What's up, everybody? My name's David Ariew. I'm a 3D generalist and
freelancer. I'm currently moving to San Diego. I've been releasing a bunch of
tutorials from my buddy, EJ Hassenfratz, Octane based tutorials, and that's been
a really fun process. It's been great to see artists doing work after watching
the tutorials, a super fun experience. So that's what I've got going on right
now, but I'm just going to jump in and show you my reel.
♪ [music] ♪
Cool, cool, cool. Thanks, you guys. All right. So you can find more of my stuff at
arievvisuals.com. So I'm going to play for you, I'm going to talk over this.
This is my space, space, space music video. This director, based out of
Virginia Beach, Derrick Borte, approached me at one point and said, "Do you want to
make a full CG four-minute music video?" And I said, "Heck, yes, I do. This is my
dream job. When do I start?"
I'm just going to play this.
♪ [music] ♪
♪ Am I in hell? Standing... ♪
I've presented this video a couple of times at GFWC 4D and at the C4D Roadshow.
And one big question that came up was how did I do my lighting and my camera work,
and that was something that I was not able to answer. It was, you know,
I wasn't familiar with my own process. I hadn't broken down for myself like what
I was actually doing to make the lighting better. I knew that I got better
on it, you know, better at it on this project, but I couldn't articulate why.
Like, why were my camera moves and lighting better? I was able to talk all
about Octane, I was able to talk, labout what the buttons did,
you know, how to create certain environments and how to create the
asteroid belt, for instance. But I wasn't able to explain, like,
kind of the creative direction side of things. So I thought for a bit and I
thought, you know, it would be cool to do a talk about how it's more possible to
become... We're becoming kind of digital cinematographers. We're,
you know... And I thought, you know, "What am I actually good at in this world?" It's
not MO graph. I don't, you know, I'm not good at making things move really
fast. I'm okay at animation but I'm not fantastic. It's definitely not,
like, espresso. I still couldn't tell you, like, what the inheritance effector does,
for instance. You know, I don't know all of that stuff. But what I do enjoy doing
is like, kind of building out a scene, texturing it, lighting it,
and art directing it, essentially. So you know, and then I thought,
you know, "What's my roots?" I started working for this production
company, the Dave Matthews Band content, for about three years. And so,
in that process, I, at one point, dropped my entire salary on
DSLRs, on steady cams, sliders, like, all of these tools, and actually
got in there and started acting like a DP. And I wasn't a very good DP, I totally,
like, tripped over things and I'm, like, I'm way better in front of a computer
than I am in the real world.
But my hope is that, you know, I can become an okay DP in this digital realm
where we've got undo buttons and we've got the ability to tweak and tweak.
So with that in mind, I'd like to share with you some of the things that I
learned, some of the ways of moving cameras, of lighting, of texturing, and
of art directing your scenes.
Yeah, you know, tips and tricks that I've picked up along the way
through this process. Because I feel like I have this kind of
weird, unique background that not a lot of motion designers have.
So I'm just going to cut this off and begin jumping in the scene with 4D,
where I'm actually comfortable, obviously, versus doing my monologue thing.
But here's my dorky self with a steady cam, probably going to trip over
this rebar here and fall flat on my face and totally destroy this equipment.
So, for instance, when I approach this music video, this was one of the first
assets I purchased and it's this very glossy, you know, fully reflective,
very sterile environment. And I wanted this, like, much more moody
environment. I wanted to create this derelict space station, right? With this
space man floating through it. And so I know that I could take this,
I liked the geometry of it, I liked the design, but I knew that I could
remap it and make it look destroyed and look a lot more moody.
All right. So this is our scene here. So I'm going to start with kind of a
controversial concept, H2Is are ruining your renders.
So let's take a look real quick. So if I fire up Octane here, now this is a
third-party GP render engine, and it makes it easy for me to art
direct and kind of show you in real time this process of lighting, of kind of
texturing and art direction. So I'll dock this in the interface here.
And I'm probably going to scootch this over here, and there we go.
So this is pretty close to the turbosquid asset, though it's not nearly as
good looking. But I've kind of gotten it in the ballpark, right?
We've got this overly well-lit, kind of glossy environment. And this is what I see
a lot of artists do. They'll jump into their HDRIs, they'll search for their
favorite HDRI. Let me show you this real quick. This is one by Maxim Ross,
you can find this on Cinerversity. So if I open this window and kind of
twirl this open here, you can see this is plausible, this could
totally work. We've got these lights, blue and cold light,
there's this kind of soft light, and a lot of it's in shadows. So we've got
kind of this dramatic look that might actually fit.
But when I turn it on, I get this. And it's not bad, but it's very evenly lit.
There's like this swath of very bright light here, and this swath of blue, dark,
you know, a little bit darker. So they'll go, "Oh, okay. Well, I just need to
rotate it." And they'll rotate it until they see, like, some shadows
and then it's still not working, it's still not intentionally designing your
lighting. And so then they'll be like, "Oh, well, you know what? This is just
too real world. This is not dramatic enough. We need,
like, a studio HDRI." So they pop in and they find something like this.
This has lots of shadow and lots of brightness. So this is going to be way
more dramatic, right? And so they turn this on and this is what happens.
It looks like a bunch of mush, right?
So I feel like HDRIs are becoming a crutch for artists, for artists getting into it.
And they're not thinking of actually intentionally designing their
lighting. It's kind of limiting. It can be a very good starting point,
but it's not necessarily enough to get that really moody dramatic look that I'm
going for. Here's a quick third example. There we go. So here, you know, this
could be really cool with these area lights. We've got this textured fall off.
Area lights, we've got this textured fall off, we've got these interesting colors
like green and yellow. And then when I turn it on, I get this blob of yellow and
blob of green, right? So not exactly what I'm going for. So let's close that out.
And I'm just going to jump to a fully textured scene. So obviously,
the textures are going to dramatically improve the lighting and make it look a
lot better, but you'll see that I'm still not where I want to be.
So let Octane load this up real quick. And so, again, this is my glossy kind of look.
And if I turn this on here, you'll see that this HDRI looks a lot better but it's
still really soft, it's still not, you know... It's baked. It's not intentionally
designing the lighting and it's never going to get me close enough to the result
that I'm after. And so you can flip through this forever, but you're never
going to get your perfect result. So, yeah, let's just stop there.
And what I like to do is I just start with a really simple Octane sky and I load in a
texture. Okay? And it's just...it's a float texture, so that's this anywhere
between black and white, and this acts as my fill lighs. So if I bring this all the
way down, we've got, like, total darkness. And I bring it all the way up,
and we've got total brightness. So I just like to use this as a base to begin with.
And the reason that I love... one reason that I love Octane is
I've got these, like, different polygons sitting here. You can see that I've got
different materials sitting on these as selection tags. So I could just
right-click on, say, select material, and you can, you know, go in and preselect
your materials like this as well. This is just so I can demo quickly.
And you can turn any geometry into a light. So I could go to "C4dOctane,"
"Blackbody Emission" here, and this is basically, like, the luminance texture in
physical render, and I can power this up. So already we've got a more interesting
light source and I can quickly do this for these other pre-selected little textures
here.
All right. So already it's looking cooler. Now, here we can just select these in the
back. These aren't necessarily lights but we could, you know, turn them into this
light banks really easily. So "C4dOctane," "Blackbody Emission," jump in here and
then make it warm, right? And then same thing goes with this here.
We could go "C4dOctane," "Blackbody Emission," you get the idea,
we'll make this one cool and then power it down a little bit. So already this is
looking a lot closer. Now, watch what happens when I bring down my overall fill
light. It's getting much more dramatic, much more what I'm after,
okay?
So the next thing to do would be to try and bring this dude out from the
background. Actually, you know, give him a silhouette against the background and,
you know, focus our eye on him. So I'm going to come over here and go
"Options," untick "Check Camera." So what this lets me do is it lets me dive out of
the camera and look around, but keep Octane updating in real time while I'm
navigating. So this is really useful for art directing your scenes if you want to
be viewing from one angle and then watching a live update from another.
So I can go... I want to drop in a targeted area light. So this will just
point at my dude once I drop him in, you know. Obviously, it just has...it's an
area light with a target tag, and so I've got my target tag here and
I drop in my guy. And there we go, it's pointing at him. And I like these
kind of long and skinny banks, and I can kind of position it really quickly and I
know that it's pointing at him because of that target tag. But if I get it too
close, it'll start to kind of get all wonky. So I can just delete the target tag
at this point and manually position it. And right now it's obviously
blowing out the scene. So I just jump into this tag and completely drop the opacity.
Okay? And I'll try to get it kind of positioned up here as a rim light.
There we go. And it's showing up a little bit, but one thing you can do is,
because this is a much smaller light than the rest of the lights in the scene,
like, for instance, this strip is a much bigger light overall,
Octane, kind of... It's got this Sample Rate. So the more...the smaller the light,
the more samples you want to give it. It's kind of this ratio where if I give it more
samples, it'll be less noisy and Octane will devote more to that light.
So now we can actually clearly see this rim light. So now I can just CTRL+drag,
duplicate it, rotate it around, and kind of put it underneath him.
So there we go. We're getting this nice rim lit look, dramatic.
What I can do is, like, I can go to Compare, Store Render Buffer.
This will save a little state here, and then I can untick these so we can see
where we've come from. Que. So that's looking a lot more interesting,
it's drawing our eye, and this is my layering process. It's, like, extremely
simple, but I think, you know, I really prefer this. I'm almost never using
HDRIs anymore in my scenes. And that's kind of a foreign concept to a lot of
people.
Next I'd like to talk a little bit about front lighting versus back lighting.
So when I first created this scene, I worked really hard in World Machine to
create this interesting texture, this intricate texture. You know, focused
on the motion blur of the asteroid belt, the character animation,
all these elements coming together. I showed it to the director and he goes,
"This looks like '90s CG trash." And I'm like, I was kind of devastated, but I
went back to the drawing board and said, "Okay, what's wrong with this
picture?" This planet is very front lit. And there's a reason why photographers
don't like on-board flashes, like, right above the cameras,
right? Because it will flatten out your face, it'll flatten out the subject, and
it's not an interesting look. We want our lights off axis or back lit or
something a little more interesting. So just going back and adding in this massive
backlight back here, totally brought this to life, added production value,
added much more quality, and really made it feel like there was an atmosphere here,
and brings out the light and shadows. So just lighting can so improve the look of
your CG in art direction.
So I find this pretty interesting. So if you look here, the light is coming from
camera right. And then still camera right. Continues. And then the next scene,
you'll see I flip it to camera left. And then it's back to camera right.
And I do this because you can cheat. Like, I don't actually care. The most
important thing is building the mood, right? You know,
in DPs, and directors will do this on set, they'll walk the lights really close
for the tight, and then on the wide they'll totally redesign the lighting.
And so ultimately, like, the continuity of motion matters, but not necessarily the
continuity of the lights. Like, people aren't necessarily going to
notice that so long as you're building this consistent mood.
All right. I'm going to fire up Octane here. So next I want to talk about the
quality of light. And by that, I mean how hard or soft your light is. And that's...
I'm talking about the shadows here. So I'm going to let this kind of load up into
Octane real quick. And I've got my dude sitting on top of the city
with some fog and a light sitting up off, you know, above the scene up here in
camera right. And what I'm going to do is the exact same trick as before.
I go to my "Options" and untick "Check Camera" so that I can...
Let me untick "Enable A/B Comparison" here so we can see what's going on so that I
can actually art direct and see my light at the same time. So if I were to go
"Compare," "Store Render Buffer" here, and just save this as a little memory,
and then scale this area light way up. Octane is physically based so that
when you're lights grow, they also get softer and much brighter.
And so if I power this down, I'm going to try to match the exposure approximately,
kind of, with this A/B Wipe here until... So we can kind of see what's
going on. This is such a handy tool in Octane to be able to compare.
But you can see that the shadows here have totally softened up, right?
But over here, I really hate what it's doing. This guy looks like a clay
renders now. It's bringing out that there's no beveled edge here.
Like this light is wrapping around a lot more but it's looking a lot junkier to me.
So in this case, you know, in some cases you want this soft light,
but in other cases where it's like a massive scene, you want this sense of
scale. Perhaps you want, like, a lot harder shadows. So I'm going to actually
scale this down and we're going to try some much harder shadows,
so maybe even more. And then I'm just going to slam this power up to something
like, I don't know... Oh, that's too much. Let's go to 5,000, 6,000. Let's try 8,000.
There we go. That's getting close enough so you can see what
I'm talking about. So you're not going to notice this as much with the dude here,
but if you look over here, the shadows are definitely sharpening up. And if you've
got like a bunch of, like, buildings that are kind of casting shadows down onto
your geometry, that might look a lot more interesting to see. Like, the shadows
of the scaffolding and see these kind of sharper shadows. They don't look like,
you know, perfectly hard shadows. That always looks CG. But we've got
enough softness that they look cool. So this is just another thing to keep in
mind. When you're art directing your scene, think about shadows,
your area lights and how big they are, what kind of shadows they're contributing
to the scene. All right. So that pretty much does it for lighting.
Next, let's talk about... So when you're an actual DP, you know,
the director is more concerned with the actors, but as a digital DP,
we actually have to potentially, you know, create the performance ourselves.
We have to be concerned with the actual, you know, CG actor,
right? And in this case, that takes a lot of freaking time and I wanted to cut
corners. So if I go to my "Window" here and I go to... I'm going to cut this out,
so I'm just going to change my layout so we can this a bit full. Full Screen,
let's do "Standard." All right. So you can see I added,
you know, all these controllers to the rig. I did what you usually do and,
you know, here's his center of gravity and he can do his little squatty dance
here and his derp, derp kind of thing, you know. Everything you'd expect from a
character rig. But, interestingly, I didn't use this at all. I went with
"Vibrate Tags," which are a procedural way of animating inside Cinema 4D. They're
just a little tag that looks like this, that's basically a wiggle expression from
after effects, right? It creates random motion. So I just went into all these
joints and, because this is the most simple character animation you can
imagine, where he's just doing this space float, he's just got the floaty arms and
legs, I was able to get away with just doing these vibrate tags. So if you look
on his arm, for instance, I've got a pretty little amplitude and a really low
frequency. So that's creating this kind of little floaty thing going on.
All right. And another thing I wanted point out real quick,
if I go to "Filter," "Joint," you can see that I was lucky that there were bunch of
other little joints along this tube here. And so it's got this sense that it's... It
gives it a little subtle sense of weightlessness because they're all moving
randomly in position. So this kind of sold that fact he's, you know, off in
space, kind of floating around, right? So I have these different
instances of, you know, slow and fast flow. Like here, he's got his kicky legs
and he's doing, you know, freaking out and going out of control. But it's as
simple as, you know, just like before, it's as simple as going in and
selecting your...some kind of joint, and then upping the amplitude and frequency.
So you can see here, we've got 44 and a frequency of 1. So much faster.
And if I blow this animation away here, so let's just stop it right about here,
I can just hold down CTRL+SHIFT and paint downwards to kind of kill this
animation. And you can see how bad this looks, right?
And this is just kind of the point of, again, cheating. Trying to get, you know,
seeing what you can get away with. Like, if you're an actual NDDP on set, you want
to sell the shot however you can. You want to not cost a ton of extra budget, but get
what you can out of it. So, in this instance, I knew that I could get away
with this crappy animation if he were spinning. Because you don't notice it
when he's spinning, right? This kind of pipes back into what I'm
going to talk about later, which has to do with relative motion.
All right. So that's all good. Let's move on. And this is about as far as I took the
technique. Here he's got an actual spine twist. And if you look at his fingers,
there's something interesting going on. He's actually kind of contracting
his fingers in a way that looks semi-realistic. And this is, again, all
done with Vibrate tags. So if I go down to his fingers here
and I select all of them except his thumb. Let's grab these four here,
I want the left ones, fingers one through four. Okay. You can see that they're all
on the same seed. Now, if they weren't on the same seed, what would happen?
So I'll set one to six, one to four, and then one to something else,
and I'll play. He's got creepy fingers. Okay. So, obviously,
we want them all on the same seed. So it's obvious why I decided to do that.
But then how did I get them to kind of cascade? Let's go back. So his pinky
got all weird, so I'm just going to revert to "Saved Here."
All right. So let's jump back in. Okay. So again, what if, on these left fingers
here, we had all the same amplitudes? Let's set 155 to the amplitude.
Okay. It looks better. They're not doing the creepy dance, but it looks a little
robotic. They're all moving the same distance. So the easy thing to do here was
to set each one with a little bit more amplitude. So on the pinky, I've got
208. On the ring finger, I've got 162. On the middle finger, 127.
And on the index finger, I've got 85. And so that creates this a little bit more
natural look. Cool? Make sense? So, yeah, again, just cheating and trying to get
away with whatever I can. And this totally freed me up to do way more lighting,
way more kind of building of the environments. And if I'd had to do
character animation, like, pure character animation on every scene,
it would've extended the timeline probably another couple of months.
Actually did do some full-key framed animation and I'll play this back for you.
He's got his floaty arms here and, hilariously, it doesn't look like that
much better. It looks pretty similar to what you just saw. Maybe there's a little
more in the wrists, but that's about it. So I was like, "Okay, I can get away with
just using vibrate tags and move on."
Next, let's talk about sense of scale. Like, what kind of digital lenses can you
use to communicate a different perception of scale? So for instance,
I've got this city that I made and you can see how to do this on my I Design Tutorial
on idesign.com or on my website. Now, I've got a lot of detail here in the little
lights in the windows. But to me this doesn't look as massive as it could.
This was shot on a wide lens and from this bird's eye view, right?
And typically, when we see bird's eye views, we are used to seeing a longer
lens because helicopters can't physically get that close to the buildings.
So in this case, it maybe doesn't look all that convincing. So what if we just,
you know, create a longer lens and see this more compressed space between
buildings? So already this, to me, looks a lot more massive. We're compressing the
buildings. This is something we're used to seeing. Whether it's just because of
how we see things in films often, it just creates this perception that this is
larger. Alternatively, what you could do is imagine yourself as a DP on set,
you know. Maybe you're flying a drone, maybe you've got a dolly. Don't Peter
Jackson it. Don't create some kind of impossible camera move,
right? Think of what is actually possible on set. You know, we would create, like,
a dolly move and we would keep it simple. So, for instance, say we wanted to shoot
this on a wild lens. We could do something like this.
And this to me looks a lot bigger because we're lower, we're almost on the street
level. And on a wider lens, the perspective is much more apparent, objects
rush past us more quickly. I really like skimming along the ground as we're on
a wide lens because you sense the motion a lot better, a lot... You know, things
appear to move much more quickly. But the one thing I don't like about this
is that the background feels static. It looks like I just took it in "After
Effects" and popped it in back there. And the reason why is because we're moving
perfectly linearly. There's no float, there's no camera shake, there's no
rotation. And we're not used to seeing that in the real world. We're used to
seeing some kind of error or some kind of pan. And so in CG, we can make things
overly perfect. And in this case, I think, even though it's technically correct,
like, if Wes Anderson were to do this, you know, and do this perfect dolly movie,
you might see this, but we don't see it very often. So what I think can help is by
adding in an additional element which could be just a simple tilt upwards.
And so here's the tilt and it feels a lot more cinematic to me.
The background is now integrated with the movement, you can see it rotating and it
feels like it's all, you know, meshed together. So what if you were to try to
shoot the same scene on a long lens? This is what I do. I would get down on the
level of the buildings and do kind of a slow track or slider shot,
whatever you want to call it, and this stacks the perspective.
It really crams the buildings together because we're on this longer lens and you
notice the parallax, you notice the shallowed up the field,
the buildings, little poles overlapping each other, and it feels massive.
Another thing that you can do that will immensely aid how, you know,
larger scenes feel, and I show you, again, if you look on idesign.com,
you can learn how to do this. In Octane, you can create Octane fog.
And this creates kind of a substance and atmosphere that we're looking through,
and we're used to seeing this in the real world. As, you know, objects recede into
the distance, they get less saturated and less contrasting.
And so we immediately buy this scale a lot more. You could also do this in C4D
with the depth pass, just with a, you know, physical render and do this in
posts. Similarly, in the long lens situation, this brings in so much more
depth.
And here another couple quick examples. So long lens look tracking, I like this look.
And then wide lens, I like to be skimming along the ground.
So you'll see in a second this really makes the, you know, perspective apparent
and makes things rush by the camera. Added a little bit of shake here to
kind of, you know, make it feel more intense, but it amps up the energy.
Especially moving in Z with the wild lens is something I enjoy doing.
And then here's, like, my signature move apparently. Like, I've done this a
bajillion times, like, across my work. You can see I like to do the classic
helicopter shot, which is just, you know, orbiting around something on a long lens.
On the, you know, last shot of Star Wars, I saw this done, the JJ one, where the
very last shot, they've got Jedis on the mountain in they're, like,
the camera's spinning around crazy and the background's blurring. And what it
does is it focuses your eye on your subject, on your pivot point,
where the background is just going wild with motion blur. So this is a look that,
apparently, I'm drawn towards and I've used a bunch of times. But it's very easy
to setup and I'll show you a little bit later how you can do that.
Let's talk about composition a little bit. So I like the composition of this shot and
here's why. It feels epic to me, it feels like you can really get a sense of how
small he is in this vast universe, you know, in space. And with this spaceship,
you don't see the whole thing. There's a little bit of mystery still, right?
This is when I'm revealing the spaceship for the first time and you're
just seeing part of the wheel. And he's receding into the distance, so he's
getting smaller. Now, if I were to do this the crappy way or just a little
more boring of a composition, I would show the entire wheel and then I would have him
coming towards camera, so you get no sense of how big he is in relation
to this object. You're kind of blowing the gag immediately. And there's also
not a clean silhouette with him against this wheel, you know.
It's just kind of a cluster.
Now, if I were to try to do this with a long lens, like, maybe this is,
like, a medium lens, like a 40 or 50, and we're kind of close.
But if I were to try to do this with a much longer lens, here's what I'd do.
I'd back the camera way up and zoom way in and do this really simple pan
downwards. Often, the simplest camera moves are the most effective.
And I should've probably lit him up a little bit more here so you can see him,
but he looks like a spec against this massive space station. I'm still not
revealing the entire space station. So that's kind of what I would do.
And then if it were a wide lens, like I've said before, I would try to skim along
whatever object it was. So here I'm like doing this rotating thing,
I'm pulling back. He's receding into the distance so you kind of get a sense
that he's small. Could have played that up a little bit better maybe,
but you get the idea that if you're on a wide lens, feels good to be,
you know, kind of close to the subject. Like, in my study cam days,
I would go in low mode, I would be on a wide lens and skimming along the ground,
and I really liked that look because you could feel the motion a lot better.
All right. So this is what I see a lot with people that just jump into Cinema4D
and animate cameras for the very first time. So they'll go to the beginning,
they'll set their key frames, go to the ends, go on forward, set their key frames
again, and then they've got to move, right? And the move is starting.
And then the move is stopping. And to me this is really distracting.
There's no motivation for this. You don't see it a lot in actual films where we're
cutting into a shot that's stopped and then it begins, and then it stops again.
It's kind of weird. We see, you know, you're cutting into a shot that's either
static or it's either... Maybe it starts
and then comes to a stop or maybe it's, you know, just already in motion.
But say I did want to motivate this kind of thing. I'm going to quickly untick this
so we can see it a bit better. I would go into my key frames here and let's do an
Edit inside of Cinema 4D. So instead of having it go into Premier and render
things out, I'm going to show you how to kind of create and edit within your scene
to check continuity and to check motion. So if I were to go to Animation,
Show Track, then go to my F Curves, what I want to do is... So I definitely don't
want to begin stopped. So what I'm going to do is just kind of set this,
like, this a-way so that we begin in motion, and then we're coming to a stop.
So I'll show you what that looks like. We're in motion and then we're easing into
a stop. And I'm going to make it shorter. So, again, right-click Animation,
Show Track, and I'm just going to kind of squish it in half. So about here
we come to a stop. And then I'll create a new camera and we'll just cut to a shot
of his face here, kind of zoom in and cut to his face, and we'll keep this one
static. So here I'm trying to, you know, generate a motivation to cut to a locked
camera. So we're right about here, we'll cut to this shot here. So I'm going to go
Shift+C, which is the Commander, and I'm going to enter in "Stage," which will
create a stage object. This is some old school thing in Cinema 4D that lets you
cut between cameras. So here I'll drop in my first camera, which is this here.
Psyche frame, go one frame forward and drop in my second camera and set another
key frame. So now we can see we're moving, we're easing in, we've stopped,
and then we cut. So to me, that makes sense, right? But, for instance,
if we have this as linear, again, the first camera, go to Show Track,
and I'm just going to set these key frames to linear, and I'll stretch this out a
little bit so that I don't get any stoppage before the move is done.
So you'll see we're moving, we're moving, we're moving, and then suddenly
we slammed into a static camera. To me that feels a little bit jarring. Not too
bad in this case, but it can feel very jarirng when you're going from this
fast motion into a hard stop. So rather than trying to sell this, the easiest
thing to do is just, you know, say screw it to the stage and this extra camera.
Let's stretch our key frames back out, and very simple linear move. But to me,
looks a lot more cinematic.
Now, what if I want to add a little more visual interest to this, right?
So again, let's just do a tilt. We'll start here. I'm just in this...
This is heading pitch and banking. I'm in the pitch. And then we go to end and I'll
tilt upwards. Super simple. And I want to make sure, like, this is the thing,
this is why people get caught up in these moves that are easing because the default,
if we go to Key Interpolation, is Spline, as we all know. And that helps with
animating objects. But for cameras, I much prefer to keep it linear. So, I mean, just
go back to my camera here and make sure that this tilt is not easing. So settle
these to linear, and let's look one more time. And this feels a lot
more cinematic to me. Because you can imagine that the DP and director set up
this dolly track and then they, you know, also had a pan and tilt head while it was
dollying. This is actually a complex thing to pull off. So it's super easy for us,
you know, CG artists to create good camera animation, which often is the most simple
and controlled.
All right. So total departure. This is a crazy camera trick that you see in,
like, Vertigo and other films. It's called a zolly or a contra zoom, and I
like to use it once in a while as an effect. It's almost like a video game
thing, like, in gears of war where the guy starts moving fast
and the camera lens widens out. So it's just animating the focal length,
it's super simple. I don't see it a lot and I like to use it as a little emphasis.
So here you can see the background receding and the whole scene
kind of stretches out a little bit. This is a project that I did,
like, three years ago. So pretty subtle, but I like to use this kind of thing in
my work once in a while. Just for funsies. And then here, this Dave Matthews Band
music video I did, like, in 2013.
He falls off a building, and then we're going to cut to this little scene
stretching out. You can feel it, it amps up. So here, when we're on this long lens
at the very beginning, you don't sense the movement as much. And then it widens out
and things get much more intense. So one more time, it goes "rrrrrpppp"
and then it feels like we're falling much faster. So that can be a fun effect.
And here I made the most biggest zolly move of all time. So here we're starting
on a very long lens and animating through to a wide lens. We're moving consistently