Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

UHDV, 240 fps and HDR with Red Epic?

Hey Jim, I thought you weren't going to "Pre-Announce" stuff anymore with dates? But, I guess "late 2012 or early 2013" is a span of potentially 12 months, so you've got time. Heh Heh.
He will never change... ! :-)
 
I can find many purposes for 120 fps or 240 fps:
[*]One can make a 240 fps video easily a 24 fps video by taking every tenth of the frames.
No, the relative shutter speed will be 18 degrees instead of 180 degrees making the motion flickery.

[*]One can make a 240 fps video easily a 60 fps video by taking every fourth of the frames.
No, the shutter speed will be 30 degrees instead of 180 degrees.

There are good methods to convert from higher frame rate to a lower frame rate, but those mentioned by you aren't among them.

[*]One can show all the 240 frames in slow motion with any display/projector. One doesn't need to have a 240 fps monitor/projector. Of course, I don't mean that the whole movie should be seen in slow motion, but some scenes could be in slow motion.
I think most punters would regard it a waste of time to watch a movie in slow motion.:Yawn:

[*]There are already lots of genuine 120 Hz monitors and projectors. If one shoots the movie at 120 fps, one could already see the movie @ 120 fps.
No, the monitor refresh rate has very little to do with movie frames per second.

[*]If one shoots the film @ 24 fps, there is no convenient way to make it 50 fps, or 60 fps or 120 fps or 240 fps or whatever. One only has 24 frames per every second. It's a bit like those efforts to convert black-and-white films into colour or 2D films into 3D films. It is possible, but it is not the real deal.
25fps and 30fps releases of 24 fps movies have been made for decades. The former by speeding up and the later by 2:3 pulldown (see Telecine)

I am also not sure what glass you would use to resolve beyond 8k with decent MTF.

Of course, it would be nice to have very high spacial and temporal resolution but, as others explained, this is not practical or even possible and not demanded. Maybe in a few year's time. Until then, enjoy RED One MX and the Epic, which are cutting edge any way you look at what these cameras produce.
 
that's not a unique look, that's what every picture looks like when someone first discovers HDR in photoshop... and squeezes everything in.

don't believe me? try a search for hdr photography

Hmmm.

Well, it does sort of look like that now that you mention it. (I followed your suggestion and looked. Thanks!) I'm not really into HDR so I probably wouldn't have considered it without someone mentioning it.

A look doesn't have to be unique or even particularly good to be useful though. I stand by my opinion, its an ugly picture of an ugly thing in an ugly place, and that can work best for some stories. Same with my opinion of the Kinkade naptime pictures.
 
Late 2012 or early 2013 you will have 9K at 48fps. Add a couple of years for what you are asking for...

Jim

I love you man.

I have to say I'm gonna be happy with 5K at 120fps for a long long time, but when I'm done with 5K ... its good to know you'll have the next thing ready for me.
 
  • One can make a 240 fps video easily a 24 fps video by taking every tenth of the frames.
  • One can make a 240 fps video easily a 60 fps video by taking every fourth of the frames.

No, not "easily".

Motion is affected by shutter speed.

If you shoot 240fps at a high shutter speed and try to make it 24p or even 60p it will be really stroby.

If you shoot 240fps at a low shutter speed 24p and 60p will look "better" but now your slow-mo will look like a crappy blurry mess.

There is no magical setting that gives you everything. You shoot for one of the other.

Don't get bogged down by trying to "future proof" or whatever. That's silly. Go with what is right for your project. Something as basic as what framerate you shoot should be decided BEFORE you shoot, not after. I recently saw a video where they shot everything at 120fps and conformed down to different framerates for different shots, and while it was "cool" the end result to me looked like shit.

If you're dealing with RED, how much more "future proof" do you need?

You know why they call it RED? Cause it's on the fucking bleeding edge of digital cinema! I cut my fingers on the keyboard just typing that... :biggrinjester:
 
Hmmm.

Well, it does sort of look like that now that you mention it. (I followed your suggestion and looked. Thanks!) I'm not really into HDR so I probably wouldn't have considered it without someone mentioning it.

A look doesn't have to be unique or even particularly good to be useful though. I stand by my opinion, its an ugly picture of an ugly thing in an ugly place, and that can work best for some stories. Same with my opinion of the Kinkade naptime pictures.

sure, i could justify somewhat similar looks (maybe not that crazy for myself), but whatever you do, just keep in mind the visual precedence, and realize that is a 'trendy' look right now.

I can understand the 8K resolution, but why would anyone need to record anything at 240fps for anything other than overcranking? Can the human eye even see that many fps? We have enough people as it is (me) clamoring for the preemptive death of 48p and 60p in cinema. 240fps for realtime capture and playback seems beyond silly.

i think the overcranking would be for slo-mo, not playback...
 
I believe Trumbull's tests when he developed ShowScan was that most people couldn't detect a difference above 100 fps, and that the fall-off in the perception of improved clarity, motion, etc. fell off quickly above 60 fps, so he settled on 60 fps as the most practical speed for ShowScan.

Remember that not only is there a reduction in shutter time at higher frame rates, therefore there is also a reduction of exposure -- your 800 ASA camera at 48 fps / 180 degree shutter becomes an effective 400 ASA, at 96 fps, an effective 200 ASA, at 192 fps, an effective 100 ASA and so on... so I don't know many people who want an effective 50 ASA camera by shooting above 200 fps all the time.
 
I believe Trumbull's tests when he developed ShowScan was that most people couldn't detect a difference above 100 fps, and that the fall-off in the perception of improved clarity, motion, etc. fell off quickly above 60 fps, so he settled on 60 fps as the most practical speed for ShowScan.

Remember that not only is there a reduction in shutter time at higher frame rates, therefore there is also a reduction of exposure -- your 800 ASA camera at 48 fps / 180 degree shutter becomes an effective 400 ASA, at 96 fps, an effective 200 ASA, at 192 fps, an effective 100 ASA and so on... so I don't know many people who want an effective 50 ASA camera by shooting above 200 fps all the time.


Wait ... what?

I think your explanation is wrong, but the result is right. I think you have a didactic point ... pardon me if I am being pedantic.

Shutter angle is a relative measurement (relative to frame rate), and shutter speed is absolute.

Shutter angle represents the fraction of the time of the fraction of a second of the frame rate. Uh ... words suck. I'll go with math.

ASC Manual 9th ed. pg615 section 2.2. eq (20)

α = shutter angle in degrees
S= frames per second
T = exposure time in seconds

T = 1/(S*360°/α)

So ...

T = 1(24 frames/s*360°/180°) =
=1/(24f/s*2)= (degrees cancel out)
1/48(frames/second)=1/48seconds/frame

More simply ... As you increase frame rate a 180 shutter angle gets faster in still photo terms 180 at 24 fps is 1/48, 180 at 48fps is 1/96, 180 at 120fps is 1/240 and so on. You are still at the same exposure index though.

So, at 24fps 90° shutter works out to 1/96 seconds per frame. 270° shutter works out to 1/32 seconds per frame.

My point being that you are actually speeding up the shutter speed as you increase frame rate using a constant shutter angle - hence shutter angle being a relative measurement.

Increasing shutter speed alters exposure intrinsically.
 
My point is that higher frame rates mean reduced exposure times per frame and thus you can think of it as having an effectively lower ASA camera (I know, the sensitivity doesn't change but what I mean by this is that the light levels needed become similar to using a lower ASA camera.) You have to compensate a stop for 48 fps, two-stops for 96 fps, three-stops for 192 fps, etc. assuming you don't change your shutter angle (and even if you do, once you go to 360 degrees, you can't open it up any more.)

I'm just saying that 240 fps is not going to become some sort of industry standard. Toss in 3D mirror rigs and their light loss, the problem becomes even worse.

I think what Trumbull was saying is that even if 60 fps is better than 24 fps, it doesn't follow that 240 fps must be much better because there are laws of diminishing returns.
 
i think the overcranking would be for slo-mo, not playback...

No, I think he wants to shoot 240fps to "future proof" to when the standard theoretical frame rate will be 240fps. Then saying he can always remove certain frames to adhere to current and possibly tomorrow's standards. Which is why I am poo poo-ing this idea. So yeah, he wants to shoot 240p for 240p playback...
 
This is from Wikipedia about testing that lead Trumbull to adopt 60 fps for Showscan:

http://en.wikipedia.org/wiki/Showscan
Trumbull also did research into frame rate, running a series of tests with 35 mm stock filmed and projected at various speeds, shown to audiences who were instrumented to biometrically test their responses. He found that as the frame rate increased, so did the viewer's emotional reaction.

Trumbull discovered that although viewers see smooth motion from film displayed at 24 frames per second (fps), the standard in motion pictures for decades, they are subconsciously still aware of the flicker. This awareness reduces the emotional impact of the film. As the speed of projection ramped up, so did the emotional response, peaking at 72 fps. After that speed, no further improvements were noted.

However, this Variety article says that now Trumbull is pushing Showscan Digital, which runs at 120 fps:
http://www.variety.com/article/VR1118035228?refcatid=13&printerfriendly=true
 
Last edited:
My point is that higher frame rates mean reduced exposure times per frame and thus you can think of it as having an effectively lower ASA camera (I know, the sensitivity doesn't change but what I mean by this is that the light levels needed become similar to using a lower ASA camera.) You have to compensate a stop for 48 fps, two-stops for 96 fps, three-stops for 192 fps, etc. assuming you don't change your shutter angle (and even if you do, once you go to 360 degrees, you can't open it up any more.)

I'm just saying that 240 fps is not going to become some sort of industry standard. Toss in 3D mirror rigs and their light loss, the problem becomes even worse.

I think what Trumbull was saying is that even if 60 fps is better than 24 fps, it doesn't follow that 240 fps must be much better because there are laws of diminishing returns.

Oh I see .. it was a didactic point! It was just more confusing to me than the accurate description.

I knew about the perceptual limits of frame rates, but I didn't know it was Trumbull who found them ... I figured it was a USAF/USN behavioral science lab somewhere.

I agree that speeds over 60fps aren't very useful as a standard shooting rate. I certainly don't think anything over 100fps is useful as a standard if you are making images for humans to watch.

I do think I'd like a camera that could run 600fps though. I'd use that to overcrank for a 60fps production of course. The frame rate I am most interested in is 48fps right now. I'd REALLY love a chance to get into screening rooms for The Hobbit and to talk to the camera crew and post staff on that show.

Another thing I'd be interested to study is what happens with high frame rates in the viewer. I mean, the real world doesn't have motion blur - that's an artifact of our imperfect perception. At what point in watching moving images do we stop perceiving the motion blur in the actual images because our brain/eye system is contributing at least the same amount of blur.

I guess I can see some applications where you'd want a camera that could operate routinely at high frame rates. Certain security/military and scientific applications mostly, situations probably of limited interest to cinematographers. Being a former scientist ... I might enjoy a camera that could shoot absurd framerates. 120000fps at 5K anyone? The world is surreal at those frame rates. For playback at 60p one second of 120000fps is 33 minutes of viewing ... so realistically I'd only use that sort of frame rate for very fast things - you know like watching light illuminate a panoramic scene. (Light would move "just" 1.5km per frame, so with a wide enough shot and a powerful enough source you could actually record light moving across the scene.) An M-16 round would take 10 minutes to get to the target on a 300 yard range.

I'm not entirely clear what the OP really wants. I get the impression that he just started spouting camera specs that he thought would allow him to reach some objective, but never really stated the objective. So ... here we are running in circles rehashing basic cinematography when probably the most useful thing is to address these hidden objectives.

If the goal is simply "future proofing", then Epic already does that. Shoot 60p for 60p with a 360º shutter. With pulldown to 24p it will look as though you shot a 144º shutter, which isn't ideal in my mind ... but probably acceptable to most viewers. playback at 48 would look like a 288º shutter- which I posit would look like smooth 24p/144º motion.

OK ... I'm babbling. G'night.
 
You want to "future-proof" something, make a good movie that people want to see again and again -- "Gone with the Wind" is future-proofed, "Singin in the Rain" is future-proofed, "The Godfather" is future-proofed, "Star Wars" is future-proofed, and they are all 24 fps movies.
 
You want to "future-proof" something, make a good movie that people want to see again and again -- "Gone with the Wind" is future-proofed, "Singin in the Rain" is future-proofed, "The Godfather" is future-proofed, "Star Wars" is future-proofed, and they are all 24 fps movies.

Amen!
 
You want to "future-proof" something, make a good movie that people want to see again and again -- "Gone with the Wind" is future-proofed, "Singin in the Rain" is future-proofed, "The Godfather" is future-proofed, "Star Wars" is future-proofed, and they are all 24 fps movies.

Of course this is true!

Most Red Users won't ever make a legendary film though ... and there is this thing the various broadcasters do. When they change technical standards there is a flurry of activity while they scour the earth looking for content to broadcast in their new technical format.

So what happens?

Old movies that are not up to the technical standard, but are, for one reason or another, extremely popular get "remastered" for the new standard.

Old movies that were made to the technical standard get shown again even if they are awful.

Old movies that are not up to the technical standard and are not extremely popular never get shown again.

So, while I agree in principle, it makes good business sense to plan for future technical standards, as best you can. Unless you can guarantee that your movie will be popular and stand the test of time - which no one can do.

Now to my mind that means shooting 5K 48fps 360º shutter 3D today, if you can afford it. Then you have all the material required for higher frame rates and/or 3D - and its trivial to deliver a 24fps 2D version as well.

Intermediate steps are shooting higher resolution, and higher frame rates. When I get my Epic I plan on shooting primarily 5K 48 fps 360º 2D material. Oh ... I expect to finish 4K 24fps 2D, and on many projects deliver in 2K or 1080, but that's how I plan to acquire.
 
Back
Top