Film Master 3.0

Product: Film Master
Version: 3.0*
Manufacturer: Digital Vision
Price: $99,000 + options

Performance & stability: n/a*
Features: 8/10
Expected return on investment: 8/10
Support: n/a*
Longevity: 8/10
Flexibility: 8/10

Pros: Some very innovative features

Cons: Seems as though it’s playing catch-up in some areas

Bottom line: Not as many bells and whistles as some of the other systems on paper, but there’s a lot of things you’ll wonder how you’d lived without

Film Master 3.0

*This review is based on demonstrations and interviews with the developers rather than first-hand experience, so no data is available for reliability or performance.

Nucoda’s solution to the digital grading problem (the problem now of course, is not how to do it, but rather, with what), has had a very loyal following. Nucoda, now a part of Digital Vision, is perhaps not as much of a household name as companies such as Filmlight and Autodesk, but on the other hand, they bring a fresh perspective into the mix, with innovative features overlooked by their rivals, and an eagerness to please their customers.

Digital Vision’s Nucoda product line includes several different systems: Data Conform (for editorial work), Film Cutter (which adds primary colour grading, editing and compositing), and DVO Image Processing (for restoration, including grain reduction, dust removal and aperture correction). Actually, they are all the same product, but with different features locked, depending on which license you buy (this in itself should make upgrading a simple process). Film Master is at the top-end of the family, with all the features of the others, plus secondary colour grading options. But so what? There are now many competing products that can make the same claim, especially considering what a relatively small industry this is (to the point where I wonder who is actually buying new systems). But Digital Vision offers several intriguing differences from the competition.

Go to any demonstrations of a digital intermediate system, and the feature everyone wants to see is the color grading capabilities. Film Master doesn’t just offer a primary color grading toolset though, it offers a choice of toolsets. In a refreshing break from tradition, Film Master gives colorists a number of different grading models to choose from. The current toolset includes models such as RGB Balance, RGB Colour Curves, Brightness Regions, Lift Gamma Gain, Hue Luminance Saturation (with separate saturation controls for each of shadow and highlight regions, hue curves and overall input and output saturation). Clearly, that’s a lot of options (and each mode is far beyond the scope of this article).

I do wonder if this much choice is something of a double-edged sword though. I have often suspected (but have never been able to get a definitive answer) that the reason for forcing users to stick to a specific grading model was that it is designed to be mathematically robust, that is, it is the least destructive way to apply operations to an image, and so reduces nasty effects like noise or clipping or banding. In practice of course, sometimes quality has to take a back seat to the creative process, and if part of that process means freedom of grading in whichever way you wish, so much the better. If I may offer another potential downside to this methodology, it is that it may slow the grading process down. When a colorist is given a limited set of options, he or she can become proficient in the system very easily, and will learn all the stumbling blocks and possibilities, thus being able to grade faster. When presented with more decisions (which grading model is the right one to use on this scene?), I wonder if this might slow the colorist down. I’m by no means suggesting that more choice is a bad thing, I would just question whether it is necessary in this case. Before anyone goes running for the hills, I should point out here that there is an option to hide different toolsets, and that the internal processing is floating point.

Film Master 3.0

Without a doubt, one of the most impressive features in the colour toolbox is the ability to "layer" effects on top of each other, using blending modes. You could for example, desaturate an image, then apply a tint to part of it, apply another tint to a new area on another layer, and so on, and in this way create colour effects that would be incredibly complex to do using a more traditional toolset. There is no limit to the number of layers that can be applied to a shot, and as an added bonus, each has its own opacity control. I’m a big fan of this feature, and would love to see it in other products.

Borrowing from Digital Vision’s hardware systems, Film Master offers a number of options to combat defects such as the motion-compensated DVO Grain and Dust. These algorithms are the product of 20 years of R&D, providing automated grain reduction and film dust removal. Additionally, there is an aperture correction control to sharpen images, but no scratch removal. Also absent from this release is a paint tool for restoring images manually, but this will hopefully be available in the future. Overall though, Film Master’s restoration set feels rather limited in controls and tools. As with systems such as Lustre, it is unlikely that you will be using this system as your primary restoration solution right now, but that won’t really matter because you’ll probably be wanting to use this system to do grading round the clock.

I’m not entirely convinced that the design of the system is specifically optimized for either speed or quality, as some of the grading procedures seem destructive when layered together (in the demo I saw, brightening an image, and then subsequently darken it on a separate layer, caused clipping, but I’ve since been told by Digital Vision that this clipping is actually an option), and rendering is required before it can be previewed. Moreover, whilst it is possible to recreate many effects that other systems can do at the push of a button, with Film Master these can take longer. In one instance I saw someone put a vignette on an image, and yet it required four separate layers to be created to produce the effect. Some sort of macro capability would be a definite requirement for future versions. In terms of performance, I don’t really like to comment on systems that I’ve not had the chance to experience first-hand in a production environment, but I can say that the overall performance seemed inferior to other systems. The answer to that, is clearly to throw more hardware at it, but that will of course mean spending more money. There is a GPU Grading mode, which utilises the video hardware of the system to apply real-time grading to shots, but many seasoned colorists will avoid this option due to potential problems, but at least Digital Vision give you the option to use the GPU on a per format basis, which might be ideal for running off tests, or for tweaking video masters.

Rendering can be done in the background while you continue grading other scenes, either on the local machine or on a farm (the DVO Grain and Dust processes can also be run this way). On the editorial side, Film Master provides multiple tracks, split-screening, automatic scene detection, and audio support. You can’t share projects, but you can easily edit the raw project files, export pull-lists and EDLs, and when you render graded images, the metadata will reference the source material. These and other "under the hood" features won’t be noticeable until you need them, but you’ll be grateful for them when you do. It’s proof, in any were needed, that Digital Vision listen to their end-users. When I first saw a Nucoda system, I figured it as "one to watch". In the past few months, many of the facilities I’ve been to have bought a Film Master system, and it seems that Digital Vision is quickly claiming its stake in the digital intermediate industry.

All in all , I am positive that you will not enjoy such a wealth of tools or options in any other system. In this way Film Master could easily be described as the Photoshop of colour grading. Ultimately then, the question is whether you need more creative freedom and choice, or if you would be better with a more streamlined system.

More information can be found at Digital Vision’s website

All reviews are based upon the principle that the hardware or software reviewed is to be used within a commercial digital intermediate environment; as such the review may not necessarily reflect the product’s intended purpose.

About the reviewer:
Jack James has been working with digital imaging technology for 10 years. He has worked within a number of digital intermediate environments since joining Cinesite (Europe) Ltd.’s Digital Lab in 2001 to work on HBO’s Band of Brothers.  He has a number of film credits, and has published the book "Digital Intermediates for Film & Video" with Focal Press.

The reviwer’s opinions are his own, and not affiliated with any third-party.

Posted: January 5th, 2006
Categories: Uncategorized
Add your comment

*/ ?>

Leave a Reply

Your email address will not be published. Required fields are marked *