The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Where's MFD headed ?

f8orbust

Active member
My guess is it’s headed squarely in the direction of SLR style cameras (XF, H6D etc.)

I don’t see any sensor manufacturer making a >=100MP CMOS sensor that will be shift friendly, nor do I see R/S coming out with a range of ultra-expensive, strongly retrofocus W/A lenses to work with current and future >=100MP+ CMOS sensors.

I could be wrong - usually am - but your thoughts are welcome (not on whether I'm usually wrong or not, but on where MFD is headed ;)).

Jim

P.S. This topic could be useful for anyone tempted to dip their toe in the world of MFD, particularly using a tech cam, so please don't hijack it and turn it into a 'everyone-will-be-shooting-35mm-within-5-years' sort of thing - let's be optimistic and assume MFD will be with us for many years to come.
 

MrSmith

Member
Mirrorless in some shape or form.
Phase to bring out a photoshop competitor.
Special editions with ostrich leather,
Signature editions from CEO's you have never heard of.
150-200mp once 35mm is at 80-100
Possibly...
 
I mean use MFD because I feel like larger formats always have a more pleasing rendering than smaller ones. I have an RX1RII and a 645z, two of the best sensors out there, but the RX1RII is not going to do for me what my 645z does no matter how many pixels it has, or how good the 35/2 is vs the 55/2.8 that I most commonly use on the Z. They're just different. I think the opposite will happen, and sensors will actually get bigger, for cheaper. Photographers WANT to shoot medium format. Even if on some technical level the A7RII and D810 are super close to a larger sensor in performance, photographers want to use the bigger format if they choose to. Until the Z, it has been reserved to a tiny tiny fraction of studio and landscape shooters in MFD. I see that changing with Sony sensors.

Serious question as I don't do this kind of work... With perspective control so achievable in post why use a tech camera at all? Like with an Alpa I understand, but why bother with a tech cam when PS will do the same thing?
 

modator

Member
Serious question as I don't do this kind of work... With perspective control so achievable in post why use a tech camera at all? Like with an Alpa I understand, but why bother with a tech cam when PS will do the same thing?
Well if You speak of making effect like toy-city photoshop can do the same with software, if you speak of making straight lines out from an angled view of a building then photoshop can do a good job sometime... and sometime not..., but if you need to recover the focus for a jewel or a macro of a watch mechanism maintaining a good color capture (not closing f45) a little shift can do something photoshop can't.

The diffraction and loose of color closing the diaphragm are unavoidable to make little things in focus... most of the high quality work in this case is made with view camera and digital back (most based on old CCD's) that for studio and at base ISO are at the edge.

Best regards, Domenico.
 

ErikKaffehr

Well-known member
Hi,

No, I don't think so. Stacking is quite useful in many situations but fails miserably in many others. Tilts are not a universal solution either. Tilts work for a single plane of focus, but anything falling outside of that will be unsharp.

Tilts are now 24x36mm territory and so are shifts, really. So MFD needs to compete in other fields.

Just to explain:

Now that we have the Sony A7xx there are great T&S options using almost any lens. There are miniature view cams from Arca and Cambo that allow a lot of shifts and tilts with the A7xx, using almost any lens. 24x36 mm may allow 12-20 mm shift, that is a lot on 24x36 mm, like 24-40 mm on MFD. And you can use it with a lot of lenses. There are also pocket size TS adapters that don't offer the flexibility of a bellows camera but still do the job nicely.

Best regards
Erik


Not even with some combination of perspective control and focus stacking?
 
Well if You speak of making effect like toy-city photoshop can do the same with software, if you speak of making straight lines out from an angled view of a building then photoshop can do a good job sometime... and sometime not..., but if you need to recover the focus for a jewel or a macro of a watch mechanism maintaining a good color capture (not closing f45) a little shift can do something photoshop can't.

The diffraction and loose of color closing the diaphragm are unavoidable to make little things in focus... most of the high quality work in this case is made with view camera and digital back (most based on old CCD's) that for studio and at base ISO are at the edge.

Best regards, Domenico.
You could do this with focus stacking too though, and at any aperture you want. I'm not sure how automated focusing stacking has gotten, though I imagine that doing in a single capture is faster...

It's just that most of the chatter I hear is about the ultra wides and their coverage. I have a hard time believing that the tech lenses are that much better than the Phase, Hasselblad or even the Pentax 28-45. Seems nuts to say that but I've used the Pentax and it's a real jewel! No movements but I can correct perspective even in LR now...
 
Hi,

No, I don't think so. Stacking is quite useful in many situations but fails miserably in many others. Tilts are not a universal solution either. Tilts work for a single plane of focus, but anything falling outside of that will be unsharp.


Best regards
Erik
Ah I am a portrait/landscape/reportage shooter so I've never had to mess with that stuff. Though i've always argued that if maximizing DoF is your game then a smaller format is another advantage given how good low-iso performance of sensors are across the board down to APS-C.
 

MrSmith

Member
We are already :OT:
Anyway. Photoshop or stacking is no substitute for the control a view camera gives you especially regarding workflow and the time spent producing imagery. Anything that hinders the process costs time and money.

I shoot still life/watches, while I use stacking a lot there is no way I'm giving up my view camera.
 

ErikKaffehr

Well-known member
Hi,

I never knew shifts could help with DoF, but I know tilts can do it. Do it to some extent.

The image below was made using tilts. The first image I made had the focus plane across the small pine in the foreground and the small rocks in the sea. The trees in the background were gravely out of focus.

Take two, I put the tilted focal point on the foreground pine and midway on the trees in background. I also used smaller apertures and shot several alternatives. The image I finally used had pretty decent focus on foreground and background. The rocks at the center were out of focus but could be helped by selective sharpening. I am pretty happy with this one. Exhibition stuff!

Artgh! It is a 24x36mm image that I should not post on an MFD forum! Well, I post it anyway as it illustrates a context.

Best regards
Erik



You could do this with focus stacking too though, and at any aperture you want. I'm not sure how automated focusing stacking has gotten, though I imagine that doing in a single capture is faster...

It's just that most of the chatter I hear is about the ultra wides and their coverage. I have a hard time believing that the tech lenses are that much better than the Phase, Hasselblad or even the Pentax 28-45. Seems nuts to say that but I've used the Pentax and it's a real jewel! No movements but I can correct perspective even in LR now...
 

jerome_m

Member
I don’t see any sensor manufacturer making a >=100MP CMOS sensor that will be shift friendly
Probably not any soon.

nor do I see R/S coming out with a range of ultra-expensive, strongly retrofocus W/A lenses to work with current and future >=100MP+ CMOS sensors.
Why a "range"? They would only need to come out with one W/A. Standard and tele lenses cause no problems with the new sensors.
 

ErikKaffehr

Well-known member
Hi,

I don't know what percentage of MFD shooters are using shifts and tilts. Those users may be a bit over represented here on GetDPI. For those shooting DSLR type camera I think there are few problems with late generation sensors.

One trend that we may see is that MFD gets lower cost of entry. Older backs are much more affordable than just a few years ago and we may see more of that coming.

On the other side, it seems that Hasselblad can sell all the VFC backs they can make as there is a large pool of Hasselblad owners who want to use their beloved system in the digital age, Perry Oosten mentioned 550 thousand bodies sold trough the years. "Older" models of Leica "S" can be bought at ridiculously low prices. That may mean new MF-users.

On the sensor side, phone camera sensors have similar issues to MFD and much of development is happening on phone cameras. Modern phone cameras have BSI, shallow wells and insulation between well. Technology that would cure a lot of problems with modern DSLR derived CMOS-backs.

Present MFD CMOS sensors are coming from Sony, except for the Leica S (typ 007). Sony going mirrorless puts them into a situation that shares many of the same problems as MFDs. Older wide angle designs are not working very well on the Sony A7 series, in part due to the cover glass and but mostly because of the beam angle.

Zeiss is a good example of that. The Loxia 50 and Loxia 35 are revamped designs for the Contax G2, with the optical calculation taking the cover glass into account. But, then new Loxia 21/2.8 is a brand new Distagon type design.

Sony obviously feels that both large sensor area and compact size are good selling argument and they need "shallow" pixel designs to make the A7 working well with compact wide angle lenses. My guess is that the present CMOS sensors are derived from 24x36 mm sensors. Next generation may be more beam angle tolerant.

Another question is how the shift to CMOS will affect the upgrade rate. Sony probably will introduce new sensors at 18 month intervalls (or so). The question is how such an update rate transfers to MFD, are the present sensors "one shot" or will we see a more frequent upgrade cycle on MFD?

There are lot of rumours Fuji or Sony may enter MFD. Would they do it with slightly larger than 24x36 sensors, or would they go full frame MFD?

Best regards
Erik


My guess is it’s headed squarely in the direction of SLR style cameras (XF, H6D etc.)

I don’t see any sensor manufacturer making a >=100MP CMOS sensor that will be shift friendly, nor do I see R/S coming out with a range of ultra-expensive, strongly retrofocus W/A lenses to work with current and future >=100MP+ CMOS sensors.

I could be wrong - usually am - but your thoughts are welcome (not on whether I'm usually wrong or not, but on where MFD is headed ;)).

Jim

P.S. This topic could be useful for anyone tempted to dip their toe in the world of MFD, particularly using a tech cam, so please don't hijack it and turn it into a 'everyone-will-be-shooting-35mm-within-5-years' sort of thing - let's be optimistic and assume MFD will be with us for many years to come.
 

modator

Member
Hi Speedographic,
Like what Erik said, also for me focus-stacking is not a good solution.

I try some time focus stacking, anyway the result have a bad look, the software searches the area on each shot that have more contrast (and with some pattern or high contrast, it fails) then compose the final image with some blur to glue each region found in focus in a single image, if focus stacking software have more sample at different distance (IE 100 frames each with a motion of 0,5 mm to cover 5cm of the depth of field) the blurred areas are very thin and sometime undetectable... but in many situation it fails generating horrible result.

This technique is oriented for shooting macro at very high enlargement ratio, another problem is the blurred background always there is a problem with it, because the software tend to add noise catched from frames thinking it's important data.
Making a very good focus stacking is not easy... tilting of some angle degree the lens can lead to a more pleasing and natural image in a fraction of the time.

To continue on the discussion, substantially the new cmos sensor are performing very similar to the smaller one used in DSLR, as said, there are many solution to do tilt and shift with DSLR and mirrorless that in some case can lead to very professional result, the era of the "CCD color supremacy slogan" is finished, some problems caused by micro-lens are still present since their introduction in MFD and DSLR too if there's the need to shift wide angle.
What remain not debatable, it's the larger size of the sensor and the resulting different COC (circle of confusion) that limit the diffraction at some stop above the limit of DSLR.. I remember at F11 with APSC was possible to spot the diffraction degradation... with the H5D I didn't see diffraction problem even at F22...

So far so good, so If You need Larger sensor I think MF is the way to take..:ROTFL:


Best regards, Domenico.


You could do this with focus stacking too though, and at any aperture you want. I'm not sure how automated focusing stacking has gotten, though I imagine that doing in a single capture is faster...

It's just that most of the chatter I hear is about the ultra wides and their coverage. I have a hard time believing that the tech lenses are that much better than the Phase, Hasselblad or even the Pentax 28-45. Seems nuts to say that but I've used the Pentax and it's a real jewel! No movements but I can correct perspective even in LR now...
 

ErikKaffehr

Well-known member
Hi,

I think that the kind of focus stacking the Phase One XF has is the best solution in most cases. Very clearly, sometimes focus stacking works but sometimes it fails horribly.

Sometimes I shot several images and do manual stacking, this works very well if the image is "layered".

Regarding diffraction it sort of affects all system similarly. On the Hasselblad I can observe loss of sharpness due to diffraction between f/5.6 and f/8, but it is a very minor effect. Stopping down to f/16 it is significant. The reality is that the more you have, the more you have to loose.

So, say that I need to stop down to f/16 on my P45+ for DoF in that case I would get the same DoF at f/11 on my A7rII. That would also apply to my A7II and to a A7sII that I don't own. Diffraction would be more on the P45+, but the pixels are also larger, so net effect would be the same. The A7II would loose less resolution than the A7rII, but the A7rII would still have a much better resolution than the A7II.

What the A7 cannot do is Phase One XF style stacking. I think Canon cameras can do it using tools like "Camranger".

The A7rII/HCam Master TSII can do a lot of nice things, like offering T&S with zooms. The lenses that allow T&S on the A7rII är generally DSLR/MFD lenses and they have low beam angles, so the beam angle sensitivity is probably not an issue. I have not observed it, but that doesn't say it doesn't exist.

Chris Barret has moved from IQ 260 to Sony A7r and uses it with the Arca Universalis and a bunch of Hasselblad V lenses. In addition he has the two Canon T&S lenses. To that comes a Canon 11-24/4 zoom that offers a lot of shift: http://forum.luminous-landscape.com/index.php?topic=93655.msg904652#msg904652

Chris makes two points, one is that there is no need for LLC with the Hasselblad/Canon lenses, keep in mind that they are made for SLR cameras. The other is that shift should be seen relative to sensor size. 12-20 mm is a lot on a small sensor. Chris says he gives up quiet a bit of image quality compared to the IQ3-100MP, but Arca/Sony/Canon combo can take images that are not possible with larger sensors.

Personally, I own a P45+ with Hasselblad Flexbody, but I didn't find the Flexbody practical in the field.

Best regards
Erik




Hi Speedographic,
Like what Erik said, also for me focus-stacking is not a good solution.

I try some time focus stacking, anyway the result have a bad look, the software searches the area on each shot that have more contrast (and with some pattern or high contrast, it fails) then compose the final image with some blur to glue each region found in focus in a single image, if focus stacking software have more sample at different distance (IE 100 frames each with a motion of 0,5 mm to cover 5cm of the depth of field) the blurred areas are very thin and sometime undetectable... but in many situation it fails generating horrible result.

This technique is oriented for shooting macro at very high enlargement ratio, another problem is the blurred background always there is a problem with it, because the software tend to add noise catched from frames thinking it's important data.
Making a very good focus stacking is not easy... tilting of some angle degree the lens can lead to a more pleasing and natural image in a fraction of the time.

To continue on the discussion, substantially the new cmos sensor are performing very similar to the smaller one used in DSLR, as said, there are many solution to do tilt and shift with DSLR and mirrorless that in some case can lead to very professional result, the era of the "CCD color supremacy slogan" is finished, some problems caused by micro-lens are still present since their introduction in MFD and DSLR too if there's the need to shift wide angle.
What remain not debatable, it's the larger size of the sensor and the resulting different COC (circle of confusion) that limit the diffraction at some stop above the limit of DSLR.. I remember at F11 with APSC was possible to spot the diffraction degradation... with the H5D I didn't see diffraction problem even at F22...

So far so good, so If You need Larger sensor I think MF is the way to take..:ROTFL:


Best regards, Domenico.
 

fotografz

Well-known member
Where is MFD headed?

It could be headed back to where it historically started ... in the studio, or special applications, or conceptual work, and institutional photography.

This forum is heavily skewed to landscape work, especially in the MF section. Inroads by super high-res cameras like the Sony A7R-II make sense for many such photographers because of the obvious size/transport/versatility advantage.

In a more controlled environment like the studio, or on location where assistants are with the photographer, the size advantage evaporates. Plus, mirrorless advantages can become a disadvantage, and MFD with LS lenses have the type of versatility needed and is unavailable with the smaller formats.

For example, I still do a fair amount of new product photography in studio and on location. New product work isn't subject to clients opting for stock photography like other catagories can be ... it is NEW. Same for lifestyle and/or location work with new product placements.

I have a photographer friend that has a large modern studio with 9 Hasselblad stations running constantly; including 2 with full movement view cameras and one station with a big old Fuji with movements. He does a massive amount of new product work for clients that are extraordinarily demanding and picky.

Point is that this forum is a microcosm ... There is a whole other world out there where MFD is both viable and even preferable ... be it rented gear, used gear or new gear ... depending on how successful the photographer may be.

As an advertising art director/creative director (now semi-retired) I continue to do a lot of free-lance consulting and campaign development for select clients. I still choose photographers to do certain work where my own photographic skills or infrastructure aren't appropriate. In the past 10 years, right up to a current campaign that I'm supervising, 35mm capture has been the exception, not the rule. While it most certainly is used when the scenario calls for it, most work is done with MFD, and all product work is MFD, (with the exception of some food work).

BTW, none of the 100s of pro photographers I've selected to work with appear on forums, have a blog, or have time to shoot comparison shots replete with charts and graphs. Most use older MFD gear unless they rent it, have tech guys keeping it running, or hire a free-lance tech person to run rented gear ... when and if they buy MFD new, it is one-on-one product demo's in their environment shooting what they shoot.

I've rarely had a discussion about digital cameras with any of them ... they discuss ideas, lighting (lots about lighting), creativity, how to sell clients a visual idea, and such.

Different world ... and a bigger world than most folks here might imagine.

- Marc

Attached are a sample of product work and one conceptual shot from a series of 10 done for a bank.
 

Attachments

modator

Member
Here's an example of Tilt with the HTS and HC-50 II (75mm resulting focal length / 59mm on 24x36 format), with only 5,6º*of tilt down 2m of floor and a box in focus... all in one shot, compared with no tilt (left)... if I have to do this with stacking it take hour and hours of retouching... hey that's used shoes !

Like Eric said, this kind of photo it can be done also with A7 / A7R's etc. off course, for me the most important thing it's that I'm satisfied of the result I get.

Regards, Domenico.

 
Top