If that is the intent, so be it. Choosing to just simply record the event would reveal the true venue size. ... unless it's the intention to conceal where something was recorded ... sounds like a spy movie.
The room (venue) is replicated in the recording ... is what I said.
Ha ha! I was once commissioned to re-align the monitoring in a prolific local advertising jingle studio. The owner of the studio, also writer and composer, could tell that his control room monitors were not doing a good job of representing what was going down, i.e. the playback didn't capture the natural timbre of the instruments being recorded. He was right, they didn't!
I spent some time with room placement and toe-in to get the sound at the mix position better behaved (engineers work along the length of the console, not just the sweet spot), and then created an EQ preference curve with suitable acoustic analysis hardware and techniques.
When all was completed, the owner played back some of the recordings he had previously made and proclaimed he didn't like the sound of them (doh!). Even today around 35 years later I instantly recognise his jingles on the radio because of the unnatural frequency characteristics caused by the mix being filtered to sound natural in the control room, which means when played back on a balanced system the inverse curve of his monitoring EQ is revealed - LOL.
The same applied of course to some degree in studios around the world in the early days, with the inverse response curve of the JBL, Tannoy, Urie, etc, monitors superimposed on recordings.
There is no such process as "simply record the event". Every recording involves myriad choices. Even recording with a stereo pair involves many choices...mic pattern, mic angle relative to each other, spacing from each other, distance from sound source, floor, ceiling, walls, people...
It's not about concealment, it's about creating a believable illusion.
You missed the point: there is no replicating the room, ever. That's why each recording is just a representation, an acceptable illusion, perhaps not even of the original, but something new and enjoyable.
You can't replicate the space, especially with two-channel stereo, because you can't replicate the original sound field with two channels. Human hearing is spatially sensitive, it can determine each sound vector (distance and direction) in a 3d space. You cannot replicate that in reproduction, and you can't capture that in any venue. Mostly, we don't even try because it works so poorly. So choices are made to create something entirely new that represents the original, or in studio recordings, is something entirely new that never existed in a space at all. And that's why, in reproduction, we have interminable discussions about things like "soundstage" and "imaging" which are about speakers, position, and mostly the acoustics of the space they are in. It's the Circle of Confusion mentioned above.
Just to head of the inevitable binaural discussion....binaural fails because it depends on recording sound using the exact HRTF of the intended listener. However, each listener has a rather individual HRTF, so recordings just generalize, which means it's always at least a partial failure. Binaural recordings are also strongly affected by headphone type and position, but since each type is so radically different, you have another failure. And binaural simply never works on a pair of speakers, so now the recording is either hobbled by a narrow target audience, or a second standard recording is made simultaneously. Either way it strongly impacts production cost with little end benefit.
Had similar experiences, but with an acceptable rate of acceptance so long as I spent time educating the user, including allowing for time for adaptation. We have much better tools today, now that high resolution time-domain measurements are affordable along with the tools to address the time-domain issues.
You mentioned UREI...the 807 used to be a go-to monitor for me, but I found that sometimes you didn't get a really good matched pair, so they didn't play nice together.
I'm so glad we're past 1/3 octave EQ and RTA.
I understand that and accept the fact that stereo playback is a compromise. However, I can still tell, from a recording, what the venue was like to be in when the recording was done. Is it going to be an exact representation, of course not but it will be good enough to create the illusion if the loudspeakers are sufficiently able to produce a reasonably accurate sound stage.
So stereo is a compromise. What else is new?
Actually, not at all. You're getting the intended illusion. We can, quite literally, change the size of the room you perceive with one control.
Speakers never, and I emphatically say "NEVER" reproduce an accurate sound stage. The reproduce an illusion. It bears no resemblance to the original, which you could only verifiy if you sat in a live venue of any size, any music type, and compared what you hear to the recording. Accuracy isn't even the intent. Because it's impossible.
Well, you say that, but then you're going for "a reasonably accurate sound stage" out of two-channel stereo. If you could understand just how much of a compromise two channel stereo is, we might move forward on this.
BTW, I differentiate "two channel stereo" from "stereo" because the word "stereo" does not mean "two".
In the 1930s, Bell Labs did some extensive testing on how many channels would be required for "accurate" reproduction of an orchestra on a stage. They concluded that a grid of hundreds of microphones, playing back on a similar grid of hundreds of speakers would be needed. Essentially, the were trying to reproduce a sound field. They also recommended that the absolute minimum number of speakers/channels for acceptable "stereo" reproduction is 3, left, center, and right. We got two because 3 wasn't as practical, not because it's better, or "because we have two ears".
Stereo: from the Greek stereós, "firm, solid". No reference to "two".
But that's not really the point. Two channel stereo (or "stereo" if you insist) can not ever replicate the original space.
We're working with what we've got jaddie.
I'm not sure which hole the rabbit went down at this point in the discussion.
(steppin' back and dusting off my pants)
I'm not sure which hole the rabbit went down at this point in the discussion.
(steppin' back and dusting off my pants)
Well, you said:
"Choosing to just simply record the event would reveal the true venue size" - not true.
"The room (venue) is replicated in the recording ... " - not true.
"I can still tell, from a recording, what the venue was like to be in when the recording was done." - rarely true, usuallly not.
I'm not just picking on you, but there are others reading this that deserve the correct information. I wen't on to explain why.
Sure, we can stop now.
Certainly in a normal size listening room there are no dimensions sufficient to support a 100ms (first) reflection.
What does Toole say? I can’t remember the numbers but none are that large. RT60 is a measure used in large spaces, that looks at 60ms.
Longest first refletion in my largish room should be on the order of 24ms.
dave
What does Toole say? I can’t remember the numbers but none are that large. RT60 is a measure used in large spaces, that looks at 60ms.
Longest first refletion in my largish room should be on the order of 24ms.
dave
It's just terminology. In acoustics, anything 100ms or less is an early reflection. Yes, there are much earlier ones, and they certainly have a different effect.
Jaddie, those are some very good posts you have put up since this morning. Thank you.
I believe under 15-20ms is considered a very early reflection and one which will have a deleterious effect on imaging.
I believe under 15-20ms is considered a very early reflection and one which will have a deleterious effect on imaging.
I got tired of looking for any audible difference for the good of the overall SQ by messing with the AVR's time delay on 5.1 channels.
I finally left everything at 0 Ms. And then, as I said, the multi-channel Yamaha receiver amplifier went to rest in the attic, waiting for a new owner.
But its price has been devalued so much! Is there now 9.1? 10.2? Oh come on, stop playing the game these guys are proposing !
In Mr. Bell's example that jaddy mentioned, how many milliseconds of delay would apply - manipulating a DSP - to each of those hundreds of speakers ? 😢
They are only theories that put into practice are impracticable, if I may use the pun.
I finally left everything at 0 Ms. And then, as I said, the multi-channel Yamaha receiver amplifier went to rest in the attic, waiting for a new owner.
But its price has been devalued so much! Is there now 9.1? 10.2? Oh come on, stop playing the game these guys are proposing !
In Mr. Bell's example that jaddy mentioned, how many milliseconds of delay would apply - manipulating a DSP - to each of those hundreds of speakers ? 😢
They are only theories that put into practice are impracticable, if I may use the pun.
Attachments
Actually anything from 1ms to 20ms is a problem. Studios run into this with so-called "near field monitors' sitting on the meter bridge of a mixing desk. Sound can reflect off the mixer.
However, we mess up terminology all the time. I really did search for the correct definition of "early reflections". But we missuse terms all the time. Even the term "near field" monitors is used incorrectly, because the definition must include both distance and frequency. So we mess with "early refelctions" too. Just the way people are.
Differences compensated for by delay settings are subtle, probably inaudible in a live room.
Their tests involved no time delay at all, as that was in the 1930s. Just the naturall time delay of positioning of mics and speakers. But it's not comparable to delay settings in an AVR, which are attempting to match your virtual speaker position to the "perfect" positioning in the mix environment. Yeah, still kind of approximate.
But in surround, you need delay in the surround speakers because early matrix soundtracks had very poor adjacent channel separation (like left > left surround), so they added 20ms to take advantage of the Haas/precidence effect to "fake" separation. It did sort of work. Did it in theaters too.
Here's another one: 5.1 isn't 5.1 channels. The .1 is supposed to be a fraction of a full bandwidth channel, but if a channel is 20kHz wide, and the LFE is only 125Hz and down, then it should be 5.00625. But that sounds terrible in marketing, so it's "creative rounding" to 5.1.
And that makes 2.0 meaningless!
And that makes 2.0 meaningless!