RSS icon Bullet (black)
  • So, About That DaVinci Thing…

    Posted on May 16th, 2010 Mike 4 comments

    Due to my work schedule, I did not attend the NAB convention in Las Vegas this year. Which, as it turns out, is a pity, because it was probably one of the most interesting gatherings in quite some time. Lots of interesting announcements on various fronts, involving some of the more significant players in our business, including Arri (with the Alexa camera line), Aaton (showing the proposed digital back for the Penelope camera), Assimilate (showing a new version of Assimilate Scratch working with Arri RAW files in real time), Filmlight (some very interesting new things coming, including integrated Red Rocket support, and support for Sony’s new software version of the SR codec, allowing for some very efficient file based workflow enhancements), and Avid (Media Composer 5, one of the most significant upgrades of that software in years). And I’ll be writing about all of these things in time. But perhaps the most significant – and potentially disruptive – announcements came from Blackmagic Design regarding their newly acquired DaVinci product line. Back in January, I predicted here that DaVinci would perhaps wind up being the most disruptive player in the software end of the business this year. I turned out to be right, but my prediction wound up understating the length to which they would go to ensure that this would be the case. By lowering their cost of entry, at least for the basic software, to a point one could almost call free, they made a statement of their intention to attempt to create a much larger market for the product, one that would potentially include everyone from hobbyists to the highest end DI facilities. And by making the product dual platform (Linux and Mac OS X – NOT Windows), they clearly hope to attract those who are using Apple’s Color product, which ships with the Final Cut Pro suite, to a much more capable and well regarded platform, as well as retain those facilities who are already using the Resolve product on Linux for high end finishing work. By keeping the versions in lock step, they are also attempting to position the Mac product as an inexpensive conform/render station for the high end users who are doing their color work on the more capable Linux version.

    In order to better analyze what all of this means, a little history is in order. DaVinci has been the premier vendor – at least in the US – of high end color correction systems for many years now, with most of the major post facilities using DaVinci systems (the basic DaVinci Unified Color Corrector at first, followed by the Renaissance, the Renaissance 888, the 2K, and the 2K Plus) for nearly all television, feature mastering, and commercial work, both in combination with telecines running film live through the DaVinci, and for “tape to tape” color work, common in television series finishing since the late 1980’s. All of the aforementioned devices were basically hardware driven. They had to be, because for many years, running live video in real time demanded that approach. In the last 5-10 years, however, commodity level hardware, combined with sophisticated software, has been able to take over a lot of that burden, minimizing the need for customized hardware to do what the DaVinci devices have done for a long time. DaVinci recognized this a few years ago, and set off to develop a “software DaVinci,” a device that would do what the 2K Plus could do, but in a resolution independent way, and with a minimum of proprietary hardware. They wanted to make the software version familiar enough to the DaVinci user base that colorists would find it simple and comfortable to move from the old platform to the new. This development effort ultimately produced the product now known as Resolve. The product was based on completely new software, running on a Linux platform, and assisted by some specific proprietary hardware to do a number of functions in real time. Some high end facilities – Company 3 perhaps being the most prominent – liked what DaVinci brought to the table, both for its capabilities when compared to the 2K Plus, and for its familiarity, its ability to do things “the DaVinci way.” About a year ago, Blackmagic Design acquired all of DaVinci’s assets and set out to remake the product as a completely software based offering, moving all of the functions done in DaVinci’s proprietary hardware boards to commodity GPU’s. A year later, that transition is now complete, and what Blackmagic was showing is really the first true pure software DaVinci product. The marketing, however, is now likely to be quite different, with Blackmagic clearly wanting to attract individual artists as well as the full facilities into the fold.

    I’ve gone through this little bit of history for a reason. During the time that DaVinci was developing what became Resolve, other players came on to the color grading scene in a major way. This was due in part to the rise of digital intermediate work, which by its nature demanded a very full featured, resolution independent, unlimited layer approach in order to be an attractive and successful alternative to traditional photochemical film timing. The 5D Colossus, which eventually became what is known today as Autodesk Lustre, led the way into this new, less limited electronic color grading world. It was followed by systems such as Nucoda (today Digital Vision) Film Master, Quantel iQ with QColor (which led to the more sophisticated color product called Pablo), Filmlight’s Baselight system, and some less costly but quite capable desktop systems, best exemplified by Assimilate Scratch and Iridas Speedgrade. Something that a number of these systems – particularly Nucoda, Quantel, Lustre, and Baselight – have in common is their ability to deal with multiple tracks of video, and conform based on more sophisticated exchange formats like AAF and Final Cut XML files. This is due in part to their development for a market that was already advancing beyond the basic “tape to tape,” single track, no effects world in which the DaVinci product line had been created. While it can certainly be argued that Resolve’s color tool set is as sophisticated and precise as any product out there, it can also be said that unlike the competing high end products, it brings little to the table in terms of enhanced workflow possibilities, at least in its current incarnation. It does not support more than one video track, it cannot conform multitrack projects, and partly as a result of that, does not easily support some of the effects that are common in modern editing approaches. It is, at its heart, a software driven “tape to tape” system. For many, especially highly experienced DaVinci trained colorists, this is a moot point. But for others, especially facilities seeking to modernize and streamline their post workflows to promote more efficiency and ultimately more flexibility, the competing systems have  a lot to recommend them. For the high end facility, it’s imperative that file based post production be facilitated by equipment that has the flexibility to understand more of what the editing systems can do and to deal with that in a more approachable way. I have little doubt that Blackmagic will ultimately move Resolve towards that goal, but in its current form, it’s not always the best fit for some facilities, even at its amazingly lower price. This is borne out by the reaction – or rather, the lack of it – of the competing players to Blackmagic’s pricing moves. While Filmlight did offer a complete entry system at a steep discount from its previous price point, the new price point is still over $90K, a bit higher than the expected total cost of a Linux based DaVinci “starter system” with comparable capabilities, which by my estimate would probably run about $80K (with full hardware comparable to the “base” Baselight). And to my knowledge, no such moves have been made by any of the other players. Perhaps everyone is still getting over the shock, and playing a waiting game for a few months to see the actual impact of Blackmagic’s moves. One thing, however, is certain: the color correction market now has its potentially disruptive product, marketed by a company with a history of such things. The next year or so will be, at the very least, interesting.

     

    4 responses to “So, About That DaVinci Thing…” RSS icon

    • Charles Taylor

      That’s interesting that Resolve only supports one video track, and can’t do a multi-track conform. That will take a lot of the interest out of the product for people looking for their first grading system, I think.

    • Seems time to revisit your thoughts about Resolve now that version 8 (shipping in June) has been announced to include a multi-track timeline. Is it now the obvious choice for anyone not already invested in the other more-expensive platforms? Or can you still justify the costs of Nucoda, Baselight, etc…?

    • If price were the only consideration, that might be the case. But it’s not. Each of the major grading programs has advantages and disadvantages, and each has specific things that appeal to specific groups of users and colorists. In DI facilities, these things don’t generally exist as standalone devices, they are part of a larger system, and as such need to fit into the specific workflows used by that facility, which are often built around needs that are specific to their particular work mix. That’s why there are choices, and that’s why the choices are all valid depending on the need. Price is only one component of a much larger and more complex picture.

    • Resolve has a large user base using the free version, here in Australia.
      All post-production tools, cameras etc are now quite cheap so I think the days of large investment in media tools is almost over. The high end world will keep developing new tools but for 80% of the market, the cheap stuff is now incredible value.

    Leave a reply