Exposing to the right is the default starting point for optimum exposure with digital sensors. It is virtually always the best choice in the more common situations in which dynamic range is no larger than that of the sensor."
So I guess it just depends on whether what you're shooting is a "common situation." If you're shooting concerts or other events where the lighting is changing by seconds and you're looking for critical moments, you don't have the luxury of ETTR. When I'm out shooting streetscapes with plenty of time, I factor in ETTR. Henri Cartier Bresson's negs would make Ansel Adams' beard whittle off his face. But, damn, they're good pics....Show more →
I totally agree. Shooting landscapes is very different from shooting concerts, wars, sport and weddings. ETTR is appropriate when possible, but don't forget that with many landscapes, like the one I show above, there is no room for ETTR because the sun is directly in the picture. You will never keep the sun under control AND retain detail in the shadows. In this situation one actually has to effectively expose to the left, and then push the shadows to retrieve them, in the sense that the great majority of pixels in the image are going to have to be darker than mid-grey, since it is so hard to cope with the very strong highlight of the sun. Those are the circumstances in which I found the 1Ds3 to be excellent at retrieving shadows. My surprise with the 5D2 is that at low isos it seems slightly less good at the trick of exposing to the left and pushing shadows in order to preserve highlights.
Don't you think your being a champion of the obvious a bit? If Brainiac spent most of his time shooting studio and landscapes, then I'd agree with you. Does he? I'm assuming he's shooting a lot of events. And within that context - yes, ETTR is a luxury. Some of Salgado's negs were hell to deal with for his developer/printer...but produced some pretty impressive pics. If you're shooting fast changing events/documentary, you've got other priorities besides ETTR. It's all context.
Sorry, I posted again there after deleting my original post.
And, Brainiac, I get where you're going about DR even in shooting landscapes with the 5dII. I know it's disappointing for you in comparison with the dIII. Just out of curiosity, are you happy with it weight/performance-wise in comparison to the dIII?
brainiac wrote:
Sorry - I didn't miss it, I just haven't had time to discuss it. I use DPP, latest version, which I think is the fairest way to assess Canon cameras since Canon produces the software. Also I am mean, and having spent much money on a camera that can produce raw files, I like to avoid spending yet more money on raw developers, when I think the camera maker should supply such software, and my experience is that the camera maker generally supplies the best available raw engine.
Fair enough, it would make sense for Canon to be able to provide the software that maximizes quality. But since this isn't a flag ship camera, it's quite possible that they spent less time optimizing the raw converter than they did for the 1DsmkIII, and perhaps that's where some of the discrepancy is coming from? I would expect that trying a few different raw converters would reveal whether the pattern noise is a product of the sensor or the conversion.
I've read most of this thread, and I have to admit I can no longer see the wood for the trees on this. I would like to run my own test on my own 5D2 to see if I can replicate the problem.
Could someone please advise me, EXACTLY please, what I need to do to check this on my own camera?
Set iso to 200. Take a normal picture that has areas of flat tone in the dark grey to black range. Process images in DPP and set contrast to minimum. You should be able to see a slight criss-cross texture in the darker shadow regions especially if you push to lighten up the shadows to gain further DR. The result from my 5D is different to what I used to see from my 1Ds3 in these circumstances.
There may be some truth to the whole issue revolving around the RAW conversion. I noticed quite different results between Firmware 1.0.6 and 1.0.7. Also I was at my friend's studio the other day and he was shooting with the MkII getting spectacular results. I asked him what converter he was using and he swore by the latest C1Pro 4.
I downloaded the free trial and sure enough it is producing exceptional conversions. I have a feeling Canon rushed the new DPP and it is not optimal as yet.
brainiac wrote:
Set iso to 200. Take a normal picture that has areas of flat tone in the dark grey to black range. Process images in DPP and set contrast to minimum. You should be able to see a slight criss-cross texture in the darker shadow regions especially if you push to lighten up the shadows to gain further DR. The result from my 5D is different to what I used to see from my 1Ds3 in these circumstances.
Ah, I had a feeling that you were doing something to push the shadows up. Not that it isn't fair to do. But contrast to a minimum is probably different than the default. Which is probably why others had problems reproducing this.
kidtexas wrote:
Ah, I had a feeling that you were doing something to push the shadows up. Not that it isn't fair to do. But contrast to a minimum is probably different than the default. Which is probably why others had problems reproducing this.
No - minimum contrast is the default in Landscape picture style, a style which comes pre-installed on the camera.
The issue here is that the 5D2 is not performing as well as the 1Ds3 in a specific way, and that was unexpected.
Andi Dietrich wrote:
So would shooting at a higher ISO improve the banding? Sounds odd, but that is what you basically say?
I don't know, but remember, my expectations are set by using a 1Ds3 for a year, so the question becomes 'does the gap between the two cameras close at higher isos?'
FashionBoy wrote:
There may be some truth to the whole issue revolving around the RAW conversion. I noticed quite different results between Firmware 1.0.6 and 1.0.7.
That is my feeling as well... FW 1.0.6 gives better IQ when used in LR 2.2 and DPP 3.5.2 (apart from the black dots and sRAW banding) IMO. Less noise at low ISO's (shadows) and virtually no magenta spots/casts.
But there is only one way to find out...
If you see excessive noise/magenta spots/casts in low ISO RAW files, please state the FW version and RAW converter used.
Do the same if you don't see excessive noise/magenta spots/casts in low ISO RAW files...
brainiac wrote:
No - minimum contrast is the default in Landscape picture style, a style which comes pre-installed on the camera.
The issue here is that the 5D2 is not performing as well as the 1Ds3 in a specific way, and that was unexpected.
I understand the issue.
It didn't look like contrast at a minimum was a default from the above screen shot of DPP, but what do I know. I don't use it. Apparently it is for Landscape mode.
However, I was referring to why others are not seeing what you are seeing, and my point still stands. Maybe if they converted their images in DPP, in Landscape mode, etc., like you are, they'd see it. Maybe not. I don't know.
brainiac wrote:
You will never keep the sun under control AND retain detail in the shadows. In this situation one actually has to effectively expose to the left, and then push the shadows to retrieve them, in the sense that the great majority of pixels in the image are going to have to be darker than mid-grey, since it is so hard to cope with the very strong highlight of the sun.
I exposed this so that the very bright area where the sun illuminates the edges of the clouds is just barely "blown" - and I bring back just enough detail using the recovery slider in PS. Technically there is a small bit of pure white there, but not much and it is appropriate for this scene.
I had to do considerable shadow recovery in several portions of this image - in the partially illuminated arroyo area part way up the ridge I had to apply curves adjustments to get a bit more contrast due to the atmospheric effect. In front of the mountains and on the playa I had to boost the shadow levels significantly. I also applied a curves adjustment to a masked area on the right side of the frame where that light beam was not as strong as I wanted for the composition. Despite having large areas of low luminosity and low contrast that had to be brought up in post, it prints nicely.
Only other explanation I can think for increased noise would be that the sensor is running hotter than spec (i.e., too much LiveView usage prior to taking the shot) or a defective component. I don't think it's reasonable to draw conclusions without trying another body.
FashionBoy, of course with Canon's QC issues, it becomes the norm to try another body, lens, copy. I don't think this should be the case at all. Lowered expectations only hurt the consumer.
FashionBoy wrote:
Only other explanation I can think for increased noise would be that the sensor is running hotter than spec (i.e., too much LiveView usage prior to taking the shot) or a defective component. I don't think it's reasonable to draw conclusions without trying another body.
As I stated earlier in the thread .... I had two bodies which exhibited the same poor low ISO quality compared to a 1DSMKIII
