JamieG Analysis

Sony is about to release a VTR that actually answers the needs of a modern Post Production House.  This has been a long time coming.  And in some ways I would like to think I had a hand in shaping its future. (see here)

A number of years ago, back when the HDCAM deck was released, I was lucky enough to go to NAB.  During this time, I was in the market for a Telecine, of which the Cinealta was my favourite.  During this time, I was introduced to the head tech developers for the VTR division. Specifically the HDCAM guys who were there to show of the new VTR.

Being a pretentious tech, I ripped the device apart explaining to them why it is completely the wrong direction to go.  In the end, they asked me to dinner that night to continue the discussion. Apart from being a Post Production Tech, I had been trained in Computer science, and Robotics and Digital Technology while at University.  Ie, I could explain it to them on a level they could better understand then the typical Post Production tech.  In the end they are computer techs more than Post production techs.  They just build what they are told to build.

I approached the idea from the stand point of being a Post Production tech, then using my IT training to actually get to the crux of what would best suit the Post Production environment.

Following is a brief example of what I tried to teach them.

Digital = Data.  A digital betacam and beyond may be all digital based video devices, but they are all restricted/bound by non upgradable codecs and interfaces. Ie proprietary DigiBeta codec or HDCAM codec, or industry standard SDI/HDSDI and YUV interfaces.

To look at the issue more holistically I told them they needed to split the VTR up into 3 subsystems.

1.       The Data Transport.  Ie store BITS (on tape) as fast and reliably as possible.  Ie a tape data backup device.

2.       The Codec transport.  Ie where you can drop in any form of compression you like, utilising newer and better ones as they are developed.  This also makes the device very future proof.  Very important when they cost $160,000.00 AUD (Similar in US dollars) per unit. This would also allow faster than real time transfers (Well as fast as the Data Transport can go) or slower then real-time transfers but STILL SPOOLABLE of uncompressed or non-standard codecs.  At the time Cineon and DPX was what I pointed at.

3.       Independent Interfaces. YUV, SDI, HDSDI where standard industry interfaces.  But what about HDMI, DVI, Ethernet, SATA, fiber-channel, and other hi-speed data transport interfaces.

Finally I told them they needed to build a subsystem around all this with quite a bit of memory to cache a lot of the data so they can effectively use the tape mechanism.  (Tapes hate writing small amounts of data at a time.)  This also lets them build the Codec transports that can spool and view data files.  For example, imagine a stream of DPX files going to tape and the HD monitor playing/spooling  them back as it does it.

At the end of the day, I was trying to convince them that the most important issue here was DATA.  How we get it, how we use it should not be dictated by the archival medium we use.

I also tried to open there eyes to the fact that MPEG2, other forms of data in conjunction with tapeless workflow or distribution technologies would reduce the need for the VTR as it was known back then. (Well I didn’t mention this too much.  Didn’t want to scare them

The conversation went on all night.  There was a reasonable language barrier, so I was hoping describing it over and over with different scenarios would help them understand the concept.

I got a lot of “Ar-so” and other “Light going on above head” type reactions.

Amazingly enough the new SRW-5800 HDCAM SR Deck appears to be very close to what I described, with a few more modem tweaks based on current technologies.

If so, I must applaud the Sony VTR development team.  This could be one of the best products to come out of Sony in a long time.

Pity I cannot afford one.