My 400D came today...

[Nill Toulme]

Well this first outing was very high contrast, and according to Capture One at least, I had a very few slightly blown highlights, but the exposures were dead on. I hear what you're saying regarding a low contrast situation where you have some wasted space on the right side of the histogram, but this was not that, and it looks to me like the little box nailed the exposures. But I'm admittedly not very sophisticated about this sort of thing.

Nill
~~
www.toulme.net

 

[WillGood]

I don't think so; that is a feature of the converter you're using; not the camera. The camera has the lowest RAW sensitivity of any existing Canon DSLR. It meters for about 120% of the stated ISO and the sensitivity of the RAW data is about 85% of the stated ISO. Many converters are already compensating for this at "0" exposure, and many don't handle RAW highlights linearly to begin with. The "Exposure" control in ACR, for example, is a bias control and a curve tool, and has no capability of rendering the highest RAW values normally when you pull the image in conversion. This is a serious defect in ACR.

Hi John
This is my first canon, I normally shoot work with kodak SLRc s - they have very wide DR. When you blow a highlight , it shows on the histogram in camera and in PhotoDesk, ACR & silkypix , as is logical. My Olympus cameras have simular histogram properties.
With the kodak image, and to some degree the E1, you can still recover highlights .

Every camera has anomalies, and the canon blows highlights in DPP when the histogram is not even close to the right. Furthermore, none of the data has been recoverable in DPP. Ive yet to run it thru Silkypix or ACR. Im just telling my own anecdotal experience here - I swear I represent no one but myself! ; )

Saying that , It seems workable by using appropriate settings, and I find myself using 400d a lot, and enjoying the handling, AF,
hi iso, and lack of artifacts produced by Kodak!

There is no noise reduction on the XTi, other than what the JPEG engine loses. The RAW data has no NR. The camera has "dark frame subtraction" as a custom function, which you can turn "On", "Off", or to "Auto". "On" is dangerous, as dark frame subtraction kicks in at one second, and one second exposures don't benefit from dark frame subtraction unless the camera is hot, even at ISO 1600. Unnecessary dark frame subtraction causes a 41% increase in noise at all exposure levels.

My kodaks have no AA filter, like MF backs and the new Leica. I only add noise reduction if needed.
So,When I first looked at the canon 400d files, it certainly looks like there is proprietary noise reduction - especially at higher iso levels. Perhaps this is just the look their AA filter produces on files, I dont know. Theres something going on....
BTW, dark subtraction is only processed at very slow shutter speeds.
Overall, I think the 400d is an incredible bang for the buck, and a real tool for a photographer....and I don't hate DPP ; )
William

 

[Sean DeMerchant]

My kodaks have no AA filter, like MF backs and the new Leica. I only add noise reduction if needed.
So,When I first looked at the canon 400d files, it certainly looks like there is proprietary noise reduction - especially at higher iso levels. Perhaps this is just the look their AA filter produces on files, I dont know. Theres something going on....

Would you be willing to share some example files shot of the same scene?

BTW, dark subtraction is only processed at very slow shutter speeds.

The XT only kicks it in at 30 seconds for ISO 100-800 and 1 second or longer at ISO 1600. Is the XTi similar?

Overall, I think the 400d is an incredible bang for the buck, and a real tool for a photographer....and I don't hate DPP ; )

DPP has nice output on occassion. I just find its thumbnails too small for first pass review.

enjoy,

Sean

 

[John Sheehy]

My kodaks have no AA filter, like MF backs and the new Leica. I only add noise reduction if needed.
So,When I first looked at the canon 400d files, it certainly looks like there is proprietary noise reduction - especially at higher iso levels. Perhaps this is just the look their AA filter produces on files, I dont know.

That's the difference. Cameras without AA filters have a sharp look, but unfortunately they capture all kinds of artifacts along with the image which can't be properly separated in PP. AA filters separate them before capture, but reduce maximum pixel-to-pixel contrast.

 

[John Sheehy]

This is my first canon, I normally shoot work with kodak SLRc s - they have very wide DR.

Do you really mean DR, or do you mean highlight headroom relative to metered values? DR can exist mostly in the shadows, or mostly in the highlights.

When you blow a highlight , it shows on the histogram in camera and in PhotoDesk, ACR & silkypix , as is logical. My Olympus cameras have simular histogram properties.
With the kodak image, and to some degree the E1, you can still recover highlights .

Every camera has anomalies, and the canon blows highlights in DPP when the histogram is not even close to the right. Furthermore, none of the data has been recoverable in DPP. Ive yet to run it thru Silkypix or ACR. Im just telling my own anecdotal experience here - I swear I represent no one but myself! ; )

The RAW data will only blow when it doesn't show on the Xti's luminance histogram when there are bright blue or to a lesser degree bright red highlights. The RGB histogram should be a bit closer, but it will show red flowers clipped (and they will be in the JPEG) when the RAW data is still far from clipping. For some reason, many converters like to blow out the red in the output to give a super-saturated look.