any analogue high frequency rolloff applied?
How much of the original HF on the tape has been lost over the last 30-40 years?

I'm just curious about whether tapes from this era actually captured much over 20kHz.

Views?
Posted on: 11 November 2011 by aysil

Interesting question about the longevity of these tapes! I am sure the highest frequency response of equipment in the 70's were compromised compared to current counterparts, but nevertheless, some of these recordings from that era are still considered as state-of-the-art of recording engineering! It depends more on the mastery of the recording engineer.

 

This probably does not mean these recordings will sound as good when they are transferred to hires files. There should be another kind of mastery in play during remastering and transfer, with many parameters. It depends on the success of the implementation.

Posted on: 12 November 2011 by Simon-in-Suffolk
Dave, interesting point. I did a quick bit of browsing and looked at the professional analogue tape master player, Fostex G16, and it's frequency range was 40Hz to 18kHz. Still looking for dynamic range... Which might be a better reason to suit Hidef!
But... I mused on this in the summer, but I feel in many way hidef is a means of reducing the distortions and compromises of digitally capturing an analogue signal and it's playback, even though the bandwidth of the analogue signal is relatively modest. Again hidef dynamic range in the master is usually beyond conventional loudspeaker technology,
Simon
Posted on: 12 November 2011 by formbypc

It sounds like you think the crucial aspect of 'Hi-Def' is presence or absence of High Frequencies..... I think there's more to it than that

Posted on: 12 November 2011 by DaveBk

I started thinking about this after looking at the Computer Audiophile thread referenced here where people were posting FFT spectrum analysis graphs of various HiDef tracks. In this, the only interest seemed to be proving that there is energy in frequency bands over 20kHz and whether its distribution looked natural or an artefact from the encoding and transfer process. My interest was also spurred on by comments made at the Tom Tom festival by Doug Graham about how the older pre CD masters represented a marvelous source for HiDef.

 

So, factors such as the much higher dynamic range of 24bit material seemed to be getting lost in the debate over higher sample frequencies.

 

Then, I started wondering what the frequency limitations of the recording process were, especially in the pre 1980's period - was there really that much captured over 20kHz? If not then the increased dynamic range could be the most compelling argument for HiDef, but I see nobody discussing 32bit material yet plenty about 176/192kHz stuff.... just a thought.

Posted on: 12 November 2011 by pcstockton

yes 24-44 brings a ton to the table over 16/44.  Difference between 24-96 and 24-192?  Not so much in my system.

 

-p

Posted on: 13 November 2011 by Simon-in-Suffolk
Dave, I guess another factor in the CD era, compression has become far more common. 16 bit transcodings for CD will tend to be compressed, where as the master won't be as processed as much. So getting access to these masters will benefit. However I suspect a master transcoded to 16/44.1 without compression will sound incredible.
What I do notice with 24 bit with classical music, I can move the volume control more to listen into the music in a way I can't with 16 bit. I guess this is unnatural but fun. It's a bit like zooming in on a Mandelbrot set.
When it comes to 32 bit I think at these depths, it's less to do with direct replay but more to do with increased accuracy for processing. Therefore the word size between DSP and the DAC or other DSP would benefit. Currently this tends to happen in individual components so I am not aware of a standardised interface for these higher word lengths.
Simon
Posted on: 14 November 2011 by AMA
Originally Posted by pcstockton:

yes 24-44 brings a ton to the table over 16/44.  Difference between 24-96 and 24-192?  Not so much in my system.

 

-p

It matches my experience so far.

But I plan to arrange a dedicated session with some 24/192 material which is claimed to be demonstrably superior to 24/96 when played through Linn KDS. We'll see.