Can anyone see the difference between the X1 and D850 images?

[BobShaw]

Did you use Phocus ?

From my experience the color from the X1D is not the same without it.

For the test shoot I didn't while there, I just shot to the card and brought home the .3FR and then expanded them in Phocus.
I didn't export out of Phocus, I just imported the .FFF files into Aperture and did basic adjustments before exporting jpg to the customer.

For the real shoot I tethered to Phocus on MBP (with Phocus Mobile on an iPad for the customer) and saved the files to an external drive.
I am importing now as we speak so will let you know of any differences.

I think using Phocus in the process does make a real difference instead of putting .3FR into Aperture or Lightroom.

[ErikKaffehr]

Hi Bernard,

What is your take on color rendition? You are known to have both H6D100c and Nikon. Capture One user, too.

Best regards
Erik

My deepest apologies. My post was just about how appreciative I am about a discussion about DSKR vs “MF” rooted in cold facts instead of magical qualities. P1 was mentioned as one example but I could have written H4D-40 just as easily.

Cheers,
Bernard

[BernardLanguillier]

What magical or in the previous comment mythical qualities are you talking about?
Here's a myth. The Sensor on even the X1D is 1.6 times the sensor size on the 35mm camera. Oh, hang on, that's actually a fact.
What other technical advantage of 60% can you get by a simple kit change?

Also every comparison of 35mm vs MF seems to be between a new 35mm and a ten year old MF. Maybe that's because MF cameras actually last 10 years (and used to cost the price of a pretty good car).

I can't speak of the D850 but I own a 35mm 50MP camera and also owned a Hasselblad H4D with a 39MP back from and H3dII on it. Unless it was sport, travel or climbing mountains i would always prefer the images from the medium format. I just sold most of the H kit and bought the X1D for less money. Although I am using the H series lenses at the moment I see myself eventually moving to XCD and maybe Canon TSE. It lighter, easier and more fun to use the X1D than the 35mm DSLR.

Last week a did a test shoot for a high end bag company. They already had all of the images on the website but were unhappy with them. They loved the colour and richness of the X1D images and I got the job. If colour accuracy and reproduction is important then it will pay for itself. They also loved being able to monitor and rate the images on the iPad while I shot them.

If you think that the 35mm is better then keep using it, please.

Hi Bob,

The magical comment was hinting at past discussions about CCD backs’ amazing dynamic range.

I know the X1D image quality is a bit better.

Since I work with the H6D-100c I am aware about the excellent colors, be it with LR or Phocus. I also think that the X1D lenses are truly outstanding.

Cheers,
Bernard

[BernardLanguillier]

i ended up with the GFX after testing the X1D vs A7RII and honestly wasnt blown away....i did prefer the GFX files to the X1D but overall handling and AF made my mind up to go with the GFX...
i have since shot with the D850, A7RIII and GFX and compared files (i wont go into handling)
when files are pushed A LOT, the GFX holds up slightly better....both the D850 and A7RIII kind of fall apart after crazy adjustments, the GFX does not....this might be subjective as well but underexposed and opened up shadows on nikon/sony turn to color noise, on the GFX those areas turn to a more monochromatic mush....which i prefer.....
a vertical 8x10 portrait crop gives you more resolution from the 33x44 sensor....
it is very very close....very close.....

I think I saw you putting your GFX for sales recently?

I hope I am not being indiscrete, but would you mind sharing the reason why you have decided to sell?

Cheers,
Bernard

[ErikKaffehr]

Hi Bernard,

Medium format makers used to make a lot noise about MFD having superior dynamic range and 16 bits data. Nothing of that was really true, since 2008. Before 2008, it was a bit different as only Canon had full size DSLRs and their sensors were lacking in DR.

If we look back in time, most comparisons were done to Canons.

Going from 24x36 to 44x33 is a small improvement. It will yield the same image quality at 160ISO as 24x36 DSLR with a Sony CMOS at 100 ISO.

The present 44x33 sensor is something like 4-5 years old, so it is probably a bit behind best of breed technology today.

DPReview did a very well researched article about small medium format compared to 24x36: https://www.dpreview.com/opinion/2341704755/thinking-about-buying-medium-format-read-this-first

I would suggest that 44x33 mm has a small advantage over 24x36. What may make a difference is that it may be that the lenses for the X1D and the GFX are truly excellent, a lot of evidence for that.

Color rendition depends foremost on white balance, raw conversion profiles and not so much on sensor. Some photographers say that the GFX has DSLR color while the X1D has MFD color. I don't have the X1D or the GFX. But, I have a P45+, and that camera definitively has MFD colors.

So, I published three images that were carefully shot under same conditions and identically processed and published a poll, asking the readers to identify the P45+.

https://forum.luminous-landscape.com/index.php?topic=123905.0

Just a few readers guessed correctly.

The comparison is available as a stacked TIFF, here: http://echophoto.dnsalias.net/ekr/Articles/Temp/Tricolore/Stacked_with_samples.tif

The top layer is the actual color of the marked patches, measured with a spectrometer.

So, do I think that there is an advantage to small medium format? Yes, I think so, but I guess it is mostly coming from the good lenses and to some extent from better color processing.

It seems that Lightroom uses part of Hasselblad's processing pipeline with the X1D. With the GFX, Lightroom is applying software corrections automagically. I think it is a natural development that camera makers cooperate with software makers like Adobe to achieve the image quality the users are expecting.

With DSLRs, I think that vendors are a bit trapped into associating large aperture lenses with high quality designs. The result is that high quality lenses for DSLRs are heavy and large.

The lenses for the X1D are relatively small and don't have very fast apertures. But, they are probably very well corrected.

Best regards
Erik

Best regards
Erik

Hi Bob,

The magical comment was hinting at past discussions about CCD backs’ amazing dynamic range.

I know the X1D image quality is a bit better.

Since I work with the H6D-100c I am aware about the excellent colors, be it with LR or Phocus. I also think that the X1D lenses are truly outstanding.

Cheers,
Bernard

[BJL]

For “portrait” shapes, in fact anything from square through 5:4 to 4:3, the size ratio depends on the short edge, so 33/24 = 1.375. A bit more than Four Thirds vs “APS-C”, a bit less than from 24x16 “APS-C” to 36x24, a bit more than from 44x33 to the largest MF sensors.

I refer you to those previous format wars to judge if this is a significant difference. Note however that whereas in most cases, smaller formats offer higher resolution (lp/mm) through smaller photo sites, DMF has tended to stay close to 36x24 in pixel size, and the coming 100MP 44x33 sensors will continue that trend. So with good lenses, the gap in “detail” might be bigger than with those smaller format comparisons. But at a far greater cost increment and with 36x24 having a big edge in other aspects of camera technology.

[eronald]

Hi Bernard,

Medium format makers used to make a lot noise about MFD having superior dynamic range and 16 bits data. Nothing of that was really true, since 2008. Before 2008, it was a bit different as only Canon had full size DSLRs and their sensors were lacking in DR.

-------snip---------
Erik

Erik,

 In this thread I'm trying to get individuals with real-world experience to talk about their own perceptions of present day equipment.
 Of course most of us understand that the H1D is mainly different in form factor and not in technology, but as everybody in Sweden knows, form does enhance function.
 Many of us on the forums are aware of the historical context and the technical issues, and you certainly more than most.

Edmund

[pschefz]

I think I saw you putting your GFX for sales recently?

I hope I am not being indiscrete, but would you mind sharing the reason why you have decided to sell?

Cheers,
Bernard

no problem at all....
i will say it again, i love the platform, i love the files, i do see a step up from a7RIII and D850 but i shoot moving people and athletes and honestly the AF is not the issue.....it is just easier for me to shoot mostly with the sony because i have to put my subjects through fewer run throughs on shoots....and in the end none of that stuff gets fine art printed.....
i have said it here many times, there is a very small difference, but for most applications the nikon/sony are the better solution and the files are so close.....
from my own tests i think the fuji is the better platform, the lenses play a huge role in this.....
i also think the discussion about color is a little silly.....all these systems have incredible color.....maybe the new phase back has the best color ever, until the next thing comes out....but from my own experience: post processing software is very important...and i honestly don't think one is necessarily better, but one might get better results with one software and someone else with another......
i also know that most people still shoot canons (i would not touch) mostly because the skintones right out of the camera and on the rear screen make clients happy.....but for a lot of people that is not an issue.....

the question is this thread is a little misleading anyway....at 100%? at 400% in a 20x30 print? my answer: probably not, yes, probably not.....

the 100mpix 33x44 sensor will be a clear step up from todays sony/nikons but we know that they will have new cameras as well....

that hasselblad comparison out now pretty much shows all one needs to see anyway......

would also like to add that i went from A7RII to GFX and came to the A7RIII at a later point....
the GFX gave me a little extra IQ with all in all similar handling (compared to the A7RII) but sony improved the handling in the A7RIII to the point where it just makes more sense to me......the very small difference in IQ is less important then faster and easier shooting in a commercial environment.....

i am looking forward to the 100mpix 33x44 sensors and actually have pretty high hopes about the new upcoming nikon mirrorless system.....not because of what that first camera might be but what the larger mount might be able to handle in terms of sensor size.....i have never really been a fan of 24x36 and hope that we can leave it behind at some point.....

[David Watson]

Erik,

 In this thread I'm trying to get individuals with real-world experience to talk about their own perceptions of present day equipment.
 Of course most of us understand that the H1D is mainly different in form factor and not in technology, but as everybody in Sweden knows, form does enhance function.
 Many of us on the forums are aware of the historical context and the technical issues, and you certainly more than most.

Edmund

Well I am not at all sure where we are with this thread with so many posters expressing their preferences.  Like Bernard I have owned and used the H6D-100 and I have to say it is a very fine camera which, through Phocus, produces tremendous image quality.  I am not sure that it is as easy to use hand held as Bernard suggest but then I usually manage to spill my coffee.

For many years I did tests between DSLR's and various Hasselblad cameras to reassure myself that i was making a sensible, if expensive decision buying a Hasselblad MFD camera.  I am now in semi retirement and have down graded from the H6D and a Nikon D810 to the X1D thinking that one camera can satisfy my needs.  You know what? - it doesn't - it is IMO neither an easy going handheld camera or a tripod mounted studio or landscape machine - in my hands. 

When I compared the 50MP sensor on my previous H6D-50C against the D810 with a good lens I couldn't honestly see the practical difference in resolution up to quite large print sizes.  I like the way that LR renders the Nikon images and found that it was not that different to the Hasselblad look from a 50MP CMOS sensor.  100MP is clearly a different proposition but I would say that the D810 held up very well. 

So to try and answer, from my own perspective, Edmunds question I would say that for a working photographer there is no financial or technical advantage in selecting the X1D system in preference to a Nikon D810/850 system.  If there are minute resolution and qualitative differences they are more than drowned out by the flexibility of the D810/850.

But then that is just my opinion although one that is based on regular use of the D810, X1D and H6D.

[BAB]

I can see at first not owning either camera how someone could honestly ask the question of which is better, however I can not see how reviewers could possibly offer the comparisons we see today except to gain hits on their website. (ar3 vs x1d)


After shooting with the H6D-100c and having most H cameras before that, shooting Nikon, Leica and Canons for different reasons each has its own preference...and all PRO's know this. Since the first CCD larger format sensors were often flat and some photographers didn't like the PP to gain the WOW exposures they quickly jumped to other tools.


Still what are you shooting, in what light, where on a mountain, at the beach, on the street, in a studio, are you alone or with help etc. Answer these types of question and test drive some of your choices and you'll have your answer its that simple. No one will tell you what is best for you unfortunately you must take the chance we get with advise or test drive and find out for yourself.




Best experience can be had by either renting or better yet ask to observe another photographer using the same system your interested in purchasing that way you will see first hand pro and cons and could quickly decide for yourself.


Even after all that the camera has to pick you by that I mean so cameras and me just aint ever goin to get along.

[BobShaw]

Medium format makers used to make a lot noise about MFD having superior dynamic range and 16 bits data. Nothing of that was really true, since 2008. .....
...
Going from 24x36 to 44x33 is a small improvement. It will yield the same image quality at 160ISO as 24x36 DSLR with a Sony CMOS at 100 ISO.
....

Well the sensor is not 16 bit but it has been proved conclusively by others on this forum that the raw file has 16 bits of live data. The camera is not just the sensor. A modern camera is just a computer that happens to have a lens attached. The Hasselblad Natural Color Solution HNCS give excellent colour accuracy and has the reproduction mode calibrating in Phocus.

The sensor is 1.6 times or 60% bigger and I see that as a significant improvement. Assuming the light is the same I would be shooting the MF at ISO 100 not 160, so I would need an ISO setting on the Sony lower that it probably has. Even little things make a difference.

The real difference is when you print. You will need to magnify a 35mm sensor 1.6 times more than the MF sensor for the same size print and that is always true in maths.

[pschefz]

Well the sensor is not 16 bit but it has been proved conclusively by others on this forum that the raw file has 16 bits of live data. The camera is not just the sensor. A modern camera is just a computer that happens to have a lens attached. The Hasselblad Natural Color Solution HNCS give excellent colour accuracy and has the reproduction mode calibrating in Phocus.

The sensor is 1.6 times or 60% bigger and I see that as a significant improvement. Assuming the light is the same I would be shooting the MF at ISO 100 not 160, so I would need an ISO setting on the Sony lower that it probably has. Even little things make a difference.

The real difference is when you print. You will need to magnify a 35mm sensor 1.6 times more than the MF sensor for the same size print and that is always true in maths.

I'm sorry but are you saying that HNCS is better or you simply prefer the way hasselblad handles color? the first people will disagree with, the second is your opinion that you are obviously entitled to as are others...

phocus is not what i would call great software, if you prefer the color you get out of it, great bbut that is the only reason to use it....
it will be interesting to see what DJI will do about phocus....

i honestly dont know where the truth in maths comes in but i doubt a well exposed file printed at 20x24 from either sony, nikon fuji or hasselblad (or phase of course) will be BETTER in any way or anyone would be able to tell which is which.....i am sure everybody would have a favorite but that would not make one better then the other.....

the obvious difference comes in when comparing any of these 40-50mpix systems with the existing 100mpix systems....

[BobShaw]

I'm sorry

The maths is simple.
Ignoring aspect ratio ..
To make a 600mm x 500mm (approx 24x16 1/12ths of Henry VIIIs foot) print from a MF sensor of 44x33 requires magnification of 660/44 x 500/33 or 206 times.
To make a 600mm x 500mm print from a 35mm sensor of 36x24 requires magnification of 660/44 x 500/33 or 347 times. So every dot and blemish is more magnified.

[douglevy]

This is an interesting conversation, and something that comes up frequently. I've NEVER been able to pin a reason why my Credo60 files look so good (Capture 1? Lenses? Sensor?) I just know they look far better than my Nikons (D4/5/850) at base ISO and the answer is probably, "all of the above". I think most working pros I know who shoot any MF system (I'm in Boston and know only 2 others, one shoots H5Dc-50, one a IQ140/XF) but both have complete Canon system as well. In today's world there's no one perfect camera - I still shoot my D5/850 far more than the H, bc it's really only a ISO 50/100 system. As I'm sure Doug will say/has said - you really have to see how they fit for how you work to make any buying decision involving this level of $$$. My .02 though is that the 850 is SIGNIFICANTLY better than the 810 in terms of color rendering, so that would not be something I'd consider. (I sold my 810 6-months prior to my 850 purchase bc I hated the color so much. The d750 suffers the same issues).

-Doug

[ErikKaffehr]

Hi Bob,

This discussion tries to do some in depth analysis, based on raw images from DPReview's testing.

I would say a 20% advantage to 44x33 mm, based on that data.

Best regards
Erik

Well the sensor is not 16 bit but it has been proved conclusively by others on this forum that the raw file has 16 bits of live data. The camera is not just the sensor. A modern camera is just a computer that happens to have a lens attached. The Hasselblad Natural Color Solution HNCS give excellent colour accuracy and has the reproduction mode calibrating in Phocus.

The sensor is 1.6 times or 60% bigger and I see that as a significant improvement. Assuming the light is the same I would be shooting the MF at ISO 100 not 160, so I would need an ISO setting on the Sony lower that it probably has. Even little things make a difference.

The real difference is when you print. You will need to magnify a 35mm sensor 1.6 times more than the MF sensor for the same size print and that is always true in maths.

[Jim Kasson]

I was just looking at the dpreview tests and the two cameras are incredibly close. Of course if Hassy gets a new 100MP sensor they won't  be.

So I was wondering about real life - what do people here think - do they see the difference?

Is a comparison of the a7RII and the GFX 50S useful to you, Edmund?

https://blog.kasson.com/gfx-50s/fuji-1102-on-gfx-otus-841-4-on-a7rii/

https://blog.kasson.com/the-last-word/gfx-vs-a7rii-visibility-of-improved-iq/

Keep in mind that the microlenses on the GFX are different from the ones on the X1D.

Jim

[eronald]

Is a comparison of the a7RII and the GFX 50S useful to you, Edmund?

https://blog.kasson.com/gfx-50s/fuji-1102-on-gfx-otus-841-4-on-a7rii/

https://blog.kasson.com/the-last-word/gfx-vs-a7rii-visibility-of-improved-iq/

Keep in mind that the microlenses on the GFX are different from the ones on the X1D.

Jim

Jim,

 If I were feeling like spouting idiocy, as I am wont, I would note that one could try to measure sensor "MTF" directly in some way. I have seen papers on this. When it comes to commercial products, it's not at all obvious that a back illuminated sensor and a classical  CMOS sensor have the same or similar MTF, in practice, microlenses would -I guess- add more low-pass icing to the layer-cake but not necessarily enough to erase the structural disparities underneath.

 Now, regarding you tests, I am seriously surprised by the extent of some of the differences I see in your pictures which would be about 2 camera generations. If Doug could show the same effect with a 50C back on a landscape image, he would have a really convincing sales argument. But of course with all that greenery etc, I don't know what can be ascribed to focus etc. or the different lenses mounted on these two excellent cameras ...In any case this is practical evidence, not a lab test.

 But what are your real-life personal opinions?

Edmund

[Jim Kasson]

Jim,

 If I were feeling like spouting idiocy, as I am wont, I would note that one could try to measure sensor "MTF" directly in some way.

Kinda like this? Not very direct, though.

https://blog.kasson.com/the-last-word/discovering-the-fujifilm-gfx-50s-microlens-size/



Jim

[Jim Kasson]

 But what are your real-life personal opinions?

For prints greater than C-size, the GFX offers subtle advantages over the a7RII. However, I think the big advantage of the GFX is the native lens lineup, and for that reason, if the subject is moving slowly enough and I don't need great portability, I will use the GFX in preference to the a7RIII (I don't use the a7RII for anything but IR anymore) and to the a9.

I also prefer the ergonomics of the  GFX to the Sony cameras, but that's a relatively small thing to me.

Jim

[eronald]

Kinda like this? Not very direct, though.

https://blog.kasson.com/the-last-word/discovering-the-fujifilm-gfx-50s-microlens-size/



Jim

Exactly my feeling. Nice work but not direct. If I'm right he uses the equation sysmtf(nu)=lensmtf(nu)*sensormtf(nu), models lensmtf(nu), measures sysmtf(nu) with mtfmapper and thus derives  an evaluation of sensormtf(nu) conditioned by a large number of model assumptions.

I think one might find a more direct method, although maybe because of the packaging of the sensor one can't. I can't find the paper I once read, I think it was written by a guy in Toulouse in remote sensing who had access to the sensor fab.

There is one thing I do wonder about - I thought the Fuji had the Sony sensor, but if it is optimised ?!

Edmund