DOF on the XL2
 

I am curious to know if the Canon XL family can easily give you good shallow depth of field, seeing as the XL2 20x (or XL1 16x) lens is quite long, shouldn't that help? Something that single body prosumer camcorders cannot easily achieve (GL2, PD150, DVX100, etc.), except through the restrictive "trick" of full zoom/iris...

Speculating here but I would think that you can get a slightly more shallow depth of field when zoomed in quite a bit. At least more than on the DVX100. I have had an Xl1 and a DVX100 and can say that the XL1 could achieve that really shallow depth of filed only when zoomed in a lot! The DVX didn't exhibit this same property even at the max zoom of 10. Neither one will get you that "prime" shallow depth of field we see in movies but I'd guess the XL series is probably your best bet in the price range.

However I have heard stories that the change in CCD area and the way the new lens focus on the CCD that in 4x3 mode the depth of field may have actually increased.....which would be the opposite of what you are asking. I don;t know about 16x9 mode.

Anyone want to elaborate?

The DVX100 and the XL1, having the same size chip, will deliver the same depth of field at a given field of view. However, the XL1 standard lens zooms in further than the DVX100, which means you can achieve more shallow depth of field due to the longer focal length for this simple reason.

The XL2 in 4:3 mode should, mathematically, deliver more depth of field for a given field of view (more in focus).

DOF is not a function of the focal length of the lens, but rather the CCD size, and lens aperture. The lenses used on the XL series have no more or less DOF than those of any other camera at the same aperture.

Although lenses with long focal lengths can give the appearance of shallow DOF, getting in close with a wide angle lens until the subject is the same size in the viewfinder gives identical DOF, at the same aperture.

Based on the specs quoted by Canon, the DOF on the XL2 using the 16:9 aspect ratio will be marginally - and probably not noticeably - greater than the XL1s, or any other 1/3 in CCD camera. Using the 4:3 aspect ratio, it will be quite significantly - and possibly noticeably - greater.

"DOF is not a function of the focal length of the lens, but rather the CCD size, and lens aperture"

That isn't correct.

Depth of Field is a function of focal length, lens aperture, and distance to the subject.

Jeff Donald wrote a good article on it for the Watchdog.

So in theory in 16x9 mode it should be close to the XL1?

However in 4x3 the DOF will be increased making it harder to get that sought after "shallow" DOF look? Is that right?

And I am aware that the DOF is determined by chip size and aperture but at the end of the day when in practical applications the XL1 can achieve the shallow look via it's zoom where the DVX can't. Example is framing an actor for a shoulder up shot......I stood about 10 feet away and zoomed in with XL1. The wooded background just bloomed and went very soft. In this case it looked like "Film quality DOF". serious out of focus background.

With the DVX I couldn't stand 10ft away as the zoom would not get tight enough to recreate the same shot composition. So I moved closer....still zoomed in all the way.....to compose the actor at about the same size in the frame. Now the background was a "little" soft. I couldn't achieve the same dramatic effect as with the XL.

Now....if I limited my XL to approximately the 10x range to compare I would expect similar results with both camera being close to the same. And obviously the DVX will win out in areas where you need the wider end of the zoom. You won't be able to shoot in really tight areas with XL series as easily.

No matter how you slice it there will be drawbacks to any camera. For me though the positives of the XL2 outweigh it's drawbacks.

<<<-- Originally posted by Luis Caffesse : "DOF is not a function of the focal length of the lens, but rather the CCD size, and lens aperture"

That isn't correct.

Depth of Field is a function of focal length, lens aperture, and distance to the subject.

Jeff Donald wrote a good article on it for the Watchdog. -->>>

CCD size is also a factor.

Luis - I stand corrected, thanks for the useful link.

Charles - our posts crossed, I wasn't trying to contradict you.

Marty - if my understanding is correct (and it appears it may not be!) - yes, in your example, if the shot is framed in the exact same way at the same aperture, I would expect DOF to be more or less identical. Does the XL lens have a wider max f-stop than the DVX?

Sounds good in theory, but doesn't work in practice. Here's a sample picture I shot with a DVX, testing that theory.
http://www.icexpo.com/dvx100/DOF-Combined.jpg

Both images were shot at f/2.8, both images were shot with identical framing (as near as I could get to identical!)

The left half of the image was shot by zooming in to max telephoto, and backing up to get the same frame. The right half of the image was shot by zooming out to full wide angle, and moving close to fill the frame.

It's obvious that the background in the telephoto shot is *much* more out of focus.

I understand the theory behind what you said, and technically you are correct, but the difference is that the field of view is changed between the two shots, so in the telephoto shot a much smaller portion of the background is optically magnified to fill the full screen. It is that magnification process that makes the shot "look" more out of focus. Think about it like a Photoshop image: if you view it at 400% size, it'll look all soft and bitty and such, but if you view it at 25% size it'll look razor sharp. Nothing changed about the image, just the way you viewed it. Same thing happens here with the background.

So yes, long telephoto is *crucial* to establishing the soft-background look. Open the iris and zoom in the lens and that's the trick for getting the shallowest DOF you can get (which is quite limited on these 1/3" CCD cameras, due to the focal length of the lenses).

Not really. Only indirectly, as CCD size dictates the focal length necessary to achieve a given field of view, and the smaller the CCD, the wider-angle you need to have to get a usable field of view.

Telephoto focal length is what gives the shallow DOF effect.

On a 1/6" CCD camera, you'd have a lens with a focal range of 2.8 to 28mm. You're not going to get shallow DOF with a 28mm lens no matter what you do. On a 35mm movie camera, the same field of view is achieved with a zoom lens of 25mm to 250mm. At its very widest the 35mm camera will exhibit DOF as shallow as the little camera at its most telephoto!

So CCD size influences the selection of the appropriate focal length for the lens, but it doesn't affect DOF directly. It only affects field of view.

"Luis - I stand corrected, thanks for the useful link."

No problem. I recently saw that someone else had posted that
link. I thought Jeff did a great job with that article.


Barry:
Thanks for the concise and clear explanation on CCD size and its
indirect influence on Depth of Field.

You managed to make a somewhat confusing issue pretty clear.

quote:

"Marty - if my understanding is correct (and it appears it may not be!) - yes, in your example, if the shot is framed in the exact same way at the same aperture, I would expect DOF to be more or less identical. Does the XL lens have a wider max f-stop than the DVX?"

First I would think that in addition to being framed the same and having the same aperture you would need to have about the same zoom factor. then I would expect the same DOF from both cameras. Zooming plays a big role in this process so 2 shots framed the same but from different distance would not have same DOF....I think.

Example to get a man to fill the screen from 5 feet away might require no zoom. He just fills the screen. But move 20 feet away and he doesn't fill the frame any more. So you zoom....say 4x for example and now he fills the frame to the same proportions as before . Only now the background will start to get soft. Even though you are at the same f-stop and same framing. So position and zoom play into it.

Unless when you say framed you mean where you stand in relation to the subject...in which case forget everything I just spurted!

:)

Also I am pretty sure that the DVX and XL series have the same max F-stop when wide. You lose a stop or so at the max zoom end of the xl but not sure on the DVX at the telephoto end.

DVX is f/1.6 at the wide end, f/2.8 at tele.

I believe the new Canon is f/1.6 at the wide, f/3.8 or something at the tele.

Barry said: "I believe the new Canon is f/1.6 at the wide, f/3.8 or something at the tele."

Is that as a result of how much further the tele end of the XL2 goes? In other words is there an inverse relationship between lowest F-stop and maximum zoom? Or is the f/3.8 determined by some other lens factors?

The DOF calculations are thrown off when the distance to subject gets too small. When a lens is focused in the near macro range, the effective aperture for DOF calculations must be increased two F stops because of light falloff and the inverse square law.

The maximum aperture is determined by the size of the physical opening and the focal length of the lens. Remember a millimeter is always a millimeter. On a 100mm lens, if the largest physical opening of the aperture is 50mm, then the lens is an F/2 lens (100 divided by 50 equals 2). If you zoom the lens from 100mm to 200mm and the physical opening stays the same size, the aperture becomes F/4 (200 divided by 50 equals 4).