I read the article on how they are generating native 6x9 from a 4x3 CCD here (thanks Chris) :

http://www.dvinfo.net/canonxl2/articles/article06.php

If I read this correctly, and understand my physics correctly, the use of a smaller portion of the CCD and the resulting 1.35x magnification factor will mean that we have an even wider depth of field than we do now, is that correct? Does have a sense of how bad that will be? The depth of field is definitley my least favorite part of dealing with the 1/3 CCDs anyway.... not helping with the film look there...

BTW, if the 3x4 image has a magnification factor of 1.35x, and you put an existing 16x lens on that camera, does that mean it will act as a 21.6x lens? Does the math work that way? Does that also mean if you put the new 20x lens on an XL1s that it will be a 14.8x lens?

What do you think?

If I understand the way lenses work correctly, the number of x represent the number of time the field of view is in telephoto, compared with the one in wide angle. so 20x is 20x, 16x is 16x, but if you put one of these on the Xl1, the lens will be a little bit more wide.

My understanding was similar except that the 16x lens (since it was originally made for the xl1) will be a little longer on the xl2 and the 20x lens (since it was made for the xl2) will be a little wider.

<<<-- Originally posted by Barry Gribble : What do you think? -->>>

I think the DOF in 16:9 mode should match the XL-1s since the full chip width is used, and that appears to be what Canon designed this camera for. Aside from that, you might expect it to behave more like a 1/4" camcorder such as the GL-2 in 4:3 mode. I think someone mentioned in another thread that the XL-2 is rated at 5 lux which is evidently less than the XL-1s and probably the result of the higher pixel count chips. (sorry if that's not correct, the info is pouring in faster than I can absorb it!)

As I just posted in the thread on CCDs, the image size of the 4:3 section on the XL2 should measure exactly 1/4", for what it's worth (I ran some simple math to calculate it).

The low light rating is almost certainly a result of this (not the pixel count) because with a smaller area you can't absorb as much light.

I checked into the lux ratings between the xl1s and xl2, and at least by my reading of it, the xl2 does appear to be an improvement in this area. The 5 lux rating refers to 1/60 sec at f1.6 on the xl2, whereas the 2 lux rating of the xl1s is at 1/8 sec. Now all things being above board here, I think that means the xl1s lux rating is actually 16. Of course your mileage may vary. Both Camera's still carry the 100 lux recommended illumination.

The pixel pitch looks slightly better than the gl2, 345k vs 410k in 1/4inch chip...so I think, given canons claim of new noise free image processing, the low light sensitivity should be pretty good. We'll all know soon.

Regarding depth of field.. I think we might see a slight increase in DOF in 4:3 mode, and possibly a slight decrease in 16:9 mode( compared to the xl1s 4:3) due to the chips higher native resolution.

Barry

If the image size of the subject needs to be a specific size (models head) then the DOF will increase on the smaller cop of the chips. As it stands now with smaller chips the head will be too large and to get a smaller head a wider angle lens will need to be used or the camera position will need to be moved back from the subject. Either or both will result in an increase of DOF.

Lux ratings are misleading because you don't have enough information to know how Canon measured the lux ratings. Lux ratings are at best a rough guide or estimation of low light performance.

As I mentioned on another thread, my measurements of the XL2 CCD sensing areas in 4:3 mode are .238-inch or a bit less than 1/4 inch. The 35% upshift in magnification with the smaller sensing area of the XL2, would give 21.6X when using the older 16X lens from the XL1.

If you used the 1.6X Canon telextender
accessory, mounted under the lens, you'd get 34.56X with the older lens and 32X with the new one. Adding the Century Precision Optics 1.6X telextender on the ends of these lenses, together with the Canon 1.6X extender, would produce
53.29X with the old lens and 51.2X with the new one.

Steve McDonald

Steve, a 16x lens remain a 16x lens on any camera. The number of X is not a magnification factor. See my first comment on this thread.

Jean-Phillipe, it is still a "relative" magnification factor right? If you look through the lens at fill wide, and then zoom in on something to full tele, a 16x lens will make that object look 16x bigger? Now I know that lenses all start at different magnifications and the CCD affects that starting magnification but the relative magnification is what the XXXx means correct?

Aaron

Yes, you are right Aaron, but the 16x lens (or the 20x for that matter) will be a 16x lens (16 times the field of view in full telephoto) on the both the XL1 or XL2. But the value of the focal lengh in terms of 35 mm equivalent will change depending of the camera due to the change of CCD size used. But I am not an expert and I can be wrong on that, in that case, correct me.

Jean-Phillippe, as you say, the 16X and 20X zoom ranges of the two lenses stay the same, regardless of the cameras on which they are used. However, since the CCD sensing area of the XL2 is smaller than that of the XL1, the magnification effect with any lens is greater when it's used on the XL2.

The zoom range and the amount of magnification a lens gives on a certain camera, are indeed two separate specifications. But, there's a positive correlation between the two figures, with any lens on amy camera.

Some persons are speaking of the zoom range and the magnification effect of a lens, as though it were like comparing apples and oranges. It seems to me more like comparing size ranges of two varieties of apples.

Steve McDonald

The zoom ratio (10X, 16X, 20X etc.) has no relationship to magnification (focal length of lens). A 4mm to 80mm lens (20X zoom ratio) would have less magnification than a 10mm to 100mm lens (10X zoom ratio). The smaller the chip the greater the magnification if subject size is to remain the same.

OK, I think we're getting confused here - and it's quite possible it's me!

Magnification is a relative comparison is it not? It's not absolute. So if I magnify something 2 times, it's twice the size. If I'm right then the "magnification" of a 4-80 *IS* more than a 10-100 because the 4-80 will magnify something that you're looking at by 20 times from full wide to full tele, but the 10-100 will only magnify it 10 times from wide to tele. Am I not right here?

Aaron

Aaron....magnification is a bad term...we all use it, but it actually refers to things like microscopes...things with a ratio of 1:1 or greater. That said magnification is not really a relative term in the sense you are talking about, a 20x microscope would magnify its subject by a specific amount relative to the viewer. A video lens that is a 20x has no such specificity to it. Its simply a term to describe the RANGE of the zoom, or the relationship between its widest setting and its longest setting. I think you are using the word magnification to refer to this relationship, and that's where you're going wrong.

If we were to apply the word magnification to a video camera lense (we'd be incorrect, but in the same sense as we would a microscope) then the longer the maximum setting, then the greater the "magnification".

Am I getting somewhere?

Barry

Ho, that's new to me Don. Can you explain this a little bit more? Is the reduction in DOF if significant?

Aaron, no, if everything is equal, the lens with the longest focal length (highest mm) produces the highest magnification. Smaller chips do not inherently have higher magnification or greater DOF. But if you choose to have the image on the smaller chip match identically the image on the larger chip, then the focal length must be changed or the distance from subject to chip plane must be changed. Changing distance to subject or the focal length changes DOF.

I follow you barry, I don't follow you don, can you explain further, is anyone else impressed by hwo quickly the xl2 forum took off... a lot of talk for a camera 99% haven't even SEEN in person!

The number of blades will not change the DOF, but will effect the Bokeh, or how out of focus highlights will appear.

Nick: that's because a lot of questions are asked and a lot can
be answered as well due to careful inspection of the specs and
other information (like the XL2 watchdog) that IS available. And
a lot of this still depends a great deal on knowledge and experience.

Like the topic we are discussing now. It is a well known fact
that video zoom factors are not the same as microscopes for
example. It is also well known that the properties of DOF will
change if you change the chip size (or the READOUT size)
without changing the lens. This also "changes" the focal length
of your lens etc.

ROb totally understand, I just think it's great that we can do all this now before most of us even touch the darn thing, quite a resource.

This also "changes" the focal length of your lens etc. No, changing chip size does not change the focal length of a lens. A millimeter is still a millimeter, nothing has changed.

Indeed, I worded this wrong. Thanks for catching this. Nothing
will change on your lens, ofcourse. What would be a better way
to describe this Jeff? Apparent focal length or something?

In most digital still photography communities the term "crop factor" has been adopted, to indicate a change in image size. The reason being is that the image isn't magnified, it's just a smaller portion of the image. Video communities don't really think in those terms and generally refer to it as a magnification factor. For example the magnification factor of EOS EF lenses on the XL1 is 7.2X.

<<<-- Originally posted by Rob Lohman : Apparent focal length or something? -->>>
The Field of View is what changes when you change the imaging size. A smaller chip will have a narrower field of view.

This is where the debate, or perhaps confusion is a better word, starts. Barry is 100% correct in his thinking, except chips (CCD, CMOS etc.) don't have Fields of View. Lenses have FOV, for example, zoom a lens and the FOV changes, not the chip. The FOV is related to the diagonal of the chip, but FOV changes with change of change of focal length. It basically boils down to FOV being a characteristic of the lens and the chip is the device which displays the FOV.

Field of View is measured from the rear nodal point to the two opposite sides of the chip, when the lens is focused at infinity and the rear nodal point is one focal length from the chip.

Yeah, it's overwhelmingly difficult to say it all and cover all bases.

What I meant was: given the same lens and all other things being the same, if you substitute in a smaller chip, you will have a narrower field of view as your resulting image. The reason being, the lens projects the same image at the same focal plane, but the smaller chip can (obviously) only record a smaller subsection of that same image. DOF and other such characteristics do not change, as those are all factors determined by the lens and are all set in stone by the time the image gets projected on to the chip. So Chip size determines how much of the lens' frame is seen. Larger chips = more of the image displayed, which means for an equivalent lens, you would see a wider field of view when using a bigger chip.

which means for an equivalent lens, you would see a wider field of view when using a bigger chip.Correct, which is why 50mm is "normal" on 35mm film, 80mm to 105mm is normal on medium format and 210mm is normal on 4x5 format. A 50mm lens would be extremely wide on a 4x5 camera.

I think we're getting nit picky here, and talking mainly semantics.

Your "perceived" FOV changes with chip size but Jeff is probably being technically correct saying the chip doesn't change the "real" FOV, the focal length does.


Aaron

<<<-- Originally posted by Aaron Koolen : Your "perceived" FOV changes with chip size but Jeff is probably being technically correct saying the chip doesn't change the "real" FOV, the focal length does. -->>>

Aaron, with all due respect, Jeff isn't being nit picky, he's being 100% factual. Would a 4x5 negative have a greater FOV than a 35mm negative? No! Why? Because the FOV of the light sensitive medium--negative, transparency, CCD--regardless of its size, is determined by the focal length of the lens. That's not being nit picky, it's being accurate.

Jay

Jay, I wasn't getting at Jeff at all. I got the feeling Barry wasn't understanding what Jeff was saying - if I was wrong Barry I apologise.

It seemed obvious to me that it wasn't the actual chip that changed the FOV, but the focal length, whereas Barry was talking about the final output looking like it had a different FOV if you changed the chip - which is also true.

I assumed that people were talking about the same thing, just different terminology and getting confused.

Aaron

Aaron, thanks for the clarification. That's much appreciated. Sometimes, depending on the length of the thread and who says what, it gets difficult to follow... sometimes the "thread" becomes a bit frayed.

Jay