An image from a virtual 6x8 medium format sensor

[Rob de Loe]

We've had lots of threads recently about stitching to create a larger virtual sensor, rather than for panoramas. Michael's recent thread in particular explored whether there are benefits to stitching to create a larger virtual sensor that go beyond just more pixels. David's is another one that started today. Clearly lots of people want a bigger sensor and think there are benefits.

By the time we reached the limit of 150 posts on Michael's thread, I think we landed on "maybe". I suspect we're not going to get past "maybe there are benefits" because we all see different things in images.

Nonetheless, I thought it would be interesting to see what would result if I put a 6x8 digital sensor behind a lens designed to cover the 6x8 film format. Perhaps results on a really big sensor will offer some insights.

The lens is a Fujinon GX 125mm f/5.6. It was designed for the old Fuji GX680 camera system, which made 6x8 images on 120 and 220 film. My Toyo VX23D has enough movement range to cover just about any medium format film size. With that lens and my Toyo, I can create a flat stitched image that is just about what you would see if you could replace the film back on a Fuji GX680 with a 6x8 digital sensor using the same sensor technology as found in a GFX 50R.

The resulting image is 11,654 x 15,539 pixels after cropping to 4:3. This is a bit larger than the area covered by an actual 6x8 frame on film. If I actually had a huge digital sensor that made 11,654 x 15,539 images, it would be 61.4mm x 81.8mm. For comparison, if I shot the image with one frame on my GFX 50R, I'd need to use a 64mm lens to get about the same angle of view. Note that I could have made it exactly the dimensions covered by 6x8 film, but it was easier to work with round numbers: +/-20mm rise and +/-15mm shift.

The image below needed 9 frames, which I flat stitched in Lightroom. Depending on the subject, 4 frames could work too -- but 9 frames almost always works. It took multiple attempts to get a final image I liked because, of course, you can only see your final image in small pieces. Composition would have been a lot easier if I'd used a 64mm lens to frame, and then switched to the GX 125/5.6.

Working with the flat-stitched file -- over 180 MB -- was a bit slow and tedious, but do-able even on my old Dell desktop.

I like the resulting image, but I don't think the benefit-cost ratio is in its favour. Yes, I could make a huge print from this file. However, it's a lot of extra work and processing, and in a field situation -- rather than a controlled studio environment -- it's harder to reliably get what you need.

So the question is: are there any other qualities that can clearly be tied to using a virtual 6x8 sensor that make it worth the effort. I'm not sure, but I'm curious what other people think.

View attachment 2e02de7a00f54381ab5c0ffd7a4dfd06.jpg
This is what a lens designed to cover 6x8 film looks like when you use it with a virtual 6x8 digital sensor. The image is built up from 9 flat-stitched images using a GFX 50R camera and a Toyo VX23D. The GX 125mm f/5.6 lens was set to f/11. For this post, I exported the file in native GFX 50R size (6,192 x 8,256 pixels); the final image cropped from the flat-stitched base file is 11,654 x 15,539 pixels.

 

[bobby350z]

Would be nice to see the full res file and the file from your 63mm f2.8.

 

[Rob de Loe]

Would be nice to see the full res file and the file from your 63mm f2.8.

If I owned a GF 63/2.8 you would have seen a comparison shot!

Unfortunately, I bought it, used it for a while and sold it.

 

[DavidMillier]

Good debating subject.

I don't see anything different from standard cameras, just a bigger file. But very happy to have people point out things I may have missed.

 

[Rob de Loe]

Good debating subject.

I don't see anything different from standard cameras, just a bigger file. But very happy to have people point out things I may have missed.

You'll notice I very carefully didn't use any phrases that had the words "medium" and "look" in them.

On one hand, I'm very empirical. I want to see what other people say exists. Evidence, repeatability, and all that....

On the other hand, I remain open to the possibility that I cannot taste, smell, see or hear things that others can. Senses are funny like that.

 

[alpshiker]

Good debating subject.

I don't see anything different from standard cameras, just a bigger file. But very happy to have people point out things I may have missed.

You'll notice I very carefully didn't use any phrases that had the words "medium" and "look" in them.

On one hand, I'm very empirical. I want to see what other people say exists. Evidence, repeatability, and all that....

On the other hand, I remain open to the possibility that I cannot taste, smell, see or hear things that others can. Senses are funny like that.

Ha ha. I guess we are all in search for the magic recipe. I haven't found it yet either to be honest. Maybe it doesn't belong entirely to the quantifiable realm of pixels and sensor sizes.

For instance, I "borrowed" a GFX100S to convince myself that it would be a worthy switch. I am currently working hard to prove to myself that it makes better pictures than the 50S, …so far unsuccessfully. But I might eventually get there if I try hard enough.

 

[DavidMillier]

Good debating subject.

I don't see anything different from standard cameras, just a bigger file. But very happy to have people point out things I may have missed.

You'll notice I very carefully didn't use any phrases that had the words "medium" and "look" in them.

On one hand, I'm very empirical. I want to see what other people say exists. Evidence, repeatability, and all that....

On the other hand, I remain open to the possibility that I cannot taste, smell, see or hear things that others can. Senses are funny like that.

I learned from debates on the Sigma forum and dabbling in the high end hifi industry that the way humans are built often makes it difficult for people to separate facts from wishes sometimes.

And it is not silly imagination, somehow the brain creates its own reality. I remember attending a course at work a few years ago a where the lecturer showed this well known optical illusion:

https://en.wikipedia.org/wiki/Checker_shadow_illusion#/media/File:Checker_shadow_illusion.svg

When he ran the animation that proved that the two squares had exactly the same tone (to applause) there was one person who refused to believe it, insisting that the animation was faked. The problem was she had too strong a belief in the infallibility of her senses and no understanding that the brain is subject to a plethora of cognitive biases. Feynman put it succinctly https://www.brainyquote.com/quotes/richard_p_feynman_137642

It is a hard lesson to learn.

--
Website: http://www.whisperingcat.co.uk/ (2022 - website rebuilt, updated and back in action)
DPReview gallery: https://www.dpreview.com/galleries/0286305481
Flickr: http://www.flickr.com/photos/davidmillier/ (very old!)

 

[Michael Floyd]

Nice one Rob, I'm glad you've started this thread as the earlier one that hit the limit ended just as I was starting to get serious. It also meandered off the point quite a bit (mostly my doing if I remember correctly. )

Where I'm at with this at the moment:

1. That is a lovely image and a perfect composition for this discussion.

2. I still don't have the answers, but I definitely have some firm thoughts and feelings around this. Everything that follows here is my opinion at this point in time, subject to change without notice. ;-)

3. I feel it's more than worthwhile to drop this link here, even though we've all read it before. Jim's brief but direct overview of Format Size and Image Quality strikes me as particularly well grounded and informed.

4. It's very difficult to come to any firm conclusions without A/B comparisons. Also because of the nature of what we are comparing with this, we are comparing lenses more than sensor sizes.

We are comparing lenses more than sensor sizes.

And this is the essential crux of this whole discussion. With ideal lenses, comparable sensor technology and resolution etc etc etc, there is no difference across format sizes. However ideal lenses don't exist.

Let's put GF lenses aside for a moment. Anyone flat shift capturing significantly larger than 44x33 is using an older lens. (Are there exceptions? probably.)

I know for certain that a 44x33 capture, or a GFX two frame stitch to make a 645 capture using a Pentax 645 film lens is significantly better by almost any image quality metric compared to an equivalent image from aps-c shot with a same era well regarded Pentax K mount lens, even when downsized to the same resolution. This is especially true for wider apertures. I don't believe that the P645 lenses are intrinsically any better than the better K-mount ones, they just cover more area.

We are getting a resolution advantage in this scenario regardless of pixel count because the K-mount lenses just aren't quite up to the task of resolving enough fine detail on an aps-c sized sensor. The larger capture in this case is taking the pressure off the lens. It simply doesn't need to be quite so 'sharp' over a given fixed sensor area to produce a significantly sharper image. This is compounded by the fact that if we are shooting for a consistent DOF, which I think is a fair assumption in most cases, we will be stopping the lens down further for the larger capture, and all lenses, especially older ones, improve on stopping down.

-

When I look at your image above I see an immensely natural and lovely presentation of the subject. No matter how close I look, or taking the image as a whole, everything is sweet and smooth. It is just simply beautiful to look at, wonderfully detailed in a totally unforced way in the focal plane, and this fades ever so gracefully and gently into the more distant areas. Do I see a tiny hint of fringing? If I do it doesn't draw attention to itself or bother me.

As I explore larger captures more, I'm finding that this kind of deliciously natural presentation is far more likely, almost the default, for larger captures. Conversely, on a little aps-c sensor things very often look a bit funky (for want of a more precise term) here and there, even using top modern Fuji-X lenses, significantly more so using adapted film lenses. FF gets noticeably better, GFX is well and truly 'good enough', although I do still feel like I achieve this kind of natural, delightful all over presentation more often with a stitch up.

All of this for adapted Pentax lenses, K-mount, P645 and P67.

-

GF lenses - My experience here is low. I have the GF45.

The obvious question is: Does the 'closer to ideal' GF lens make up for the relatively smaller capture area? I don't know, is my answer. Ha!

My thinking is that if you're aiming for all over sharpness, front to back, side to side - probably you're better off (or at least as well off) just shooting GF lenses. (Tilt/swing/shift aside)

Moving into seriously wobbly and subjective territory, if we're talking transitions and OOF rendering.... Ah, well I've seen some OOF areas from GF lenses that strike me as a bit challenging, and I'm often delightfully pleased with what comes out of my older Pentax lenses, especially as the capture grows. So all bets are off and I'm certainly not the person to make any call on this. Practically speaking, for me, the point is moot, as I'm not going to go out and pick up the 110/2 or probably any other GF lens. I'm here with the 50R because my lenses of choice pointed me in this direction, and for these lenses I think that it's clear that, for me at least, bigger captures are better in observable ways. My task at this stage is to drill down on where is 'good enough' for which circumstances, and where is the crossover line marked 'more trouble than it's worth'.

-

One last point and entirely aside from image quality. I'm quite enjoying the process of shooting for a stitch, either two frames on a simple shift adaptor, or with the Vertex. I do think that the process influences the results in ways that are not measurable. Obviously this is highly subjective.

-

I would love to be able to put up a great A/B comparison here to close, that shows clearly what I'm seeing. Unfortunately I haven't got anything yet that I think is convincing enough (well, not from GFX compared to something larger - the aps-c >GFX wide aperture comparisons are blatant.). It will come in time. Or perhaps it won't, and I will eat my words.

 

[DavidMillier]

Now I have my vertex 645 square crop virtual sensor and a couple of test prints on A2 paper, I'm about to put the format war to a critical test: 100MP 645 vs 20mp 1" bridge camera on 16" square prints. I'm actually going to take my print outside with me to ensure I frame the bridge camera properly

 

[Doppler9000]

Nice one Rob, I'm glad you've started this thread as the earlier one that hit the limit ended just as I was starting to get serious. It also meandered off the point quite a bit (mostly my doing if I remember correctly. )

Where I'm at with this at the moment:

1. That is a lovely image and a perfect composition for this discussion.
2. I still don't have the answers, but I definitely have some firm thoughts and feelings around this. Everything that follows here is my opinion at this point in time, subject to change without notice. ;-)
3. I feel it's more than worthwhile to drop this link here, even though we've all read it before. Jim's brief but direct overview of Format Size and Image Quality strikes me as particularly well grounded and informed.
4. It's very difficult to come to any firm conclusions without A/B comparisons. Also because of the nature of what we are comparing with this, we are comparing lenses more than sensor sizes.

We are comparing lenses more than sensor sizes.

And this is the essential crux of this whole discussion. With ideal lenses, comparable sensor technology and resolution etc etc etc, there is no difference across format sizes. However ideal lenses don't exist.

Let's put GF lenses aside for a moment. Anyone flat shift capturing significantly larger than 44x33 is using an older lens. (Are there exceptions? probably.)

The Schneider Digitar and Rodenstock Digaron lenses cover enough to shift, some more than others.

The 100mm Digaron, for example, is pretty impressive…





I know Rob thinks very highly of his APO Digitar 90mm.

These lenses need a bellows or helicoid to focus, of course.

I know for certain that a 44x33 capture, or a GFX two frame stitch to make a 645 capture using a Pentax 645 film lens is significantly better by almost any image quality metric compared to an equivalent image from aps-c shot with a same era well regarded Pentax K mount lens, even when downsized to the same resolution. This is especially true for wider apertures. I don't believe that the P645 lenses are intrinsically any better than the better K-mount ones, they just cover more area.

We are getting a resolution advantage in this scenario regardless of pixel count because the K-mount lenses just aren't quite up to the task of resolving enough fine detail on an aps-c sized sensor. The larger capture in this case is taking the pressure off the lens. It simply doesn't need to be quite so 'sharp' over a given fixed sensor area to produce a significantly sharper image. This is compounded by the fact that if we are shooting for a consistent DOF, which I think is a fair assumption in most cases, we will be stopping the lens down further for the larger capture, and all lenses, especially older ones, improve on stopping down.

-

When I look at your image above I see an immensely natural and lovely presentation of the subject. No matter how close I look, or taking the image as a whole, everything is sweet and smooth. It is just simply beautiful to look at, wonderfully detailed in a totally unforced way in the focal plane, and this fades ever so gracefully and gently into the more distant areas. Do I see a tiny hint of fringing? If I do it doesn't draw attention to itself or bother me.

As I explore larger captures more, I'm finding that this kind of deliciously natural presentation is far more likely, almost the default, for larger captures. Conversely, on a little aps-c sensor things very often look a bit funky (for want of a more precise term) here and there, even using top modern Fuji-X lenses, significantly more so using adapted film lenses. FF gets noticeably better, GFX is well and truly 'good enough', although I do still feel like I achieve this kind of natural, delightful all over presentation more often with a stitch up.

All of this for adapted Pentax lenses, K-mount, P645 and P67.

-

GF lenses - My experience here is low. I have the GF45.

The obvious question is: Does the 'closer to ideal' GF lens make up for the relatively smaller capture area? I don't know, is my answer. Ha!

My thinking is that if you're aiming for all over sharpness, front to back, side to side - probably you're better off (or at least as well off) just shooting GF lenses. (Tilt/swing/shift aside)

Moving into seriously wobbly and subjective territory, if we're talking transitions and OOF rendering.... Ah, well I've seen some OOF areas from GF lenses that strike me as a bit challenging, and I'm often delightfully pleased with what comes out of my older Pentax lenses, especially as the capture grows. So all bets are off and I'm certainly not the person to make any call on this. Practically speaking, for me, the point is moot, as I'm not going to go out and pick up the 110/2 or probably any other GF lens. I'm here with the 50R because my lenses of choice pointed me in this direction, and for these lenses I think that it's clear that, for me at least, bigger captures are better in observable ways. My task at this stage is to drill down on where is 'good enough' for which circumstances, and where is the crossover line marked 'more trouble than it's worth'.

-

One last point and entirely aside from image quality. I'm quite enjoying the process of shooting for a stitch, either two frames on a simple shift adaptor, or with the Vertex. I do think that the process influences the results in ways that are not measurable. Obviously this is highly subjective.

-

I would love to be able to put up a great A/B comparison here to close, that shows clearly what I'm seeing. Unfortunately I haven't got anything yet that I think is convincing enough (well, not from GFX compared to something larger - the aps-c >GFX wide aperture comparisons are blatant.). It will come in time. Or perhaps it won't, and I will eat my words.

 

[DavidMillier]

Well, that was a bit of non competition.

It goes to show that to judge these things properly you need to do a side by side competition, rather than looking at images in isolation. I was initially a bit underwhelmed by the vertex test prints, but comparing them at 6" viewing distance to a 13MP crop from my FZ1000 1" sensor bridge camera puts things into perspective.

They look amazing.

 

[alpshiker]

Well, that was a bit of non competition.

It goes to show that to judge these things properly you need to do a side by side competition, rather than looking at images in isolation. I was initially a bit underwhelmed by the vertex test prints, but comparing them at 6" viewing distance to a 13MP crop from my FZ1000 1" sensor bridge camera puts things into perspective.

They look amazing.

Glad you noticed, David, for this leaves us with some prospect !

But as you noted, stitching works from the Vertex are just an aspect of a whole different approach to photography, and for the results to be noticed, there needs to be more than just a higher resolution. It forces the photographer to think and invest in more creativity. In some cases, it is possibly a counter productive approach which has the benefit to introduce a part of fortuitance, which can work in favour of the final work or not. What I like in it is that I have little idea of how good the result will be until everything is assembled and edited. Often disappointing, but it's part of the game. Kind of similar longing as when we sent the sheet films to the lab with some anxious expectations. As Michael and Rob emphasized, it is above all a fun approach.

 

[DavidMillier]

Much closer this time. The prints have a similar overall look and feel unlike the 1" sensor which was breaking up from too much enlargement and visibly noisy. But the 100MP stitch is better: more fine detail and micro-contrast on distant foliage, the surface of stonework is more detailed, while the single frame looks a little air-brushed in the finest details.

This shows me that my assumption that input data in excess of 240ppi yield no visible improvement in matte surface prints is wrong. Printers are sharper than I had supposed based on my previous experience with lower pixel counts.

I'm not seeing any "special MF" qualities in these early test prints, though. The advantage is solely limited to detail and micro-contrast. If you take a couple of paces back, the overall look of the print is the same.

Nonetheless, this is an reassuring early test that I haven't thrown my mental energy (and cash) at something that yielded no visible improvement. Well done to Fotodiox for developing this series of stitch adaptors. Realistically, I don't have the cash for a 645 150MP Phase One outfit, but the vertex will do. I should buy more Pentax lenses

--
Website: http://www.whisperingcat.co.uk/ (2022 - website rebuilt, updated and back in action)
DPReview gallery: https://www.dpreview.com/galleries/0286305481
Flickr: http://www.flickr.com/photos/davidmillier/ (very old!)

 

[Rob de Loe]

Good debating subject.

I don't see anything different from standard cameras, just a bigger file. But very happy to have people point out things I may have missed.

You'll notice I very carefully didn't use any phrases that had the words "medium" and "look" in them.

On one hand, I'm very empirical. I want to see what other people say exists. Evidence, repeatability, and all that....

On the other hand, I remain open to the possibility that I cannot taste, smell, see or hear things that others can. Senses are funny like that.

I learned from debates on the Sigma forum and dabbling in the high end hifi industry that the way humans are built often makes it difficult for people to separate facts from wishes sometimes.

And it is not silly imagination, somehow the brain creates its own reality. I remember attending a course at work a few years ago a where the lecturer showed this well known optical illusion:

https://en.wikipedia.org/wiki/Checker_shadow_illusion#/media/File:Checker_shadow_illusion.svg

When he ran the animation that proved that the two squares had exactly the same tone (to applause) there was one person who refused to believe it, insisting that the animation was faked. The problem was she had too strong a belief in the infallibility of her senses and no understanding that the brain is subject to a plethora of cognitive biases. Feynman put it succinctly https://www.brainyquote.com/quotes/richard_p_feynman_137642

It is a hard lesson to learn.

That's a great example. You either believe in double-blind testing, or you don't...

 

[Rob de Loe]

Nice one Rob, I'm glad you've started this thread as the earlier one that hit the limit ended just as I was starting to get serious. It also meandered off the point quite a bit (mostly my doing if I remember correctly. )

Where I'm at with this at the moment:

1. That is a lovely image and a perfect composition for this discussion.

2. I still don't have the answers, but I definitely have some firm thoughts and feelings around this. Everything that follows here is my opinion at this point in time, subject to change without notice. ;-)

3. I feel it's more than worthwhile to drop this link here, even though we've all read it before. Jim's brief but direct overview of Format Size and Image Quality strikes me as particularly well grounded and informed.

4. It's very difficult to come to any firm conclusions without A/B comparisons. Also because of the nature of what we are comparing with this, we are comparing lenses more than sensor sizes.

We are comparing lenses more than sensor sizes.

And this is the essential crux of this whole discussion. With ideal lenses, comparable sensor technology and resolution etc etc etc, there is no difference across format sizes. However ideal lenses don't exist.

Let's put GF lenses aside for a moment. Anyone flat shift capturing significantly larger than 44x33 is using an older lens. (Are there exceptions? probably.)

I know for certain that a 44x33 capture, or a GFX two frame stitch to make a 645 capture using a Pentax 645 film lens is significantly better by almost any image quality metric compared to an equivalent image from aps-c shot with a same era well regarded Pentax K mount lens, even when downsized to the same resolution. This is especially true for wider apertures. I don't believe that the P645 lenses are intrinsically any better than the better K-mount ones, they just cover more area.

We are getting a resolution advantage in this scenario regardless of pixel count because the K-mount lenses just aren't quite up to the task of resolving enough fine detail on an aps-c sized sensor. The larger capture in this case is taking the pressure off the lens. It simply doesn't need to be quite so 'sharp' over a given fixed sensor area to produce a significantly sharper image. This is compounded by the fact that if we are shooting for a consistent DOF, which I think is a fair assumption in most cases, we will be stopping the lens down further for the larger capture, and all lenses, especially older ones, improve on stopping down.

-

When I look at your image above I see an immensely natural and lovely presentation of the subject. No matter how close I look, or taking the image as a whole, everything is sweet and smooth. It is just simply beautiful to look at, wonderfully detailed in a totally unforced way in the focal plane, and this fades ever so gracefully and gently into the more distant areas. Do I see a tiny hint of fringing? If I do it doesn't draw attention to itself or bother me.

As I explore larger captures more, I'm finding that this kind of deliciously natural presentation is far more likely, almost the default, for larger captures. Conversely, on a little aps-c sensor things very often look a bit funky (for want of a more precise term) here and there, even using top modern Fuji-X lenses, significantly more so using adapted film lenses. FF gets noticeably better, GFX is well and truly 'good enough', although I do still feel like I achieve this kind of natural, delightful all over presentation more often with a stitch up.

All of this for adapted Pentax lenses, K-mount, P645 and P67.

-

GF lenses - My experience here is low. I have the GF45.

The obvious question is: Does the 'closer to ideal' GF lens make up for the relatively smaller capture area? I don't know, is my answer. Ha!

My thinking is that if you're aiming for all over sharpness, front to back, side to side - probably you're better off (or at least as well off) just shooting GF lenses. (Tilt/swing/shift aside)

Moving into seriously wobbly and subjective territory, if we're talking transitions and OOF rendering.... Ah, well I've seen some OOF areas from GF lenses that strike me as a bit challenging, and I'm often delightfully pleased with what comes out of my older Pentax lenses, especially as the capture grows. So all bets are off and I'm certainly not the person to make any call on this. Practically speaking, for me, the point is moot, as I'm not going to go out and pick up the 110/2 or probably any other GF lens. I'm here with the 50R because my lenses of choice pointed me in this direction, and for these lenses I think that it's clear that, for me at least, bigger captures are better in observable ways. My task at this stage is to drill down on where is 'good enough' for which circumstances, and where is the crossover line marked 'more trouble than it's worth'.

-

One last point and entirely aside from image quality. I'm quite enjoying the process of shooting for a stitch, either two frames on a simple shift adaptor, or with the Vertex. I do think that the process influences the results in ways that are not measurable. Obviously this is highly subjective.

-

I would love to be able to put up a great A/B comparison here to close, that shows clearly what I'm seeing. Unfortunately I haven't got anything yet that I think is convincing enough (well, not from GFX compared to something larger - the aps-c >GFX wide aperture comparisons are blatant.). It will come in time. Or perhaps it won't, and I will eat my words.

I think you're quite right about the role of lenses. I've noticed this particular lens does have a nice transition from in focus to out of focus. It's evident in a single capture on 33mm x 44mm too. Is it enhanced in a large assemblage like this? I don't know.

The A/B comparison point is a good one, but there are so many variables that it's always possible for someone who wants to argue one position over the other to find nits to pick.

I will note though that many people will tell you that they can "pick the medium format camera" from photographs made with different cameras on film. If they can do it reliably (an open question...) then there must be something they're seeing that is "real". My guess is it's all the things that have been rehearsed on this forum over the years (type of lens, size of negative, etc.)

We can also get practical and pragmatic about this: if we're shooting purely for our own personal pleasure, then by all means go into the rabbit hole. However, if we're shooting for an audience of "normal" people, I really think there's little profit to being in the rabbit hole. Most people simply can't tell the difference.

A related practical issue is how much tolerance one has for failure. If I'm just messing around, I'm happy to fail all over the place and try again. But if I'm in the field working (limited budget, limited time, etc.) I have a very low tolerance for failure. I need to get that image because I might not be able to come back, or it might not be available again. Stitching makes me nervous in that context.

 

[JimKasson]

I learned from debates on the Sigma forum and dabbling in the high end hifi industry that the way humans are built often makes it difficult for people to separate facts from wishes sometimes.

There are time-tested methods for testing that cleave the two. All it takes is the knowledge and the will to use them.

 

[DavidMillier]

I learned from debates on the Sigma forum and dabbling in the high end hifi industry that the way humans are built often makes it difficult for people to separate facts from wishes sometimes.

There are time-tested methods for testing that cleave the two. All it takes is the knowledge and the will to use them.

Some (most?) people prefer to believe their comforting wishes over unpleasant facts.

 

[JimKasson]

I learned from debates on the Sigma forum and dabbling in the high end hifi industry that the way humans are built often makes it difficult for people to separate facts from wishes sometimes.

There are time-tested methods for testing that cleave the two. All it takes is the knowledge and the will to use them.

Some (most?) people prefer to believe their comforting wishes over unpleasant facts.

And they wish to avoid the intellectual challenges and hard work of finding out for themselves.

--
Website: http://www.whisperingcat.co.uk/ (2022 - website rebuilt, updated and back in action)
DPReview gallery: https://www.dpreview.com/galleries/0286305481
Flickr: http://www.flickr.com/photos/davidmillier/ (very old!)

 

[alpshiker]

I think you're quite right about the role of lenses. I've noticed this particular lens does have a nice transition from in focus to out of focus. It's evident in a single capture on 33mm x 44mm too. Is it enhanced in a large assemblage like this? I don't know.

Maybe the same taken with a little more distance and transition, revealing details in the background, would show more of the qualities of the lens?

The A/B comparison point is a good one, but there are so many variables that it's always possible for someone who wants to argue one position over the other to find nits to pick.

I will note though that many people will tell you that they can "pick the medium format camera" from photographs made with different cameras on film. If they can do it reliably (an open question...) then there must be something they're seeing that is "real". My guess is it's all the things that have been rehearsed on this forum over the years (type of lens, size of negative, etc.)

In the emulsion film era, it was relatively easy to pick images taken on a film larger than 135mm, simply by looking at the grain-sharpness ratio if I say it right. Sensors definition and high bit sampling have reached such levels today that the cards are shuffled.

We can also get practical and pragmatic about this: if we're shooting purely for our own personal pleasure, then by all means go into the rabbit hole. However, if we're shooting for an audience of "normal" people, I really think there's little profit to being in the rabbit hole. Most people simply can't tell the difference.

I guess that not so many professional photographers who have to work hard to pay their mortgage have the time to visit this forum, let alone waste time on non fast profit making exotic devices!

A related practical issue is how much tolerance one has for failure. If I'm just messing around, I'm happy to fail all over the place and try again. But if I'm in the field working (limited budget, limited time, etc.) I have a very low tolerance for failure. I need to get that image because I might not be able to come back, or it might not be available again. Stitching makes me nervous in that context.

I think this sums it well. That's also why people who are bound to results would not bother using non GF lenses. This could be us if we had to travel to a once in a lifetime destination. An amateur : French lit. 'lover [of something]') is generally considered a person who pursues an avocation independent from their source of income (Wikipedia).

 

[JimKasson]

...people who are bound to results would not bother using non GF lenses.

I don't buy this at all.

• There is no GF lens that's nearly as good for 6x8 and smaller scanning as the 105/5.6 HR-Digaron.
• There is no GF lens that's nearly as good for 6x8 and smaller scanning as the appropriate 120/5.6 SK industrial macro lenses.
• There is no GF lens at all close to the 180/5.6 HR Digaron.
• There are currently no T/S GF lenses at all.
• There is no GF 11 mm fisheye.
• There are no GF lenses longer than 250mm.
• There are no Petzval GF lenses.

And that's just a few off the top of my head. There are lots of reasons to use non GF lenses when you're looking for results.