New Announcements in the World of Hardware

There have been a few interesting product announcements this week, presumably due to the IBC show inEurope.  Most of them relate to the development of new tools for 4K workflows, which I expect will be the pattern for a while to come.

AJA announced a new 4K interface, the Corvid Ultra, which is based on the Riker technology they were showing off at NAB 2011.  It has a new scaling engine called TruScale that I hope to see in other products in the near future.  They also started shipping their T-TAP, which was announced at NAB 2012.
Read the rest of this entry »

NVidia Announces Kepler for Professionals

NVidia finally announced the first of their professional lineup of products, based on their next generation Kepler processor.  The Quadro K5000 is a high-end professional graphics card, while they also announced thr Telsa K10 and K20, which are dedicated GPU processing cards.  They were announced earlier this month during SIGGRAPH, but won’t be released for another month or two.

Besides having a terrible identifier in K5000 for product differentiation, the new Quadro looks to offer some useful new features.  It supports PCIe 3.0 for faster transfer of data between the system and the GPU, which is becoming more important with CUDA accelerated processing needing data to be sent back into the system instead of to the display output.  Speaking of display outputs, the new Kepler products support 4 separate monitors from a single card.  This will be advantageous for those planning to use it to replace a dedicated video I/O card for fullscreen monitoring, but still wanting two UI displays.  This was the one feature that AMD’s cards had over the Quadros up to this point.  The K5000 will also be able to output 4K over a single DisplayPort cable, which will become more relevent once 4K displays become more widely available.

On the Tesla front, the K10 is basically a dual chip version of the same processor that is in the K5000, while the K20 will be based on a whole new GPU design.  Both of these designs are departures from the previous Telsa products, which were basically Quadro cards with the display outputs removed.  The new K10 appears to be about four times as powerful as the existing C2075 cards, at least on paper.  They also announced the next generation of their Maximus program, marketing Telsa cards in parallel with Quadro cards, but we will have to wait for software developer to really leverage those capabilities.

Hopefully I will be able to offer a more in-depth look at the Quadro K5000, and its processing performance, once it is released to the public.

Adobe’s Creative Suite 6 Released Today

After being announced at NAB last month, Adobe has released CS6 to the public today.  Most of the new features are well known at this point, but I am most looking forward to the anticipated increases in performance and stability, now that their native 64bit engine as had more time to fully mature.  There are some cool new developments in this version to highlight though.
Read the rest of this entry »

DI Finishing for Act of Valor

The biggest challenge we faced in the DI finishing process for Act of Valor, was preparing for three different aspect ratios for our deliverable, without sacrificing resolution.  All of our footage was shot 1.78 (16×9) and we were required to deliver 2.39, 178, and 1.33 versions of the film.  The normal way to do this is to letterbox and reposition for finishing in 2.39, and then use that master to make the final 1.78 and 1.33 copies by cropping out the excess on the sides.  The problem is that you lose a nearly half of your image resolution in that process, which we couldn’t afford to do with the 5D source material, if we wanted a crisp looking 16×9 version for Blu-Ray and other 1080p distribution.

Our solution was to do our entire post process on the full 16×9 image, with software letterboxing applied for monitoring.  This allowed us to maintain the full scope of our image throughout the post process, but required some creative project management.  All reframing had to be deferred until after we split out the versions for different aspect ratios.  That meant that all titles had to be applied after that point as well, so they wouldn’t get repositioned out of sight in 2.39.  And any changes we made to the movie after we split into different versions would need to be made to each version, and carefully tracked.
Read the rest of this entry »

Visual Effects on Act of Valor

At Bandito Brothers, we have always prided ourselves in being authentic and real.  This is one of the values that led Scott and Mouse to cast real Navy SEALS for the main roles in Act of Valor, and doing things like using live ammo in certain scenes.  So at first glance, it would seem surprising that we had hundreds of visual effects shots in the movie.  About half of them were in response to issues that resulted from shooting with the 5D, but we also did have to add things like blood hits, and do fixes like painting out crew members.  Additionally, we had a fair bit of motion graphics work, both overlaid graphics for the audience, and screen replacement for briefing information.
Read the rest of this entry »

Shooting 4K for HD Delivery

So this year at NAB, we continue to see more 4K acquisition tools being developed, but nearly all content is distributed and viewed by its audience at 1080p or below, and will be for the foreseeable future.  So do we really need 4K, if we aren’t finishing to a 4K DI?  Having 4K resolution available offers some interesting workflow options, but people need to avoid getting caught up in 4K output aspect.  Almost all 4K cameras are single sensor devices, with a bayer pattern to differentiate colors.  While a good demosaicing algorithm can interpolate the detail at full resolution, it will never be to the same level as a three chip camera.  Instead, if you treat each 4 sensor block as a single pixel, you only end up with half of the resolution on paper, but you end up with the best of both worlds.  You have a camera with a single large sensor, which better simulates the optical response of traditional film, and you have dedicated sensor locations for each color.  This is the principle behind the Canon C300 having a QuadHD sensor, and only recording full color 1080p.  A more ideal solution is to record the RAW single channel QuadHD or 4K, and treat it as half that resolution, like the “1/2Res-Fast” decoding option in Red’s software.  There are ways to encode that data over 3G SDI that involve using the alpha channel space to carry the extra green data, effectively giving you 8:4:4 color in a sense, if measured from that half res perspective.  This workflow has been enabled by Red for a while, but Canon’s new tools force this route.  While it is happening blind to the user, understanding the underlying ideas can allow users to better leverage the capabilities.
Read the rest of this entry »

NAB 2012 – Part 2

So I have had a bit more time to get browse the show, and while I haven’t come across anything that totally amazed me, there are a variety of little things that have caught my attention.  GoPro’s Hero2 now shoots 24p, and has some advanced color profile options, both of which will make it much more suitable for my company to potentially use.  But I think we are pretty satisfied with Canon DSLRs as our lightweight cameras for now.
Read the rest of this entry »

NAB 2012 – Part 1

So NAB is here again.  4K and Thunderbolt seem to be the most popular topics at the show, with 3D hardly being advertised.  There are a lot of new products that have been announced over the last few days.  I have only gotten through the lower South Hall so far, but I will hit the rest of the show over the next couple of days.  (My series of articles on Act of Valor will be continued once NAB is over.)
Read the rest of this entry »