As I read the "First Impressions" article about this new camera, many things
seems to be upgraded, which of course I'am pleased to se but:
as I looked at the lovely photo of Cody and especially the 100% detail,
I have to say that I'am not all that impressed, at least on my screen it
looks like it's "watercolored" ?

As this camera uses 2 Digic III prosessors it posess a lot of computal power,
which to my eyes looks like it's been used on some heavy noise reduction ?
I hope my imression is just result of compression for the web, if not then I
surely hope that you can get the raw files without noisereduction as I would
like to have control over the noise reduction.

Worst case scenario: The sensor itself has som form of built in "hard" reduction
that can't be altered by firmware.

Hope the'y (Canon) got it right,
Lefa

One thought struck me... Note that all the samples were from RAW files. I wonder if perhaps there isn't some optimization to be made for this camera that isn't accounted for in the RAW converter that Michael used? For instance, ACR might work right now, but the MkIII isn't officially supported by 4.0, either. Its possible that a different RAW converter would do a different job on the files... ???

The "painterly" effect in the 100% crops (as shown in the pic of Michael's dog) remind me of ACR conversions with my 10D at 100%.... I wonder what DPP would look like (not like I'm planning on using it in my work flow, but....)...??? Hmmm....

The look of the dog, is also very typical for the 70-300 DO.

Canon wrote a white paper on Canon's CMOS sensors, including the purposes and technique of noise reduction in their CMOS sensors. It is on the Canon website.

Frankly, this generalizing on the basis of a single photo seems pretty far off the mark to me. A combination of plane of focus, f stop, presence of wind, subject movement, and last but not least, the conversion to jpeg may be responsible for the look of the photo.
There is also the fact that the camera is pre-production. I wouldn't draw any conclusion from this one photo or one camera.

George Deliz

I hope Michael doesn't mind me embedding the image here.

I too find this very lacking in detail, and it looks very much like the result of typical heavy noise reduction. Of course we can't make any firm conclusions about this camera from one image but so far I am suspecting that Canon employs "in-camera noise reduction" to achieve low apparent noise. The transitions aren't smooth. The flower shots have more detail but the noise still looks suppressed. Still very useful, don't get me wrong!

Uhh, humm, .....

It's a part of the image that's OUT OF FOCUS. Deliberately so to show noise, or the lack of it.

Has nothing to do with the lens used. Really.

Michael

Just as a start I would like to state that I have never used the 1D MKII nor the 70-300mm for that matter. Thus, I rely and respect your estimation on the noise level of the 1D MKII. Nevertheless, after reading the thread, I felt compelled to point that the lens used does indeed affect the appearance of out of focus parts of the image.

I am not talking about bokeh per se, but the interplay of image processing algorithms and bokeh. Depending on the lens used, even when focal length and aperture are the same, the slide from a sharp part of the image to an OOF part will be different. The lens leaves the shape of it's aperture and optics visible on all parts of the image. If the bokeh of a lens isn't nice and smooth, the shape of the aperture comes apparent on contrasted parts of the OOF area. One might think of this as a kind of a brush that leaves it's shape behind on everything that it touches. Furthermore, if the noise reduction algorithm of the RAW processor has a nasty structure to it (think of the median filter), it might exacerbate the "imprint" left by the aperture on the image. As I have no experience on the lens used nor the RAW program used for processing, I feel uncertain about what I am seeing on that particular image. It's nice that you have the different ISO versions of the flower photo for us to see, but the crop of the photo in question (Cody in the Shade) loses some of it's informational value because there is no "noiseless" comparison photo. I for one would have needed this second input to evaluate where the artifacts were coming from.

Anyway, even though in danger of falling off of a pixel peeping cliff, I was triggered to respond because I felt "nothing" was quite a strong word for this occasion.

Thank you for a great site,
SD