denoir wrote:
Anyway as for the difference, does it matter if it's software or hardware? With ACR the difference was even larger.
it matters if you want to discuss crop vs FF rather than 7D vs 5D (which isn't of much interest to me since i own neither and don't picture myself buying either). out of curiosity, at what output size does the shot from the 7D surpass the cropped shot from the 5D? i'm assuming at some larger size the 5D shot will look worse due to lower pixel density.
carstenw wrote:
The X1 looks a little more purple/blue in the 5D2 shot. Is it reflecting something?
The background looks like a tad more Black Level slider would pull the 7D up, but you say that this is not your experience?
A contrast adjustment would probably do the 7D shot some good, but it wouldn't change the local features nor would it help with the color separation. In my experience the 5DII definitely has an edge there and that on average its rendering is significantly better. It took me quite a long time to admit it as while I was still using Canon AF glass I was very convinced that the 7D was the better camera in every respect (it is except for IQ).
To put it in a vague terms I feel that when I take a shot with the 7D I usually get an image that accurately represents reality. When I do the same with a 5DII I instead get an image that has an artistic quality to it more than just representing the real thing. I know that there are a bunch of factors there such as vignetting, thinner DOF for the same framing etc The colors however are a significant part of it. Since I've switched to almost only using Zeiss glass I almost exclusively use my 5DII. I still keep the 7D and a bunch of Canon AF glass, "just in case" but in practice I'm not using them at all.
sebboh wrote:
it matters if you want to discuss crop vs FF rather than 7D vs 5D (which isn't of much interest to me since i own neither and don't picture myself buying either). out of curiosity, at what output size does the shot from the 7D surpass the cropped shot from the 5D? i'm assuming at some larger size the 5D shot will look worse due to lower pixel density.
You mean when cropped? No idea. I don't usually crop my 5DII images by a factor of exactly 1.6
oh, well. i've always been curious at what print sizes those extra pixels really make a difference vs the trade off of having bigger pixels. there probably isn't a fixed point though as the lens and sensor act as a group and lens resolution beyond the nyquist frequency probably still helps (unless AA filters mostly prevent this?).
This would be far more informative with Zeiss glass. With the 58/1.2 the lens is the limiting factor not the sensor.
Luka, do you have shots to compare resolution of Zeiss glass with Rokkor 58mm 1.2? I had Canon 85mm 1.2, Zeiss 85mm 1.2 (50 Anniversary), and Rokkor 58mm 1.2. If my memory is correct, Rokkor resolution is close to the big boys. Color rendition is different though. I think your shot was taken at 1/6 sec and was wondering if the resolution would improve at higher shutter speed (to minimize camera shake -- even on tripod).
sebboh wrote:
i feel like i've helped drive this pretty far off topic now and i have no new pictures to post, so i'll dig through my flickr archives for old 50/1.2 shots (i may even have to resort to cat shots ): http://farm3.static.flickr.com/2745/4081851709_762a03e52e_b.jpg
denoir wrote:
The difference is pretty big in terms of color rendering and micro contrast. I very much doubt you could make them look similar with PP. At least I could not.
I wouldn't underestimate yourself that badly. It's just a couple of slider nudges.
Bifurcator, I'm afraid that I wouldn't call those shots very similar. The blacks have been blown on the 7D shot and the highlights compressed (i.e too much contrast added). You have tried to fix the local contrast of the Leica logo by sharpening and done a fairly good job, but image is way over sharpened now. The color rendering of the blur is quite different as well.
And really there's about the same difference between the FF and My Edit as there is between the two you posted. Just leapfrogged in one direction so now the differences are the opposite.
Here's the two you posted:
Here the lightest one is the crop sensor
There's a massive amount of difference here and it could be due to almost anything including reflected light off the Leica camera, different table angles - so a massive difference in bloom and bounced light, in camera processing differences, etc. When you change the angles like this it becomes inadequate for comparison IMO.
Here's the differences between my edit of the crop sensor and your original FF:
Here the lightest one is the FF sensor. And the Crop sensor is the one with the better histogram.
About the same amount of differences but now the crop-sensor-edit has passed the FF in the same directions as the differences between your originals - (where the crop sensor had less contrast than the FF it now has a tad more, etc.) The Red dots are the blown whites (notice more in the FF image) and the GIF compression decided to kill the blue ones signifying the blown blacks - but there were like 18 in the FF original and 12 in the crop sensor edit. The original crop sensor shot you posted has zero.
--
PS, yeah, I'm calibrated perfectly for the signals and displays I deal with.
On my NEC 2689WUxi, 7D with post-processing by Bifurcator looks harsh and artificial. My monitor is calibrated by NEC tools. This version is no go for me.
I sold my Rokkor and this thread made me buy it again. It arrived yesterday and I did a quick comparison with my Zeiss 85mm 1.2 Jahre 50 year anniversary. I have to say that this Rokkor at 1/10 the cost of the Zeiss is doing very well.
Both photos were taken at f1.2 wide open. They were processed with C1 5.2, all default options. No post processing at all.
cavewalker wrote:
sebboh: i prefer the 50mm 1.2 The sharpness is slightly better and the bokeh have some kind of Noctilux 0,95 style (smooth and semicircles)
yeah, i really liked that for some situations with the 50/1.2. my MD 50/1.7 seems to do the same thing but doesn't blur near as much so it ends up looking too busy at times. the MC 50/1.4 (same optical design as the 58/1.2 according to rokkor files) seems a bit better behaved in that respect, which some people will view as good others won't.
):
