Wow cinematic question

Hey guys , so im gonna keep this one breef.

I´ve been watching alot of wow cinimatics “behind the scenes” and one thing i notice is from the first wow cinematic to the warlods of dreanor , when they show how it looks in the viewport for example when he pours up the demonblood in the coop , they only have the cut of hands and the cup and blood in the scene. and in the first wow cinimatic when we see a nightelv running in the woods and turning in to an animal , its only the nighelfs legs that is in the viewport.

I get that this might save som computer power , but how is it superior to have it all in the scene directly ? , isent they shadows and reflection easier to maintain if its all in the scene ?

and why do they green screen the characters ? , i mean when i render in blender, ( funny rim :smiley: ) i have all the textures on the character and well its not green.

I have been wondering about this for a long time, and get it might be a long answer for this. but plz give it a try ! cheers guys.

You should probably include some links, so people can see what you are looking at when you come up with your questions. I suspect, though, that the framing (hands and blood; only legs) is an artistic decision, rather than a technical one.

hey ye sorry ! :stuck_out_tongue: here is the link

in 15:03 u have the hand for example

Doesn’t necessarily mean that was what is rendered. In that particular case I’d guess it was a hand-only model used as a fluid simulation prop (no reason to sim obstacles on the entire body when only the hands are actually involved), but there isn’t enough info there to say for sure what that shot represents.

The general reason for all of this though is just to keep performance up by only working on what you need. Sim only affects the hands? Sim against the hands, then swap in the full model once the sim is cached. Don’t need the textures or highest subdiv level to get the animation right? Turn it down in the viewport. Don’t need to see the decorations in the back of the environment while laying out the scene? Hide them on another layer. In a lot of cases, it’s not only easier on the computer, it’s also easier on you. Textures can make it hard to see if your mesh itself is detailed enough, hence clay renders. Walls can block your when laying out objects in the middle of the room, so you hide them. You get the idea.

Good idea, J_the_Ninja, except the final shot was an extreme close up of the hand dipping a goblet into a bucket of green blood, then pouring the blood back into the bucket. I didn’t watch the whole hour and a half video at the link, just the minute or so where they talked about the hand shot, and a bit more, but it didn’t seem to me that crew was overly concerned about the workload on their hardware.

I didn’t see the green screen part (and I’m not going to spend an hour and a half watching the whole thing to find it) but it’s possible that they have some compositing software that works with green screen, since that’s been an industry standard for putting live actors in front of arbitrary backgrounds for quite a while. And since they aren’t really putting ‘live actors’ in front of the background, but images of live actors, it works the same for images of CG characters, too.

I wonder, rybergs, whether this is idle curiosity on your part, or do you have an actual Blender specific question in mind you want an answer for? It’s kind of hard to tell.

hey orinoco , well :slight_smile: the question i guess is a little big. But i dont get why they choose to render the guy in green and later on put the textures and stuff on , why not render with texture directly.

But the main question is since they are defiding there scenes so much how can they get reflections, like if the fire bolts are rendered from another layer, and then throw composition blend them together , how can the reflection and the lighting still effect the ground on another layer, is that possible in blender ( i mean render layer )

that what it boils down to I guess.

becouse if I render a ground in one render layer and then the fire , the ground dont have the lighting effect from the fireballs . and say i have a lake or something like that i would want the reflection. , And one more question ! :slight_smile: is it possible to render one renderlayer in black and one transparent in one render ? , the transparent button is not specific of layer what I can see and noticed.

Sorry about all the questions , but im really hopeping to get some answers : ) cheers and thx for your time

here is a link , https://www.youtube.com/watch?v=VDj9vimokdg

look at 03:20 its the character but only the legs and some of the body are showing. its a 4 min video…

okay done with the spamming today :slight_smile:

Ok, what you are watching is a big budget studio making these kick ass animations with a large team, probably on software that has either been developed in-house or has been so extensively modified over the years (Blizzard has been around for a long time) that it might as well have been custom made.

Why do they do some of the things you’ve noticed they do, like have the girl running headless and texturless through the forest? Because it’s a test of the body motion, to see whether the gesture reads the way the director wants it to read, and putting her head on her shoulders, with it’s attendant hair simulation, would either a) distract from the body, because the hair isn’t simulated, or b) slow things down tremendously when rendering.

Can Blender do these things? Yes, for the most part, it can, and has been able to for a long time. Is it done the same way? No, probably not. Is it as easy as the Blizzard guys make it look? Trust me, it isn’t easy for them, either, and they’ve been at it for years.

Making an animation is a multi-layered multi-stage production, especially when working with a team. You go step by step, and make changes along the way, a lot of changes. So at each stage, you want to make sure that you do only the minimum needed to see what is going on, and that any needed changes will be as simple to make as possible, since you are going to make some.

To get a bit into the Blender ‘how to’, you need to look at layers, render layers, visibility of objects, compositing and nodes. If you have a specific question about a Blender project you are working on, you’ll get a lot more detailed and useful answers by posing those questions, rather than asking about a Blizzard making of video. If you have general questions about how Blender works, then I’d suggest you fire up Blender and do some tests of various ideas using simple objects and see how it works for yourself.

Best of luck. :smiley:

hey agien Orinoco , thx for the replay! :slight_smile: yes ofc , and thx for the time :! )

@Orinoco and @J_the_Ninja are correct: You don’t include geometry in simulations (hair, liquid,etc.) that isn’t part of the sim. Simulations take a long time in most cases and you don’t want things in your calculations that would slow it down even further.

hey safetyman , thx for the answer , ye I got that :slight_smile:

So now I have the knowledge to make my own blizzard cinimatic , im just saying keep an eye up ! :wink: