Simon Ubsdell
6
COW Coins
Forum Replies Created
-
[Tim Wilson] “Which still leaves my question — do any of you know where Shake’s tech wound up in the Apple ecosystem?”
Well, quite! Apart from optical flow, which subsequently made an appearance in Motion, I can’t see anything that fits the bill, despite the fact that we hear this notion repeated pretty frequently. And even optical flow – did Apple really need to “borrow” this from Shake? I don’t know enough about the commercial history of this, but I’m far from convinced that it was a proprietary Shake feature that couldn’t have come from somewhere else.
Simon Ubsdell
tokyo-uk.com -
[Bill Davis] “Sigh. “
Perhaps we can conclude that if you don’t understand the need for this feature, then the chances are that your personal editing requirements tend toward the modest.
Simon Ubsdell
tokyo-uk.com -
Thank you for your fascinating reply. Lots of really interesting points in there, but I think you are perhaps missing an essential nuance or two.
Film language is much less about physical geometry than it is about “engagement”, and engagement is a direct function of “distance” both physical and psychological. Because distance is a sliding scale, so is the effect of any shot or sequence of shots. (I’ll explain this in a moment.)
You seem to talk as though there were a Platonically existing reverse shot that it is possible to pin down and measure, with the camera being placed exactly here and looking at exactly this, but my contention would be that there is a continuum in what we are psychologically prepared to accept as a reverse shot and it extends a long way in both directions. But what is important to understand is that the concept is grounded in a simple perceptual model.
Let’s step away from reverse shots for a moment, because I think they are a special case of a much more basic principle which we can understand better if we focus on what’s happening in a single first of all.
The viewer is “engaged” with the character he is looking at any one time to the exact extent that the actor’s eyeline relates to the lens.
It goes without saying that if a character has his/her back to us, we are completely disengaged.
If we move around so that we are at right angles to the eyeline, things have obviously improved but we are still disengaged to a considerable degree. We feel this viewpoint as an objective one – we are “looking on” much more than we are empathising. This viewpoint is typically dispassionate.
If we continue moving round, we engage more and more. An obvious reason for this is that we can now see into the actor’s eyes and this is a really huge factor in creating engagement – we have evolved to be incredibly sensitive to the messages we get from another person’s eyes. Unless you’re a sociopath, it’s physically impossible not to be absorbed when you can see into someone’s eyes.
If we move all the way around, until we are looking directly into the actor’s eyes, we have full engagement, and it is surprisingly disturbing especially in a narrative context, in that instead of “viewing” the scene, we become implicated in it.
Obviously a clever film-maker is going to manipulate the angle of the eyeline to the camera differently at different moments to enhance or reduce the engagement we are feeling. Clearly there is no ideal, right answer to the choice of this angle or the degree of engagement that results from it – that’s obviously absurd. It’s all about making a choice that is right for that particular moment. Obviously there will be times when disengagement can be used just as effectively as engagement. A film is not a steady state experience – it’s an ebb and flow of states and great film-makers are masters of the ebb and flow. (What’s really depressing incidentally is to see mere aesthetics trump this fundamental principle – pretty shots are meaningless, and very frequently utterly counter-productive, if they are not subservient to the engagement principle.) Note that I am absolutely not saying that engagement is good and disengagement is bad – both are essential, and to some degree you can’t have one without the other. But that’s a discussion for another time. (Also I should point out that of course actors turn their heads, which means the the eyeline is not fixed by the camera alone, but let’s not over-complicate for now, but merely note that the actor can shift the level of engagement within the shot and this can be a very powerful factor. If you really want complication, remember that the camera too can change its position …)
What I think this discussion of the eyeline angle and engagement shows us is that the camera puts us the viewer in the scene, not just physically but psychologically. Sticking with our single for a moment, it’s obvious that both the choice of lens and the distance of the camera from the subject are going to be factors that we will want to manipulate to enhance or reduce engagement. A lens and camera position that excludes the environment and isolates the actor will create greater engagement than a wider lens that allows us to be aware of what’s happening around the character. It’s a well-worn trope to use a long lens and shallow focus in order to put us inside the character’s thoughts (which usually also involves an eyeline that is close to the lens). Incidentally I don’t think we experience the actual distance of the camera in this instance (which will often be quite considerable) – what we feel is the sensation of being inside the character’s personal space and it’s this psychological “distance” that we feel most strongly, not the geometric distance. (There is the “voyeur factor” that a long lens can introduce sometimes, but again let’s not over-complicate.)
So what does all that mean for our reverse shot discussion? I think the important point is that we have established that our engagement is predicated on a sensation of being physically present in the space of the scene. We have also seen that the canny film-maker knowingly varies the “distance” we have from the actors/action, this distance being both a physical and psychological thing and not necessarily dependent on physical geometry.
So I come back to my original contention that in essence we perceive the shot/reverse shot method to conform to the “tennis match” model. In other words, it stems from the natural sensation of standing/sitting between two people and looking from one to the other as they interact, with a neurologically-attested ellipsis for the head movement that relocates our gaze. There is nothing odd or unnatural about this – we know exactly what it feels like to stand/sit between two people who are interacting with each other but not with us, and we know exactly how we redirect our attention, depending on what we want to hear and/or see (not always the same thing).
Now let’s think about this in terms of the eyeline/engagement factor. We can go from both actors looking right down the lens (very rare but not unheard of – check out Tokyo Story and be amazed!), all the way round to behind either of the actor’s heads. Clearly this is a continuum. There is no cut-off point at which the single of actor A looking straight down the lens ceases to be a single and becomes a reverse shot. Which leads us to the slightly counter-intuitive realisation that a reverse shot is just a (developed) single of actor A … But what it also shows is that there isn’t a definitive set-up for a reverse shot. I would suggest that as film language has evolved, we have become used to how the continuum stretches what it would be physically possible for us to see as an observer within the actual space. As long as we stay on the right side of the eyeline, we accept that it’s possible to have a viewpoint anywhere on the arc between the two actors – because if we were quick enough on our feet or our necks were long and supple enough, we could probably just about get there physically in time to redirect our gaze.
I just don’t think it’s psychologically plausible that we would accept two discrete and unrelated viewpoints as being possible, as you seem to be suggesting. I think you’d have to do a lot of explaining as to how that might have come about if it were true.
The point about this principle being a continuum is that again the knowing film-maker is going to adapt the degree of engagement to achieve specific results at specific moments. Tony Zhou’s fascinating video essay on the Coens’ use of reverse shots illustrates this nicely. There is a point that he doesn’t mention which I think is interesting, and that’s how the camera placement is such that it disengages us entirely from actor B (the off camera actor) and puts us in the personal space of actor A. I think part of the reason for this is that the Coens’ characters typically “talk past each other” rather than to each other, something that all the best dialogue achieves to some degree. The characters are shown as being “disengaged” from each other – Barton Fink is a magnificent example of this but all the films show it to a greater or lesser degree. They are isolated in their own worlds of desire, ambition, delusion, longing, etc., and we are the only ones who can really hear what they are saying. And we hear it because of where the Coens have physically located us within the space. (Actor B just “doesn’t get it” the way we get it.)
What of the opposite of the Coens’ method here, namely the over the shoulder reverse? Interestingly, in the light of what we have discussed, an over the shoulder in many ways maximises engagement because we are moving around to a point where actor A’s eyeline is closest to the lens. However, because of the principle of “distance”, the fact that actor B’s shoulder is part of the shot makes us much more aware of B’s presence in the scene, so our engagement with actor A is subtly different. We are now “siding” with actor B in our view of actor A. Again, it’s important to recognise how significant is our perception of our own physical presence within the space. Literally and metaphorically we are “standing shoulder to shoulder” with actor B, we are “on his/her side”. A clever film-makers will use this for the psychological value that is carries with it – sadly, you don’t see that happening too often and again the choice is often banally aesthetic rather than freighted with intention.
So it may be that we are not disagreeing as much as all that, but I do think it’s very important to stress the psychology of film language as against the mere geometry of it. Yes, the film-maker can use staging and editing effects that are actively disruptive of our engagement but it’s best if this is done knowingly – it’s even a legitimate tactic to provocatively draw attention to the filmic artifice, but again I would stress that disengagement itself is a sliding scale and we can be acutely conscious of it, or only just barely perceive it or anywhere in between. It’s all good if you now what you’re doing. If the reverse shot were as disruptive as you seem to be claiming, I’m pretty sure we would feel it – instead of which, most of the time we are barely aware at all that a cut has taken place because in our brain all we’ve done is turn our head. Yes, of course, it can be used disruptively for creative purposes (and often is by accident because the film-maker isn’t actually in control of what s/he is doing), but it’s not fundamentally unnatural – indeed the reverse. As it were.
https://www.youtube.com/watch?v=R65wTHVUCGk
Simon Ubsdell
tokyo-uk.comSome contents or functionalities here are not available due to your cookie preferences!This happens because the functionality/content marked as “Google Youtube” uses cookies that you choosed to keep disabled. In order to view this content or use this functionality, please enable cookies: click here to open your cookie preferences.
-
[Charlie Austin] “FWIW, it’s not X2Pro in this case, it’s Media Composer.”
Thanks for clarifying, Charlie. That makes sense. It wasn’t clear to me that this was the route.
Coincidentally we recently delivered some GOT trailers and spots to HBO and they were asking for mixes and stems not the original laid audio as it seems they did in this case, hence my confusion here. I was assuming the issue was getting from FCPX via X2Pro into ProTools, not taking an AAF from X2Pro straight into Media Composer – which is obviously not what it’s designed to do, as you say.
Simon Ubsdell
tokyo-uk.com -
[Shawn Miller] “I feel as if this might be harder to define now than it was say, ten years ago.”
I think everything you say there is exactly right – “motion graphics” is a much wider and more sophisticated category than it’s ever been and only becoming more so all the time.
However, it’s important to stress (and you of course know this) that the gulf between even the highest end motion graphics and “true” VFX is getting wider as well. High end VFX shots are vastly more complicated in every sense than the kind of thing we’re talking about here and the complexity keeps getting significantly greater. Unfortunately there has been a general blurring of definitions that is quite confusing and often misleading – so you get movies with pretty basic “motion graphics type” effects shots claiming 1000+ VFX in a bid to claim parity with genuine VFX heavy shows. It helps with the marketing buzz but it’s a bit cheeky some of the time …
Simon Ubsdell
tokyo-uk.com -
[Andrew Kimery] “Anytime you transfer an edit from one NLE to another you are going to lose things, especially to/from Avid since it’s the only one that uses AAF as opposed to XML.
“Yes, agreed – I was just quoting selectively for journalistic effect 😉
However, it remains that the X2Pro export didn’t work for the audio – which should be a matter for concern. I have great respect for what the guys at Marquis have achieved and generally it’s solid enough, but at the same time I’ve had too many failed exports just like this to be entirely happy with the workflow. And this was a really pretty short and simple timeline …
Simon Ubsdell
tokyo-uk.com -
Of course, one of the key takeaways from this article is this:
“When it came to sending across the timeline there were quite a lot of hoops to jump through and settings to get exactly right. But very basically, I exported an XML to Resolve, and an AAF from Resolve to Avid. This bought across the picture (though not any speed ramps).
With audio I tried to send as an AAF to Avid via X2Pro, but it didn’t like it (I suspect mixed sample rates and more than 24 mono tracks was the issue). So I consolidated all the audio files and used a spreadsheet from Producers Best Friend to rebuild the audio manually.
I wouldn’t particularly recommend this as a workflow, It was a little bit of a pain.”
Simon Ubsdell
tokyo-uk.com -
[Shawn Miller] “Have you considered using Syntheyes for tracking? I believe it has an export module for Motion – for $299.00, I don’t think you could do better.”
Totally agree – it’s one of the very best 3D trackers there is.
Included exporters for After Effects (2- & 3-D, with plugin effects matching SynthEyes’s distortion model for AE CC/CS6; PixelBenders for AE CS5.x); Alembic ABC (1.5 Ogawa format); AutoCAD DXF; Bentley Microstation; Biovision BVH (import also); Blender (various ongoing versions); Carrara; Cinema 4D; COLLADA; Combustion (2- & 3-D); Electric Image; Filmbox FBX; FLAIR motion control cameras; Flame (2- & 3-D); Fusion 5 (2- & 3-D); Fusion 7 (very full-featured 3-D & 2-D planar corner pin); Hash Animation:Master; Houdini; Inferno (2- & 3-D); Lidar XYZ; Lightwave; MAXscript (3ds max, 3D Studio VIZ); Maya; Mistika; MDD animated mesh vertices; Modo; Motion(2- & 3-D); Nuke (3-D & 2-D planar corner pin); OBJ meshes; Particle Illusion; PC2 Point Caches; PhotoScan; Poser; Realsoft 3D; Shake (2- & 3-D); Smoke2008 (2- & 3-D); Softimage dotXSI; toxik (pre2009); trueSpace; Vue 5 & 6 Infinite;
Simon Ubsdell
tokyo-uk.com -
[Marcus Moore] “And I think that (to me) is Motion’s real achilles heel. As great as Ae is, a lot of the most advanced functionality comes from 3rd party support from Trapcode, Mocha, Video Copilot ect…”
You’re quite right about third party support for Motion – I think it’s very hard for developers to justify Motion-specific products, sadly. However I should point out that Mocha is available for Motion:
Simon Ubsdell
tokyo-uk.com -