Fuji GF 30mm Tilt Shift and Rodenstock 32mm HR Digaron-W Lens Comparison

[Paul Spinnler]

I think I was unclear, there exists for example Contax 645 to GFX AF adapters and Canon EF to GFX adapters that can autofocus.

Separately, I did reverse engineer the Mamiya/Phase lens protocol, which is just an extension of the same SPI bus used to interface Phase backs. I never did anything with the project though, and autofocus would be a hole thing since they are screw drives lenses. I can do the CSE/EE, just the mechanical engineering is a bit beyond me for something like that.

Alpa with their FPS has a pretty similar setup to the Pico it seems, though I prefer the alpa setup with their swappable lens modules. (Also, why did they stop making the Rollei 6000 adapter? Was it because DW photo is the worst to deal with? )

The FPS is great, but it was conceived in the era before mirrorless and the key mount to include nowadays (FFD being the problem so far with 3rd party cameras) would be the GFX one. There are so many adapters out there for Fuji so if you build a bridge to Fuji you open the door to many other things. The FPS construction might not allow for GFX, but I am not sure.

For example, I saw that there are I think 0.8x reduction adapters which increase speed from Pentax 6x7 to GFX – imagine how cool that is to use the 105 Takumar 2.4 focal reduced on an IQ4.

So if the FFD is 26.7 for Fuji it mens there is only 4.5mm to integrate the GFX mount, probably not enough for a smart adapter and not easy for a dumb adapter either.

That would literally be a dream if they made an aperture control adapter which can accept Fuji G.

The big benefit of the bellows type design is that a) you can mount camera bodies with handgrips while still being able to do X/Y shifting and b) you can get closer FFD wise as you work with two carriers on a rail instead of a body frame in-between.

That's the key to the mirrorless world and if they build it in a smart way one could even adapte Nikon Z – I am talking about the Nikkor 58 0.95 ... which is sharp wide open!

This said, the Pico with its own shutter and carrier / bellows system will be a dream tool to adapt any lens onto anything. A real tool for 2024 and beyond ... its basically FPS reloaded in a typical Arca kind of way – engineering-wise built in the smartest way, but as a classic bellows cam. Reminds me of the genius in the R system which mounted the helical into the body, thereby lowering lens costs for everyone by making tilt a body feature and by moving the helical out of the lens onto the body as well.

Here they take it one step further – everything except the optics is moved onto the body which means any lens can be adapted if you dont have an in-between body frame. You don't even need a tube for copal lenses as you have rails and move standards apart as much as you like ... genius.

 

[Paul Spinnler]

In case it is in future useful:

View attachment 209782

Refer also to this thread for context, from which the above image is sourced, especially @f8orbust and @GrahamWelland's discussion on page 2.

Thanks so much, that's really good info. 22.2mm – noted!

 

[diggles]

I'd be very curious to see how a P1 BE35 stacks up against the TS30.

Around a year and a half ago, I came across a post where a photographer adapted the P1 BR35mm lens to use it on his Sony camera. I thought if it could be adapted for Sony, it might also work with my GFX camera. The adapter I used was the Fotodiox Mamiya 645 to GFX pro tilt-shift.

Similar to the Canon TS lenses, the aperture on the BR lenses can be stuck by setting the aperture and disengaging the lens. To do this, I purchased an old P1 camera on eBay for $500.

Here's what I discovered: When you stay within a 10mm shift range, which covers roughly a 70mm image circle, the lens performs exceptionally well, and Capture One's distortion corrections work effectively. However, when you shift more than 10mm, the resulting image exhibits complex distortion that cannot be adequately corrected using Capture One's distortion correction tools.

I believe the reason the C1 distortion corrections didn't work well beyond the 70mm image circle is because the profile was not originally designed to correct for shifting. This is understandable because the lens itself was not designed for shifting.

 

[Paul Spinnler]

That's very interesting, thanks – you are right, on the XF it is always fixed ... how much further could you go when shifting before hitting the IC? 15, 20mm? Or just 10?

 

[guphotography]

If 30mm ts can be adapted to Arca/Alpa, it would seriously make 44x33 BIS sensor a strong contender against IQ4, especially from value proposition.

I thought I was set hard on IQ4 if I were ever move on from IQ3 tri, but it may no longer be the case.

 

[Paul Spinnler]

I think we should not forget that P1 most likely has also an upgrade cycle coming. I would be very surprised to not see a highly desirable IQ5 given that they've underscored commitment to photography by essentially consistently releasing new stuff.

We will have a 90 tilt announcement, a secret product – most likely a large XT, to bridge 2024 and then in 2025 time could be ripe for a true successor to the IQ4.

Its just sth to bear in mind – all of a sudden they have this analogue colour science 250 megapixel high DR back with 3x battery live, EVF port, etc. and you're like sad to have sold all the top end stuff for some TS lenses in smaller formats.

It seems to me most likely that we'll see an IQ5 and when it comes it will be interesting to see what Phase has been cooking in the meantime.

I wouldn't be surprised to see another step up in DR, which to me would be the absolute game changer.

The case of Arri is instructive here – their Alexa sensor tech from the basic DR side has been the same for a decade; it was just so good that technology plateaued for ten years; now that they've brought out the Alexa 35 they significantly raised the bar again for years to come. It literally beats color neg film in terms of DR – very difficult to get it to clip under normal shooting conditions ... so the IQ4 was and is just so good that it now just takes time to step it up. But that a step up will come, I wouldn't doubt.

45k list gives them a lot of leeway to put in high end components other manufacturers just can't even with economies of scale. So interesting times ahead.

Wouldn't write off the King yet, lol.

 

[marc aurel]

I think we should not forget that P1 most likely has also an upgrade cycle coming. I would be very surprised to not see a highly desirable IQ5 given that they've underscored commitment to photography by essentially consistently releasing new stuff.

We will have a 90 tilt announcement, a secret product – most likely a large XT, to bridge 2024 and then in 2025 time could be ripe for a true successor to the IQ4.

Its just sth to bear in mind – all of a sudden they have this analogue colour science 250 megapixel high DR back with 3x battery live, EVF port, etc. and you're like sad to have sold all the top end stuff for some TS lenses in smaller formats.

It seems to me most likely that we'll see an IQ5 and when it comes it will be interesting to see what Phase has been cooking in the meantime.

I wouldn't be surprised to see another step up in DR, which to me would be the absolute game changer.

Just wouldn't write off the King yet, lol.

Well, if the GF 30mm TS could be adapted to an Arca / Alpa camera – it would not be necessary to decide between both worlds. In an ideal future we will be able to adapt and use it on a 54x40mm sensor (be it one of the current ones or a possible one in the future). The Digaron 32mm has an image circle 90mm. The image circle of the Fuji lens is at least 85mm. That is what Fuji claims and what we can see when used on a 44x33mm sensor. We don't know if it even has a larger image circle that would be only visible on a view camera.

 

[Paul Spinnler]

You can already adapt GFX lenses to an IQ4:

Camera Body for Phase One IQ3 100, IQ4 Digital Back & Any MF Lens w/GF Adapter | eBay

Instead, it relies on live view mode and the electronic shutter of the IQ3 100(Needed Firmware update #3) or IQ4 series digital back for capturing images. If any digital backs that supports live view and electric shutter should also function well with this body.

www.ebay.com

This should actually do the trick as you can shift in lens, right? If you can fix the aperture by mounting it first on a Fuji body to say F8-11 you are already also there apeture wist – its not that you constantly need to shift between apertures in architectural photography ...

Only problem are then the distortion aspects where you'd need to manually type in the rise in C1. You can nudge the values up and down until it "fits"?

Mhh – thinking this through. The IQ4 will have a bigger coverage of the IC and since C1's distortion correction tool does not allow for a sensor input you get into problems when shooting the TS30 on an IQ4 and need to correct distortion.

Looks like you're stuck with Fuji once you cross the TS30 bridge.

 

[diggles]

Looks like you're stuck with Fuji once you cross the TS30 bridge.

Well, in this case, it's not just a saying – it really is greener on the other side!

 

[Doppler9000]

I wouldn't be surprised to see another step up in DR, which to me would be the absolute game changer.

I believe that we are pretty much flat-lined on maximum dynamic range per unit area.

Jack Hogan and Bill Claff have expressed skepticism about the ARRI claims.

 

[Doppler9000]

You can already adapt GFX lenses to an IQ4:

Camera Body for Phase One IQ3 100, IQ4 Digital Back & Any MF Lens w/GF Adapter | eBay

Instead, it relies on live view mode and the electronic shutter of the IQ3 100(Needed Firmware update #3) or IQ4 series digital back for capturing images. If any digital backs that supports live view and electric shutter should also function well with this body.

www.ebay.com

This should actually do the trick as you can shift in lens, right? If you can fix the aperture by mounting it first on a Fuji body to say F8-11 you are already also there apeture wist – its not that you constantly need to shift between apertures in architectural photography ...

Only problem are then the distortion aspects where you'd need to manually type in the rise in C1. You can nudge the values up and down until it "fits"?

Mhh – thinking this through. The IQ4 will have a bigger coverage of the IC and since C1's distortion correction tool does not allow for a sensor input you get into problems when shooting the TS30 on an IQ4 and need to correct distortion.

Looks like you're stuck with Fuji once you cross the TS30 bridge.

The Fuji GFX lenses are focus-by-wire, so I think they are a non-starter with this type of adapter. I believe the adapter serves more of a “universal” intermediate function for a second mirrorless adapter, though manual GFX-mount lenses would also work. Mirex used the Canon EF mount in a similar way.

Maybe Techart, Fringer, Metabones, Fotodiox will make a smart adapter.

 

[Paul Spinnler]

I shot with the IQ4 and Portra 400 side by side a lot and there is still a quite evident gap between film and digital in the highlights.

I have no idea who Mr Hogan is or what the purpose of introducing a random doubt statement of his is, but the dynamic range gap from Arri is evident in all reviews of the camera and you also see it in the footage. It is very nice how it captures the gamut of any scene.

It is a significant advancement and the camera is actually difficult to clip short of blasting it and having a wide open shutter - the IQ4 is also great, but there is still room there.

Let’s see what they bring to the party ….

 

[marc aurel]

The Fuji GFX lenses are focus-by-wire, so I think they are a non-starter with this type of adapter. I believe the adapter serves more of a “universal” intermediate function for a second mirrorless adapter, though manual GFX-mount lenses would also work. Mirex used the Canon EF mount in a similar way.

Maybe Techart, Fringer, Metabones, Fotodiox will make a smart adapter.

The GF TS lenses are an exception in Fujis lineup. They are pure manual focus, not focus by wire.
But aperture is controlled electronically. So you would be stuck with a preset aperture. At least until someone releases a smart adapter.

 

[Doppler9000]

I have no idea who Mr Hogan is or what the purpose of introducing a random doubt statement of his is, but the dynamic range gap from Arri is evident in all reviews of the camera and you also see it in the footage. It is very nice how it captures the gamut of any scene.

It is a significant advancement and the camera is actually difficult to clip short of blasting it and having a wide open shutter - the IQ4 is also great, but there is still room there.

Let’s see what they bring to the party ….

First, I want to be clear that I am not making a multi-quote personal attack as some others seem to enjoy - I am interested in informing the discussion.

Bill Claff runs the Photons to Photos site - an excellent dynamic range resource. Jack Hogan is a scientist with a deep background in the relevant issues here.

We are at the practical limits of what the current materials will allow with regard to dynamic range. If someone is claiming substantially higher dynamic range results, then one ought to look more carefully at the underlying data, etc.

 

[Doppler9000]

The GF TS lenses are an exception in Fujis lineup. They are pure manual focus, not focus by wire.
But aperture is controlled electronically. So you would be stuck with a preset aperture. At least until someone releases a smart adapter.

Thanks for clarifying, Marc.

 

[diggles]

That's very interesting, thanks – you are right, on the XF it is always fixed ... how much further could you go when shifting before hitting the IC? 15, 20mm? Or just 10?

It's been a couple of years, but I believe it was about 15mm of shift before vignetting.

 

[jng]

First, I want to be clear that I am not making a multi-quote personal attack as some others seem to enjoy - I am interested in informing the discussion.

I, for one, appreciate seeing the citations. Proper attribution is de rigueur in the scientific (and other scholarly) literature, and for good reason. It not only gives credit where credit is due, but also provides the reader with information and context that allow them to make critical judgments about the veracity of claims being made and delve deeper if they so desire. Of course, framing and context are important to maintain productive and civil discourse, which (still) sets GetDPI apart from other forums. Just sayin'...

John

 

[Paul Spinnler]

First, I want to be clear that I am not making a multi-quote personal attack as some others seem to enjoy - I am interested in informing the discussion.

Bill Claff runs the Photons to Photos site - an excellent dynamic range resource. Jack Hogan is a scientist with a deep background in the relevant issues here.

We are at the practical limits of what the current materials will allow with regard to dynamic range. If someone is claiming substantially higher dynamic range results, then one ought to look more carefully at the underlying data, etc.

On the citation and multi quote point.

The problem is this: one can take apart any quote by doing desktop research always; for example, one can google whether we are at the practical limits of dynamic range or even use GPT 4.5 and the reply and add some citations, etc. and then, based on one’s goals break it down to a full long form argument. That in turn can be replied to in a similar manner. The internet is so vast and truth is so elusive that if you sit down on a computer and work an answer it can become this endless discussion.

The other person is just a cross forum troll with too much time and I have refrained from taking apart every statement because in the end it is a waste of time and I am busy with multiple projects so I often time do not have the energy and time to come up with a perfect, scientifically backed up post or the willingness to spend time debunking or making claims … one needs to be able to discuss without having to resort to a scientific citing.

Re dynamic range - I come from personal experience and I have scanned and worked with film extensively and as well with the IQ4; so when I sometimes even do side by side with the same lens and aperture I can clearly see that the film completely covers the gamut of the scene in the highlights while the IQ4 has blown out areas. So that’s what I see and where I think the back has not yet reached the same quality as the best color neg film.

So to me it is now pointless to start a theoretical long form discussion of how spurious Arris claims are based on datapoints - I can see the IQ4 doesn’t reach Portra level highlight retention.

This said, clearly we should be able to back up what we say and not just say things, but even if you find an inconsistency it can stem from the fact that people meant the right thing, but from memory for example didn’t represent sth correctly for a lack of time - I often just write on my IPhone for example when travelling.

So look, I am very curious what the next IQ5 back will bring to the table, maybe it’ll be something in the DR department, maybe not; I feel that there is still room, that’s all …

 

[Doppler9000]

On the citation and multi quote point.

The problem is this: one can take apart any quote by doing desktop research always; for example, one can google whether we are at the practical limits of dynamic range or even use GPT 4.5 and the reply and add some citations, etc. and then, based on one’s goals break it down to a full long form argument. That in turn can be replied to in a similar manner. The internet is so vast and truth is so elusive that if you sit down on a computer and work an answer it can become this endless discussion.

Re dynamic range - I come from personal experience and I have scanned and worked with film extensively and as well with the IQ4; so when I sometimes even do side by side with the same lens and aperture I can clearly see that the film completely covers the gamut of the scene in the highlights while the IQ4 has blown out areas. So that’s what I see and where I think the back has not yet reached the same quality as the best color neg film.

So to me it is now pointless to start a theoretical long form discussion of how spurious Arris claims are based on datapoints - I can see the IQ4 doesn’t reach Portra level highlight retention.

This said, clearly we should be able to back up what we say and not just say things, but even if you find an inconsistency it can stem from the fact that people meant the right thing, but from memory for example didn’t represent sth correctly for a lack of time - I often just write on my IPhone for example when travelling.

So look, I am very curious what the next IQ5 back will bring to the table, maybe it’ll be something in the DR department, maybe not; I feel that there is still room, that’s all …

I will assure you that I did not go and google dynamic range in order to find citations to attempt to make you look bad, I was simply pointing out the realities of where we are, with regard to the topic. I did not and do not wish to offend you personally.

I made no reference to a theoretical, long-form discussion on the claims made by ARRI, using datapoints - I simply referenced the views of two people who know a lot more about the subject of dynamic range than most.

If a smallish company like ARRI makes a claim, after a decade-long plateau in unit dynamic range, that they have increased performance almost three-fold, the claim ought to invite significant inquiry.

 

[Paul Spinnler]

I will assure you that I did not go and google dynamic range in order to find citations to attempt to make you look bad, I was simply pointing out the realities of where we are, with regard to the topic. I did not and do not wish to offend you personally.

I didn't have that impression. No worries.

But I do think that you have some aspects wrong here - Arri is an absolute leader in the field of cinema cameras with over 1600 employees and making half a billion of revenues and recognized in the cinema world as the de facto gold standard in both colour science and dynamic range. I wouldn't exactly call them small. I think you missed something here. In fact they dominate the field on blockbuster featured basis and within the cinema communities there have been countless deep dives on its dynamic range with extensive testing way exceeding what photo people would typically do (ie shooting some charts).

With all due respect for DPReview forum gurus – Arri has won technical Oscars https://www.arri.com/en/company/pre...ri-wins-academy-award-for-alexa-camera-system) not for nothing or by making spurious technical claims. They've been in business for over a hundred years. They are actually quite reliable it is my understanding compared to Red which typically exaggerates claims, especially with regard to dynamic range.

I personally spent considerable resources on the study of film and colours and worked mostly with cinema people who are a lot more advanced than photo people because there's real money in film still and because colour grading motion pictures is more complex and highly sought after by cinematographers; while in photo it has become a lot harder to justify the expensive systems from a pure professional standpoint.

In any case – as someone having shot analoge and digital extensively, I can only say the IQ4 is not there yet – its my personal view based on side-by-side comparisons – make of that what you want and I am just curious if we see further advances there one day.