Jump to content

Cloverfield


Recommended Posts

Didn't work on it myself....

 

I did read an article about it, though, that said that they shot the VFX shots on a much higher-end camera than the one that they wanted it to look like it was shot on - something like a Viper or Genesis.... And then degraded the VFX shots in post to match the rest of the footage.

Link to comment
Share on other sites

I did read an article about it, though, that said that they shot the VFX shots on a much higher-end camera than the one that they wanted it to look like it was shot on - something like a Viper or Genesis.... And then degraded the VFX shots in post to match the rest of the footage.

 

 

The VFX plates were shot on the CineAlta F23 and then degraded to match the rest of the footage.

 

I hear that the rest of the movie was shot on an HVX but I can't find anywhere that confirms that

Link to comment
Share on other sites

The VFX plates were shot on the CineAlta F23 and then degraded to match the rest of the footage.

 

Ah - okay - thanks - I knew it was on something somewhat better than how it was meant to look...!

 

I hear that the rest of the movie was shot on an HVX but I can't find anywhere that confirms that

 

That's what I'd heard too...

Link to comment
Share on other sites

Hey Guys... thanks for replying to the post...

 

But my original question was more about the R&D they did to comp the CGI with video.

 

Im mostly interested in the Motion Blur. The way I see it the renders HAD to be HDR (open EXR or smtg) because of the way the bright sources of light leave a streak as the camera shakes... HOWEVER... its not just that... These streaks have a very video edge to them... very unlike the thing's ud get from a str8 3D render with motion blur.

 

I should prolly post on a VFX forum.

Edited by Benjamin_Lussier
Link to comment
Share on other sites

Most VFX studios use EXR for their renders these days, that's pretty much a given. I expect that they had several different shake nodes (or scripts) designed to emulate certain image characteristics of the HVX-200 such as noise, pixel-shifting, 4:2:2 compression, edge-enhancement, etc on the rendered CG elements.

 

The way to add those aspects to the image is to figure out how they work in real-life first, like for example 4:2:2 compression is easy to fake by converting an RGB image into an YCrCb image and then apply a 2 pixel mosaic to the Cr and Cb channels and then convert the YCrCb back into a RGB image. Noise or gain may be something you either sample or overlay from reference footage - ie. by shooting 30 seconds of black using different amounts of gain should cover you.

 

If you looked at the CG elements by themselves you'd see an image that looks fairly clean (even the motion blur), it's much easier in cases like this to downgrade the image in the composite.

Link to comment
Share on other sites

Will I get shot for saying this? I hope not....

 

Do check out (and ask this) on vfxtalk.com (it's mainly about the 2D side of things over there) - there are a number of people from both DNeg and Tippett who will, I'm sure, be able to help you out on this.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...