Thursday, September 11, 2014

Perception and Failed Illusions

Perception and Failed Illusions


Does a muffled band in a rehearsal room down the street sound more 'live' to you than a high def recording played through high quality stereo? Or have you listened to a recording and thought "whoa shit, I thought that was band, but it was just a recording played through speakers?"  Is it just live music that defies realistic reproduction or do these same perceptions apply to voices, fireworks and just about everything else we try and recreate through loudspeakers?

It seems there is something about a live band playing, even muffled and far away that divulges its aliveness. A certain rawness, dynamics, dispersion from the sources, perhaps multiple source locations. If these observations are taken to be correct or at least in the realm of close, what is especially interesting is that frequency response appears to have little to do with whether we perceive something to be real-time live. Comparing the muffled band a block away versus direct exposure to Hi-Fi wherein clearly the Hi-Fi has a distinct advantage in presenting a wider and less altered frequency response, yet somehow the live band sounds more real.

Quite puzzling indeed and perhaps this perspective lends some credibility to why so many audio people have such an affinity for audio gear that exhibits less than flat response and less that technically ideal sonic characteristics. Perhaps, just perhaps we are focusing on the wrong aspects as a starting point and that is why sonic realism is so illusive.

Oh, and just a thought on making sound reproduction more real. What if rather than focusing on frequency response and accuracy of the reproducer we instead focus on eliminating introducing aspects that identically alter all the instruments? For example; we can significantly compress a single instrument without undesirable side effects yet even minor compression on the whole mix is readily audible. Think FM radio versus the CD or when the PA vendor puts a left/limiter on the PA. Looking at the dispersion of instrument, cymbals have a 360 degree radiation pattern whereas a kick drum has more of a figure 8 radiation pattern.  Guitar rigs can be figure 8 or somewhat cardioid pointed forward. Each of these instruments have vastly differing and unrelated frequency responses that can be readily altered  by choosing a differing cymbal manufacturer, switching to a felt kick beater changing the guitar cab type, yet none of those changes alter their "liveness".

Frequency response can be altered before or after the sound reproduction process so it is important for things sounding desirable but has little to do with things sound real or live.

So based on this I am thinking I can add the concept:

"Nowhere in nature do multiple unrelated sound sources offer identical dispersion patterns, yet in pro audio we do it all the time"

to the two other fundamental concepts I have been distilling, which are:

"Nowhere in nature do multiple unrelated sounds radiate from a single point in space, yet in audio reproduction we do it all the time."

and

"Nowhere in nature does the exact same sound radiate from multiple points in space but in audio reproduction we do it all the time."

With the underlying idea being that the more we stray from what occurs naturally, the more illusive realism will be.

Ok and on another note, I have been putting together a small playback studio with some spare and older bits of gear from Rat.  I have a lot of old multitrack DA88 tapes from various past tours and am looking to do some multi-channel mix downs for use in sound seminars and demos and such. The tapes have all been backed up to ProTools but rather than buy or tie up one of Rat's Pro Tools rigs and since we have 5 DA98/88/38 series machines laying around, I figured why not bring them home and use them.

What I quickly found is that they were very hungry and had a overwhelming urge to eat tapes. I cleaned the heads and ran the error rate option and all 5 machines were all over the place so I opened them up to see what I could find. What was happening was the tape tension seemed to be low and excess loose tape was building up in front of the pinch roller. For some reason they lacked tape tension.

After a bit poking around and found some good info here that helped a bit and got me in the right area

http://www.tangible-technology.com/machines/springfling.html

And thought I would share what I found was the exact same issue with all 5 machines. The tape tension arm shown below was no longer moving freely



When you press play, the arm should wing to be on the tape as shown here. but instead was barely moving


So, with a very small tweezers and exacto knife, I removed the two plastic split washers


Which are very small, do not drop them in the machine!


Removed the tension spring


Removed the tension arm by gently rotating back and forth with a needle nose pliers. I snapped a wood q-tip in half and soaked it in some WD40 and used it to clean out the pivot hole. Then used alcohol on a q-tip to clean the pivot shaft. Then put a drop of some multi purpose ultra lube on the pivot shaft. Dont put any lube on the tension shaft that contact the tape.


 And then reassembled it all back together and everything worked perfectly and all the error rates dropped to zero. 


And since there is more to life than pure audio. a few adventure pics to lighten the load. There was a most wonderful week of amazing waves that came to visit our southern California shores


but not without consequences for the unprepared and when the waves get big enough to break things


It can become a spectator sport as well




But when it looks so good I just can not resist


DR





5 comments:

  1. Related to what you're talking about, I think a lot about why at a live show, it's usually pretty easy to tell if you're listening to a mic'd vocal (or instrument) or to playback. I think the reason has to do with how feedback alters the quality of the sound at high gain levels, even before a mic starts howling.

    Any microphone in an acoustic system with a loudspeaker will hear itself. In other words, you always get the original sound, plus a delayed and frequency altered copy of the sound, and of that one and of that one and so on. So, first of all this sort of obviously creates its own 'reverb' tail, on top of the sound of the room picked up by the mic. But secondly, we know that a sound mixed with a delayed copy of itself will create comb filtering. And that should get more pronounced as the two signals become more equal in level, which happens as the gain goes up, until the system starts howling.

    So, I think one way to make sounds that are played back seem more like they are being amplified live would be to run them through some processing, an eq'd delay, to try and recreate that.

    ReplyDelete
  2. what are all the jobs in the entertainment industry?
    HOMEBOY RUS - COUNTIN MONEY

    ReplyDelete
  3. thanks… I have been bookmarking them for a while now and just decided to create a post to provide them to others…
    Patented one to many RC based wrist band control

    ReplyDelete
  4. I have to say this has been probably the most helpful posts for me. Please keep it up. I cant wait to read whats next.
    best brand for power bank

    ReplyDelete
  5. I have read many blogs in the net but have never come across such a well written blog. Good work keep it up
    best brand for power bank

    ReplyDelete