John Curl's Blowtorch preamplifier part II

[Chris Hornbeck]

Originally Posted by bear

My listening room is likely close to 16 bit depth on winter nights, with the heat off... it's at times scary quiet... assuming you work back from peak SPL, yep think it makes that for the noise floor.

Wish I had my ears minus 20 years or so...

Fair enough, but let me put my concerns another way. All storage systems are both bandwidth and dynamic range limited. Perhaps we can all agree that adding zeros beyond *some* point in unimportant.

For you (or least your younger self) more bandwidth than 22 KHz would help. For most adults, not so much.

A good seat in an unamplified orchestral setting gives peaks of 105 dB SPL, and this should be reproduced cleanly for "realism". 16 bits down is 10 dB SPL, and most folks have to hold their breath to get that quiet. (People are noisy machines in this context.) For typical peak-to-average ratios this (105 dB SPL peak) is an averaged flat 85 dB SPL, mastering level and, to me, very loud.

So for most adults, 16 bits is a comfortable margin.

Thanks,
Chris

[morinix]

A woman at 40yo 22khz more likely. I would have to see tests.

[Chris Hornbeck]

Originally Posted by RNMarsh

The lack of low level resolution is a big deal for digital. The specs for most DAC chips fudge on this with thd et al .... with weighting for noise and rating at full output. Only a few rate the performance at -60 level. The AD1955 is better at -60 than most and so i use that one and it is awfully good.... especially with direct master 24/96/192 downloads with wav file. Now things are starting to get interesting! Stilll need better mic (soundfield) technique. Anyway for audio, the low level resolution may be the last bastion for analog.

Most ADC and DAC conversions these days run at small bit depths and high oversampling rates, which are the same at all storage rates. Only the calculatin' is different, usually.

How is low level resolution different from one sampling rate or storage bit depth to another?

Thanks,
Chris

[fas42]

Originally Posted by RNMarsh

The lack of low level resolution is a big deal for digital. The specs for most DAC chips fudge on this with thd et al .... with weighting for noise and rating at full output. Only a few rate the performance at -60 level. The AD1955 is better at -60 than most and so i use that one and it is awfully good.

Huuhh? I never get this thinking - all reproduction systems have increasing noise and distortion with falling level, there's no way to get around this. And you don't need to worry about it, because the distortion is always below audible levels, unless you run your system at PA sound levels. Vinyl is an example of a replay mechanism that has truly appalling levels of measured distortion, but a lot of people still reckon it's pretty nice to listen to ...

Frank

[RNMarsh]

Originally Posted by SY

So in your world, "resolution" and "noise" have the same meaning? And analog recorders have better noise floors?

Deja vu....

?? In MY world ?!? You're painting with the wrong brush --

Now why would I/anyone hear them as the same meaning?! In analog, I can hear at low recorded levels details in music and the random/hiss background. And, i can hear no random noise but distorted music at low recorded levels in digital. No mystery to that.

[miklos]

Originally Posted by Chris Hornbeck

Who has even a 16 bit listening room? What adults can hear 22 KHz?

Thanks,
Chris

I'm up to 14kHz now and going strong.

[Chris Hornbeck]

Originally Posted by RNMarsh

In analog, I can hear at low levels details in music and the random/hiss background. And, i can hear no random noise and distorted music at low playback levels in digital. No mystery to that.

It's important for us to not give the impression that digital storage has an inherent low level distortion, "stair steps", or missing information below the smallest bit. It's surprising how common these ideas still are, so it's extra important for influential folks like you be "careful with your phaseology". There's no fundamental reason that digital storage, properly dithered, must have *any* low level artifacts, so you could say there's a practical mystery in the implementation.

Thanks,
Chris

[john curl]

People here have many opinions. They are not MY opinions based on my experience in the audio field for such a long time. They are not my colleagues' opinions, either. I don't know how to address the situation.
Today, I got a phone call from Jim Borgornio (sp). He has been amazingly successful in making pretty darn good electronics recently.
The rest here, have a lot to learn, before you can teach me anything much.

[fas42]

Originally Posted by Chris Hornbeck

There's no fundamental reason that digital storage, properly dithered, must have *any* low level artifacts, so you could say there's a practical mystery in the implementation.

Thanks,
Chris

There are no audible artifacts with digital as a format, it's always the playback implementation that's the problem child. I've done this exercise over and over again, of listening to digital attentuated to ridiculously low levels and have always been impressed at the detail that is preserved.

But, I've heard plenty of poor digital playback, that sounds like it has been kneecapped, the detail below a certain level completely evaporates, disappears into thin air. This is faulty playback!!

Frank

[RNMarsh]

Originally Posted by Chris Hornbeck

It's important for us to not give the impression that digital storage has an inherent low level distortion, "stair steps", or missing information below the smallest bit. It's surprising how common these ideas still are, so it's extra important for influential folks like you be "careful with your phaseology". There's no fundamental reason that digital storage, properly dithered, must have *any* low level artifacts, so you could say there's a practical mystery in the implementation.

Thanks,
Chris

yes, you are quit correct and again I wasnt being technical in my listening mode description - (maybe I go in and out of right-left brain )
digital is getting better and better. In fact, I am going towards measuring harmonics of analog designs with a digital ADC/DAC system!

Never-the Less, when replayed back levels are very low and fine detail structure in music isnt resolved... CD 16/44.1 hasnt been good enough and sounded as described. I attribute that loss of detail in music at low recorded levels to the fewer bits being used at low levels in some of the popular playback technologies. Maybe it was elsewhere -- like a bad digital volume control. Then there is/was 1-bit technology and others. I am talking about playback - the AD1955 IS a DAC..... compare its -60 thd et al if you can get the info from others. Cleanness over a wide dynamic range is where analog had a good bit (sic) to offer. Ok so kill me. -RNM