DVX100AP, screen grabs and DOF test

Re: DVX100AP, screen grabs and DOF test

DOF is influenced by lens aperture and by focal length, and by distance to subject -- nothing else. The more telephoto you are, and the more wide-open your lens is, and the closer you are to your subject, the shallower the DOF will be.

As Jaron said, selecting a higher shutter speed (thus less exposure time) may force your lens to open up a bit, and it is that wider aperture that would cause a shallower DOF -- but at the expense of making your motion look more staccato.

A DVX100 will never be able to match the shallow-DOF capability of the PD150/PD170 or even the XL1 -- those cameras have longer lenses, and the more telephoto you go, the shallower the DOF. The DVX100/DVX100A have a maximum telephoto of 45mm, whereas the PD150 goes to 72mm and the XL1 goes all the way to 88mm -- the longer the lens, the shallower the DOF.

So the rules of shallow DOF are simple: use ND filters to open your iris as much as you can. Then zoom in as far as you can. Then get as close to your subject as you can. And, if at all possible, put as much distance between your subject and your background as you can.
 
Re: DVX100AP, screen grabs and DOF test

[quote author=Barry_Green link=board=Clips;num=1074062499;start=0#20 date=01/19/04 at 00:18:41]
So the rules of shallow DOF are simple: use ND filters to open your iris as much as you can.  Then zoom in as far as you can.  Then get as close to your subject as you can.  And, if at all possible, put as much distance between your subject and your background as you can.  [/quote]

"Shiiiit, nizzle. That's all you had to say"
 
Re: DVX100AP, screen grabs and DOF test

[quote author=Barry_Green link=board=Clips;num=1074062499;start=0#20 date=01/19/04 at 00:18:41]DOF is influenced by lens aperture and by focal length, and by distance to subject -- nothing else. [/quote]

Barry, not to nitpick, cause you are right, those are the 3 things that influence DOF in a given format. (such as 1/3" video)  But there are also those pesky variables of shooting format, field of view (FOV) and circles of confusion!   ::)

This is why 35mm has a shallower DOF than 16mm, and video is even less.  Based on some of your posts that I have read in the past, I'm sure you probably know and understand this, but what the hell, it's worth bringing up!  Here's an example...

You want a nice, shallow focus CU of your talent.  On a 35mm camera, we'll set the standard as this:

Subject distance= 5 feet.
Stop= T2
Focal length= 85mm.

This setup will give you a very shallow DOF, only about 2 inches worth!  That's crisp eyes and fuzzy ears man!

Here's where it gets interesting/confusing.  Assuming that the light level (shooting T stop) is the same as the 35mm example, and to try and match that framing as close as possible, the following would be necessary with a 16mm camera:

Subject distance= 5 feet
Stop= T2
Focal length= 41mm

This example means that the 16mm FOV is as close as possible to the 35mm FOV.  The subject size in frame is identical, as is the angle of view, roughly 1'3" wide and 11" high at the subject distance. (Both camera aspect ratios of 1.37:1 Academy)  But because of the difference in size of the optical targets, a shorter focal length lens is required for the 16mm camera to achieve this.  The resulting DOF in the 16mm camera is extended to 4 inches.  

Smaller target size= smaller circle of confusion, or the largest on-film circle that you see as a point.  Anything larger is seen as a small circle, and is therefore perceived as out of focus.

Of course you can compensate to a certain extent to mimic the DOF, but you will always have limitations.  You may not be able to get the shallow look you're after without some EXTREME image magnification, i.e. really tight close ups.  That will also affect your FOV, cause at a longer lens, you may have the same subject size in frame, but you lose background info cause you're actually seeing less of it with a narrower FOV.  One could also compensate by opening up the stop and changing the lighting levels or ND'ing the lens, but you're still limited.  For the purpose of outlining the inherent difference in the DOF characteristics of each format however, the same exposure was used for both formats in my examples.  For truly shallow DOF, 35mm is the way to go.

By the way, all the figures in my examples were computed with the use of David Eubank's PCAM Depth-of-Field software for Palm OS.  You can download it here:

http://www.davideubank.com/

Christopher Ratledge
Iatse 600 AC
 
Re: DVX100AP, screen grabs and DOF test

Christopher, you're exactly right.

The main issue I harp on is people think that the sensor size matters when calculating DOF, and it doesn't. The sensor size dictates what focal length will deliver the proper field of view, and it is the FOCAL LENGTH that is changing the DOF.

So a DVX100 has a lens with a 4.5 - 45mm zoom range.
The equivalent fov on a 16mm camera may be 10mm-100mm.
And on 35mm it's 25mm-250mm.

It's the lens focal length that's responsible for the shallower DOF. There's no way a DVX at 45mm is going to deliver the DOF of a 35mm camera at 250mm! Yet the FOV will be approximately the same.

As far as the circle of confusion, that's really a measurement that's not specifically dependent on the shooting format, but also on your expected viewing size. "28 Days Later" is a great example -- looks pretty sharp on a television, but totally fell apart when blown up to the big screen. So the CoC as quoted in the ACM is referencing a theatrical projection, and therefore on the 16mm format it's quoted as a smaller CoC, because the 16mm will get magnified more to fill the big screen. The end goal is to get a pixel to be about the same size when on the big screen -- and because S16 gets blown up optically to about 3x its size to match 35mm, it needs a CoC about 3x as small. But, when not going to the large screen, those tight CoC's are less important in the calculations. If you ARE going to the big screen, it's extremely important to get it right -- and, coincidentally, that will give your small-format camera SHALLOWER DOF, because proper focus will be in such a small pinpoint range!

But I agree with everything you wrote, you're right on the money.
 
Re: DVX100AP, screen grabs and DOF test

Hey, brother nerd, I'm just glad to meet someone who "gets" it! :) So many people insist that sensor size determines DOF, and I'm like... dude, the lens projects an image -- it doesn't know what it's projecting onto, it could be a 16mm film frame, it could be a 2/3" CCD, it could be a flat wall, it could be an Agus35 piece of frosted CD... how would the lens know? It doesn't. Sensor size dictates the field of view, and the field of view dictates what the appropriate focal length needs to be, and focal length dictates DOF.

Anyway, to recap the rules, for those who are still awake (with or without smoking pocket protectors): open up the lens iris as much as you can, zoom in as far as you can, and move as close as you can, and separate the subject from the background as far as you can -- do that and you'll get the shallowest DOF the DVX is capable of.
 
Re: DVX100AP, screen grabs and DOF test

Whoa John, that's too much...

Anyway, I've been experimenting with this a bit further too, and come to an interesting conclusion...

Getting mathematical and technical, there really is no such thing as DOF, because a lens can only focus on one infinitely-thin plane at any one point. What we consider DOF is the amount in front of and behind that plane that is "acceptably" sharp. Every atom in front of and behind that plane is progressively more and more out of focus -- only the atoms on the exact plane are actually "in focus". But, we set a certain threshold of tolerance (i.e., the Circle of Confusion) before we declare that something has now crossed the line into being "out of focus".

(another handy rule to remember: the depth of field isn't equal in front & behind, it extends to about 1/3 in front and 2/3 behind the subject).

Now, technically, for any given position of camera to subject, the aperture is really the only thing that determines the DOF. Putting a picture under a microscope and blowing it up to equivalent sizes, an image from a 4.5mm lens and an image from a 45mm lens, same subject, same distance to subject, etc., but not same framing (obviously, because of the FOV differences between the focal lengths) will yield the same depth of field! This is a tough one to grasp because it doesn't seem like it should be so, but it's where the concept of the CoC comes into play -- the wide-angle image will have to be magnified so much to match the framing/fov of the telephoto image, that the pixels that were formerly "in focus" (i.e., rendered at a size smaller than the acceptable CoC) will now be blown up to exceed the dimensions of the CoC, and thus yield shallower focus.

BUT:

For any given circumstance, the telephoto image will SEEM to have shallower DOF. Because the background is magnified more, due to the optical compression properties of the telephoto lens, it will LOOK more out of focus -- even though it's just as out of focus (optically) as the wide-angle shot, the magnification makes it more obviously out of focus. And, when it all comes down to it, how it LOOKS is more important than what degree of technical accuracy it has.

Close focus has similar properties to telephoto zooming -- the closer you are to your subject, the shallower the perceived DOF -- but, they have an inverse relationship. To get close, you usually have to zoom out to a wider angle to get an acceptable fov. When zoomed all the way in, you usually have to be further away. So the two seem to contradict each other. However, the telephoto effect is a more pronounced visible determiner of the shallow DOF effect -- except under macro conditions, the wide angle counteracts the benefits of close focus.

So, the #1 rule is -- zoom all the way in. Telephoto seems to have more of a practical effect on shallow DOF than aperture does. A full-wide-angle shot at f/2 doesn't look dramatically more shallow than a full-wide-angle shot at f/8 -- but a 45mm shot will be much more shallow than a 4.5mm shot. So zoom all the way in first.

1) Telephoto is the key. Try to frame your shot so you can be zoomed all the way in to 45mm (or, better yet, get the tele-extender from Century and be zoomed into 72mm). Change your shot to allow you to get full telephoto. Shoot through a (clean) window, etc. Choose your shooting locations such that you can back up and zoom in.

2) Distance between subject and background. This simple rule can yield dramatic 35mm-style shallow DOF shots. Obviously not possible in all circumstances (like an interior interview), but whenever you can, get the background far, far away from your subject. If you're shooting outdoors, frame so that the background is a building or tree across the street, not right behind the subject. Pick your backgrounds carefully and you can really create the 35mm look.

3) Aperture. Wide-open is best. Use the ND filters to get the maximum wide aperture you can. But, remember telephoto comes first, so even though the DVX loses 1.5 stops of light over the course of its zoom range, 45mm/f2.8 is much better for shallow DOF than 4.5mm/f1.6.

4) Fake it with shallow FOREGROUNDS. Keep in mind that the DOF extends for 1/3 in front of the subject, and 2/3 behind. So, it's a lot easier to get something out of focus in front of the subject than it is to get it out of focus behind them. You can set up your shot so there's something in the frame that's out of focus in front of your subject, and it'll still give a lot of depth to your shot. Maybe they're standing a good distance behind a chain-link fence, or they're sitting at a table and way in the foreground there's an out-of-focus teapot, etc... this shallow-foreground/sharp-subject technique gives you in-focus/out-of-focus depth, but it's much easier to create than the sharp-subject, shallow-background look.
 
Re: DVX100AP, screen grabs and DOF test

... your going to get carple tunnel barry!   good good info though. :)

anyhow..  with your findings on  the distance to subject contradicting the zoom factor. this is a point that i have long since coveted. so here some losing pointers for other peeps experimenting with shallow dof:

what the above means is that the shallow dof/selective focus techniques are not always limited to large areas. when in a cramped space, instead of trying to knock out a wall to get more tele, youcan simply let the lense go wider, and pull focus more and get in closer, and acheive the same result.  what i do is vary which technique i use based on the amount of room i have , and the frame size im looking for. and generally i blend the two. by getting as close as possible to pull focus.. #2 zoom in as much as i can while keeping the framing im looking for. iris wide open. ..

the reason that pulling focus doesnt seem to affect the dof so apparently is that  pulling the focus affects the dof on a curve, vs. zooming in/magnifying, and opening the iris both affect the dof proportionately relative from wide to tele, and closed to open. but if you are pulling focus to create a shallow dof, the dof will get shallower on a curve by being barley noticeably smaller at all around 99 to 80 and extremely noticeably smaller between  40 to 20 and below.
the key to effectively utilizing the focal length to achieve shallow dof rather than zoom is to understand this curve. really what this means is that even if you are using the zoom technique, if you pull focus as much as possible while maintaining sharpness on the subject, you dof effect will be better, becaus the focus will drop off much after the subject so in general i place my subject around the back 1/3 of the dof, instead of the center focal point(about 2/3 of th way forward from the back). now if anyone wants to take some time to experiment. what you will find is this ... pending your distance to subject either the zoom, or focus pull will have a greater effect on the dof, zooming as i mentioned above is always relative.. you are essentially just magnifying the picture to the point that you can see that things arent really in focus .. you could theoretically continue zooming in until only one atom was in focus because as barry pointed out the lense can only focus on one infinitly thin plane at a distance from the lense. the same for opening the iris..it allows more light in which allows for more refraction, and blurriness thus making things appear more out of focus. however, focal distance shortens the dof by a total differnt manner. ..oh no here it goes again. think of your focal point/plane as the center of an x.. and the are that is in focus is only where the lines are crossing and anywhere where the lines are really close together. now if you streched that x out from left to right really wide, the lines willstill only cross at one little infinitly small point, but the area where they are close together is pretty large.. therefore you are increasing the dof(the area which is acceptably sharp to the eye) . and if you shorten the x to really thin. the dof would be really tiny. but it works on a curve  from long to short it progressively effects the dof more and more . anyhow .. if you use this technique it allows for shally dof in small places because youll notice that you would have to be fairly close to the subject instead of at a distance for tele, but the dof drops off so quick that you still get that shallow dof look..even if you stay wider on the lense. and learning to use a combination of the 2 can really give you some great results..  simply put if your zooming in to get shallow dof . pull focus as much as possible, and if your pulling focus to get shallow dof, still zoom in as much as possible all while keeping good framing. it really does about equal out as to which methoid you use, but alot of times knowing how to do it both ways determines whether you can pull it off in a certain spot or not.  
 
Re: DVX100AP, screen grabs and DOF test

lol, really i am a monkey in the madison zoo that stole someones laptop about a year ago.
 
You must log in or register to reply here.
Back
Top