— ROGERS

Archive
Tag "RED One"

lens.jpgThere is quite an interesting (old) post on Indie 4k about Birger Engineering’s 35mm stills-lens mount adapter for the RED One. The interesting bit is not the lens mount itself, but that you will be able to control aperture and focus from the camera, as well as getting “metadata” (much like the Cooke /i Technology), so your focus, aperture and lens type are recorded at the same time as the footage. The lens and the body, like, talk!

It is a screw-in/dust free job (similar to the already offered Canon FD and Nikon N adapters), so not a super-neat-o-pop-on-and-off solution. Not that you’d probably want that anyway.

Of course, pulling focus might be a bit of a problem with a stills lens—but that can be solved by using the related Birger follow focus disc. Or if you are cheap or otherwise inclined, with a third party addition such as the RedRock microLensGears which just clamp around the lens.

From reduser.net:

The Canon EOS EF Mount, designed by Birger Engineering, will enable the use of Canon EOS EF lenses on the RED ONE. Focus and iris will be controlled via the RED SuperGrip or by Birger’s own wired or wireless interface. Focus control is in 4,096 steps from CF to infinity. The Lens mount also requires a $75 LEMO cable. The follow focus disc is $300 and the required LEMO cable to connect it is also $75. There will be a wireless controller with an LCD and two control surfaces [a slider for iris and knob for focus] available for $600.

Available (or soon to be available) mounts are:

Canon EF-S (control iris & focus w/ “EFâ€� & “EF-Sâ€� lenses; available 15 December)
Sigma SA (control iris & focus; available 15 January)
FourThirds (control iris and focus; available 15 February)
Nikon F (control iris with “Dâ€� & “Gâ€� lenses, focus with “AF-Sâ€� lenses; available15 March)

Whichever way, it would seem like a pretty cheap way of expanding the lenses at hand, especially if you don’t have a case of Cookes sitting next to you. Considering RED branded lenses are rumoured to be re-stamped (and re-housed) Sigmas, perhaps there is not too much optical consideration required?

However, there is a question of what crop-ratio this lens adapter gives you. Of course, it will be the same as a 35mm motion camera will give you, and not be what a stills camera will give you. This is because the RED sensor (24.4×13.7mm) is much the same as a lot of digital SLRs (for example, Canon’s 40D is 22.2×14.8 mm). Which is also much the same size as a Super 35 frame (funny that). A 35mm stills image is larger (36mmx24).

36mm ÷ 24.4 = 1.475

The disparity between the digital sensor size, and 35mm negative size means you get ~1.475x “sensor crop“. Or in other words, if you have a 50mm Canon EOS lens, it will need to be multiplied by 1.475x to get the equivalent “35mm stills” field of view (in this case, 74mm).

As I mentioned, this won’t make a lot of difference, or be particularly enlightening if you are used to shooting (super) 35mm motion—because the result is the same.

Rather, it is just something to bear in mind if you are going to go the stills lens route. What you see through the lens in the RED or a 35mm motion camera, won’t be the same as what you see through a standard-sized stills camera. So your wacky 10mm lens is going to turn into a rather sensible 15mm (okay, okay, a 14.75mm [10mm x 1.475])…

Maybe it was time to trade in that Hawaiian shirt for something neat and black anyway…

Read More

FRAMES AND TIMECODE

With everyone getting into a lather over DI, it is worth bearing in mind that when you have material on tape (or Quicktime) and edit it offline using timecode… then your data (DI material) needs to be matched back to that same timecode when you do the online conform.

You would think that this is obvious, but after hearing a recent horror story about an established post company failing to do that, I think it is worth spelling out a couple of techniques which will save a lot of eye-matching pain.

First, the easy way. The DPX file format has a place to save the timecode information inside it. The downside is that a lot of programs ignore this. Stop here if you are lucky enough to have one that does.

Second, the brute force method. First a bit of hoary theory, from me:

Think of timecode as just a frame counter, which uses the format of time, like a digital clock, to count. The lowest increment is a frame; then seconds; minutes; and hours.

Just like you can calculate the number of seconds in an hour (60 secs x 60 minutes); you can calculate the number of frames in an hour (e.g. 24 frames* x 60 secs x 60 minutes). Therefore, you can express a timecode as a number of frames.

For example 02:10:10:04 (24 fps) = 187444 frames.

So then, if you were to put that into a typical DPX sequence:

myfilm.00187444.dpx = myfilm (frame 02:10:10:04)

*The only important thing to remember is what frame rate you are dealing with (30 or 25 or 24 etc).

A lot of compositing and DI packages are able to translate frame numbers into timecode (RED Cine does this, Flame, Shake, After Effects, Film Master all do it too). It makes the transition between tape or Quicktime and sequences of images very easy. It also means you are not ripping your hair out when it comes to doing a conform.

Of course, this isn’t a new way of working, nor is it rocket science. If you are acquiring on film and scanning, or using one of those new-fangled digital cameras, it is all the same.

Don’t panic.

And if that seems like incomprehensible rubbish, it probably is. If you are still with me, then try this.

Read More

WORKFLOW

Another day of Red queries. I’ve heard that some rental houses are requiring a “RED Technician” to go with their cameras. Sounds like more FUD. This is a great pity, because at the core of it, there is no more to a RED than your average video camera… now I’ll be accused of gross simplification… How about this: the camera can do a lot, but you don’t have to think about all those things all the time.

Okay, I understand productions wanting to be cautious, but really, all that is changing is that the clapper loader is unloading and loading “film” data with a laptop. If someone can’t cope with that, then God knows how they ever managed film. Previewing a clip (aside from the realtime video streaming out of the camera) is just a matter of double-clicking a quicktime. Not a big deal.

I know there will be problems with firmware and the like, but what camera hasn’t had its share of bugs?

The on-set workflow can be scaled up and down in complexity. Bottom of the ladder would be to take the data and copy to a disk via a laptop. Further up would be some sort of little on-set network between Camera department and the playback person. Mix to taste, and don’t forget to backup.

I think I should make some sort of diagram.

Read More

Well, that’s it. Having seen the RED camera footage start streaming through the door, I’ll say this to the detractors and the fan boys: it’s a bloody digital camera, get over it. Yes, it is better than Genesis (which is just one of these with a little Panavision sex). I’ll justify the Genesis statement another day. No, the RED is not “better” than film.

At ~$17k USD for a body (the Sony equivalent is $116k USD list), it is going to gouge out a big niche in the market. Think of the impact of cheaper digital stills cameras. And, you know what? It is exciting.

If you are able to expose correctly (don’t get me started), then it will turn out great results. It—and other cameras like it—will rule. In Sydney, the few that are available for rent are working their little sensors off. Good luck getting one. The new Scarlet camera—considering the little anyone knows is only that it will be cheaper and smaller—will rule as well. Say hello to a new breed of DP’s, just like the new photographers that have been born of the digital stills industry.

The image quality of the RED One is so close to a RAW digital still, that there is not question that it will be successful. Panavision, Sony, ARRI, et al should be shitting their pants, or at least coming up with Plan B.

Anyway, enough of the pontificating.

Under the current RED compression and storage scheme, which is writing to 8GB Compact Flash cards, I thought that the image quality would be pretty awful, especially when pulled apart. It isn’t. At least it is no worse than the artefacts you would see on other video formats.

Red deconstruction

As you can see in the above image, the noisiest channel is blue. Red and green survive pretty well. I know there are new implementations of the compression coming out seemingly daily, but it is interesting to see where they are at now. The HLS image is probably the most telling of how the compression is working and what you have to colour grade with, where the RGB will make a big difference if you are doing visual effects (hint: shoot greenscreen). The RAW image sits in that strange not-quite-log space of RAW digital stills. There is plenty toe and shoulder (for a digital image) to play with.

In short, it is impressive. Considering that it will only get better, I think this should be an intriguing year.

Read More