I think Paul is talking about back handling, ie how the histogram shows. At the raw level all backs behave in the same way a simple linear filling of a channel and than clip straight off (non-linear highlight response is indeed coming to sensors, but it's not in our cameras yet).
Most highlight "blinkies" show blinking a bit early, for example based on a calculated luminance for the set white balance, rather than show exact raw clipping. The P45+ has more than one stop more sensitive green channel than the red in daylight, while the Dalsas have more balanced RGB sensitivity in daylight (which also counts for the older 7.2um Dalsa I'm using). Maybe this difference make the P45+ show blinkies later than the Dalsa backs? An ideally exposed red channel will likely mean a clipped green channel on the P45+, while the "daylight-balanced" Dalsa can expose all channels well (simplified of course, depends on what colors you have in the scene etc).
This difference in relative sensitivity might also affect how highlight reconstruction works in Capture One, I don't know. I'm sure Pauls observations are valid, he has a lot of experience with this product and others, but he's talking about handling in the Capture One workflow rather than what's exactly happening on the raw level.
Just having more noisy shadows could also make it appear that you have less highlights as you'd want to expose further to the right.
I think it's okay to use the "noisy" sensors, but be aware that 1) I use grad filters and 2) I have a shooting style which prefer softer light conditions and often not any visible sky at all and 3) my post-processing style is such that I don't push shadows that much.
It's also important to learn by testing how the highlight blinkies and histogram works. I know I can push a little past the blinkies on my Aptus and still not clip. Rather than a white-balance dependent luminance level I'd prefer to have true raw blinkies with red=all channels clipped, orange=two channels clipped and yellow=one channel clipped, but I guess manufacturers consider that to be too user-unfriendly...
Hi Torger:
Actually I was referring to the fact that the P45+ would blow a highlight period. On the LCD screen, if you even saw a blinking highlight, then when I got back to work the file in post that would be 255-255-255, totally non recoverable and at times pink. So on a nice day, with a partly cloudy sky, if I exposed such that I saw any blinking highlight, (as I remember they were blinking in red, but it may have been black), I knew that I had to work up a different exposure until I found one where I had no highlight warning, this totally of course meant my shadows were worthless and underexposed as much as 1 to 1.5 stops. So the solution was a exposure bracketing. Working with water, in the 1 to 5 second range was the same issue even worse.
As Bart and and Eric have mentioned, C1 opens the raw files with seemingly a push if you use the "film curve", and you could go back to the linear curve. This would help in about 30% of the shots, however I never liked the output from the linear curve and always used the film curve.
If you took the shot that was correctly exposed for highlights, thus considerably underexposed, then the shadow recovery was gone, no details, tons of noise and smearing. These were all shot at iso 50 or 100 on the P45+ with the most current firmware that was available at the time.
Now take this solution to a tech camera, where you are forced to add the mandatory LCC frame and start bracketing for exposure you ware talking a ton of frames for 1 image and even more if you are working with shifts. To me way too much work in post. Yes I agree by chip design the 39MP Kodak was more tech camera friendly in relation to shifts, as you did not see the same issues the Dalsa chips have, but I still feel the Dalsa chip is the overall better solution when all aspects of the exposure are taken into consideration. Dalsa allows a for a much better highlight recovery, even with a lot of the sky blinking with highlight warnings. Thus your shadow areas have much more life and less noise.
Eric has shown many times his comparison of an A99 and P45+ shot where he attempted to recovery the shadows, in his straight forward test, you can see just how poorly the P45+ was able to handle any type of push. Thus again, the shadow area had to be exposed directly which would result in a blown highlight if there were any parts of the image that would be involved, water, sky etc.
Now also add to the fact that with a tech camera, you are currently forced to a copol shutter solution in 95% of the lenses in use. The Copol offers no 1/3 or 1/2 shutter speeds, so this problem will be even worse. At least with the DF body, you had the ability to get the in between shutter speeds, and having these can make a huge difference, as there is quite a bit of light between 1/15 and 1/30, but if you could have 1/20th or 1/25th you might have had better results.
I am not trying to trash the Kodak chip, just trying to point out what I feel are real world issues and limitations. In a post on getdpi you mentioned, the Hasselbald users of both the 39 and 50MP Kodaks seem to have less issues, some of this may be because I believe most of them are not on tech camera solutions. Phase One has always had an easier back to move to tech cameras, be it the P45+ or a Dalsa chip.
I also strongly don't feel that the long exposure capabilities of the P45+ are worth what they were in 2008. Back then the P45+ was the only solution as even 35mm DSLR"s couldn't really get to 1 hour with any clean exposures and they didn't begin to have 39MP. This has changed so much to the opposite, as pretty much any modern 35mm DSLR from Sony, Canon or Nikon can easily make an equal or cleaner 1 hour exposure than the P45+, and with Nikon have almost the same MP's. For exposures in the 1" to 15 sec range the DSLR I feel is superior as you can turn off the dark frame exposure and still have a very clean image. Don't forget, which many seem to, that the temperature range of the P45+ for 1 hours exposures is 69 degrees F and much less than 100% humidity. If you tried to use a P45+ in Arkansas or anywhere in the south of the US in late spring or summer outdoors, you would not be able to get anywhere close to 1 hours, more like 15 minutes.
Purchases of MF backs, is so much more difficult since you just can't walk into a local camera store and try one out, and you may also need to use some pretty specialized camera equipment. I know just how friggin hard it is for a shop like mine to get a rental on any Phase One back due to the insurance, net it's impossible, so if I want to see or try out a product, I have to travel to either NY city, (not going to happen) or Atlanta, a bit more possible. So for the OP, I do hope you can find someone in your area that either has one of the products you are interested in or a good dealer with demo capabilities.
Paul
