It is an interesting thought - but one totally at odds with what I feel about the way we respond to audio visual images.
From persistence of vision to the ability to perceive images from low density pixels our brains seem to ADD to what we see not remove. The insistence of our brains to try to create a narrative from disparate images cut to music, the ability to follow and connect edited scenes leaping from location to location, to respond emotionally at all to what's on a flickering screen goes far beyond the level of light stimulating photo receptors.
I think the truth of our audio visual system is that it was designed for survival. The high density of rods at the corner of eyes being over responsive to flicker being a defensive mechanism against subtle movement at the periphery of our vision. Our peak audio response being somewhat tuned to an adult male scream - anything making a grown man scream being more worthy of our attention than a high pitched baby's cry.
That necessity of having to experience the world around us to stay alive is what allows us to experience films and tv. That we blink when we turn our heads to reduce image overload lends to us not being jarred by cuts. That the fovea at the center of our vision creates a real world hot spot combined with lower resolution at the edges of our eye makes vignettes such a well received effect.
But let's not to overlook the importance of sound as an alert mechanism for unknown danger or music with its bass and drum rhythms providing a replacement for the comfort of a mother's heartbeat in the womb. I could go all Ray Manzarek here but music/sound is literally the soundtrack to our lives and is a very strong memory trigger emotionally at least as evocative as looking at an old photograph.
So what of compression? I don't think it serves to say we throw away unimportant things, I think its more that we respond to change, it's intellectually stimulating and from a survival point of view its important to pay attention. This is why fast cut shows are more exciting, its why MTV captured a generation. To similarly extend the metaphor our eyes dart around to continually refresh the cones, as rhodopsin is bleached by light we have to move the hotspots around to prevent visual fade, this random motion is 100% opposite what long GOP likes.
Electrically? I'm not buying that either. Showscan tests put electrodes on people's heads and showed a wide variety of scenes. There was as much brain stimulation/activity looking at a slowly panning scene as there was in an action scene. What did increase brain activity was increasing frame rate. Up to about 60 fps where it plateaued. My personally opinion is that was the limit then based on mains frequency flicker and current tv frame rates and I would love to see a modern test in a world with higher refresh common now in computer games. Modern fighter pilots have been shown to perceive around 200 fps and computer testing suggests 186 fps is optimal for HD. Yes I think we do focus on things that change in a scene but I doubt that we are compressing the data flow, we are simply not thinking about things that offer no threat of danger.
Lastly why 555nm peak response? Peak green is about the same as a sunny field, so seeing that would be ideal to perceive change against the scenery - pretty handy to quickly see that sabre tooth tiger starting to charge.
Exactly, and I'm extending the comparison a bit by questioning how
editing and style of editing might play into this. The type of visual
content we encounter in the natural world lends itself very well to long
GOP compression schemes. The visual content we create in edited video,
where we might average an edit every 5 seconds (or more) is less
efficient to compress. So if what Khalid told me about how our brains
process visual information is true, highly edited material would put
more stress on the parts of the brain that process vision. My question
then becomes, is this a limiting factor on our ability to perceive a
sense of realism in the type of programming we might expect to watch on
4k or 8k monitors?
I don't know the answer. If I was like my friend Khalid I would drop
what I'm doing, go back to school and do the research. I don't have his
brain or his discipline. --J.B.
blafarm@yahoo.com [Avid-L2] wrote:
>
>
> That's a very interesting concept. Also, it sounds an awful lot like
> how long GOP compression works.
>
>
> ---In Avid-L2@yahoogroups.com, <jb30343@...> wrote :
>
> A random thought on the subject.
>
> I had a very interesting conversation today with an electrical engineer
> friend. Part of his current research deals with the way human (and I
> suppose other animal) brains process visual information. Assuming that
> his references are correct, it more or less boils down to an exercise in
> data compression. My non electrical engineer understanding is that
> signals from our eyes are fed to our brain. Simultaneously, the brain
> creates a signal representing what it expects to see. The two signals
> are compared and the visual signal is discarded except for the parts
> that are different from the brain's predicted signal. Our conversation
> drifted to other things but it makes me wonder how much the act of
> editing might disrupt the processing of visual information and if the
> realism of a display might be limited by factors other than resolution,
> dynamic range, 3D, etc. --J.B.
>
>
>
>
>
Posted by: Mike Parsons <mikeparsons.tv@gmail.com>
| Reply via web post | • | Reply to sender | • | Reply to group | • | Start a New Topic | • | Messages in this topic (25) |
No comments:
Post a Comment