When you adjust the "brightness adjustment" in DPP, it appears to just add an offset (based on viewing the histogram display in the "RAW Adjustment Tab"). But that "offset histogram" is pushed over to the right, moving with respect to the tone curve which remains fixed. So the effect of the tone curve is applied differently too if you see what I mean.
Thus, if you view the RGB histogram (which shows the result of the conversion), the histogram is stretched as you increase this "brightness adjustment" on the RAW tab.
I only make adjustments on the RAW tab in DPP and then save the converted file and open it in Photoshop for other adjustments. But I don't need to make brightness adjustments of any kind at that point because I've done all of that with the RAW adjustments in DPP.
In the RAW adjustment tab of DPP, another very valuable tool is to simply grab and drag the white point and the black point right on the histogram itself.
To me, this is a very intuitive way to make adjustments to contrast and brightness level at the same time so I almost never use the "brightness adjustment" control in DPP. Instead, if I need to brighten a picture, I just drag the white point down as needed.
I think that a lot of people fail to realize that the white and black points can be moved in DPP, and that is what leads to them not liking it as well as other programs. For me, however, that ability to drag the white and black points, right on the histogram, is fantastic and is something I use all of the time.
I find making the adjustments right on the histogram just seems intuitive with respect to the math that's describing how the input data will be mapped to the output file. But it's a matter of personal preference, I think.
Still, if Canon would put visible "handles" on the white and black point lines so that it was obvious that you can drag the white and black points, I think they'd make a lot of new friends for the program. I can't help thinking that most people don't even realize that you can move the "ends" of the chart.
I like being able to make an adjustment on the RAW tab and then check the RGB tab to see what the "output" histogram looks like after that adjustment. It'd be neat if you could see both at the same time, though.
I still want someone to analyze the RAW data from this type of test to see what the actual RAW exposure difference is. I can't help but to think that the tone curve applied by DPP for the 40D shots is enough different to what it applies to 20D shots that this factor alone might be responsible for the apparent difference in brightness we see. So I hesitate to put too much faith in analysis of the overall brightness of the converted pictures.
But then again, this IS what we're getting for our output, using their software set to its defaults, so it may well be fair to say that the 40D is exposing about 0.5 EV lower than the 20D in this test.
Thanks for running that analysis on those.
I should try something similar on the full frames but I suspect I'd get the same results.
You should install and play with DPP. It's actually a very good RAW converter IMO and you'd be able to see the two histogram displays that it provides and see how it shows a 2 stop better available DR (on the RAW histogram) when you load up a 40D file than it does when you load up a 20D file. This is part of why I think the tone curves must be different.
The next thing I want to do is to try out the "highlight priority" mode and see if that opens up the "grayed out" upper two stops on the RAW histogram in DPP.
--
Jim H.
Thus, if you view the RGB histogram (which shows the result of the conversion), the histogram is stretched as you increase this "brightness adjustment" on the RAW tab.
I only make adjustments on the RAW tab in DPP and then save the converted file and open it in Photoshop for other adjustments. But I don't need to make brightness adjustments of any kind at that point because I've done all of that with the RAW adjustments in DPP.
In the RAW adjustment tab of DPP, another very valuable tool is to simply grab and drag the white point and the black point right on the histogram itself.
To me, this is a very intuitive way to make adjustments to contrast and brightness level at the same time so I almost never use the "brightness adjustment" control in DPP. Instead, if I need to brighten a picture, I just drag the white point down as needed.
I think that a lot of people fail to realize that the white and black points can be moved in DPP, and that is what leads to them not liking it as well as other programs. For me, however, that ability to drag the white and black points, right on the histogram, is fantastic and is something I use all of the time.
I find making the adjustments right on the histogram just seems intuitive with respect to the math that's describing how the input data will be mapped to the output file. But it's a matter of personal preference, I think.
Still, if Canon would put visible "handles" on the white and black point lines so that it was obvious that you can drag the white and black points, I think they'd make a lot of new friends for the program. I can't help thinking that most people don't even realize that you can move the "ends" of the chart.
I like being able to make an adjustment on the RAW tab and then check the RGB tab to see what the "output" histogram looks like after that adjustment. It'd be neat if you could see both at the same time, though.
I still want someone to analyze the RAW data from this type of test to see what the actual RAW exposure difference is. I can't help but to think that the tone curve applied by DPP for the 40D shots is enough different to what it applies to 20D shots that this factor alone might be responsible for the apparent difference in brightness we see. So I hesitate to put too much faith in analysis of the overall brightness of the converted pictures.
But then again, this IS what we're getting for our output, using their software set to its defaults, so it may well be fair to say that the 40D is exposing about 0.5 EV lower than the 20D in this test.
Thanks for running that analysis on those.
I should try something similar on the full frames but I suspect I'd get the same results.
You should install and play with DPP. It's actually a very good RAW converter IMO and you'd be able to see the two histogram displays that it provides and see how it shows a 2 stop better available DR (on the RAW histogram) when you load up a 40D file than it does when you load up a 20D file. This is part of why I think the tone curves must be different.
The next thing I want to do is to try out the "highlight priority" mode and see if that opens up the "grayed out" upper two stops on the RAW histogram in DPP.
--
Jim H.
) 
