agrumpyoldsod wrote:
I hate to do this but here we go -- take Matt G's recent "comparison" vid of the X2D and GFX 100s. He shared a bunch of RAW files from each camera and when we looked at them we discovered the shooting settings were vastly different for every set up - massively different ISO/Aperture/Shutter speeds - the night shoot of HK were vastly different. Vast differences in white balance AND since one body was shot with an XCD V 55mm f/2.5 and the other a GF-63mm f/2.8 R WR as a result the portraits and street shots were also very different.
NONE of us had the development profiles Matt used and we all came up with very different looks to each other and his.
So-- while I prefer hands on comparisons - the comparison shots need to be made using EXACTLY the same shooting settings, lighting and field of view -- fairly obviously the portrait shots with the 55 should have been closer than those taken with the 63. Sorry Matt but yours were not. ...Show more →
Should the GFX match to the X2D, or the other way around? 😁
That's what I call a real world usage comparison. I experienced recently, shooting with a7c and M-P240 for a comparison, and I can match the Sony to the Leica pretty closely. But when I tried to shoot only with the Sony for a week or two, using profiles I created, tweaking the white balance from what I remembered about the Leica. I thought I was doing pretty well, until I brought out the Leica to shoot again, another painful reality check.
Interesting to note, the X2D did caught a lot of attentions, even my friend who's a big fan of Leica, said to me it's been a long time since a camera excite him so much. I wasn't affected to say the least, cause it's simply out of my price range, not just the camera, but needing to upgrade my macbook is part of the package. 😳
Maknof wrote:
I knew AF is single point, but compared with other full frame MILC, it miss the point more often, and sometimes the motor still search the point.
It was Montecchio Emilia 44.7008785,10.4454211, if a tourist come here, it is a nice place, not big place but only 2 bucks for the ticket and the guide is free
There is even a nice photo shop, the one that that own the 13 nearly grand (tax included) of this stuff i used, as demo.
As you can see in restored mechanism, focus was a little back (center spot)
About slow exposure times i am uploading 3 more shots in same folder, 1/10th iso 100 and 1/20th iso 6400 and 1600 it will take at least 60 minutes cause here there is no fiber and trees cover OTA waves. ...Show more →
Thanks I luv this part of Italy - particularly when it is not too hot.
jeffersoncasey wrote:
Should the GFX match to the X2D, or the other way around? 😁
One would assume based on the "fact" the sensor is "the same" the images should both match each other -- the "but" they don't.
PhotonstoPhotos have just "added" their sensor measurements for the X2D-100c to their site - so one can see that there are performance differences between the outputs from each body. The GFX 100 having a significant dual gain boost at ISO 503 (setting) and above, whereas the X2D does not appear to have a similar step. Based on their measurements the X2D appears to have slightly better Photographic Dynamic Range Shadow Improvement -vs- ISO Settings results than the GFX 100s
agrumpyoldsod wrote:
One would assume based on the "fact" the sensor is "the same" the images should both match each other -- the "but" they don't.
PhotonstoPhotos have just "added" their sensor measurements for the X2D-100c to their site - so one can see that there are performance differences between the outputs from each body. The GFX 100 having a significant dual gain boost at ISO 503 (setting) and above, whereas the X2D does not appear to have a similar step. Based on their measurements the X2D appears to have slightly better Photographic Dynamic Range Shadow Improvement -vs- ISO Settings results than the GFX 100s
...Show more →
I believe there are so many variables - processors and electronics, processing software and calibration, color filter and filter stacks, and most importantly the lenses. I dunno, side by side without matching shows you what's the closest result you'll get from the camera. If GFX or X2D needed a higher ISO or slower shutter speed to achieve the same exposure, so be it, one will be using that way, and there's no way you know if you have a "better" camera without a A/B all the time.
If I want to match everything for my own comparison of a7c and M-P240, the M always under expose by 2/3 if I match the setting. Does that mean I should ditch the Leica? I wished, and tried hard to do so. But everytime I just love the output of the Leica more than the Sony even with all cons considered. IMHO there are more meaningful things than figures and test charts. Or else people won't be shooting films anymore, no?
agrumpyoldsod wrote:
One would assume based on the "fact" the sensor is "the same" the images should both match each other -- the "but" they don't.
PhotonstoPhotos have just "added" their sensor measurements for the X2D-100c to their site - so one can see that there are performance differences between the outputs from each body. The GFX 100 having a significant dual gain boost at ISO 503 (setting) and above, whereas the X2D does not appear to have a similar step. Based on their measurements the X2D appears to have slightly better Photographic Dynamic Range Shadow Improvement -vs- ISO Settings results than the GFX 100s
X2D has a "significant" dual gain boost at ISO 200. It seems to be the same amount of boost as GFX100/S has. It only occurs earlier. The spike is not as pronounced because Hasselblad only allows ISO setting in full stops.
Once X2D's graph is moved half a stop to the right (adjust for the difference in ISO), the GFX100/S and X2D graphs are very similar, except that X2D has a bit of an advantage at ISO200.
agrumpyoldsod wrote:
One would assume based on the "fact" the sensor is "the same" the images should both match each other -- the "but" they don't.
PhotonstoPhotos have just "added" their sensor measurements for the X2D-100c to their site - so one can see that there are performance differences between the outputs from each body. The GFX 100 having a significant dual gain boost at ISO 503 (setting) and above, whereas the X2D does not appear to have a similar step. Based on their measurements the X2D appears to have slightly better Photographic Dynamic Range Shadow Improvement -vs- ISO Settings results than the GFX 100s
It looks like the colour filter array may be different between the GFX and the Hasselblad, according the threads on DP review.
Even with the older X1D I found the Hasselblad colours not easy to match perfectly. You'd get it close for one shot and the next would be off again. Hasselblad load a huge colour and calibration data file and each sensor is hardware calibrated. There's something going on that's actually not that easy to replicate.
However, doing a comparison and not using the same exposures is just daft. At least match the histograms as best as possible......
flash wrote:
It looks like the colour filter array may be different between the GFX and the Hasselblad, according the threads on DP review.
Gordon
There are still far to many folk on DPR who think they are engineers and since they have no actual data that is why many of the threads they seem to take over change from use to to become tedious and boring. Thankfully FM is much more user friendly.
It would be a surprise if the actual sensor design and construction were different -- but to be honest who cares.
I luv the camera and its images are cleaner than the X1D-ii/X1D.
flash wrote:
It looks like the colour filter array may be different between the GFX and the Hasselblad, according the threads on DP review.
Even with the older X1D I found the Hasselblad colours not easy to match perfectly. You'd get it close for one shot and the next would be off again. Hasselblad load a huge colour and calibration data file and each sensor is hardware calibrated. There's something going on that's actually not that easy to replicate.
However, doing a comparison and not using the same exposures is just daft. At least match the histograms as best as possible......
Gordon
Yes. The sensor is one thing, and what is before (lens, filter array,...) and after (electronics and in camera image processing) is another. GFX100S and X2D do have the same sensor, but the cameras and their rendering are completely different.
I am very happy with the HNCS and the colours my "old" X1D II provides, when compared with all the digital cameras I've had (Canon, Leica,..). And I haven't observed any difficulty of colour matching between pictures. But, as an amateur, I am maybe not demanding at the same level as yours.
flash wrote:
It looks like the colour filter array may be different between the GFX and the Hasselblad, according the threads on DP review.
Even with the older X1D I found the Hasselblad colours not easy to match perfectly. You'd get it close for one shot and the next would be off again. Hasselblad load a huge colour and calibration data file and each sensor is hardware calibrated. There's something going on that's actually not that easy to replicate.
However, doing a comparison and not using the same exposures is just daft. At least match the histograms as best as possible......
Gordon
The problem is not with the measurements but that many people misuse the numbers published on P2P without proper understanding what they mean. They use the numbers rather for support than for illumination, FujiAddict being an example.
SrMi wrote:
The problem is not with the measurements but that many people misuse the numbers published on P2P without proper understanding what they mean. They use the numbers rather for support than for illumination, FujiAddict being an example.
Sorry. I should have been clear. I was responding to Matt Granger not using the same exposure in his testing. And yes, it was a reply to a post about P2P.
This is why I need more coffee in the morning, or is that less?
tenmangu81 wrote:
Yes. The sensor is one thing, and what is before (lens, filter array,...) and after (electronics and in camera image processing) is another. GFX100S and X2D do have the same sensor, but the cameras and their rendering are completely different.
I am very happy with the HNCS and the colours my "old" X1D II provides, when compared with all the digital cameras I've had (Canon, Leica,..). And I haven't observed any difficulty of colour matching between pictures. But, as an amateur, I am maybe not demanding at the same level as yours.
I think one can take two images taken side by side and get them pretty close. That's what I'm currently doing as I test my X2D against my GFX. Lens rendering aside of course.
What I haven't been really successful at is creating a profile for the X1DII and other cameras that makes them look similar all the time, on import. I can get a Sony and a Canon reasonably close. Or Canon and Leica. I did a lot of that when I shot weddings as it made life easier if things were equalised on import. But not the X1D or XTrans Fuji files to anything else. M9 files were the same as well but that was a CCD sensor. The Fuji I get but I beleive that because Hasselblad adds a huge amount of calibration data at startup each raw file is slightly different and so a generic caliubration that works for one shot may not for another. This is of course pure speculation.
As well as each sensor being individually calibrated to the body, Hasselblad use a huge amount of corrections for lenses and exposure times As well as subject based algorithms. They measure long exposure noise off the edges of the sensor as the exposure is happening so you don't need (and they don't offer) LENR. People say HB don't do software but when it comes to this part they do it better than anyone else.
Not that I really care how they do it. Could be voodoo for all I care. I just like the colours.
Yes Gordon, I love the HB colours too. Home made sensor calibration is a very difficult job, and, if we want to be strict and precise, we must make an ICC camera profile for various temperatures. Crazy. I had a customized profile for my Leica, but actually, it didn't work quite well at all temperatures.
The way HB do it seems to be completely different. I've seen a video about the way they process, but I don't remember well where. It should be easy to find out, but it's late in Paris (France) now and if I have time I'll try to find the link tomorrow. But they say it works for about all temperatures. Yes, it's a sort of voodoo. We just have to trust....
Robert
Sauseschritt wrote:
Well, if color accuracy is so important to some - then why not use a Colorchecker ?
Like the various models sold by various companies, or really you just can print one out with a calibrated printer.
It's not about colour accuracy. It's about colour consistancy. From the linked article....
"The human eye automatically enhances details and contrast. If a camera simply restores accurate colour values in technical terms, it would be monotonous and boring, nowhere near what the human eye actually perceives. In order to take care of these subtleties, HNCS also optimises tone and contrast."
For me, Hasselblads colour is like exposure. Sure, you can fix it in post, sort of, sometimes but it'll often look better if you get it right in camera. HB's approach of calibrating the whole system to work as one from sensor to lens, is another part of a colour managed workflow.
I know there's a group who think you can shoot a color checker and make a profile and magically all files look the same. But in practice that's not how it works. You'd need to take a colour checker shot with every lens you use, at every aperture in every type of light at every ISO. Even then your sensor isn't hardware calibrated. If you could calibrate then you'd end up with files that are accurate. But no one wants accurate. They want pleasing. Hasselblad does all this for you. Set your white balance and away you go.
Now you could hate HB's colour science and that'd make things difficult. Mostly though, people seem to rave about how Hasselblad does colours.
Every Sony, Canon, Leica and Nikon I've even owned has had a different colour signature for different cameras and different lenses. Sony is by far the worst offender. If I add a light in studio, I'd need to do a Color Checker shot. And again if I change lenses or aperture. No thanks. Not going to happen. Olympus, Fujifilm and Hasselblad seem to do this the best.
I don't beleive you can perfectly colour match different camera systems 100% of the time. Different CFA, processors, corrections baked into raw files.... the list goes on. We might get close enough in an individual test not to be able to see a difference but there will be one. Colour science is a thing. And a colour checker won't get around it.
I’m excepting a rather large firmware update in the near future. Maybe before the holidays. Too many small escapements to let go for long. Hopefully before my copy comes available.
BrandonSi wrote:
I am surprised they didn't release it with aspect ratios.. you'd got to think that's one of the easiest things to implement. I guess you have to stop somewhere to ship, but that's a surprising one to leave out.
RomanMF wrote:
Can't wait for this to hit the rental market. I definitely want to take one up with me to Yosemite. I would rent one for my White Sands trip but that's a little too dangerous, lol.
Hope it's not that dangerous, can't wait to go back to White Sands with the X2D.
Looks like MG is going down the X2D purchase path, even after the most confusing set of review videos ever compiled. He’s bought the zoom he thinks is too big and heavy while he waits for a body. But it looks like he finally got the memo about how good it is optically.
Latest info is end of Jan for my 90V, although my XCD90 works extremely well on the X2D. It’ll be nice to have the extra speed and hopefully a few grams off but the AF and IQ of the current lens is really good on the new camera.
Currently I have two lenses enjoying a holiday to Sweden. The 30mm, I dropped!!!! Only a couple of feet but onto a hard wood floor. The 135 is acting weird. Exposure drops as you stop down. At f16 images are 3 stops darker than they should be, in any mode including manual. Can’t explain it but it’s on all my cameras. Maybe since the firmware upgrade?? Anyway, hopefully not too long and too expensive.
I’ve been doing a LOT of side by side shooting with my GFX100S. Apart from some colour difference and the fact both seem to report ISO differently, IQ is very similar. More lens dependant than anything else. Same with AF. Although the GFX has faster AF with more lenses than the X2D.
EVF, IBIS, handling and menus are noticeably better on the X2D. Flippy screen and a complete feature set on the GFX (ratios, focus modes etc). Basically I choose the GFX with zooms and the X2D with primes, most of the time. Simply because the GFX has them. If the X2D had more zooms I’m not sure I’d use the GFX anymore. It’s not that the X2D is better. It’s just more fun. It feels *special*.
