This article???

Foto64 · Feb 7, 2019

They never said they matched though in some cases they will but they said they are very close which as you can see in the images taken side by side they are very very close

Rick Knepper · Feb 8, 2019

I don't understand the title but...

Foto64 said:
They never said they matched though in some cases they will but they said they are very close which as you can see in the images taken side by side they are very very close

KEY TAKEAWAYS (according to the article)

GFX 50R offers greater IQ potential thanks to larger sensor and less resolution demand on lenses
The best full-frame sensors can offer very similar image quality, thanks to a similar number of pixels and lower base ISO allowing comparable levels of light capture

This last point regarding similar image quality? Seems like a contradiction of the first point. Take a look at the image below. The isolated area clearly shows a much higher level of resolution in the GFX file. Imagine this difference across the frame. This will make a difference on large paper prints, deep crops and viewing on large monitors from 18" away as many of us do.

I find it disingenuous that only under-exposed images were presented in the comparisons since the Z7's only strength seems to be ISO 64. I downloaded the RAWs and found the embedded settings to be curious although these may have been Adobe defaults. Hopefully, they were not the settings DPR used for the comparison.

--
Once you've done fifty, anything less is iffy.

GodSpeaks · Feb 8, 2019

moimoi said:
I am curious what your thoughts are on that recent article?

https://www.dpreview.com/reviews/fu...tm_medium=marquee&utm_campaign=traffic_source

I feel dpreview is pushing the readers to believe that FF output files can possibly match the GFX output files. Personally, I found this to be complete garbage.

I am lucky that I could afford one of the GFX system (GFX50R), and compared to my FF 5DIV system (which is a great camera/system), they are not within the same league at all (even with the best Canon lenses).

Anyway, I am intrigued...spell it out

Cheers,

Max

... who has owned and shot MF film AND digital I can concur that MFD and FF at about the same MP are very equivalent, regardless of what the MF priests say.

... and I have shot under good light studio conditions.

--
The greatest of mankind's criminals are those who delude themselves into thinking they have done 'the right thing.'
- Rayna Butler

moimoi · Feb 8, 2019

GodSpeaks said:
... who has owned and shot MF film AND digital I can concur that MFD and FF at about the same MP are very equivalent, regardless of what the MF priests say.

... and I have shot under good light studio conditions.

I would like to believe it, but I want to see real shots from FF camera. So, please show me a real shot (no studio shot please) that was taken with a 46-50 MP FF camera and with a 50 mm that can resolve details as good as this:

(100 % crop has been provided in the link below)

https://www.dpreview.com/forums/post/62278252

Cheers,

Max

--
http://www.maximesiegler.com/
https://www.instagram.com/maxsiegler645/
https://www.facebook.com/maxime.siegler.photography
And back to Analog Photography as well, oh well!
"All we have discovered is that it starts with a single individual - always a child - and then spreads explosively, like the formation of crystals round the first nucleus in a saturated solution." Arthur C. Clarke (Childhood's End, 1953)

Foto64 · Feb 8, 2019

It would be good to see more examples portraits and such but notice even in your own example they said similar. Similar is not exact. The point is they are so close that for many the extra cost and lack of lenses and AF performance is something people really need to think about. Why is that such a scary concept for GFX owners.

moimoi · Feb 8, 2019

Foto64 said:
It would be good to see more examples portraits and such but notice even in your own example they said similar. Similar is not exact. The point is they are so close that for many the extra cost and lack of lenses and AF performance is something people really need to think about. Why is that such a scary concept for GFX owners.

Not a scary concept. But that article was really silly, and it actually makes me question two things:

(i) how dpr tests camera?

(ii) what was the original intent from dpr with that article? Do they want to convince people buying more FF cameras?

I owned the GFX50R and I can tell you that the GF files are a step above everything I have seen from FF (and I am a FF user for quite some time).

--
http://www.maximesiegler.com/
https://www.instagram.com/maxsiegler645/
https://www.facebook.com/maxime.siegler.photography
And back to Analog Photography as well, oh well!
"All we have discovered is that it starts with a single individual - always a child - and then spreads explosively, like the formation of crystals round the first nucleus in a saturated solution." Arthur C. Clarke (Childhood's End, 1953)

Foto64 · Feb 8, 2019

There you go again nobody is saying that GFX image quality is not better than FF. The question is and has been shown in the test photos taken side by side. Is it good enough to warrant putting up with all the down sides.

Rick Knepper · Feb 8, 2019

Foto64 said:
It would be good to see more examples portraits and such but notice even in your own example they said similar. Similar is not exact. The point is they are so close that for many the extra cost and lack of lenses and AF performance is something people really need to think about. Why is that such a scary concept for GFX owners.

That difference is hardly similar to me.

Jared Willson · Feb 8, 2019

moimoi said:
I am curious what your thoughts are on that recent article?

https://www.dpreview.com/reviews/fu...tm_medium=marquee&utm_campaign=traffic_source

I feel dpreview is pushing the readers to believe that FF output files can possibly match the GFX output files. Personally, I found this to be complete garbage.

I am lucky that I could afford one of the GFX system (GFX50R), and compared to my FF 5DIV system (which is a great camera/system), they are not within the same league at all (even with the best Canon lenses).

Anyway, I am intrigued...spell it out

Cheers,

Max

No mystery or magic on this. A medium format camera is going to have some clear benefits in terms of image quality:

At a given megapixel count, you won't be pushing medium format lenses as hard, so the lens itself should out-resolve full frame as long as the lens is made to the same level of correction; that is, as long as the lens has the same Strehl in both systems for a given frequency of light, you will get more resolution on a larger format chip.
Assuming the same technology of CMOS chips, the larger chip at a given megapixel count will have an advantage in terms of full well capacity and thus signal to noise ratio and dynamic range

Since the chip in, for example, the Nikon Z7 is a generation or two newer than the chip in the GFX cameras, point number two is going to be mitigated. I don't have any concerns about that. Hopefully if you are buying into 44x33mm medium format, whether Fuji or Pentax or Hasselblad, you are expecting to periodically upgrade camera bodies while maintaining lenses from one camera generation to the next. At any given point in time, either system might have the more advanced/newer chip, thus widening or shrinking the performance gains from medium format.

As to my first point, it is clear from the level of moiré in the 44x33mm images I have seen and taken that the limiting factor on resolution is substantially the sensor, not the lenses. So for now, again, the differences between full frame and medium format are small, but with 100 megapixel chips that will no longer be true. Of course, whether or not people see benefits from 100 megapixel chips in their final product is a whole different discussion.

So, are full frame and 44x33mm medium format very similar in terms of technical image quality? Absolutely. Will they remain that close? A lot of why they differences are so small right now is just the generation and megapixel count of the chips being used. Will the next generation of medium format be substantially better than the current generation of full frame? Almost certainly. Will that matter in terms of end results? Each photographer will have to answer that question for himself/herself.

- Jared

Rick Knepper · Feb 8, 2019

Foto64 said:
There you go again nobody is saying that GFX image quality is not better than FF. The question is and has been shown in the test photos taken side by side. Is it good enough to warrant putting up with all the down sides.

I am trying to think of a downside for my usage. Help me out.

Richard Feynman · Feb 8, 2019

Rick Knepper said:
Foto64 said:

It would be good to see more examples portraits and such but notice even in your own example they said similar. Similar is not exact. The point is they are so close that for many the extra cost and lack of lenses and AF performance is something people really need to think about. Why is that such a scary concept for GFX owners.

Click to expand...

That difference is hardly similar to me.

Doesn't that simple sentence pretty much sum up the entire discussion? The difference is real, whether it's significant depends on the person.

At the same time we have this other discussion taking place, which is can the differences be seen in down sampled images? I think they probably cannot, at least between FF and MF, though I'm not sure. That assumes best case scenario for each.

KingTing · Feb 8, 2019

NAwlins Contrarian said:
NAwlins Contrarian said:

I am curious what your thoughts are on that recent article?

https://www.dpreview.com/reviews/fu...tm_medium=marquee&utm_campaign=traffic_source

I feel dpreview is pushing the readers to believe that FF output files can possibly match the GFX output files. Personally, I found this to be complete garbage.

I am lucky that I could afford one of the GFX system (GFX50R), and compared to my FF 5DIV system (which is a great camera/system), they are not within the same league at all (even with the best Canon lenses).

Click to expand...

Some photographers (like audiophiles and others) have a streak of believing in magic hardware. IMO we ought to try to be accurate and honest about the areas where the GFX has real image-quality advantages--and the areas where it does not. What we think you see or hear is not too reliable, because we're biased; the relevant question is what we could tell apart in double-blind testing. To be clear, what follows is not about many other things that affect overall quality of results (e.g., focus capability and accuracy, ergonomics, etc.).

The GFX appears to benefit from very good lenses, and gains sharpness at the expense of artifacts with its gapped sensor microlenses.

The GFX also has a modest advantage in noise at a given gain--but that is basically offset by the system's on the whole slower-maximum-aperture lenses and need to stop down more to get the same depth of field.

In terms of maximum dynamic range,* the GFX only has a very small / minimal advantage over the best current FF cameras (Nikon D850 and Z7 and Sony A7R III). Why? The GFX sensor has only about 1.67x the area of FF; all else being equal, that would mean about a three-quarter-stop (0.74 EV) advantage. But all else is not equal. The best current FF cameras have BSI sensors whose basic architecture is about three years newer than the GFX's sensor's basic architecture. And for whatever the merits of the 5D Mk. IV, in terms of dynamic range, it is not in the same league as the best current FF cameras. Says who? DxO has not tested a GFX or a Z7, but using the similarly-sensored Pentax 645Z as a proxy for a GFX** and comparing it to a D850, a 5D Mk. IV, and an A7R III, the D850, 645Z, and A7R III are essentially equal in maximum dynamic range, and all three are a little over 1 stop (1.11 to 1.22 EV) better than the 5D Mk. IV, after equalizing for the same output size ("Print" not "Screen'):

Nikon D850 14.81 EV

Pentax 645Z 14.73 EV

Sony A7R Mk. III 14.70 EV

Canon 5D Mk. IV 13.59 EV

See https://www.dxomark.com/Cameras/Com...5D-Mark-IV-versus-Pentax-645Z___1177_1106_951 and https://www.dxomark.com/Cameras/Sony/A7R-III---Measurements.

Don't believe DxO? Photos to Photos reports about the same about maximum dynamic range spread (albeit using a different measure of dynamic range):

Fuji GFX-50S 11.90 EV

Fuji GFX-50R 11.86 EV

Sony A7R Mk. III 11.65 EV

Nikon D850 11.63 EV

Nikon Z7 11.56 EV

Canon 5D Mk. IV 10.83 EV

See http://photonstophotos.net/Charts/PDR.htm. By those measure, the two GFX models have essentially identical maximum dynamic range, but only about 1/4-stop (0.21 to 0.44 EV) better than the Nikon D850, Sony A7R Mk. III, and Nikon Z7, and the Canon 5D Mk. IV is just over 1 stop (1.03 to 1.07 EV) behind the Fujis and just over 3/4 stop (0.73 to 0.82 EV) behind the Nikons and Sony.

*By maximum dynamic range, I mean dynamic range at the setting that gives the most of it, i.e., the setting you use where there's no issue of needing to reduce the shutter speed.

**This is actually giving a boost to the GFX because its sensor's gapped microlenses cost it a little in noise and dynamic range. Otherwise, both use Sony 51 MP CMOS sensors of the same size and basic architecture.

Honestly I don't see why people are too surprised at at the findings, the real difference for FF vs MF is when we compare the new GFX100 against the recent FFs.

MF will always remain a specialised tool and let's see if GFX100 can make the gap wider with the new sensor and better AF, and not the obsession with pixel count or 1:1 zoom into a hair detail difference

laukkart · Feb 8, 2019

Richard Feynman said:
At the same time we have this other discussion taking place, which is can the differences be seen in down sampled images? I think they probably cannot, at least between FF and MF, though I'm not sure. That assumes best case scenario for each.

The answer is yes, you can see.

Lets look into the studio scene comparison tool, Z7 vs. GFX and add A6500 vs. A7iii for comparison, using the Print as image size, which is a 8MP downscale.

It is rather clear how Z7 does not manage to render those thread tones in shadow areas (if you cannot see it, change the ISO to 3200 and then 1600):

Loading…

www.dpreview.com

And here you can observe the softness on all smaller sensors vs. GFX:

Loading…

www.dpreview.com

And here the label has smoother color transition into the shadow on GFX and A7iii - more clear at ISO3200 (one can turn to the ISO 12800 to observe on this label the advances the dual gain has brought for low light, as A7iii does best job then):

Loading…

www.dpreview.com

And here the coins are rather acceptable only on GFX:

Loading…

www.dpreview.com

As we may observe, the GFX manages to produce the smoothness of the tones of A7iii, while exceeding the details of Z7. Where as Z7 lacks in tones, A7iii lacks in detail. That is the "power" of higher resolution and bigger pixel pitch, even when downscaled.

We can also observe that if we shoot ISO100 - ISO800, downscale to 8MP and do not postprocess, the A6500 will do just fine. So we can conclude that both MF and FF are waste of money and APS-C is the way to go - if we'd go by the hegemony of these forums. Lucky me that I own the best gear possible - a 24mp APS-C. /sarcasm

As a side note, it will be interesting to see how GFX100 will do in these, as its pixel pitch will be significantly smaller than on GFX50S. The "oversampling" of added resolution and dual gain will help, but how much?

Erik Kaffehr · Feb 8, 2019

Hi,

Doing comparisons at high ISOs is pretty much nonsense, IMHO.

Why do you shoot high ISOs:

To limit the effects of camera shake.
Limit subject motion

Let's say you are shooting with the 110/2 on the GFX. On the Sony A7rII I have I could use an 85/1.4. So, I could use ISO 1600 instead of ISO 3200.

Taking in body image stabilisation into account I could shoot at say 400 ISO.

Very clearly, the GFX or any other 44x33 body using a Sony sensor would have a small or significant advantage over 24x36 using Sony sensors at base ISO. But once you are not shooting base ISO on tripod, maximum apertures and image stabilisation also come into play.

Now, I am the on tripod at base ISO kind of guy, those who would benefit most from the GFX.

But, shooting in handheld conditions, in the real world, that may be a different thing.

Best regards

Erik

--
Erik Kaffehr
Website: http://echophoto.dnsalias.net
Magic uses to disappear in controlled experiments…
Gallery: http://echophoto.smugmug.com
Articles: http://echophoto.dnsalias.net/ekr/index.php/photoarticles

laukkart · Feb 8, 2019

Erik Kaffehr said:
Hi,

Doing comparisons at high ISOs is pretty much nonsense, IMHO.

Why do you shoot high ISOs:

To limit the effects of camera shake.

Limit subject motion

Let's say you are shooting with the 110/2 on the GFX. On the Sony A7rII I have I could use an 85/1.4. So, I could use ISO 1600 instead of ISO 3200.

Taking in body image stabilisation into account I could shoot at say 400 ISO.

Very clearly, the GFX or any other 44x33 body using a Sony sensor would have a small or significant advantage over 24x36 using Sony sensors at base ISO. But once you are not shooting base ISO on tripod, maximum apertures and image stabilisation also come into play.

Now, I am the on tripod at base ISO kind of guy, those who would benefit most from the GFX.

But, shooting in handheld conditions, in the real world, that may be a different thing.

Best regards

Erik

If you read my earlier posts, you may notice that I talk about postprocessing headroom, which is commonly neglegted in these debates, although it is highly important thing for majority of photographers.

High ISO samples are good way to study and demonstrate that without actually spending time on postprocessing. When you push and edit your files, you are often generating the same situation onto shadow areas as you would do in shooting higher ISO.

For example, for on location portrait shooter, the way colors and detail behave in shadow areas during postprocessing is highly interesting and important factor.

Thus, to me watching a SOOC settings baked JPEG shot at base ISO is rather nonsense. It does not reflect the reality.

One can also use many f/1.4 lenses with GFX successfully. But I rather eliminate the lens question, because then we would have to consider things like across the frame sharpenss, LoCA, each individual use case demands for DOF or lack of it, and so on.

Richard Feynman · Feb 8, 2019

laukkart said:
Richard Feynman said:

At the same time we have this other discussion taking place, which is can the differences be seen in down sampled images? I think they probably cannot, at least between FF and MF, though I'm not sure. That assumes best case scenario for each.

Click to expand...

The answer is yes, you can see.

Lets look into the studio scene comparison tool, Z7 vs. GFX and add A6500 vs. A7iii for comparison, using the Print as image size, which is a 8MP downscale.

It is rather clear how Z7 does not manage to render those thread tones in shadow areas (if you cannot see it, change the ISO to 3200 and then 1600):

https://www.dpreview.com/reviews/im...1&x=-0.31643518518518504&y=0.5467703065606303

And here you can observe the softness on all smaller sensors vs. GFX:

https://www.dpreview.com/reviews/im...1&x=0.3281658766901824&y=-0.10912793068475737

And here the label has smoother color transition into the shadow on GFX and A7iii - more clear at ISO3200 (one can turn to the ISO 12800 to observe on this label the advances the dual gain has brought for low light, as A7iii does best job then):

https://www.dpreview.com/reviews/im...=1&x=0.1135095164609056&y=-1.0812224141239686

And here the coins are rather acceptable only on GFX:

https://www.dpreview.com/reviews/im...&x=-0.19272486772486755&y=-0.6372415365185705

As we may observe, the GFX manages to produce the smoothness of the tones of A7iii, while exceeding the details of Z7. Where as Z7 lacks in tones, A7iii lacks in detail. That is the "power" of higher resolution and bigger pixel pitch, even when downscaled.

We can also observe that if we shoot ISO100 - ISO800, downscale to 8MP and do not postprocess, the A6500 will do just fine. So we can conclude that both MF and FF are waste of money and APS-C is the way to go - if we'd go by the hegemony of these forums. Lucky me that I own the best gear possible - a 24mp APS-C. /sarcasm

As a side note, it will be interesting to see how GFX100 will do in these, as its pixel pitch will be significantly smaller than on GFX50S. The "oversampling" of added resolution and dual gain will help, but how much?

Thank you for your input.

With regard to your final paragraph, there are some comparisons out there between FF and MF that include the 100 PM Phase One. The differences between the Phase One and the other FF/MF cameras dwarfed the differences between the Fuji MF and Nikon/Sony/Canon FF. So my guess would be the answer is "a lot". At that point the debate will turn to "who needs that resolution anyhow unless you want to print billboards"

.

JimKasson · Feb 8, 2019

laukkart said:
If you read my earlier posts, you may notice that I talk about postprocessing headroom, which is commonly neglegted in these debates, although it is highly important thing for majority of photographers.

High ISO samples are good way to study and demonstrate that without actually spending time on postprocessing.

In the GFX 50S and GFX 50R, high ISO brightness increases are not performed in-camera, but in post-production.

Jim

Erik Kaffehr · Feb 8, 2019

laukkart said:
Erik Kaffehr said:

Hi,

Doing comparisons at high ISOs is pretty much nonsense, IMHO.

Why do you shoot high ISOs:

To limit the effects of camera shake.

Limit subject motion

Let's say you are shooting with the 110/2 on the GFX. On the Sony A7rII I have I could use an 85/1.4. So, I could use ISO 1600 instead of ISO 3200.

Taking in body image stabilisation into account I could shoot at say 400 ISO.

Very clearly, the GFX or any other 44x33 body using a Sony sensor would have a small or significant advantage over 24x36 using Sony sensors at base ISO. But once you are not shooting base ISO on tripod, maximum apertures and image stabilisation also come into play.

Now, I am the on tripod at base ISO kind of guy, those who would benefit most from the GFX.

But, shooting in handheld conditions, in the real world, that may be a different thing.

Best regards

Erik

Click to expand...

If you read my earlier posts, you may notice that I talk about postprocessing headroom, which is commonly neglegted in these debates, although it is highly important thing for majority of photographers.

High ISO samples are good way to study and demonstrate that without actually spending time on postprocessing. When you push and edit your files, you are often generating the same situation onto shadow areas as you would do in shooting higher ISO.

For example, for on location portrait shooter, the way colors and detail behave in shadow areas during postprocessing is highly interesting and important factor.

Hi,

The problem with your approach is that ignores reality.

Just as an example, Sony and Nikon cameras switch analogue gain between 400 ISO and 800 ISO.

So, shooting on any of those systems at 800 ISO will give different results than shooting at 400 ISO.

As a matter of fact, DPReview offers a lot of raw images underexposed compared to nominal, but they can be hard to find.

Almost any image at DPReview, you can download in raw and do any evaluation you want.

So, stop the ranting and do some proper work...

Best regards

Erik

laukkart · Feb 8, 2019

JimKasson said:
laukkart said:

If you read my earlier posts, you may notice that I talk about postprocessing headroom, which is commonly neglegted in these debates, although it is highly important thing for majority of photographers.

High ISO samples are good way to study and demonstrate that without actually spending time on postprocessing.

Click to expand...

In the GFX 50S and GFX 50R, high ISO brightness increases are not performed in-camera, but in post-production.

Jim

Yes, I am aware of that and I mentioned it in my earlier comment.

But I do not see that how it would change the comparison, since I concentrated on the shadow areas. To my experience, pushing lower gain file vs. lifting the actual gain on same camera usually results to quite close to the same quality in the shadow areas.

Greg7579 · Feb 8, 2019

Don't worry Max. Photographers who actually shoot w the GFX know that this notion is total BS.

This article???

Forum Enthusiast

Forum Pro

Forum Pro

Veteran Member

Forum Enthusiast

Veteran Member

Forum Enthusiast

Forum Pro

Senior Member

Forum Pro

Leading Member

Forum Enthusiast

Member

Veteran Member

Member

Leading Member

Community Leader

Veteran Member

Member

Forum Pro

Similar threads

Keyboard shortcuts