|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
65.19.76.104
In Reply to: RE: Cable article link posted by jrlaudio on October 17, 2013 at 21:20:37
I have recordings made at 96/24 on microphones that extend well beyond 40 kHz and I play them back on speakers that extend to 40 kHz.
Frequencies in the range of 20 to 100 kHz at normal levels will not create audible IM distortion through properly designed amplifiers and transducers.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Follow Ups:
I agree that amplifiers can pass these ranges within their "power bandwidth". However, not always so in terms frequency response or phase response stability. Additionally in other electronics these signals can be inherently unstable and cause adverse interaction when seen strictly in terms of amplitude and phase.Also these frequencies, if recorded at high enough levels, can have an adverse effect upon accurate metering, whether dBm or dBFS scaled. They can also impact the performance of the AD/DA converters, regardless of bit depth or sample rates. Case in point, at one session on an SSL console we were seeing a substantial signal on the console channel for an electric guitar amp. Unknown to us a bad tube was self-oscillating ultra-phonically and the microphone was passing this signal to the SSL. Of course the SSL "super analogue" circuits were passing this along to the HD I/O of Pro-Tools. It was higher than the guitar level being played through the amp. The meter stayed fixed at one level even when the guitar was being played. However on the Pro-Tools HDX interface this was not showing on the output of the converter, but was showing on the input. The recorded signal was terrible, due to massive error correction going on at the a/d converter or possibly simple overload. All the while what we were hearing sounded normal from the "super analogue" output from the console through our near-fields and the amplifier itself sounded fine acoustically. Switching in a 22kHz low-pass filter on the SSL is what allowed us to discover the problem. Obviously we couldn't hear it. We rolled through the tubes in the amp, found the offending tube, replaced it and moved on. Problem solved.
Having these frequencies in a recoding may also lead to instability issues and wasted power in playback of the end product, especially in consumer level playback equipment. And this can certainly lead to IM artifacts within the "normal" audio band. Especially in tweeters that cannot produce these frequencies. In certain cases this can actually lead to unnecessary and completely avoidable heating and subsequent failure of those drivers.
The reason why the response of professional amplifiers and microphones is extended out to these frequency extremes is not to capture or reproduce them, it is to extend any instability out to frequencies beyond the normal working passband therefore reducing band edge interaction within these devices themselves. However having wider power bandwidth on an amplifier should never infer stable performance above the normal stable 20kHz frequency response. It only assures stability at that limit.
These and other reasons are generally why these frequencies are filtered and not because we feel they do not exist, it is simply good engineering practice. Even this statement is not absolute since one can extend this higher, based upon the requirement of the recording. One case being ultrasonic modulation of audio frequencies that occur in pipe organs that effect timbre of certain stop combinations. However these conditions are rare and specific.
Edits: 10/20/13 10/20/13 10/20/13 10/20/13
I am not interested in junk.
If audio electronics can't properly pass signals that affect human hearing then they can not be said to be high quality. The same can be said for recordings. Amplifiers should be unconditionally stable with all signals into all rated loads. If they aren't its poor design and/or poor manufacturing quality control.
A similar comment applies to DACs that clip when driven by analog signals with inter sample peaks. They are simply not competently designed, or perhaps they are dishonestly designed so that they measure better on common tests.
I am not interested in cheap junk or crappy recordings. There are enough of those in the world, but they do not provide an excuse for poor design. As to recording engineers that deliberately produce audible clipping distortion: they should be summarily executed.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
Again I entirely agree, as you can see by the type of gear I work with routinely. However, in terms of good engineering practice it makes no sense to record frequencies as high as 60kHz and pass them on to the mastering house, or even the end user. This is junk.However, the reason for having recording gear with this capability is actually to insure the highest possible stability within the bandwidth normally associated with audio. However that does not mean it is good practice to use all that extended bandwidth, if fact due to instability near any band limit no matter how wide it is poor practice to do so.
Personally, I will make the leap and state that I do not buy into the idea that ultrasonics in isolation have an effect on a listener in an acoustic performance. However as I mentioned, they can have a timbral interaction with frequencies that lie within the normal audio band, such as in an orchestra or pipe organ. You cannot perceive those ultra-high frequencies per se, however you can hear the effects of its interaction with others. And that can be recorded even if filtered, since it is an acoustic phenomenon and these effects from interaction fall within the normal audio band pass.
I suspect the "hard hitting" engineers you refer to don't understand the differences between dBu, dBm, dBVU and dBFS, or even dBTP, a very common problem. Where I have a real problem these days is in brick wall mastering, ARGH!
Edits: 10/20/13 10/20/13 10/20/13 10/20/13
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: