Another year – another NAB exhibition. A lot of fun stuff to see. Plenty of innovation and advances, but no single “shocker” like last year’s introduction of the Blackmagic Cinema Camera. Here are some observations based on this past week in Las Vegas.
Yes, 4K was all over. I was a bit surprised that many of the pieces for a complete end-to-end solution are in place. The term 4K refers to the horizontal pixel width of the image, but two common specs are used – the DCI (film) standard of 4096 and the UltraHD (aka QuadHD) standard of 3840. Both are “4K”. Forgotten in the discussion is frame rate. Many displays were showing higher frame rates, such as 4K at 60fps. 120fps is also being discussed.
4K (and higher) cameras were there from Canon, Sony, RED, JVC, GoPro and now Blackmagic Design. Stereo3D was there, too, in pockets; but, it’s all but dead (again). 4K, though, will have legs. The TV sets and distribution methods are coming into position and this is a nonintrusive experience for the viewer. SD to HD was an obvious “in your face” difference. 4K is noticeably better, but not as much as SD to HD. More like 720p versus 1080p. This means that consumer prices will have to continue to drop (as they will) for 4K to really catch hold, except for special venue applications. Right now, it’s pretty obvious how gorgeous 4K is when standing a few feet away from an 84” screen, but few folks can afford that yet.
Interestingly enough, you can even do live 4K broadcasts, using 4K cameras and production products from Astro Designs. This will have value in live venues like sporting events and large corporate meetings. A new factor – “region of interest” – comes into play. This means you can shoot 4K and then scale/crop the portion of the image that interests you. Naturally there was also 8K by NHK and also Quantel. Both have been on the forefront of HD and then 4K. Quantel was demonstrating 8K (downsampled to a 4K monitor) just to show their systems have the headroom for the future.
ARRI did not have a 4K camera, but the 4 x 3 sensor of the ALEXA XT model features 2880 x 2160 photosites. When you use an anamorphic 2:1 lens and record ARRIRAW, you effectively end up with an unsqueezed image of 5760 x 2160 pixels. Downsample that to a widescreen 2.4:1 image inside a 4096 DCI frame and you have visually similar results as with a Sony or RED camera delivering in 4K. This was demonstrated in the booth and the results were quite pleasing. The ALEXA looked a bit softer than comparable displays at the Sony and RED booths, but most cinematographers would probably opt for the ARRI image, since it appears a lot closer to the look of scanned film at 4K. Part of this is inherent with ARRI’s sensor array, which includes optical filtering in-camera. Sony was showing clips from the upcoming Oblivion feature film, which was shot with an F65. To many attendees these clips looked almost too crisp.
In practical terms, most commercial, corporate, television or indie film users of 4K cameras want an easy workflow. If that’s your goal, then the best “true” 4K paths are to shoot with the Canon C500 or the Sony F55. The C500 can be paired with the (now shipping) AJA KiPro Quad to record 4K ProRes files. The Sony records in the XAVC codec (a variant of AVC-Intra). Both are ready to edit (importer plug-ins may be required) without conversions.
You can also record ARRI 2K ProRes in an ALEXA or use one of the various raw workflows (RED, Canon, Blackmagic, Sony, ARRI). Raw is nice, but adds extra steps to the process – often with little benefit over log-profile recording to an encoded file format.
With the shake-up that Apple’s introduction of Final Cut Pro X has brought to the market, brand dominance has been up for grabs. Apple wasn’t officially at the show, but did have some off-site presence, as well as a few staffers at demo pods. For example, they were showing the XAVC integration in an area of the Sony booth. FCP X was well-represented as part of other displays all over the floor. An interesting metric I noticed, was that all press covering the show on video, were cutting their reports on laptops using FCP X. That is a sweet spot for use of the application. No new FCP X news (beyond the features released with 10.0.8) was announced.
Adobe is currently the most aggressive in trying to earn the hearts of editors. The “next” versions of Premiere Pro, SpeedGrade, Audition and After Effects have a ton of features that respond to customer requests and will speed workflows. Adobe’s main stage demos were packed and the general consensus of most editors discussing a move away from FCP 7 (and even Avid) was a move to Adobe. In early press, Adobe mentioned working with the Coen brothers, who have committed to cutting their next film with Premiere.
The big push was for Adobe Anywhere – their answer for cloud-based editing. Although a very interesting product, it will compete in the same space as Quantel Qtube and Avid Interplay Sphere. These are enterprise solutions that require servers, storage, software and support. While it’s an interesting technology, it will tend to be of more interest to larger news operations and educational facilities than smaller post shops.
Avid came on with Media Composer 7 at a new price, with Symphony as an add-on option to Media Composer. The biggest features were the ability to edit with larger-than-HD video sources (output is still limited to HD), LUT support, improved media management of AMA files and background transcoding using managed folders (watch folders). In addition, Pro Tools goes to 11, with a new video engine – it can natively run Avid sequences from AAF imports – and faster-than-real-time bounce. The MC background transcode and the PT11 bounce will be time savers for Avid users and that translates into money saved.
Avid Interplay Sphere (announced last year) now works on Macs, but its main benefit is remote editing for stations that have invested in Interplay solutions. Avid is also bundling packages of ISIS storage, Interplay asset management and seats of Media Composer at even lower price points. Although still premium solutions, they are finally in a range that may be attractive to some small edit facilities and broadcasters, given that it includes installation and support.
The other NLE players include Avid DS (not shown), Quantel Pablo Rio, Autodesk Smoke 2013, Grass Valley EDIUS, Sony Vegas, Media 100 (not shown) and Lightworks. Most of these have no bearing in my market. Smoke 2013 is getting traction. Autodesk is working to get user feedback to improve the application, as it moves deeper into a market segment that is new to them. EditShare is forging ahead with Lightworks on the Mac. It looked pretty solid at the show, but expect something that’s ready for users towards the end of the year. It’s got the film credits to back it up, so a free (or near free) Mac version should shake things up even further.
One interesting addition to the market is DaVinci Resolve 10 gaining editing features. Right now the editing bells-and-whistles are still rudimentary, though all of the standard functions are there. Plus there are titles, speed changes with optical flow and a plug-in API (OpenFX). You can already apply GenArts Sapphire filters to your clips. These are applied in the color correction timeline as nodes, rather than effects added to an editing timeline. This means the Sapphire filters can be baked into any clip renders. The positioning of Resolve 10 is as an online editing tool. That means conforming, titling and trims/tweaks after grading. You now have even greater editing capabilities at the grading stage without having to return to an NLE. Ultimately the best synergy will be between FCP X and Resolve. Together the two apps make for a very interesting package and Apple seems to be working closely with Blackmagic Design to make this happen. Ironically the editing mode page looks a lot like FCP X would have looked with tracks and dual viewers.
I was reading John Buck’s Timeline on the plane. Even though we think of the linear days as having been dominated by CMX, the reality was that there were many systems, including Mach One, Epic, ISC, Strassner, Convergence, Datatron, Sony, RCA and Ampex. In Hollywood, the TV industry was split among them, which is why a common interchange standard of the EDL was developed. For awhile, Avid became the dominant tool in the nonlinear era, but the truth is that hasn’t always been the norm – nor should it be. The design dilemma of engineering versus creative was a factor from the beginning of video editing. Should a system be simple enough that producers, directors and non-technical editors can run it? Sound familiar?
When I look at the show I am struck at how one makes their buying choices. To use the dreaded car analogy, FCP X is the sports car and Avid is the truck. But the sports car is a temperamental Ferrari that does some things very well , but isn’t appropriate for others. The truck is a Tundra with all the built-in, office-on-the-road niceties.
If I were a facility manager, making a purchase for a large scale facility, it would probably still be Avid. It’s the safe bet – the “you don’t get fired for buying IBM” bet. Their innovations at the show were conservative, but meet the practical needs of their current customers. There simply is no other system with a proven track record across all types of productions that scales from one user to massive installations. But offering conservative innovation isn’t a growth strategy. You don’t get new users that way. Media Composer has become truly complex in ways that only veteran users can accept and that has to change fast.
Apple FCP X is the wild card, of course. Apple is playing the long game looking for the next generation of users. If FCP X weren’t an Apple product, it would receive the same level of attention as Vegas Pro, at best. Also a great tool with a passionate user base, but nothing that has the potential of dominating market share. The trouble is Apple gets in its own way due to corporate secrecy. I’ve been using FCP X for awhile and it certainly is a professional product. But to use it effectively, you have to change your workflow. In a multi-editor, multi-production facility, this means changing a lot of practices and retraining staff. It also means augmenting the software with a host of other applications to fix the short-comings.
Broadening the appeal of FCP X beyond the one-man-band operations may be tough for that reason. It’s too non-standard and no one has any idea of where it’s headed. On the other hand, as an editor who’s willing to deal with new challenges, I like the fast, creative cutting performance of FCP X. This makes it a great offline editing tool in my book. I find a “start in X, finish in Resolve” approach quite intriguing.
Right now, Adobe feels like the horse to beat. They have the ear of the users and an outreach reminiscent of when Apple was in the early FCP “legacy” era. Adobe is working hard to build a community and the interoperability between applications is the best in the industry. They are only hampered by the past indifference towards Premiere that many pro users have. But that seems to be changing, with many new converts. Although Premiere Pro “next” feels like FCP 7.5, that appears to be what users really want. The direction, at least, feels right. Apple may have been “skating to where the puck will be”, but it could be that no one is following or the puck simply wasn’t going there in the first place.
©2013 Oliver Peters