The "rumor" is a 22 megapixel full frame camera, but chances are it'll cost $10,000+ for the body. But I've also heard a rumor that Canon won't release a camera like that until they do an update on their lenses. The problems with digital is that it shows more lens defects than film cameras do. And Canon isn't alone in their thinking. There's a reason that Nikon has banked so heavily on their smaller sensor cameras that only use the center of the lens.

But I really don't know anything and everying I say is pure speculation.

Larry,

Are you sure that is the reasoning behind Nikon's use of APS sensor chips? Isn't it that Kodak and Nikon never got along very well and Kodak was the only source available to Nikon to get full-frame sensor chips? Nikon is too small a company to manufacture its own digital sensor chips, so is dependent on other for their dSLR's, like just about every other camera brand except Sony of course.

As for Canon lenses showing deficiencies with full-frame sensor like the 5D, I used the 16-35mm L f/2.8 with the camera when I tested the camera for my report in Shutterbug. The lens performed OK, but not as finely as the older 20-35 L f/2.8 from my personal inventory, but at 16mm the acute angle the image strikes the sensor at the outer limits of coverage ( the outside portion of the image close to the frame edge) the sensor shows its fault characteristics by recording linear chromatic fringing, a not uncommon artifact with any digital sensor shooting with an extreme wide angle of view.

Anyway, the original question was regarding the mid-range semi-pro cameras. In my humble opinion Canon really is not pressed by the competition in this category, and has a firm grasp on the #1 world sales position. Once the competition makes a move with a new generation mid-range that at all threatens the price/performance position of the Canon 5D, then Canon will pull its next version off the shelf and into the market.

Well I said speculation, didn't I. It'll be interesting to see what a year from now brings.

Quote:
The "rumor" is a 22 megapixel full frame camera, but chances are it'll cost $10,000+ for the body. But I've also heard a rumor that Canon won't release a camera like that until they do an update on their lenses...

So you mean that I have the option now of either procrastinating over buying a digital SLR because Canon may eventually release a better body, or because they also may update their lenses? I thought it was bad enough when I only had the dilemma of buying an expensive body that may soon be superceeded as an excuse!

I presume from your helpful comments that the lens/sensor issues affect wide angle lenses such as the 16-35mm at its wider angle settings. Does the problem disappear at, say 18mm or 20mm?

Does a similar problem affect the 10-22mm zoom fitted to the EOS400D or other part-frame sensor models?

Many thanks for the advice!

>>I presume from your helpful comments that the lens/sensor issues affect wide angle lenses such as the 16-35mm at its wider angle settings. Does the problem disappear at, say 18mm or 20mm?

Does a similar problem affect the 10-22mm zoom fitted to the EOS400D or other part-frame sensor models? <<

To answer that specific set of questions, it has been my experience that the artifact of chromatic image fringing has been seen at the outer edges of the picture frame with both APS sensor size as well as full frame. This only occurs when there are lines and sharply contrasted detail in the image's outer extremities when the lens is ultra wide , in the range of 16 to 20mm with a full frame and 10-12mm with an APS size sensor chip.

This color fringing of lines and contrasty detail can be cleaned up readily with color noise filters that are a part of applications like Adobe Camera Raw. So it is not a serious problem because a clean image can result with just a little processing care.

Quote:
>>... it has been my experience that the artifact of chromatic image fringing has been seen at the outer edges of the picture frame with both APS sensor size as well as full frame. This only occurs when there are lines and sharply contrasted detail in the image's outer extremities when the lens is ultra wide , in the range of 16 to 20mm with a full frame and 10-12mm with an APS size sensor chip.

This color fringing of lines and contrasty detail can be cleaned up readily with color noise filters that are a part of applications like Adobe Camera Raw. So it is not a serious problem because a clean image can result with just a little processing care.

This is interesting and reassuring.

Can I ask for people's opinion on an unfair comparison then? Given that I am starting from scratch (digitally), how much of a difference would I see in the end result between:
a) the Canon EOS-400D with the 10-22mm f3.5-4.5
versus
b) the Canon EOS-5D with the 16-35mm f2.8L

It may be an unfair comparison as the latter combination is roughly 3 times the price (US$1,400 versus nearly US$4,000), but both would fulfil my needs, which is equivalent to 20 to 30mm range in the 35mm format (longer focal lengths appear easier to fulfil). I have looked but cannot find any side-by-side or even comparable reviews of the two lenses, and would very much appreciate comments from those who have had access to these optics/bodies.

At what point in print size would one see a difference between these two approaches, and just how big a difference would it be?

Any comments most gratefully received, and a Happy New Year to all reading this!

>>Can I ask for people's opinion on an unfair comparison then? Given that I am starting from scratch (digitally), how much of a difference would I see in the end result between:
a) the Canon EOS-400D with the 10-22mm f3.5-4.5
versus
b) the Canon EOS-5D with the 16-35mm f2.8L<<

Probably the differences will be subtle, like the fact the sensor site size of the 5D is larger than the 400D so each cell gathers more light, and when the light level is low, in shadows or due to low-light long exposures the 5D should produce cleaner values in shadows and dark parts of low-light photographs.

But there is a difference in the lenses involved too, although when I tested the Canon digital 10-22mm with an earlier Canon its performance was very good, although some image edge chromatic fringing was evident set a 10mm focal length in some images made. And the Canon 16-35mm f/2.8 L in my experience is not as critically sharp as the older 20-35mm version of that lens. ( I shoot my 5D with the older 20-35mm f/2.8 L Canon lens and find the results in large (16x24 inch image) prints rival or surpass what I previously obtained on ISO 100 slide film that I have since scanned.

BTW when I did a test and review of the Canon EOS 5D as published in Shutterbug last year, I shot many of the images with the 16-35mm f/2.8 L (and a few with my personal 20-35mm f/2.8 l) and used the Canon 10-22mm in a previous article, so have image files from both lenses, however the latter were made with an 8MPX Canon dSLR.