Friday, November 28, 2008

RED camera - early test data







All images © George Monteiro, all rights reserved

I'm not a videographer. But, I am a big fan of any means to get a better still image, and that includes using video to do so. Which these days, includes technology introduced by the RED system. http://www.red.com/

There has been a lot of chatter on various photography user groups about the potential of the RED digital cameras, in their current iteration and in terms of new products predicted. I'd tell you more about it, but truthfully don't know much more than I read on the web. This from RED's website: Typical high-end HD camcorders have 2.1M pixel sensors and record with 3:1:1 color sub-sampled video at up to 30fps. RED offers the Mysterium ™ Super 35mm cine sized (24.4×13.7mm) sensor, which provides 4K (up to 30 fps), 3K (up to 60 fps) and 2K (up to 120 fps) capture, and all this with wide dynamic range and color space in 12 bit native RAW. At 4K, that’s more than 5 times the amount of information available every second and a vastly superior recording quality. In addition, you get the same breathtaking Depth of Field and selective focus as found in film cameras using equivalent 35mm P/L mount lenses.

Of course, most are talking about the RED as competition to 35mm film for cinema projects, or video that massively out-resolves existing high definition technology. But, there is also the camp that contends that it won't be long until, for some types of photography, single frames from the RED will be competitive with the capture quality from still cameras. Imagine a sports shooter covering the 100-meter dash at the Olympics. Put the RED on a tripod and let it roll. Send the clip back to an editor and let them decide what the iconic, decisive frozen-moment-in-time might be. At the same time, there is video clip that can be used for the client's website. RED for web is overkill perhaps, but it will be done. As bandwidth and hard-drives increase in speed and capacity, it all seems very plausible.

Given all the excitement and hype, I've been very eager to see the current state of the art in RED image capture, and when my friend George Monteiro (from Sea-Cam video productions) stopped by my studio recently (he was down to do a test shoot underwater in Key Largo), I asked him to e-mail me a few sample JPGs from the day. Obviously, within the context of a blog you'll never be able to decipher image quality variables, but when I dug into the files in Photoshop I made a few basic deductions:

1. RED topside - The shot of the covered bridge is quite impressive. It was transmitted as a small JPG, but opened in Photoshop as a 24MB, 8-bit file. That's about the size file I would expect from a 10-12 megapixel digital still camera. Not necessarily all the detail I'd expect to see from a 12-megapixel camera, but considering this is a still frame from a video, amazing. It held detail quite well in the 100% enlargement. See the screengrab from Actual pixels in Photoshop. Considering the context, really a significant achievement in technology.

2. RED underwater - Here's a few of George's comments in his post to me. Here are a few test stills I pulled from yesterday's shoot. Please don't judge them for composition most are from the middle or end of a tracking or pan shot. But they will show you the native resolution of the red in 4K mode at 30FPS with a wide open shutter (1/30th of a second) so you will see motion blur in the close fast moving fish. The images have been compressed as Jpegs to about a half meg each. They were shot in natural light with a UR Pro with a dome port using the 18mm setting on the wide angle Red zoom lens ... I color corrected them for maximum dynamic range in RED Cine and used the various white balance features to achieve what I thought was agood balance. They may be a little contrasty and over saturated but this was my first attempt with underwater footage with RED

He made the other significant comment that it was all shot at 1/30th of a second. I asked why 1/30th, immediately thinking back to the very old analog days when the Pentax 6x7 I bought was essentially DOA for underwater use because it only would synch with strobe at 1/30th second and slower. 1/30th was way too slow for most things I shot on the reef, and only acceptable with wide angle shots with models, or shots in low ambient light. He explained, logically enough, that choosing faster shutter speeds made the video less "fluid" and more choppy. Faster shutter speeds would be better for freezing the action of moving fish in a still frame grab, but may not be the perfect solution for optimizing video. Apparently, that will be a consideration when choosing shoot parameters primarily for video (motion) versus primarily for extracting stills.

George was dialing in a new dome, and in looking at the JPGs I see he probably missed focusing on the virtual image correctly, as the underwater shots aren't as sharp as the topside shots he showed me. However, finding the exact nodal point for a zoom lens is a complicated matter, and getting it right the first time would be a lucky thing indeed. Still, if the camera renders a sharp still frame topside, there is no reason it won't do the same underwater with the right port and port extension. The RED zoom focuses very close, so I doubt that it will need a diopter to focus on the virtual image, once the focus "sweet spot'' is determined for the dome of choice. George already has a plan to improve the result in his next dives.

The same rules that apply to minimizing optical aberrations with a housed still camera will apply to the housings for the RED. For moving pictures there is probably some latitude for smearing corners, but the higher the camera resolution, the more optical flaws will be evident. And, when those optical flaws are frozen, in a still frame, they are ever more obvious.

From this it is clear that RED bears strong potential for use underwater, but the housing manufacturers will have to get very serious about dome performance if these images are to hold up to publishing standards, competitive with existing still technology. Still, there will be some subjects that might never be captured any other way, and for these the RED will be marvelous.

Of course, technological convergence cuts both ways. As RED creates video technology that encroaches on the potential of still imaging, Canon has just introduced their 5DII still digital camera, capable of capturing high definition video, separate or even simultaneous with a 21-megapixel still digital photo. And as RED gets ready to introduce their new modular system, you can bet that our traditional camera manufacturers are pursuing revolutionary upgrades of their own.

These are fascinating times to be a shooter, for sure.

2 comments:

Call Box said...

Wow- this is very cool. It's nice to see uses other than motion picture photography for the RED camera. Got it on the blog.

Valencia said...

I recently came accross your blog and have been reading along. I thought I would leave my first comment. I dont know what to say except that I have enjoyed reading. Nice blog. I will keep visiting this blog very often.


Joyce

http://www.shunmigraine.com