UnFramed Visual Effects

Pending downtime

Hugh Macdonald — Wed, 26 Mar 2008 16:13:20 +0000

Yes, it’s been a while. And no, this isn’t a proper update, I’m afraid….

It’s just a quick note to let anyone who’s keeping an eye on the site know that it’ll be going down for a bit. I’ve got to move hosts, and I haven’t even started to look for a new one yet. So it’ll probably take me a while to get things back up and running.

It will happen. At some point. Keep an eye on the RSS feed if you want to know when - I’ll post something as and when we get up and running again.

Colour space, and the many pitfalls

Hugh Macdonald — Thu, 03 May 2007 21:55:48 +0000

In compositing, colour issues are one of the things that give a lot of people the most hassle. They’re the kind of thing that a lot of people can cope fine not understanding entirely, but having a better understanding of the issues surrounding colour can really help your understanding of what is going on, both in your comp and what would happen in the real world.

Disclaimer: I am no colour scientist. If I’ve presented something here as fact and am completely off the mark on it, please do let me know.

Most of this post will be about colour spaces (that’s Log/Lin for most of you), but I’ll briefly talk about colour gamuts at the end.

In film, we’re usually given log plates. The log colour space was designed to replicate how the grains in celluloid change as light falls on them. For double the amount of light hitting the film, the log value in the image increases by the same amount (approximately 0.088, although the exact number appears to vary depending on who you ask).

Linear (or Linear Light, but I’ll call it Linear), on the other hand, is more related to the actual light hitting the film, rather than how the film reacts. If a pixel in a linear is twice the value of another pixel, then this is because that pixel had twice the amount of light hitting it.

Both Log and Linear are mathematical spaces. The values mean something, but don’t necessarily look good if you view an image on a monitor.

Some examples might be nice….

For all of these, I’m using Marcie (that link will give you a 2k Log image from Kodak)

Click on any image for a larger version

Marcie in Log:

Marcie in Linear:

Neither of these look as they would if you saw this piece of film in a cinema. As I said before, they are both mathematical colour spaces, and so not necessarily for viewing. Shake, to use a tool I know as an example, comes with the Truelight node. This node is for converting an image from Log or Linear into an image that actually represents something that we would see, in most examples, if it were put on film and projected in a cinema. The Truelight node ought to have information about your monitor, but does come with a few pre-set display profiles. If you’re playing with this in Shake, have a look at it through the Truelight node. There are different profiles for the different types of film that you may be printing it to.

For example, Marcie, when printed onto Fuji 3510 film, would look like this (if you are viewing it on a Sony HD monitor, which I’m not either):

One area that often confuses people is that a lot of people who don’t work on film say that they are working in “Linear”. I beg to differ. I say that they are working in Gamma-Corrected Linear. (I mentioned “Linear Light” earlier - this is to differentiate it from Linear as called by people who work on TV/Video). Gamma-Corrected Linear is, as it’s name applies, Linear, with a gamma curve applied to it to make it more viewable. This is another viewing colour space, and, in my opinion, shouldn’t be worked in (although often is).

Marcie in Gamma-Corrected Linear, with a gamma of 2.2, looks like this:

What I’m focussing on here, though, is why we work in Log or Linear.

When we composite, we are dealing with light. We are not, with one exception, thinking about what is happening inside the actual celluloid itself. So, as Linear is directly related to the amount of light, it makes sense to do our compositing in Linear.

The one exception is adding film grain. Because this is actually happening inside the film, this is always done in Log.

One other possible exception is grading. Grading can be done in either Log or Linear. Earlier I mentioned that twice the amount of light hitting the film meant 0.088 extra in log? This is why grading in Log should always be done using Add nodes (or your tool of choice’s equivalent). You want to brighten it up by a stop? Add 0.088. Darken by a stop? Subtract 0.088. Add half a stop of red? Add 0.044 to the red channel.

If you’re grading in Linear, it’s Mult that you should be using. Instead of Adding 0.088, Mult by 2.0. Instead of subtracting 0.088, Mult by 0.5.

If anyone’s got more thoughts on why grading should be done in either Log or Linear, please do let me know - I’d love to get other peoples’ thoughts on this one….

So, other than grading and graining, why should everything be done in Linear? Because, as I mentioned before, and I’m sure you know already, most of compositing (the “making it look real” part, anyway) is about replicating how light would interact were the scene you’re creating real. You’ve got something behind glass, and a reflection on the glass? In real life, what you’d see would be the light coming through the glass added to the light reflecting off it. So, in Linear, you can just add the two things together. (Screen, in case you’re wondering, is a non-accurate simulation of this, but gives a reasonable result when used in Gamma-Corrected Linear)

Working with CG also generally requires linear. 3D renderers will be calculating the light in linear, and, while they often (well, Maya’s software rendered does, at least) add a gamma curve on, for proper integration with an environment, linear is the best way to go.

That’s about it - if there’s anything I’ve not been clear on, please do let me know in the comments, or (probably more likely) if there’s something that you think I’ve misunderstood myself, please do similarly.

New project - time-slice shot.

Hugh Macdonald — Wed, 14 Mar 2007 12:31:10 +0000

I’ve got another upcoming project which has been on my plate for a while, but I haven’t had the chance to say much about it until now.

You may remember a breakdown I posted last year…. It was a particularly rough and crappy attempt at a time-slice shot. It’s finally time to do this again, but properly.

I learnt a lot of lessons looking back on last time, and, this time around, I have every confidence that this one will be done properly, with care, and will give a stunning final result.

The shot in question involves a woman throwing a glass of wine over her husband in the middle of an argument in a restaurant. It starts as an over the shoulder shot of her as she throws the wine. The camera follows the wine as it flies through the air, ending on a reverse angle shot on him as it hits him in the face.

I’m going to be keeping this blog updated with news and progress as we go through pre-production and into the shoot.

Unlike the last time, this shoot will be doing with an array of approximately 30 digital stills cameras. This will involve a lot more planning in terms of positioning and triggering them.

The current plan involves the following:

Previz - I need to know the exact positioning for each of the stills cameras, plus what point in actual time each camera needs to go off at. Because of the potential complexity of the shoot, I’m going to be doing a few options, ranging from a relatively simple “fire them all at the same time” shot through to one where time will appear to slow down before suddenly jerking back to motion.
Sourcing the cameras - I have plans for this - more when I know. Eventually, I just need about 30 identical stills cameras. More news soon, hopefully.
Triggering the cameras - I have a friend who is currently building me a triggering system for the cameras. This will allow me to set each camera off with a pre-determined delay on a per-camera basis. This will allow me to have much more control over the apparent movement in the final shot. This is being worked on at the moment - once it’s finished, I’ll post more details on it.
Assistants - I’m going to need a lot of them. I’ve not started looking yet, as we don’t have a definite shoot date yet, but with that many cameras, I want to have enough people to keep a close eye on all of them - I don’t want to discover later that there are cameras that haven’t been going off properly or have the wrong settings.

For the moment, that’s about it - we’re hoping to shoot this in a month or two’s time - it’s giving us lots of time to get everything organised in advance - we all want this to work well, and, as a low-budget project, we’ve got the luxury of being able to take time over it.

Keep an eye on the blog for more info - I’ve created a category specifically for the time-slice posts - you can see all of them here.

Recent down-time

Hugh Macdonald — Sun, 11 Feb 2007 13:54:52 +0000

Ouch… that was all down for a little longer than I had planned.

We should be back up for the forseeable future now - the recent down-time was a combination of a server move that didn’t quite go as smoothly as planned, combined with christmas and a very busy period at work.

I’ll be back on track soon, I promise! In the mean-time, those of you in the US can keep an eye out for some of my recent work by watching episode 9 of the second series of Rome…

Other films I worked on that are coming out soon:
Amazing Grace (released end-February (US) or end-March (UK))
Sunshine (released mid-March (US) or mid-April (UK))

Shake Tip - Water Droplets

Hugh Macdonald — Thu, 26 Oct 2006 22:42:04 +0000

With the aim of properly getting back into keeping this thing updated, I’m trying to hit the ground running…..

So, after a little (ahem) break, I’m going to do my best to bring you weekly shake tips.

My aim with these is to try to introduce you to various tools that you may not have used much before, or to a different way of using some of the common tools…. If you’re after tips on some of the more common things that you may encounter, then I’d highly recommend Aruna’s tips of the week (which cover stuff like keying, grain matching, black points and roto)

Anyway…. Today’s tip is to do with generating water drops. It’s something that I put together at work recently (but don’t worry, work people, I recreated these examples from scratch outside of work time!) Hopefully, if you’ve not used them before, this tip will give you some idea of the power that Convolve and DisplaceX give you when used together…

The final result of this will be this: (click for quicktime (2.3Mb))

Water drops stuck to a window will be doing two things to the light that’s hitting them. They will be refracting the light coming from whatever is on the other side of the window, and they will be reflecting the light from this side of the window off their surface.

The first thing we need to generate is a matte for the water droplets. For this, I used the Fractal Noise macro from FXShare. (On the page, it says it’s for Shake 2.x, Windows only. Don’t believe it - that’s a lie - it works quite happily on 4.1, and, I’m sure, anything in between)

So…. The FractalNoise macro. Make sure it’s outputting in float, and then expand and clamp it to generate your drops image. Of course, you could always use your own painted drops map - it would work just as well… Or even an animated map with dribbling drops.

In this example, I’m going to be using this as my drops map:

These are very even drops. If you look at real drops on a window (okay, so I was bored on the bus the other day, and the window had a perfect example of what I’m recreating here), you’ll see that they are bulging at the bottom, due to gravity. To recreate that, I blurred the image, moved it down a little, and then multiplied that by the original. This is what gives the softer falloff at the top.

Next come the aforementioned Convolve nodes. Convolving is the act of using a group of pixels to decide what the new pixel is going to be. Blurs are a form of convolve, as are edge detects. The Convolve node in Shake has a number of functions that it can do, and you can even add your own. We’re going to use two of the in-built ones - SobelV and SobelH. Sobel is a type of 1D edge detect. Each pixel’s new value is the value of the old pixel one to the left minus the value of the old pixel one to the right (with SobelV, that is - with SobelH, it’s bottom and top)

What I’ve done to get the next image is a reasonable blur (to get rid of any flat section in the middle of the drops), and then a SobelV and a SobelH. I’ve then copied the green channel from the SobelH, and set the blue channel to 0.

Because we’ve been working in float, there are also negative numbers in there - the red, for example, is +ve to the left of the drops, and -ve to the right of the drops. We’re going to use this to decide how far away to displace the background image from.

Next, the DisplaceX. Firstly, I created two local parameters - xStrength and yStrength. These control how far the displacement can be. Then, with the background in the first input, and the image above in the left, and setting the parameters to be:

xExpr: min(max(x + (r * strengthX), 1), width)

yExpr: min(max(y + (g * strengthY), 1), height)

The min() and max() functions are to stop black bits from appearing in the drops near the edge - if it’s trying to get a pixel from outside the image, it just gets the edge pixels. The result of this is:

It’s not quite the effect that we’re after, but, because of the blur before the Convolves, there drops don’t touch the glass with hard edges - they fall off too softly. This is easily fixed by SwitchMatte-ing with the original drops alpha, and then putting this over the background.

Almost there with the drops. The only problem is that a lot of them disappear in the areas where the background is relatively samey. The noticable point in this example being the trees. What it’s missing is some extra highlights from the light in the sky.

For this, it really depends where the drops are on the imaginary pane of glass. If they are on our side of the glass, then we’d have a strong highlight on the top, where they’re reflecting the sky, and a softer light on the underside. If they’re on the other side of the glass (which is what I’m going for here), we’re really just adding to the refractions with a bright highlight on the bottom edge, and a softer one on the top.

For this, it’s a couple of Emboss nodes, with the elevation set to 0, and the azimuth being 180 degrees different between them. Don’t forget to clamp them, otherwise we’re going to get negative values, which we don’t want here. Blur and fade one, and blur the other a little less, add them together, and SwitchMatte them with the original drops, and we have our highlights.

For the final version, I actually added these to the refractions before their respective SwitchMattes, and then just did one SwitchMatte after the addition.

And this, blurred a little to take off the sharp edges, and over the background, gives our final movie: (click for quicktime (2.3Mb))

You can download the Shake script that I did this with from here. (don’t forget to download and install that Fractal Noise macro)

Thanks to Pete Wallington for letting me use this clip for this example. It’s from a short film called “The Long Kiss Goodbye” that I did the compositing on (so you may well be seeing this clip again…)

VFX Job Descriptions

Hugh Macdonald — Wed, 25 Oct 2006 11:40:41 +0000

Someone on VFXTalk asked recently about the various different jobs involved in visual effects. I spent a while typing out a longish answer, and, after posting it, I did a quick google for other resources on this. I didn’t find anything, so I thought I’d chuck my post up on here for your edification… It’s not so low-budget (I’d be very impressed at a low-budget film that managed to fill all of these roles!), but it’s something that might interest some of you (but also probably won’t be anything new to a lot of you)

Production

VFX Producer: Overall responsible for the business and scheduling side of the VFX.
VFX Coordinator: Works with the VFX Producer - often responsible for the artists on a specific sequence.

Supervision

VFX Supervisor: Overall in charge of the look and implementation of the shots. Will be on set to make sure that things are shot correctly. Will be the one to talk to the clients about what is needed. On bigger films, there is often a client VFX Sup, who works on the entire film and oversees the whole thing, and a VFX studio VFX Sup who will actually work with the artists. (The client VFX Sup only really deals with the director and the VFX studio VFX Sups)
CG Supervisor: Responsible for the 3D side of things - all of the 3D artists, and making sure that the compositors get what they need to make the shots work.
2D Supervisor: Looks after all of the compositors, and the 2D only shots, as well as the integration of the CG into the shots.

3D Artists

PreViz: This is a pre-production job rather than a post production job. Creates relatively rough animation so that the director can make decisions on camera angles before he gets onto set. Often used for situations where there’s going to be a lot of VFX, it also gives the actors an idea of what the shot is going to look like, even if they’re just working on a greenscreen set.
Matchmove: Tracks the live action camera move in 3D, giving a 3D camera that the CG scene will be viewed from.
Modeller: Creates 3D models of objects/characters
Texture Painter: Creates the 2D images which will be layed over the 3D models to give them colour and texture. There are usually many textures for each part of an object, defining things like the basic colour, the displacement, areas where there’s more specular reflection, etc. Not necessarily a 3D job, but it’s more tied in with the 3D side of things than the 2D side, so I’ve included it in the 3D section.
Rigger: Creates the rig (skeleton) to go inside the model so that it is easier to move bits of it without having to edit it veryex by vertex.
Animator: Takes the rigged model and gives it movement.
Technical Director (TD): A number of the more technical of the 3D jobs are classed as TDs. This is nothing to do with directing in the usual movie sense of the word.
Effects TD: Often particles - dust, snow, rain, etc. Also cloth and fluids. Bigger companies may have people who specialise in cloth, or fluids, etc.
Lighting TD: Adds lights to the scene. This is the final artist in the 3D pipeline. Often will deal with rendering too. Or the rendering will be done by a…
Render TD: Makes sure that the scenes that the lighter finishes with will render properly. This may involve optimising the scene to make sure that it will render as quickly as possible (for example, deleting all of the geometry that won’t be seen by the camera)

2D Artists

Roto Artist: Cuts out objects, often frame by frame (although interpolation is allowed). Often used for situations where a green/bluescreen was not feasable for cost/time/intelligence reasons. The roto artist may also be called upon to do rig and tracking marker removals, and other clean-up work.
Matte Painter: Creates 2D background paintings that are usually used by the compositors, but may also be passed back to 3D to be rendered in multiple layers
Compositor: Takes the rendered CG, and the live action plate, and any other 2D elements and puts them all together, producing (hopefully!) a result that looks as though it was all shot at the same time.

R & D

Shader Writer: Writes shaders, which tell light how to interract with the different parts of the model, and how to use the textures that the texture painter has created.
Pipeline Programmer: Writes the pipeline software. This is software that doesn’t actually create images that will appear on the screen, but that helps the artists workflow, and overall improves efficiency.
General R&D: Writing anything else that the artists may need. Anything from plugins for Shake or Maya through to crowd generation tools.

I hope I haven’t missed any jobs here - if you think that I have, please do let me know in the comments, and I’ll add them in.

Comments and moderation…

Hugh Macdonald — Wed, 25 Oct 2006 10:10:03 +0000

After such a long hiatus, I came back to find the front lawn covered in garbage. Fortunately, I’d turned comment moderation on for potential spam, and I had a queue of over 2000 comments. Needles to say, I didn’t go through them all one by one, so if anyone did have a comment that they posted that never appeared, I’m sorry, it’s almost certainly gone….

I’m going to try to get some better spam protection going at some point… Until then, if you’ve commented before, you’ll be fine, otherwise I’ll have to okay your comment before it’ll appear….

Back on the road again!

Hugh Macdonald — Tue, 24 Oct 2006 19:29:00 +0000

Right….. after far too long of a hiatus, I think it’s finally time to get things going on here again.

What has been going on, I hear you cry… (or is that just the voices in my head? It’s hard to tell sometimes!)

I’ve got a number of low-budget shows that I’ve worked on recently that I’ve got breakdowns planned for - I’m going to spread them out a bit. Also, a few little Shake tricks that I probably ought to turn into macros, but, for the moment, I’ll just write up descriptions of how to put them together.

So, more (hopefully) interesting stuff coming up - do let me know in the comments if there’s anything specific that you’d like to hear about.
(this would have been in about April some time…). A couple of weeks ago, I left MPC (after almost 3 years) and moved to Rainmaker UK.

So, requests in the comments, please - let me know that you’re still reading!

Updates…. what’s been going on?

Hugh Macdonald — Thu, 09 Mar 2006 13:08:38 +0000

It’s been far too long, and to those of you who’ve been checking the site on a vaguely regular basis, I apologise (I can’t believe you’re still here!)

A few updates…..

I’ve got a couple of shot breakdowns to post - one more from Enter The Zombie, and one from Darkness, which I mentioned in my last post… The Darkness one will have to wait until the episode that it’s in has been released, but that’s only a couple of weeks away.

I’ve been very busy in the last 6 months or so, but only partly with VFX work…. One of my sidelines, which has been getting a little more busy recently, is music photography… I won’t go on about this, but if you’re interested in that side of things, do check out my photography website…

As far as work itself goes, I’m currently compositing on Amazing Grace. If you want to know more about the project, do check out the IMDb page.

Anyway, watch this space for a few more breakdowns, and hopefully a return to more VFX-related blogging!

New Projects

Hugh Macdonald — Tue, 04 Oct 2005 12:11:56 +0000

Why do I always seem to get involved in projects that fall by the way-side? Okay, so it’s not all the time (the Enter The Zombie premiere was a few weeks ago), but the two I’ve been going to do that have had a more substantial amount of VFX work have both been put on indefinite hold….

Ah well - I’ve got a few more things lined up for the next few months….

James (director of “Enter The Zombie” and “Draw”) is directing a monthly series of 5 minute shorts. The VFX count is currently on 1 shot, but it should be a fun interesting one to do.

I’ve also got myself involved with a group of london film-makers, one of whom is doing a dramatic short which requires a couple of invisible VFX shots (involving text replacement on a gravestone)

Both of these are a bit of a way off, and I won’t be posting anything before they are released, but there should be something to keep people going…