DoF and bokeh Sony A7rII vs P45+ - A quick and dirty comparison…

[ErikKaffehr]

Hi,

A couple of weeks ago there was a vivid discussion about DoF and bokeh differences between MFD and FF 135. We would expect the difference to correspond to a bit under two stops, corresponding to the ratio of format diagonals.

As I happen to have a Sony A7rII with a 90/2.8G macro and a P45+ with a Macro Planar 120/4, I felt I should do a practical test. These are the first results.

The first attachment shows the difference between the Sony A7rII at f/2.8 and the P45+ at f/5.6 in the area of focus. The second one shows the second cup in the rear row.

The third one shows a LensAlign target. Here it is clear that the Planar 120/4 is a bit back focused. That lens is hard to focus exactly, it doesn't snap into focus clearly and focusing throw is a bit on the short side. Smallest move on the focusing ring changes focus +/- 0.5 mm on the focusing ring for optimal focus I would say…

Regarding the target, intentions:

- Main subjects are the cups. Shooting distance around 1.8 m subject spread around 10 cm on the cups.

- The Swarowsky crystals are intended to demonstrate bokeh

- The LensAlign ruler is intended to show axial chroma and give some quantitative measurement of defocus bokeh.

- The ruler is intended to measure magnification, but it was moved to expected focus line in the second (present) shoot and it is now the line focused upon.

Commentary:

In this case I don't feel we see anything near a three stop difference, what I see is that the difference is pretty close to the expected difference of around 1.5 stops.

Fortunately, the two lenses are close regarding FOV (Field Of View) and the sensors are close in resolution 39 vs 42 MP, even closer if aspect ratio is taken into account.


Would anyone be interested in the raw files, they are available here: http://echophoto.dnsalias.net/ekr/Articles/Shoots/DOFComparison/

Edit: I realised the opportunity of some improvements, so the subjects were reshot. Basic findings still apply.

Best regards
Erik

[eronald]

Hi,

In this case I don't feel we see anything near a three stop difference, what I see is that the difference is pretty close to the expected difference of around 1.5 stops.

Erik

I agree.

E.

[synn]

I am not an expert at all these measuring festivities, but I am not seeing a point of reference where focus is spot on, from where the viewer can see how far the DoF spreads. There is a ruler and there is a cup, but there is no indication on how far the DoF spreads from the ruler to the cup. I would most definitely choose a different test scene for a test like this.

Also, I am sure many have mentioned this before, but Erik, your P45* images are far softer than they have any right to be.

[rueyloon]

Do you think the Sony is a good enough replacement ? I've been thinking of upgrading my aptus65, that or just ditch the whole system for a Sony.

[ctz]

Do you think the Sony is a good enough replacement ? I've been thinking of upgrading my aptus65, that or just ditch the whole system for a Sony.

I have put my trusty P45+ in a drawer a month ago, when I got the A7R II.
Never looked back. The IQ is better on Sony, even indeed not day and night better.
But Live View, High ISO, autofocus, FPS: NO CONTEST.

I had used P45+ on Sinar P2, now I use Sony on an Arca M, same lenses, Rodenstocks 55, 105mm, etc.
I'll try to post some examples, but I have no time right now.

[fdisilvestro]

In this case I don't feel we see anything near a three stop difference, what I see is that the difference is pretty close to the expected difference of around 1.5 stops.

Agreed, thanks for providing the images

I am not an expert at all these measuring festivities, but I am not seeing a point of reference where focus is spot on, from where the viewer can see how far the DoF spreads. There is a ruler and there is a cup, but there is no indication on how far the DoF spreads from the ruler to the cup. I would most definitely choose a different test scene for a test like this.

In the link provided by Erik you can find the whole images where you can see the focused area. Having said that, I agree that the setup can be improved.

Regards

[torger]

Do you think the Sony is a good enough replacement ? I've been thinking of upgrading my aptus65, that or just ditch the whole system for a Sony.

It depends on your use case. How are you using your Aptus65 today? If you like your SLR camera body and lenses you can't attach a Sony to it. If you do product photography in studio on some sort of view camera and don't need wide angle shots, you have a good case for replacing it with Sony. On the other hand if you need to get adapters perhaps a new body anyway etc the Sony upgrade can be more expensive than getting a second hand MFDB, if you aim at a milder upgrade like a P40+ or Aptus-II 8.

[ErikKaffehr]

Hi,

This is a study of DoF so sharpness at optimum focus point should not be of major interest. A low level of sharpening was applied to both images. The raw images are available for download, so you can see how the five cups are ordered.

But I realised I can improve on the test and will post new images soon, but that will not affect the results.

Best regards
Erik

I am not an expert at all these measuring festivities, but I am not seeing a point of reference where focus is spot on, from where the viewer can see how far the DoF spreads. There is a ruler and there is a cup, but there is no indication on how far the DoF spreads from the ruler to the cup. I would most definitely choose a different test scene for a test like this.

Also, I am sure many have mentioned this before, but Erik, your P45* images are far softer than they have any right to be.

[ErikKaffehr]

Hi,

I did some improvements to the setup. Moved the ruler two cm forward, so it is in the intended line of focus. It's intention is solely to be able to measure magnification, BTW. There is also a LensAlign ruler in the image, but that was almost perpendicular to lens axis. In the next series of images it will have a larger tilt.

Best regards
Erik

Agreed, thanks for providing the images

In the link provided by Erik you can find the whole images where you can see the focused area. Having said that, I agree that the setup can be improved.

Regards

[Ken R]

Im not gonna start questioning the setup. I am assuming camera/subject position and focus point is the same and also equivalent angle of view between both systems.

So Erik you would say there is about a stop difference? (to get similar dof in both systems)

[Bart_van_der_Wolf]

Im not gonna start questioning the setup. I am assuming camera/subject position and focus point is the same and also equivalent angle of view between both systems.

So Erik you would say there is about a stop difference? (to get similar dof in both systems)

Which would be in line with the calculations ..., no surprise.

Cheers,
Bart

[ErikKaffehr]

Hi Ken,

It is hard to say. Theory says about 1.5 stop and I would say may images are consistent with that. There are some rendering differences. The Sony lens has a tendency to double contours on structures in the front of the focus point. Behind focus blur may be a bit smoother with the Sony. The Sony has circular aperture while the Planar has a hexagonal or pentagonal shape. Focusing the Planar is not easy, it doesn't snap in and out of focus.

It is the second version of the Macro Planar I have and I guess that I have gone trough something like 8-9 Hasselblad lenses until I settled on my collection, so I don't think I have experienced nine bad lenses.

What I clearly don't see is a three stops difference or any magic properties.

I have made very few direct comparisons on the Blad vs. the Sony A7rII. What I have observed was that the Sony was doing well with most of my old lenses. Of the actual comparisons I shot between the P45+ and the Sony A7rII, it is 2:1 for the Hasselblad. The three shots were Distagon 40/4CF vs Canon 24 TSE LII, Planar 120/4 Macro vs Sony 90/2.8 G Macro (on test charts) and Planar 100/3.5 vs the Sony 90/2.8G Macro.

My take right now is that the systems are close. The Sony has live view, zoom lenses and flexibility. So when it matters I will go with the Sony. But, I enjoy shooting with the Hasselblad, so I am going to keep it. Well, unless someone is giving me 15 kUSD for a Hasselblad/P45+ kit...

Best regards
Erik

Im not gonna start questioning the setup. I am assuming camera/subject position and focus point is the same and also equivalent angle of view between both systems.

So Erik you would say there is about a stop difference? (to get similar dof in both systems)

[ErikKaffehr]

Hi,

I don't know. Your mileage may vary. What I have seen is that the Hasselblad/ELD555  P45+ combo still can yield sharper images than my Sony A7rII at least with the best lenses under optimal conditions. The Hasselblad is difficult to focus (I am using a 3X monocular). The Sony has excellent AF, covering most of the image. The Hasselblad needs mirror lockup while the Sony eliminates vibrations by large.

I will keep the Blad, unless someone wants to buy it for very decent money, but the Sony's will be my workhorse.

Best regards
Erik

Do you think the Sony is a good enough replacement ? I've been thinking of upgrading my aptus65, that or just ditch the whole system for a Sony.

[rueyloon]

I'm a generalist photographer, different cameras for different types of shoots and I enjoy my work, having different gears keeps me challenged and refreshed.
The 5d MK III is the work horse, literally pays for itself every month. The MF set is when I'm in the mood or a client that asked for a MF camera shoot. (museums etc...)

Because of the current development in image quality, I felt things like shadow noise will stand out as a defect to clients. I don't want to be caught out in a situation where
while I'm shooting with a MF, and a client realises that any modern DSLR can have less noise.

I'm feeling paranoid because in the last shoot that I did, the client specifically picked out that the image is noisy because it was a very dark painting.
Another way I can solve this problem is to upgrade my flash instead so I can get more light into the museum....

Hi,

I don't know. Your mileage may vary. What I have seen is that the Hasselblad/ELD555  P45+ combo still can yield sharper images than my Sony A7rII at least with the best lenses under optimal conditions. The Hasselblad is difficult to focus (I am using a 3X monocular). The Sony has excellent AF, covering most of the image. The Hasselblad needs mirror lockup while the Sony eliminates vibrations by large.

I will keep the Blad, unless someone wants to buy it for very decent money, but the Sony's will be my workhorse.

Best regards
Erik

[torger]

I'm feeling paranoid because in the last shoot that I did, the client specifically picked out that the image is noisy because it was a very dark painting.
Another way I can solve this problem is to upgrade my flash instead so I can get more light into the museum....

Increasing the flash power seems like the right solution to me. Even if clients pixel peep I doubt they would care about the difference in noise between even an Aptus 65 and a Sony if the shot is well exposed. A painting can be dark, but it can't really have that wide dynamic range so with flash that allows ETTR exposure noise shouldn't be a problem. If I would guess that client just saw that there was noise in an absolute sense and didn't think about comparison with any other gear. Few clients have any specific knowledge in photographic gear.

[Bart_van_der_Wolf]

Another way I can solve this problem is to upgrade my flash instead so I can get more light into the museum....

Hi,

Or you can just take the average (or median) of multiple shots with the current flash, if noise reduction software (used in moderation) can't solve it.

Cheers,
Bart

[Bo Dez]

While the differences are there, I would not be in a rush to invest heavily in medium format right now.

[synn]

I'm a generalist photographer, different cameras for different types of shoots and I enjoy my work, having different gears keeps me challenged and refreshed.
The 5d MK III is the work horse, literally pays for itself every month. The MF set is when I'm in the mood or a client that asked for a MF camera shoot. (museums etc...)

Because of the current development in image quality, I felt things like shadow noise will stand out as a defect to clients. I don't want to be caught out in a situation where
while I'm shooting with a MF, and a client realises that any modern DSLR can have less noise.

I'm feeling paranoid because in the last shoot that I did, the client specifically picked out that the image is noisy because it was a very dark painting.
Another way I can solve this problem is to upgrade my flash instead so I can get more light into the museum....

As I mentioned in the other thread, try upgrading to the latest version of C1P if you havent already. That by itself makes a big difference to files form older backs.

I am also wondering if the concerns from your client are about the final output files or about what he/ she sees on the tethered laptop at the time of shooting.

If it is the former, try using a good noise reduction plug in (And also the NR controls in the latest C1P) to see if you can achieve the desired results. if it is the latter, that's a bit unfortunate as they are not judging your work based on the final output, but add a more aggressive NR setting for capture time to satisfy their peeping needs. Work on the files in a more structured way when you are editing.

Lastly, Adding more light is always a good idea. CCD backs love light and doing ETTR might just solve your problem.

Investing in new gear is always a big commitment and my policy is to only do it if youa re absolutely certain your current kit (HArdware and software) cannot achieve what you are aiming for.

[ErikKaffehr]

Hi,

I am pretty sure that modern CMOS sensors have cleaner deep shadows than MFDs, with the new CMOS backs as a shining exception.

But, even my elderly P45+ has little problems in most cases and I don't think you should ever get problems on paintings. Admittedly, I have been told that I have a pretty good P45+ sample.

Most of the noise should come from "shot noise", unless you underexpose, so adding more light is a good proposal. Ideally, exposure would be ETTR and I would strongly suggest that you try a tool like RawDigger on your raw files. That tool shows the original histogram in the raw image.

I would also guess that if you need resolution, a P65+ back with an excellent lens holds advantage over a 36-42 MP sensor with an excellent lens.

Best regards
Erik

As I mentioned in the other thread, try upgrading to the latest version of C1P if you havent already. That by itself makes a big difference to files form older backs.

I am also wondering if the concerns from your client are about the final output files or about what he/ she sees on the tethered laptop at the time of shooting.

If it is the former, try using a good noise reduction plug in (And also the NR controls in the latest C1P) to see if you can achieve the desired results. if it is the latter, that's a bit unfortunate as they are not judging your work based on the final output, but add a more aggressive NR setting for capture time to satisfy their peeping needs. Work on the files in a more structured way when you are editing.

Lastly, Adding more light is always a good idea. CCD backs love light and doing ETTR might just solve your problem.

Investing in new gear is always a big commitment and my policy is to only do it if youa re absolutely certain your current kit (HArdware and software) cannot achieve what you are aiming for.