Jump to content


Photo

How are CGI effects actually added to film? (im mean literally to the celluloid)


  • Please log in to reply
6 replies to this topic

#1 James Millward

James Millward
  • Basic Members
  • PipPip
  • 20 posts
  • Student

Posted 09 May 2009 - 04:29 AM

I have wondered this for a while now.

Presumably, the each cell of film has to be scanned like I would do with a photograph?

If so, what res do they scan at? 2K if so, does that mean that when watching a film in the theatre, I am only seeing 2k? (not that not matters as such, jusy curious).

Now, when the shot is complete, how is it transfered back to film? Is it printed, like you or I would print the photgraph in the earlier example? If so, what kind of machines do this? again 2K?

Im also interested in how this was done in the early days of CGI (say T2 1991 ish). What res were they able to capure and print at in those days?

Thanks guys
James
  • 0

#2 K Borowski

K Borowski
  • Basic Members
  • PipPipPipPip
  • 3905 posts
  • Camera Operator
  • I.A.T.S.E. Local # 600 Eastern Region

Posted 09 May 2009 - 06:41 AM

Film is usually scanned at 2K (though most people agree that it takes a 4K scan to really get all of the images off of a modern, well-exposed negative, especially slower stocks like 5201).

CGI/DI images are then literally lasered or CRT imaged back onto film with a device called a film recorder. These devices have been around since the 1970s and are still in use today in much the same manner as the past.

They have just gotten far faster and far higher in resolution as their work has become literally ubiquitous throughout the industry.

But the digital workflow in use in the early 1990s, while slower, uncommon, and far more expensive than that employed today, was essentially the same.
  • 0

#3 Chris D Walker

Chris D Walker
  • Basic Members
  • PipPipPip
  • 108 posts
  • Other
  • Cornwall, United Kingdom

Posted 09 May 2009 - 07:07 AM

I know the resolution of the effects in T2 were done at HD (near 2K) because James Cameron had the whole film scanned and stored as 1080 PsF files for when the technology in the home reached the point it's at now. I also remember having either read or heard that the visual effects for Jurassic Park were done near 1.5K.

Most films have visual effects rendered at 2K, the same resolution that most films now go through for a digital intermediate process. Just as examples Jarhead went through a 4K DI but had the 2K visual effects 'blown up' to save on storage space requirements and rendering times, while I, Robot had a 2K DI with 2K effects, both of which were blown up to 4K for recording back to film.

I'm not sure about the early days but modern film scanners like the ArriScan and Northlight can capture film frames as oversampled 6K, 10-bit Log files which can be filtered to 4K, 2K or HD for effects and DI's.

To record a finished effects shot back to celluloid back in the late 80's and early 90's the film would have rested on the glass of a special CRT (cathode ray tube) that would project the final rendered image. However, this didn't wield enough information so later one frame would be exposed three times for each of the primary colours, effectively doubling the visible resolution. There was even a process where the glass of the CRT was taken off so the photons made direct contact with the film in three passes: red, green and blue. Now, visual effects and whole films are output to film via laser recorders.

Hope those were the answers you were chasing after.
  • 0

#4 David Mullen ASC

David Mullen ASC
  • Sustaining Members
  • 19769 posts
  • Cinematographer
  • Los Angeles

Posted 09 May 2009 - 09:54 AM

"T2" was one of the first movies at ILM to have all their final effects finished digitally and then recorded back to film. Before that, it was more of a hodgepodge of compositing techniques. The CGI water snake in "The Abyss" for example, was recorded to film as a separate element and then composited into the scene using an optical printer.

Effects plate scanning is often higher than 2K -- for example, a VistaVision plate may be scanned at 4K and higher so that they have some flexibility in resizing the image. Final compositing though is often done at 2K.
  • 0

#5 James Millward

James Millward
  • Basic Members
  • PipPip
  • 20 posts
  • Student

Posted 10 May 2009 - 02:52 PM

Many thanks for the replys.

I am really interested in the technology that was used in the early days of CGI.

What processing power created the shots we see in say the Abyss, T2?

Were these Workstations along the lines of SGI or onyx? What were the specs of these machines?


Is CGI today done on specialist workstations (im talking ILM Level), or are they just very high spec PC's

Really interested i this guys

James
  • 0

#6 Scott Fritzshall

Scott Fritzshall
  • Basic Members
  • PipPipPipPip
  • 584 posts
  • Other
  • Los Angeles

Posted 10 May 2009 - 03:11 PM

Yeah, back then the CG would have been done on SGI workstations. Today all of the software runs on PCs, so the computers they're using are more or less the same as what you're using at home. If they're doing 3d stuff, they might have a workstation graphics card, which accelerates the interactions and makes the display a bit more precise- these are just off-the-shelf parts that cost a few thousand. Other than that, the only real difference between what they're using and what you're using is that they have large render farms in order to process vast numbers of complex scenes much faster. Sometimes they'll also steal idle processor cycles from other users' workstations on the network.

I don't remember the exact specs on my workstation at work, but it's got 8GB of RAM and dual processors- not sure if they're the Intel Xeon ones or just Core2Duo. Basically it's probably twice as fast as my PC at home, and to tell you the truth, I've got one of the newest machines in the company- most people use computers that are way slower than the one I've got. I'm not a 3d guy though, I'm a compositor; which means I work mostly with 2d elements, and my work generally isn't as processing intensive as the 3d stuff.
  • 0

#7 dan kessler

dan kessler
  • Basic Members
  • PipPipPip
  • 172 posts
  • Other

Posted 16 May 2009 - 01:18 AM

Here's a brief overview of the typical fx pipeline:

The camera footage, or background plates, are scanned in.

The 3d animators and effects artists will model, animate and track their elements to match the plates.

The cg lighters will light the elements to match the plates.

The cg elements are then rendered, usually as multiple layers, i.e., shadows, diffuse lit, spec lit, etc.

The rendered elements then go to the compositor, who performs the final task of blending them
seamlessly into the background plates.

The final composite is then rendered out, and that is the image that goes finally to the film recorder.
  • 0


FJS International, LLC

Abel Cine

Wooden Camera

Metropolis Post

Technodolly

rebotnix Technologies

The Slider

Willys Widgets

Opal

CineLab

CineTape

Glidecam

Media Blackout - Custom Cables and AKS

Rig Wheels Passport

Broadcast Solutions Inc

Tai Audio

Paralinx LLC

Visual Products

Gamma Ray Digital Inc

Ritter Battery

Aerial Filmworks

Aerial Filmworks

Media Blackout - Custom Cables and AKS

Rig Wheels Passport

Broadcast Solutions Inc

FJS International, LLC

CineTape

rebotnix Technologies

Ritter Battery

Tai Audio

Abel Cine

Paralinx LLC

Wooden Camera

CineLab

Visual Products

Metropolis Post

Opal

Willys Widgets

Glidecam

The Slider

Gamma Ray Digital Inc

Technodolly