Transients and frequencies above 20 kHz

If there wasn't already a lowpass filter in the digital chain, I'd probably would want to add one just to keep all the ultrasonic garbage out of the recording process.

 

I see what you mean. A "Transients and frequencies below 20 kHz" thread would actually be more suited to humans. Problem is, I don't know how many humans post here.

 

At least for me this is a more interesting question regarding digital than the one you posed. Yes, note onsets have a high proportion of noise components (transients) to the sound and less perceptible notes (tones) which are the periodic element. Believe it or not, these transients are what allow the ear to rapidly distinguish different instruments as they provide tone color.

The question I would have is how do interpolation algorithms and the need to deal with quantization noise affect the accurate reproduction of these instrument transients? This has been the crux of my dissatisfaction with digital audio sonics.

I should add the point that I am not sure that sampling rate and # of bits are the whole issue here. I have recently been investigating the digital converters used by Soundstream and Denon in the 70s before Sony Redbook was established. To my ear the Denon and Soundstream Telarc LPs are noticeably more realistic sounding than Redbook LPs of the 80s.

 

But was that due to their digital components or their analog components? Because no two analog decks are going to sound exactly the same, either...

 

I agree that since we can't restore the recording chains of Denon and Telarc and run scientific comparisons there is a certain amount of guesswork. However I would direct you to my comparison of two Telarc LPs from the 70s vs one from the 80s. Both were done by the same recording team using similar techniques and mic'ing. The difference is that one was recorded to Sony Redbook and the other recorded to Soundstream.

http://forums.stevehoffman.tv/threa...-than-sony-cd-pcm.574771/reply?quote=14913315

 

Right, but you realize that there are analog components in all A/D converters, right? And that those are going to sound different as well, even between two Redbook devices.

We also have no idea what other equipment was in the recording chain on those later Telarc recordings. Were they using digital reverbs that didn't even exist 8 years before? Equalizers? Compressors? What kind of mixing board? What kind of EQ? Also, it sounds like the two recordings you're comparing were recorded in completely different halls. Not exactly a valid comparison - a hall will have its own distinct sound, an orchestra will have its own distinct sound, different types of music (classical vs. pop) will have their own distinct blend of instruments and resulting sound, etc. etc. It's apples and oranges.

 

I bought myself some new, fancy, very expensive gear last week. Everything measures perfectly straight from 5-50000Hz. Now I finally can hear every electrical impulse & high pitched error that was hidden in these old master tapes. (I even think I finally heard the mythical, subsonic rumbling from meat loafs belly, listening to the hypnotic heaven can wait.) I feel relaxed now. I finally found peace...

 

Again I agree but if you read my evaluation the hall sonics or any added EQ (if any were done) are the least of the problems noted.

 

The problem is, both of those things can dramatically impact the quality of the overall sound, including the problems you noted. As could the analog stages of the equipment used - both the A/D converter as well as anything upstream.

 

Sunspot I can only do what's possible currently. Since the recording team and set up, mics, preamps etc were the same I have to defer to their expertise in getting the best sound possible. If they had great analog stages or other equipment that smoked 80s gear why didn't they continue to use them? People use old mics decades after discontinuance. But I can't rule out strange things.

 

I agree, what components are affecting the sound cannot be but a guesswork.

 

Music is sooo much better when you can enjoy it, without worrying if your equipment "could be better".

 

Denon DN-023R (1972 - 13-bit/47.25kHz)



Soundstream Digital Tape Recorder (1977 - 16-bit/50kHz) [upped from 37kHz which proved insufficient]




SONY PCM-1610 (1981 - 16-bit/44.1 kHz, successor to the PCM-1600 first available in 1978)
Brochure


DECCA Digital Audio Recorder (mid 70s - started out at 16-bit/48kHz)
[later incarnation pictured below]


The Decca Digital Audio Recorder »

 

I do believe you hear differences - the question is why are they there differences and what do they prove?

The only way you could compare "pure analogue" to "192 PCM" is to have a recorded tape, a tape machine and a transparent AD/DA converter and setup a proper test. But then again - if you have a transparent AD/DA, by definition you can not hear a difference between the analogue waves on both sides of the converter, which means that you can not compare in practice two different formats and all you are comparing is two reproduction systems and that doesn't say much about the formats themselves

Furthermore, even if you hear differences between 192 and the same 192 properly downsampled to 96, it may just mean that your dac plays either of these formats better, which may be due to various reasons, neither of which is inherent to the format being played.

 

It would be interesting to see an FFT of a live performance to see how much information/power actually exists above 20 kHz.

imho it would below the s/n ratio of any recording or playback equipment. So it may be the 'thing' that differentiates live from recorded music I doubt it can be captured on a recording or seperate one method from another because that information is unlikely present/preserved/captured.

 

Some of it's not below the S/N ratio - we can see that on high-sample rate copies of old analog masters. But a lot of that ultrasonic content is also obvious noise - some if it pretty loud - and heaven knows what impact that's gonna have on the audible signal when pumped thru your home stereo.

Probably not a good one, though...

 

I always thought bad PA equipment, B.O. and obnoxious people were the 'thing' that differentiates live from recorded music.

 

Thanks for the info on the Decca Digital recorder. I was familiar with the others. The drawback is that they had to do sample rate conversion to get it consistent with Redbook. Back in the 80s the rate converters were awful. Quite frankly I was never impressed with Decca digital LPs. They are almost all opaque sounding with closed in treble. The two that impress me by the LP sonics are the 034R Denon and the 50k Soundstream.

 

http://www.aes.org/aeshc/pdf/fine_dawn-of-digital.pdf

 

Thanks. I'm aware of it. See thread
Denon PCM Encoding in 1970s. Is it different than Sony CD PCM? »

 

Because whether these analogue stages sounded better or not is subjective to the artist, engineer and the listener, it certainly wasn't transparent and YMMV. Artists use all sorts of things to get a certain sound and sometimes a certain sound can be achieved by happenstance through the equipment chain imparting its own character to the sound. However the general trend since the dawn of recorded music has trying to get the most accurate recording/processing equipment leaving the artistic side of sound to the artist and mixing/mastering engineers.

 

I understand but the point discussed between sunspot and myself concerned the same team producing a much inferior recording using a different (Redbook) digital recorder. Sunspot raised the point that even though the recording team was the same and the recording chain up to the recorder was the same, analog components might have been different/poorer in recording/mastering and spoiled the sound rather than the digital recorder doing so. My response was since it was the same team why would they use inferior analog stage components simply because they were newer? I concede that we can't interrogate the engineers or recreate their hearing test but it is a reasonable point.

 

Yup. And I'd add that in my experience with live music, stuff with that dreaded '80s Sound actually sounds a lot more like live music - especially live amplified music in a live space - than most stuff from earlier decades (including a bunch of well-regarded recordings). I think recording equipment by the late '80s was far more successful at accurately capturing what was going on in the studios, and to some degree engineering techniques had evolved to favor that live, somewhat dry sound (the same happened in classical).

I'd also say I often find that sound fatiguing, even if it is flawlessly accurate.

It's the same reason why many favor tube-based equipment, even if it's measurably not anywhere near as accurate as transistor-based equipment. The distortions and limitations of that equipment actually makes a lot of music more enjoyable. I just question whether the playback chain is where that really belongs - the master itself should be produced to deliver a pleasurable playback experience (and I've always thought that should be a more calculated process, not something reliant on random distortion from certain pieces of kit, but YMMV).