3.21.2017

Best Article I've Read on DP Review in Years. A debunking of some MF mythology. A nod to current full frame camera tech.

https://www.dpreview.com/opinion/2341704755/thinking-about-buying-a-fujifilm-gfx-50s-read-this-first

I'm so used to seeing advertorial writing on DPReview that I was a bit amazed to see this reasonably well written article that calls into question whether the investment(?) in a (smaller than) medium format camera, such as the new Fuji, is really going to deliver the things you might think are exclusive to the larger format.

It would be nice to see more writing like this and less gushing about sponsored Canon topics. Give it a read and see if you agree.


14 comments:

Rufus said...

I read it. A good read, well written.

But..

The article basically highlights the fact that the relative lack of appropriate lenses to really exploit the merits of the format is the issue. Whats more, it says this about a system which is 3 months old when being compared to a system which is decades old.

So while accurate, it seems a bit harsh to me. MF has its benefits but it is no panacea as you would know, given than 1' sensors can do a great job for the tasks you put them to.

Fuji launching MF is a wise move. Camera sales are plunging. This is Fuji's way of making fewer cameras for niche purposes and selling them for a lot more money. Makes sense to me. And sales of the camera suggests it will work for Fuji.

Kirk Tuck said...

Good point, Rufus. It remains to be seen whether or not the Fuji will sell. Hasselblad made MF cameras for years and rarely made fast lenses. Let's see if Fuji takes a different road.

Michael Matthews said...

I'm not competent to question the author's accuracy in explanation. The problem I have is that the samples (way clever, that test shot with moveable selection and changeable settings) seem to contradict the point being made. Each comparison, to my eye via iPad, appears to show the larger sensor providing a visibly better image.

tnargs said...

I think his points are well made and well demonstrated. Certainly I can see the FF beating the MF in the setups presented.

But (as an amusing aside) is the article saying that if I buy a 5DS *and* a D810 *and* an A7rII, *and* three sets of fast FF glass from 28 to 90mm, I can beat one Fuji MF? ;)

Anonymous said...

Yes, it is all about the lenses! Which is why Olympus was spot on when they introduced the 25mm F1.2 lens.

If you use that lens on an Olympus Pen F at ISO 80 the resulting image should be very similar to using the Fuji GFX at ISO 640 with the 63mm lens!

so you can get the look of medium format even on micro four thirds!

ODL Designs said...

Hey Kirk,
I chimed in on that post as I see this type of writing a lot on equipment...

SO he basically says the Sony 42mp sensor has similar noise... while showing a grey square. If you look a an area of fine detail you can see that not only is the fuji showing less chroma noise, but it is also showing more detail...

He then moves onto Nikon to say, and see the Nikon has as much DR... It might have a fraction less... but there is is... beaten for DR.

Then we move onto Canon and he talks about detail... and see he states, Canon has as much detail.

He then talks about lenses for a bit.

Then I ask myself... Okay, that was the most cherry picked set of comparisons I have ever seen.

Another way of looking at this is that the fuji has ALL of the best sensor characteristics of the three leading FF technologies.

It has more detail and less noise then the highest resolution (canon)
It has more detail and less noise than the second highest resolution (sony)
It has more detail and less resolution than the third highest resolution (nikon).

It beats or matches all three for DR

You lose a stop or so in aperture size in a few focal lengths.

His conclusion, you might not want to buy this because the lenses just are not there... but... It has just been released, and if Fuji's Xmount is anything to go by, they are not going to rest on their lens laurels but will push out to maintain an advantage with new lenses.

Of course, having said all that... I am not about to spend 10K on a body and 2 lenses :) I just dont like these rediculous cross mount comparisons as if FF is one brand or mount.

AndrewW said...

One other point, and one you have made Kirk, is that the larger format with a longer lens for the same field of view generally gives a faster ('better'?) falloff in the transition from focus to out of focus. So the 'equivalent' depth of field is not the full picture.

Wally said...

I read the article before you post. The only thing I took issue with dealt with fixation on having huge apertures f1.4 & f1.8 which would make medium format lenses both bigger more expensive and more expensive. All of this is true. What I see when looking at YouTube is how much of the fashion and commercial imaging industry is looking for more depth of field and will shoot at f8 and above not lower! Part of the argument is that full frame and those sports shooter f1.4 lenses are needed and not available. This smacks of the bright aperture sports journalism requirements and marketing hype camera companies push on the market.

Chris said...

Well, I essentially feel the same way about these MF cameras as the author of the article. You end up paying a lot more for an incomplete system, with no secondhand market, for dubious real world improvement. I think the trend, if anything,is towards smaller sensors, and these so-called MF cameras are not much larger than 35mm anyway. I predict they will remain specialist products, so they do not warrant the amount of coverage and apparent excitement by keen amateurs. If you need it then you know it, but as an aspiration for the keen amateur: I see them as a money sink for marginal benefit.

amolitor said...

I think his analysis of DoF relative to aperture is wrong. He's fixated on the idea that it scales with the crop factor, and I don't think it does. DoF drops off (at equivalent blah blah) more quickly as the format goes up than simply the crop factor.

At these numbers the difference isn't *that* great, but a basic mistake like that definitely creates a little bit of an odor about the rest of the piece.

But in general, I agree, and more generally there's just *never* been that much daylight between 645 and 35mm formats, and the Sony 50mp sensor is only half of a 645. So, yeah. Digital "MF" has always been a bit of a scam.

Maybe get the GFX if what you really want is Fuji handling, service, whatever, mated to a lot of of megapixels? That's a rationale that actually makes sense.

Anonymous said...

I'm a bit bemused by the sensor size obsession on photography forums. On the odd occasion that I've queried what has been holding back people from getting the results they're after with their existing kit there's been a consistent response that I'm missing the point. It seems that the theoretical limits of the equipment is more important than what it's used for.

Anyway, I welcome the fuji mf direction and hope it is successful. I'd love to use my bronica lenses on an affordable 6*6 sensor back.

In the meantime I'll stick with film.

And my olympus pen. Hopefully the sensor size turf war will migrate to ff vs mf and leave the m43 stuff in relative peace for a while.
Mark

Anonymous said...

Refreshing.

Anyone else had enough of the bought and paid for Fuji endorsees?

TMJ said...

The Fuji GFX and Hasselblad X1D strike me as both being 'aspirational' cameras. Rather as Leica cameras, (sadly), have become.

Fuji probably aren't too worried about sales as the parent company is massive and the photographic offshoot minor. Hasselblad probably are worried as it is only a camera company with some poor decision making over the last decade or so.

Ravi Bindra said...

Maybe just my limited understanding, but the article seems to be confused.

It is the size of the photocells that determines shot noise from photons (and having developed photomultipliers for single photon counting, shot noise is something I used to know about).

A photograph made at f4, does not care if the lens has a maximum aperture of f0.95 or 2.8. the light hitting the sensor is the same.