cinematography Can't get this aperture right.

[harmonica44]

I want to do shots like this, like the shot 15 seconds into the video:

http://www.youtube.com/watch?v=ik_XxZEfjJo

As you can see Bond and the Armorer, are both in focus and M sees to be in focus as well. My camera operator collaborator tells me, and if he goes higher than f8, he starts to get aperture diffraction. But f8 is not high enough to put everyone in focus. I have done some tests with my T2i, but I too cannot get 3 different people in the room into focus at f8.

 

[knightly]

Hyper focal is turning the focus ring to as far away as it will go from the lens... then moving the camea backwards until what you want to have in shot is in that acceptable focus distance range that the calculator will show you. At infinity, everything from the shortest distance to the horizon and beyond will be in focus... this is how you achieve deep focus without fancy lenses. Physics! Simple optical physics. No magic, no fancy lens tricks, no voodoo potography... just physics. If you don't have a physics mind, tell your DP what you want with a sketch and a reference photo, ask the budget needed to achieve that shot, then make it happen... otherwise, read every thing you can get your hands on over and over again and test those concepts until you have a deep understanding of everything that is happening so you can make a logical leap to developing new techniques based on reality rather than hearsay. This is how these shots get made. Roger Deakins doesn't ask questions on this forum... he grabs a camera and some lights and figures it out... then falls back on years upon years of doing these exercises over and over again until they're part of his subconscious mind allowing for inspiration to shove multiple techniques together when needed for a new effect.

 

[WalterB]

I showed that Dr. No scene to a DP collaborator. He said that that scene was recreated using computer effects, cause they wouldn't have been able to get that deep DOF. Obviously he doesn't know what he is talking about.

Maybe it's even CGI!
http://en.wikipedia.org/wiki/Dr._No_(film)

For the record Dr. No was released in 1962.
Years before they put a man on the moon using far less than 60kB of computer capacity...

 

[TheArtist]

Hyper focal is turning the focus ring to as far away as it will go from the lens... then moving the camea backwards until what you want to have in shot is in that acceptable focus distance range that the calculator will show you.

Man, no, that's not hyperfocal.

Hyperfocal is a certain finite distance. When you focus on THAT distance (and not as far as the lens allows which would be infinity (or beyond if the lens is not perfectly calibrated for infty)) you have acceptable focus from half that distance to infinity.

 

[knightly]

Yes, but the practical application is to turn the focus ring to the end, then move the camera back... then everything past that point will be in focus... you can argue theory with me all day, but practical application is what is needed for the particular audience here - then working that practical knowledge into theoretical understanding.

 

[TheArtist]

Yes, but the practical application is to turn the focus ring to the end, then move the camera back... then everything past that point will be in focus... you can argue theory with me all day, but practical application is what is needed for the particular audience here - then working that practical knowledge into theoretical understanding.

Do you have a source for this ?

It doesn't make any sense for me... If you turn the ring all the way, you're focusing on infty (if you're lucky, most of the time, everything will be blurry because you'll be past infty). And by definition, "infty" is independent of the position of the camera.

Your "technique" simply cannot work. There is no good reason why moving the camera back would suddenly make everything in focus from HFD/2 to Infinity.

 

[ItDonnedOnMe]

Yes, but the practical application is to turn the focus ring to the end, then move the camera back...

I've got to agree with TheArtist on this one, that isn't at all what the term hyperfocal refers to, in theory or in practice. Go punch in some numbers on the DOF calculator and it'll show you your hyperfocal distance:

http://www.dofmaster.com/dofjs.html

So for instance, a 50mm on a 7D at f/5.6 will have a hyperfocal distance of 76 feet. Go to f/11 and the hyperfocal distance is 38 feet. Neither one of those is focused at infinity.

In fact the only combination where focusing at infinity makes any sense with that lens option is at apertures under f/2, where your hyperfocal distance gets into the hundreds of feet. In a practical application though that's pretty useless except for landscapes, as hyperfocal only gives you half the distance to the focal point in focus. So if you're subject isn't over 100 feet from the camera it doesn't really matter.

The practical application is understanding that at f/16 your hyperfocal distance is 27 feet, allowing you to get both a person 15 feet away and the distant mountains in focus at the same time. In that situation, focusing at infinity would only give you the mountains. Focusing on the person would only give you them in focus. Focusing 12 feet beyond the person gives you both of them in focus - that's the practical part.

 

[knightly]

http://www.dofmaster.com/hyperfocal.html

For the intended audience (who is elsewhere still working on the physics of light), the point is that at whatever other settings, spinning the focus to the infinity stop, then moving the camera back will get them their intended results.

Yes, adding complexity to the description will be more accurate, but will end up with more frustrated posts as our more advanced users assume a level of knowledge that the current audience (OP) hasn't yet attained. Simplify. Show the technique in a way that will get used in camera right now... when the next round of questions come up about how the f/stop will affect their hyperfocal distances, we can get more detailed.

We can cite calculators and sources -- but the bottom level understanding is that if the subject is far enough away from the camera, they and everything behind them will be in focus... to that end, spin the focus to infinity -- then move the camera away from the subject until they're in focus... everything behind them should now be in focus.

 

[knightly]

For the record, I'm looking at the barrel of my Tokina 80-200mm and the distance markings seem to go from 6m > 10m > 15m > inf in relatively short order putting it at right around 30m or so accounting for the somewhat logarithmic nature of the numbers.

Yes, it's far away and semi pointless, but others are asking H44 elsewhere to rent out grip trucks for a test shoot and hire a full budget production staff for their first couple of shorts... these are all ridiculous requests, mine's approachable given enough space and will get the desired results without any added cost to the production.

By all means H44, if the calculator makes it all click for you, start exploring the finer points of how your lens changes affect the hyperfocal distance to get your long DoF shots you're looking for... if not, spin the focus wheel to infinity, and move the camera back from your actor. It's no less ridiculous than some of the stuff we used to have to do to get shallow depth of field with my 1/3" chip XL1s... zoom into a close up as far as the lens would allow with the aperture wide open and move the camera 30' back to get the frame we were looking for. We had to setup wireless relays to get the audio into the camera on those shots.

Theory is great, helps make decision making quick (and you guys are spot on with your descriptions)... but the theory needs to be contextualized with practical experience... and that's currently lacking here. So my goal in my horribly oversimplified explanation is to give a single concrete exercise that will start this seed germinating so that further explanation will have a foothold.

We have examples that this knowledge isn't sticking dating back 2-3 years with a thread about split diopter (followed not too long ago by a second thread about split diopter asking mostly the same questions) and a few other threads asking the same fundamental question... how do we get long DoF... with all the same answers over and over again, hence TheArtist's hyperfocal art nouveau poem earlier in this thread.

 

[David.rhsc]

Guys,

Trying to get h44 to dial to the mathematical hyperfocal distance for a given lens/fstop/focal plane size on a lens ring where the witness marks are so far apart isn't going to happen. The theoretical discussion is fine and all, but the fact of the matter is that the two actors who are in focus in the frame at which the guy entering the door stands next to Bond at the desk are simply not that far apart in terms of focal plane.

Here is how to make this happen:

Block the scene, keeping the actors on a similar plane as in the sample shot you are trying to mimic.

Measure distance to near actor, measure distance to far actor.

Use focus calculator to determine the split focal point. Have someone stand in or hold a focus chart/newspaper/call sheet, whatever at that exact distance and focus to it. Confirm by eye because you are working with still lenses and do not have good witness marks. Split focus is basically the distance at which to dial focus and keep objects at distance A and B within acceptable focus. Conceptually speaking, hyperfocal distance is the extreme end of a "split focus" shot, but you are simply fine-tuning the split point to the needs of your shot.

A more simple way: just shoot a 24mm or wider lens on your camera at a stop somewhere between 5.6 and 8, keep your actors about 10' feet from the lens and no farther than 2' difference in the distance from each actor to the camera. In otherwords, if actor A is 10 feet from the lens, then actor B cannot be any farther than 12 feet, and should not be any nearer than 9 feet.

I'm looking at FilmCalc on my phone, using a 7D for reference (should be similar enough to your t#i), and at a 5.6-8 on a 24mm you should easily pull of a shot like the one you linked. Unless you are trying to keep someone sharp as they walk up to someone else who is sharp, in which case that is completely different.

 

[TheArtist]

http://www.dofmaster.com/hyperfocal.html

For the intended audience (who is elsewhere still working on the physics of light), the point is that at whatever other settings, spinning the focus to the infinity stop, then moving the camera back will get them their intended results.

Yes, adding complexity to the description will be more accurate, but will end up with more frustrated posts as our more advanced users assume a level of knowledge that the current audience (OP) hasn't yet attained. Simplify. Show the technique in a way that will get used in camera right now... when the next round of questions come up about how the f/stop will affect their hyperfocal distances, we can get more detailed.

We can cite calculators and sources -- but the bottom level understanding is that if the subject is far enough away from the camera, they and everything behind them will be in focus... to that end, spin the focus to infinity -- then move the camera away from the subject until they're in focus... everything behind them should now be in focus.

Yes, but that's not hyperfocal.

You're talking about putting everything in the shot far away from the camera. Then yes, focusing on Infinity will make everything in focus.

But again, that's not AT ALL a case of hyperfocal. So stop this nonsense of "practical" vs "theory". Hyperfocal is leagues away of what you're describing and allows you to have everything in focus from a few inches/feet in front of your camera to Infty. Not just Infty. It has the merit of not taking a shit on your composition as your "technique" suggests.

 

[ItDonnedOnMe]

For the intended audience (who is elsewhere still working on the physics of light), the point is that at whatever other settings, spinning the focus to the infinity stop, then moving the camera back will get them their intended results.

Here's the thing - my intended audience for these responses isn't necessarily who you assume it is. I assume a lot of people are reading these threads, many of whom have similar questions and may be ready to actually learn and understand how this stuff works. So my goal is to put out accurate, useful information and resources to help them do that.

And I personally don't see any way in which hyperfocal distance can be effectively understood or applied without taking aperture into account. Spin your lens to infinity at f/2.8 and you'll have to back your camera up nearly 150 feet. Do the same at f/16 and you'll only need to back up 25 feet. The former is essentially useless for all except maybe establishing shots, while the latter is only slightly more useful for general blocking. Not understanding the principle just means you can't take advantage of it in any practical way. It won't get them their intended results, it'll get them only a very wide shot with everything in the distance more or less in focus and everything close to the lens completely out of focus.

And how does that make anything easier for your intended audience? Doesn't it in fact lead directly to the next question?

"How do they shoot close-ups like this [youtube link]? When I focus at infinity and back my camera up until everyone is in focus they're all so tiny in the frame I can't even see their faces. Do you think a distributor will accept a movie without close-ups? My collaborators said maybe I could zoom in in post to make their faces bigger. That makes everything too soft though. I tried a sharpness filter and it's a little better but I think maybe I need a better filter. What sharpening filter is best? Is there a sharpening filter I can put on the lens?"

 

[TheArtist]

"How do they shoot close-ups like this [youtube link]? When I focus at infinity and back my camera up until everyone is in focus they're all so tiny in the frame I can't even see their faces. Do you think a distributor will accept a movie without close-ups? My collaborators said maybe I could zoom in in post to make their faces bigger. That makes everything too soft though. I tried a sharpness filter and it's a little better but I think maybe I need a better filter. What sharpening filter is best? Is there a sharpening filter I can put on the lens?"

We should really start an H44 contest .

 

[rayw]

We should really start an H44 contest .

Dear God.

It'd be like winning the ugliest wife contest. Are we really winners?

 

[knightly]

And here's why the horrible oversimplification (and others' frustrations):

source: http://www.indietalk.com/search.php?searchid=2591936&pp=25&page=1

These are threads started by the OP asking about how ISO, Aperture, Focal Length and Shutter Speed work...

http://www.indietalk.com/showthread.php?t=47858
http://www.indietalk.com/showthread.php?t=47379
http://www.indietalk.com/showthread.php?t=43827
http://www.indietalk.com/showthread.php?t=41949
http://www.indietalk.com/showthread.php?t=37687
http://www.indietalk.com/showthread.php?t=40193
http://www.indietalk.com/showthread.php?t=40732
http://www.indietalk.com/showthread.php?t=39091
http://www.indietalk.com/showthread.php?t=37032
http://www.indietalk.com/showthread.php?t=35349
http://www.indietalk.com/showthread.php?t=35458
http://www.indietalk.com/showthread.php?t=33910

they date back a few years and have generated quite a bit of traffic, lots of new members here searching for answers to the same, and not a little bit of frustration from others in the community who feel as if they've answered the same questions repeatedly for the past few years... the information isn't sticking, so the current thrust of education has split, some are using the straight theory (apparently over the current level of the audience; H44), and others, myself included are teaching the skills from a "here is a single exercise to take your camera out right now and test." standpoint. Both approaches are true and correct; my exercise here will work in the intended manner.

As a related example, teaching someone about complementary colors doesn't need to delve into frequency harmonics to be a useful teaching exercise, that comes later, once the foundation is built. Educational strategies are near and dear to my heart and my history... moreso even than cinematography. In my past professional career (IT), I've had to teach users how to configure security on a rack of servers, how to alter their code to produce regionally differentiated publications automatically, and how to use a mouse... I guarantee that that last one would have been unsuccessful to the grad student learning had I delved into the electronics inside the device.

i.e. : here's how to make a phone call:

 

[knightly]

I'll definitely concede that your approach is much better for a broader audience and much more complete.

My answer here is specific and targeted to get a single exercise underway by an individual to use as a starting point to build from.

 

[TheArtist]

I'll definitely concede that your approach is much better for a broader audience and much more complete.

My answer here is specific and targeted to get a single exercise underway by an individual to use as a starting point to build from.

We got that. We're just saying that what you are proposing is not a practical exercise of hyperfocal which is a different thing (not only in Theory but also in Practice).

As for this particular individual... I have no more idea on how to deal with him. I feel like he hasn't learned anything in his YEARS here. I wonder how he was able to learn English to start with.

 

[CraigL]

i.e. : here's how to make a phone call:

But... where are the buttons??!! How can I dial? And where do I speak into?

=)

CraigL

 

[knightly]

...what you are proposing is not a practical exercise of hyperfocal which is a different thing (not only in Theory but also in Practice).

My goal was to get across the idea that there is a point in front of the lens at some distance beyond which everything is in focus in the fewest possible steps. Then explain why and expound on that concept.

Start at definition 2 and work into def 1 :http://en.wikipedia.org/wiki/Hyperfocal_distance

 

[8salacious9]

can anyone confirm if H44 has started a thread on how to turn on a camera? I don't mean sexually either...