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!
Leica 006 VS Fuji GFX 50
- Thread starter rollsman44
- Start date
MGrayson
Subscriber and Workshop Member
Sorry to address neither the 006 nor the GFX 50, but I've just received an 007. While I think the 006 is the most fun you can have with your clothes on, I'm coming to appreciate some of the advantages of the 007.
The first is Live View. Yes, the 006 has the best OVF on earth, but you can't always get your face up to it. Live View lets you see what you're aiming at, and allows for magnified MF or CDAF at any point in the frame. This is neither as fast nor as easy as with dedicated mirrorless cameras, but it also DOES still offer the OVF. For landscape, I'd say it's the best of both worlds.
The other advantage is that it turns the 30-90 zoom into an actually useful lens. What this lens lacks is low f-number and pleasing bokeh. The primes have those in spades. But if you want an "everything in focus" picture - not uncommon in landscapes - it works well. As others have said, from 30-60, it's dead sharp at all apertures. Longer than that, it needs a bit of stopping down. 90mm f/8 is splendid. This lens is a nightmare on the 006. I was about to sell it. But on the 007, it is a viable walk-around, and a fine detailed landscape lens.
Don't get me wrong, it's no replacement for the 24/45/100 trinity (or any other set of S primes you like), but it's a serious option.
Best,
Matt
(Oh, and all seven of my S lenses worked without failing when mounted on the 007 for the first time.)
The first is Live View. Yes, the 006 has the best OVF on earth, but you can't always get your face up to it. Live View lets you see what you're aiming at, and allows for magnified MF or CDAF at any point in the frame. This is neither as fast nor as easy as with dedicated mirrorless cameras, but it also DOES still offer the OVF. For landscape, I'd say it's the best of both worlds.
The other advantage is that it turns the 30-90 zoom into an actually useful lens. What this lens lacks is low f-number and pleasing bokeh. The primes have those in spades. But if you want an "everything in focus" picture - not uncommon in landscapes - it works well. As others have said, from 30-60, it's dead sharp at all apertures. Longer than that, it needs a bit of stopping down. 90mm f/8 is splendid. This lens is a nightmare on the 006. I was about to sell it. But on the 007, it is a viable walk-around, and a fine detailed landscape lens.
Don't get me wrong, it's no replacement for the 24/45/100 trinity (or any other set of S primes you like), but it's a serious option.
Best,
Matt
(Oh, and all seven of my S lenses worked without failing when mounted on the 007 for the first time.)
ErikKaffehr
Well-known member
Hi Bernard,
You have some issues with your math. It is quite true that low frequency detail dominates visual perception more like 20 lp/mm than higher frequencies. But there is not a factor of two between the A7rII and the P45+. It is more like 45%. The sensor area of the P45+ is 49 x 39 = 1813 square mm, while the Sony has a surface area of 24 x 36 = 864 mm, so the sensor area is 2.1 times. But resolution is a linear quantity, so it is lp/mm. So, the factor in resolution is sqrt (2.1) -> 1.45. So you would compare the 30 lp/mm on the P45 with 43.5 lp/mm on the Sony A7rII. If you check my curves, you will see that the Planar 100/3.5 has 70% MTF at 30 lp/mm. the Sony is around the same level at 43.5 lp/mm.
Here is a practical example:
P45+ at actual pixels (Planar 120/4 CF at f/11)
Sony A7rII at actual pixels (Sony A7rII, Sony 90/2.8G at f/8)
The two f-stops are essentially equivalent, they give the same DoF.
The artefacts I have shown with the P45+ are low frequency detail. So, you don't need to print very large to be able to see them. Nyquist is the same as the lp/mm. So, if you print an image at 180 lp/inch, you will be at the Nyquist limit. But, the important fact is that you want to have low MTF at Nyquist, it doesn't relate to print size.
So, what happen when you reduce pixel size? The pixel aperture will get smaller, and that increases the MTF of the sensor. The MTF of the lens at Nyquist will be reduced, and that reduces aliasing. So you end up with a sharper image with less artefacts.
In the sample images I have posted. it was quiet obvious. The IQ3100 MP showed no aliasing at all. That indicates that it transfers very little MTF at it's pixel size, which is around 4.5 microns. The Hasselblad X1D shows more aliasing than the Canon 5DsR? Why, because it has a sharper lens.
So, if you have a sharper lens, you would need smaller pixels to have an optimal match.
The reason I compare the Planar 100/3.5 with the Sony 90/2.8G lens is that I happen to own both, and just to say, the Planar 100/3.5 is a very good lens. It is as good as the HC 100/2.2. Send me a Leica S type 007 with a couple of lenses and I will be glad to shoot demo images with it.
That said, I am not in the lens testing business. Using test images from say DPReview has some advantage as they try to limit experimental errors. Just as an example, you can use a focusing rail for bracketing focusing in the lab, but it would not be very practical at longer distances.
Best regards
Erik
You have some issues with your math. It is quite true that low frequency detail dominates visual perception more like 20 lp/mm than higher frequencies. But there is not a factor of two between the A7rII and the P45+. It is more like 45%. The sensor area of the P45+ is 49 x 39 = 1813 square mm, while the Sony has a surface area of 24 x 36 = 864 mm, so the sensor area is 2.1 times. But resolution is a linear quantity, so it is lp/mm. So, the factor in resolution is sqrt (2.1) -> 1.45. So you would compare the 30 lp/mm on the P45 with 43.5 lp/mm on the Sony A7rII. If you check my curves, you will see that the Planar 100/3.5 has 70% MTF at 30 lp/mm. the Sony is around the same level at 43.5 lp/mm.
Here is a practical example:
P45+ at actual pixels (Planar 120/4 CF at f/11)
Sony A7rII at actual pixels (Sony A7rII, Sony 90/2.8G at f/8)
The two f-stops are essentially equivalent, they give the same DoF.
The artefacts I have shown with the P45+ are low frequency detail. So, you don't need to print very large to be able to see them. Nyquist is the same as the lp/mm. So, if you print an image at 180 lp/inch, you will be at the Nyquist limit. But, the important fact is that you want to have low MTF at Nyquist, it doesn't relate to print size.
So, what happen when you reduce pixel size? The pixel aperture will get smaller, and that increases the MTF of the sensor. The MTF of the lens at Nyquist will be reduced, and that reduces aliasing. So you end up with a sharper image with less artefacts.
In the sample images I have posted. it was quiet obvious. The IQ3100 MP showed no aliasing at all. That indicates that it transfers very little MTF at it's pixel size, which is around 4.5 microns. The Hasselblad X1D shows more aliasing than the Canon 5DsR? Why, because it has a sharper lens.
So, if you have a sharper lens, you would need smaller pixels to have an optimal match.
The reason I compare the Planar 100/3.5 with the Sony 90/2.8G lens is that I happen to own both, and just to say, the Planar 100/3.5 is a very good lens. It is as good as the HC 100/2.2. Send me a Leica S type 007 with a couple of lenses and I will be glad to shoot demo images with it.
That said, I am not in the lens testing business. Using test images from say DPReview has some advantage as they try to limit experimental errors. Just as an example, you can use a focusing rail for bracketing focusing in the lab, but it would not be very practical at longer distances.
Best regards
Erik
When i said «2014 sensor, futur is already gone» this was just a joke, because i felt you were maybe too exalted about your camera ....
But now, if you say : 2017 is the futur....then i might laugh again
ErikKaffehr
Well-known member
Hi,
My take is that a 37.5 MP pictures camera takes 37.5 weather the model has been updated to a higher resolution version or not.
It can also be said that that the difference between 37.5MP and 50 MP is not that large. It depends also on which aspect ratio a photographer prefers.
Smaller pixels have at least two advantages:
If we shoot stopped down beyond f/11, diffraction is the limiting factor. But even at smaller apertures, smaller pixels yield an advantage:
So, indications are that increasing resolution, within reasonable limits, is always beneficial.
In all probability, it is not more expensive to make a 100 MP sensor of say 44x33 mm size than an 50 MP sensor of the same size. The designs effort may be more expensive. Production may also be more expensive, if it needs more layers.
To me, the GFX 50 is a no brainer. It is the most affordable small medium format camera designed for the 44x33 mm sensor. That said, I am pretty sure that the 100 MP version will be a much more optimal and optically better balanced solution. So, if we are on a limited budget, it may make some sense to wait for it. But, it may be a long wait.
Weather Leica S2 or Fuji GFX is another question. I have no doubt the Leica S2 is capable of making very fine pictures. So is my Hasselblad 555/ELD and Phase One P45+ combo. My experience with the Hassy/P45+ combo is:
Best regards
Erik
My take is that a 37.5 MP pictures camera takes 37.5 weather the model has been updated to a higher resolution version or not.
It can also be said that that the difference between 37.5MP and 50 MP is not that large. It depends also on which aspect ratio a photographer prefers.
Smaller pixels have at least two advantages:
- They will deliver higher fine detail contrast.
- And they will create less sampling artefacts.
- In theory, smaller pixels give a small loss of (engineering) dynamic range, but photography is often more affected by shot noise (Poisson distribution of photons sampled) than by read noise.
- Vignetting and cross talk increase with smaller pixels. But neither is a problem on a DSLR design, like the S2. Backside illumination may help a bit on both.
If we shoot stopped down beyond f/11, diffraction is the limiting factor. But even at smaller apertures, smaller pixels yield an advantage:
- The pixel aperture is smaller and that increases microcontrast.
- Having more megapixels is helpful in sharpening.
So, indications are that increasing resolution, within reasonable limits, is always beneficial.
In all probability, it is not more expensive to make a 100 MP sensor of say 44x33 mm size than an 50 MP sensor of the same size. The designs effort may be more expensive. Production may also be more expensive, if it needs more layers.
To me, the GFX 50 is a no brainer. It is the most affordable small medium format camera designed for the 44x33 mm sensor. That said, I am pretty sure that the 100 MP version will be a much more optimal and optically better balanced solution. So, if we are on a limited budget, it may make some sense to wait for it. But, it may be a long wait.
Weather Leica S2 or Fuji GFX is another question. I have no doubt the Leica S2 is capable of making very fine pictures. So is my Hasselblad 555/ELD and Phase One P45+ combo. My experience with the Hassy/P45+ combo is:
- Focusing is difficult.
- The system is a bit limited in DR.
- Aliasing is a very frequent problem and does not play well with things I shoot.
- Few Hassy pictures make it to the wall.
- To much focus on technique when shooting an image.
- The Hasselblad I use with primes, while the 24x36 systems I almost exclusively use with zooms. Zooms give me the liberty of choosing the best point of view. The primes I use with the Hassy are somewhat limiting. With the Hassy I would very often stitch.
- Stitching gives optimal results, mostly.
- But, stitching also means that I postpone a great part of the creative process until processing the images. Some spontaneity have been lost.
- Shooting MFD, I raise the ribbon.
Best regards
Erik
When i said «2014 sensor, futur is already gone» this was just a joke, because i felt you were maybe too exalted about your camera ....
But now, if you say : 2017 is the futur....then i might laugh again
Last edited:
Erik -
You’ve long been unhappy with your Hassy/p45 setup, for reasons you have made clear many times. My own experiences with MFDigital gear run the other way, as it gives this user much pleasure.
I agree that focus and alignment are key. And we know older systems are often stretched to their limits. As your leading frustration is difficulty with focusing, why not try another system or setup and see how it feels? Just rent something else, give it a whirl. You might be surprised.
Geoff
Erik,
I acknowledge that my "back-of-the-envelope" maths were off by a factor of √2 . The main points still stand.
You are using a 50-year-old medium format lens as your reference, and comparing it to modern designs. The Planar 3.5/100 was used on the Apollo Moon missions, so it was certainly a good lens at the time, but it shouldn't be your first choice today for maximum resolution or contrast. It's a very good choice for other reasons.
You are very preoccupied by response at the Nyquist frequency. I think that may be a symptom of printing too large, or looking too close (in print or on screen). There's an old English idiom: "Can't see the forest for the trees," which applies (in my opinion).
There is an artistic tradition of exploring the point at which a medium breaks down, which may be what you are doing. The movie Blow-Up comes to mind, where a greatly expanded photographic detail takes-on an aesthetic (and narrative) quality which was not evident in the original image.
It has been my experience that medium format imaging offers greater micro-contrast in finished prints, when comparing prints of similar size.
I acknowledge that my "back-of-the-envelope" maths were off by a factor of √2 . The main points still stand.
You are using a 50-year-old medium format lens as your reference, and comparing it to modern designs. The Planar 3.5/100 was used on the Apollo Moon missions, so it was certainly a good lens at the time, but it shouldn't be your first choice today for maximum resolution or contrast. It's a very good choice for other reasons.
You are very preoccupied by response at the Nyquist frequency. I think that may be a symptom of printing too large, or looking too close (in print or on screen). There's an old English idiom: "Can't see the forest for the trees," which applies (in my opinion).
There is an artistic tradition of exploring the point at which a medium breaks down, which may be what you are doing. The movie Blow-Up comes to mind, where a greatly expanded photographic detail takes-on an aesthetic (and narrative) quality which was not evident in the original image.
It has been my experience that medium format imaging offers greater micro-contrast in finished prints, when comparing prints of similar size.
I made the choice of digital back for the flexibility across camera systems. I use the digital back for the XF and two Arca systems, the RM3D and 4x5 Large format setup. The flexibility of lens combinations allows you choose which lens to draw with and not be stuck to the prefix menu of a single camera system.
On the large format setup, not only do I use an excellent 4x5 Macro lense, but also a Cooke PS945 and a couple of very old 1910 lenses. These are in addition to a 1950's Kodak Ektar glamour lens a Fuji lens etc. The back can change over time, but the analogue portion of the gear can be used for decades. You can pretty well future proof your gear setup.
On the large format setup, not only do I use an excellent 4x5 Macro lense, but also a Cooke PS945 and a couple of very old 1910 lenses. These are in addition to a 1950's Kodak Ektar glamour lens a Fuji lens etc. The back can change over time, but the analogue portion of the gear can be used for decades. You can pretty well future proof your gear setup.
ErikKaffehr
Well-known member
An interesting megapixels vs. detail comparison
Hi Bernard,
Aliasing and Nyquist don't depend on print size. Also, modern printers print at 300 to 720 PPI, internally. I do occasionally print at 70x100 cm. A 70x100 cm print at 300 PPI corresponds to 97 megapixels. So, that is no excessive size. My point is that anything you see in the image will be present in a print, at least using today's high resolution printers.
The samples i have shown illustrate that the "50 year" old lens produces plenty of aliasing artefacts. Replacing it with a new design, would just increase it.
The point I make is easily illustrated. Check the image below:
The top left image is Hasselblad X1D, with the very good 90 mm lens. On the right is the Canon 5DsR with the cheap 85/1.8 lens, bottom row is Leica M-10 with the Apo Summicron 90/2 and on the right the Sony A7rIII with a cheap Sony 85/1.8 lens. All images are downscaled to 24 MP, so they match the Leica M10. So Leica M is native resolution, while the others would be downscaled.
What I see here is that the X1D produces the clearest image, due to it's sensor size combined with extremely good optics. But, colour artefacts show up early. The Canon image is less crisp, but it has far less colour artefacts. The reason is that it has smaller pixels that are a better match for the sensor.
The Leica M-10 image is not very crisp, although it is the only image at native resolution. But, it has a lot of colour artefacts. The Sony A7rIII image is on par with the Canon.
Leica seems to consider 24 MP enough for 24x36 mm. The samples I show clearly indicate that it is not the case.
Leica also seems to think that 37.5 MP is good enough for medium format. The P45+ I have was regarded king of medium format in 2007, it is still highly regarded, due to it's long exposure capability. But it has been supplanted by 60, 80 and 100 MP systems.
I do actually think that medium format needs to go 100 MP to really play it's benefits, at 50 MP it will produce a lot of artefacts when put in optimal use.
If you stop down to f/11 to f/16 on a Leica S/S2 you will get rid of most artefacts, but you will also loose sharpness.
Also, it seems that at least some photographers have issues with getting pin-point accurate focus on the Leica S. Diglloyd reported that issue on at least four different bodies and in one case confirmed it with a Leica representative. Another poster here, "Chrismuc" indicated the same problem. He switched to the Fuji GFX and he is a happy camper with the Fuji.
Focusing issues may mask the large pixels/coarse sampling problem.
It is quite possible that Leica will present a 70 or 100 MP version of the Leica S at Photokina or something like an 50 MP version of the Leica M. In my humble opinion that is something they need to do.
But, it seems that Leica is in a state of denial about the benefits of sensor pixel sizes matching the capabilities of their lenses.
One aspect that should not be forgotten is that the plane of accurate focus is very thin, say 1cm at 100 cm distance. That means that many subjects will be more or less out of focus. So, out of focus rendition may be more important than in focus rendition.
Brandon Dube, a young optical engineer who writes a lot about lenses, gives a lot of credit to Leica for designing lenses that render well. it means that they may leave some third degree aberrations undercorrected so that it will play better with fifth degree aberrations, he is probably right. I am not an optical engineer, my field used to be reactor physics.
For folks like me, who shoot landscapes and architecture, out of focus rendition may play a lesser role. Landscape is often at near infinity and architecture is often essentially flat.
I may add that I am bit sensitive to colour aliasing, one reason is that I live near the seashore. The one thing that always show colour aliasing is sailboat rigs. Those are thin and made of steel. If I shoot a landscape with a sailboat in it, the P45+ will show a lot of fake colour.
Another problem is that rippled water will very often have severe aliasing artefacts. When I was shooting both 39MP Hasselblad and the 24MP Sony Alpha 99, I would not really be able to tell them apart in A2 size prints, except for one thing. I could look for aliasing artefacts. With the 39MP Hasselblad, some aliasing could always be found.
Interestingly, the A7rII (42MP) also produces some aliasing, which the A99 not really did. the A99 had a proper OLP filter while the A7rII does not.
Just to say, my Zeiss Planar 100/3.5 CF was made 1984, and the 120/4 CFi was made 1998, so those lenses are nowhere like fifty years old.
Best regards
Erik
Hi Bernard,
Aliasing and Nyquist don't depend on print size. Also, modern printers print at 300 to 720 PPI, internally. I do occasionally print at 70x100 cm. A 70x100 cm print at 300 PPI corresponds to 97 megapixels. So, that is no excessive size. My point is that anything you see in the image will be present in a print, at least using today's high resolution printers.
The samples i have shown illustrate that the "50 year" old lens produces plenty of aliasing artefacts. Replacing it with a new design, would just increase it.
The point I make is easily illustrated. Check the image below:
The top left image is Hasselblad X1D, with the very good 90 mm lens. On the right is the Canon 5DsR with the cheap 85/1.8 lens, bottom row is Leica M-10 with the Apo Summicron 90/2 and on the right the Sony A7rIII with a cheap Sony 85/1.8 lens. All images are downscaled to 24 MP, so they match the Leica M10. So Leica M is native resolution, while the others would be downscaled.
What I see here is that the X1D produces the clearest image, due to it's sensor size combined with extremely good optics. But, colour artefacts show up early. The Canon image is less crisp, but it has far less colour artefacts. The reason is that it has smaller pixels that are a better match for the sensor.
The Leica M-10 image is not very crisp, although it is the only image at native resolution. But, it has a lot of colour artefacts. The Sony A7rIII image is on par with the Canon.
Leica seems to consider 24 MP enough for 24x36 mm. The samples I show clearly indicate that it is not the case.
Leica also seems to think that 37.5 MP is good enough for medium format. The P45+ I have was regarded king of medium format in 2007, it is still highly regarded, due to it's long exposure capability. But it has been supplanted by 60, 80 and 100 MP systems.
I do actually think that medium format needs to go 100 MP to really play it's benefits, at 50 MP it will produce a lot of artefacts when put in optimal use.
If you stop down to f/11 to f/16 on a Leica S/S2 you will get rid of most artefacts, but you will also loose sharpness.
Also, it seems that at least some photographers have issues with getting pin-point accurate focus on the Leica S. Diglloyd reported that issue on at least four different bodies and in one case confirmed it with a Leica representative. Another poster here, "Chrismuc" indicated the same problem. He switched to the Fuji GFX and he is a happy camper with the Fuji.
Focusing issues may mask the large pixels/coarse sampling problem.
It is quite possible that Leica will present a 70 or 100 MP version of the Leica S at Photokina or something like an 50 MP version of the Leica M. In my humble opinion that is something they need to do.
But, it seems that Leica is in a state of denial about the benefits of sensor pixel sizes matching the capabilities of their lenses.
One aspect that should not be forgotten is that the plane of accurate focus is very thin, say 1cm at 100 cm distance. That means that many subjects will be more or less out of focus. So, out of focus rendition may be more important than in focus rendition.
Brandon Dube, a young optical engineer who writes a lot about lenses, gives a lot of credit to Leica for designing lenses that render well. it means that they may leave some third degree aberrations undercorrected so that it will play better with fifth degree aberrations, he is probably right. I am not an optical engineer, my field used to be reactor physics.
For folks like me, who shoot landscapes and architecture, out of focus rendition may play a lesser role. Landscape is often at near infinity and architecture is often essentially flat.
I may add that I am bit sensitive to colour aliasing, one reason is that I live near the seashore. The one thing that always show colour aliasing is sailboat rigs. Those are thin and made of steel. If I shoot a landscape with a sailboat in it, the P45+ will show a lot of fake colour.
Another problem is that rippled water will very often have severe aliasing artefacts. When I was shooting both 39MP Hasselblad and the 24MP Sony Alpha 99, I would not really be able to tell them apart in A2 size prints, except for one thing. I could look for aliasing artefacts. With the 39MP Hasselblad, some aliasing could always be found.
Interestingly, the A7rII (42MP) also produces some aliasing, which the A99 not really did. the A99 had a proper OLP filter while the A7rII does not.
Just to say, my Zeiss Planar 100/3.5 CF was made 1984, and the 120/4 CFi was made 1998, so those lenses are nowhere like fifty years old.
Best regards
Erik
ErikKaffehr
Well-known member
Hi Geoff,
I use the Hasselblad with a 3X monocular from Zeiss, that gives me a focusing precision like +/- 2 cm at 2m. The major issue is that it does not have a usable magnified live view.
It is not that I am unhappy with the Hasselblad. The focusing is an issue, but using the Zeiss Monocular helps a lot. If I have something easily focused in the subject, like a flagpole, I can achieve pinpoint accurate focus.
The GFX, which is discussed here, has probably the best focusing system of any camera. Yes, it has an issue with focus shift, but you can focus it stopped down with actual pixels view in the viewfinder. That is a great advantage.
Which MF system would I buy to have accurate focus?
I have the Hasselblad 555/ELD, that camera was enhanced for digital backs and has tighter tolerances than other models.
I would add that one of the reasons that the Hasselblad is difficult to focus is the amount of barrel rotation. I can put a marker on the lens barrel and focus five times. The marking is within 1 mm, but sharpness differs significantly.
It is quite possible that some people don't regard money as a finite asset, but I do. Discussing what you get for you money is a bit relevant. For the price of an Hasselblad H6D50c I with 5 lenses I can make 6-10 perfectly fine trips to US-national parks or Iceland from here in Sweden.
The final point is tolerances. Hasselblad's criteria on film/flange alignment used to be 0.03 mm, that is 30 microns. Phase One's criteria used to be 0.012 mm, that is 12 microns. So tolerance on the system is +/- 42 microns. If you use it at f/2.8, the corresponding CoC is 0.015 mm and that corresponds to 3.8 pixels. So, factory tolerances reduce my 39 MP to around 10 MP. This excludes alignment on lens.
Back in 2009, it seems that problems were really bad in MFD alignment accuracy:
https://www.josephholmes.com/news/2009/4/5/medium-format-digital-camera-optical-precision
Hopefully it is better now. But, those issues were not really discussed than and they are not really discussed now.
Joseph Holmes actually wrote a companion article to the quoted one, discussing achieving optimum quality in MFD:
https://www.josephholmes.com/news/2009/4/9/getting-top-quality-from-medium-format
In that article he suggested using the 3X Zeiss monocular I also use.
The simplest ways to achieve accurate focus is to:
With live view or CDAF, you are actually using the sensor for focusing and it eliminates all tolerances in the point of focus. It doesn't eliminate a tilted focal plane or field curvature, but it at least guarantee correct focus at the focusing point. That is not bad for a starter.
The GFX has all that. It also seems to have excellent lenses and 100 MP is around the corner.
BTW, I was thinking about this thread regarding focusing issues on the Leica:
https://www.getdpi.com/forum/medium...60441-leica-007-warning-potential-buyers.html
As far as I recall, he got a lot of support from Leica and dealer but his issues were not resolved.
Best regards
Erik
I use the Hasselblad with a 3X monocular from Zeiss, that gives me a focusing precision like +/- 2 cm at 2m. The major issue is that it does not have a usable magnified live view.
It is not that I am unhappy with the Hasselblad. The focusing is an issue, but using the Zeiss Monocular helps a lot. If I have something easily focused in the subject, like a flagpole, I can achieve pinpoint accurate focus.
The GFX, which is discussed here, has probably the best focusing system of any camera. Yes, it has an issue with focus shift, but you can focus it stopped down with actual pixels view in the viewfinder. That is a great advantage.
Which MF system would I buy to have accurate focus?
I have the Hasselblad 555/ELD, that camera was enhanced for digital backs and has tighter tolerances than other models.
- Diglloyd has reported that it is not possible to focus the Pentax 645D reliably.
- Diglloyd also found that at least four samples of the Leica S2/S could not achieve correct focus. he tested four samples.
- 'Chrismuc' indicated that he found the Leica S inaccurate in focus, here on this thread. He now shoots the GFX and is extremely both with it and it's lenses.
- I don't know if the focusing system on the Phase One systems is accurate.
- I recall another poster discussing AF accuracy on Leica S - he has simply given up on his Leica system. His system was serviced a couple of times.
I would add that one of the reasons that the Hasselblad is difficult to focus is the amount of barrel rotation. I can put a marker on the lens barrel and focus five times. The marking is within 1 mm, but sharpness differs significantly.
It is quite possible that some people don't regard money as a finite asset, but I do. Discussing what you get for you money is a bit relevant. For the price of an Hasselblad H6D50c I with 5 lenses I can make 6-10 perfectly fine trips to US-national parks or Iceland from here in Sweden.
The final point is tolerances. Hasselblad's criteria on film/flange alignment used to be 0.03 mm, that is 30 microns. Phase One's criteria used to be 0.012 mm, that is 12 microns. So tolerance on the system is +/- 42 microns. If you use it at f/2.8, the corresponding CoC is 0.015 mm and that corresponds to 3.8 pixels. So, factory tolerances reduce my 39 MP to around 10 MP. This excludes alignment on lens.
Back in 2009, it seems that problems were really bad in MFD alignment accuracy:
https://www.josephholmes.com/news/2009/4/5/medium-format-digital-camera-optical-precision
Hopefully it is better now. But, those issues were not really discussed than and they are not really discussed now.
Joseph Holmes actually wrote a companion article to the quoted one, discussing achieving optimum quality in MFD:
https://www.josephholmes.com/news/2009/4/9/getting-top-quality-from-medium-format
In that article he suggested using the 3X Zeiss monocular I also use.
The simplest ways to achieve accurate focus is to:
- Use live view with maximum magnification or contrast detecting AF.
- In either case, focus at shooting aperture, so focus shift when stopping down is avoided
- Jim Kasson has found that peaking is very usable with magnified live view. My lenses are not sharp enough for that.
With live view or CDAF, you are actually using the sensor for focusing and it eliminates all tolerances in the point of focus. It doesn't eliminate a tilted focal plane or field curvature, but it at least guarantee correct focus at the focusing point. That is not bad for a starter.
The GFX has all that. It also seems to have excellent lenses and 100 MP is around the corner.
BTW, I was thinking about this thread regarding focusing issues on the Leica:
https://www.getdpi.com/forum/medium...60441-leica-007-warning-potential-buyers.html
As far as I recall, he got a lot of support from Leica and dealer but his issues were not resolved.
Best regards
Erik
ErikKaffehr
Well-known member
A small demo of focusing accuracy...
Hi,
I am a strong believer in using magnified live view for accurate focus. Here is a nifty example of how useful it can be.
In this case, I prepared a comparison of bokeh on my Hasselblad 555/ELD and Sony A7rII using equivalent apertures.(*)
To do this, I had two puppies positioned in and out of focus. Obviously, I needed to have the same point of focus in both images. I wanted to focus on the nearest eye on the puppy in front. The Hasselblad I used with a 3X-loupe I normally use. I could not focus on the eye, because there is not enough detail, so I focused on strain of 'fur' right in front of it. So, I came up with this:
The whole image was like this:
Now, I shot the same image on the A7rII using magnified live view. So I needed to decide which strain of 'fur' i would focus on:
For the final posting I reshot the A7rII image, with what I thought was the focus of the P45+ shot:
So, what I see is that doing focusing at magnified live view gives you a significant advantage in both accuracy and selectivity. I can not really talk about AF-accuracy, as I normally use manual focus when I am shooting on tripod, which I prefer to do.
The other point I seldom miss to make is that large pixels are more prone to aliasing than small ones. This is really a consequence of math, but it seems that many posters don't believe it. Here is a nice sample illustrating the issue:
It was shot on the P45+ with a Distagon 40/CF in front. That combo is good enough to create a lot of aliasing near axis.
Just to say, even the A7rII I normally shoot can show aliasing issues. Personally, I feel that a 42 MP sensor still needs a OLP (anti aliasing) filter when combined with decent quality lenses. Or, with other words, Sony needs a 70-100 MP sensor on the 24x36 mm cameras, at least if they want to sell a new camera to this photographer.
Best regards
Erik
(*) https://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care
Hi,
I am a strong believer in using magnified live view for accurate focus. Here is a nifty example of how useful it can be.
In this case, I prepared a comparison of bokeh on my Hasselblad 555/ELD and Sony A7rII using equivalent apertures.(*)
To do this, I had two puppies positioned in and out of focus. Obviously, I needed to have the same point of focus in both images. I wanted to focus on the nearest eye on the puppy in front. The Hasselblad I used with a 3X-loupe I normally use. I could not focus on the eye, because there is not enough detail, so I focused on strain of 'fur' right in front of it. So, I came up with this:
The whole image was like this:
Now, I shot the same image on the A7rII using magnified live view. So I needed to decide which strain of 'fur' i would focus on:
For the final posting I reshot the A7rII image, with what I thought was the focus of the P45+ shot:
So, what I see is that doing focusing at magnified live view gives you a significant advantage in both accuracy and selectivity. I can not really talk about AF-accuracy, as I normally use manual focus when I am shooting on tripod, which I prefer to do.
The other point I seldom miss to make is that large pixels are more prone to aliasing than small ones. This is really a consequence of math, but it seems that many posters don't believe it. Here is a nice sample illustrating the issue:
It was shot on the P45+ with a Distagon 40/CF in front. That combo is good enough to create a lot of aliasing near axis.
Just to say, even the A7rII I normally shoot can show aliasing issues. Personally, I feel that a 42 MP sensor still needs a OLP (anti aliasing) filter when combined with decent quality lenses. Or, with other words, Sony needs a 70-100 MP sensor on the 24x36 mm cameras, at least if they want to sell a new camera to this photographer.
Best regards
Erik
(*) https://www.dpreview.com/articles/2666934640/what-is-equivalence-and-why-should-i-care
Attachments
There are many terrific things about the GFX. The manual focusing system using magnified live view is about the worst I have ever used. The magnified live view is very coarse and it was impossible for me to tell when I achieved ideal focus on my chose spot. Strange, because the specs of the EVF appear to be quite good. On paper, better than the X1D.
Have you actually focused the GFX and compared it to a Sony A7RII or the X1D?
MGrayson
Subscriber and Workshop Member
Two of my earliest captures with the Leica S(006).
It was the perfectly natural rendering of ordinary objects that first drew me to the S. (Duck decoys, if I recall correctly....)
My daughter's skin would show up very mottled red on most systems. The S captured her quite naturally.
Best,
Matt
ErikKaffehr
Well-known member
Hi Howard,
Now, I did neither use the GFX or the X1D. But, I communicate frequently with Jim Kasson who is very happy with the GFX. He uses magnified live view for focusing, combined with peaking. Jim also shoots Sony A7rII/A9/A7rIII, Leica M, Hasselblad H3D and some other stuff. Great scientist.
Personally, I feel that magnified live view is about the most important feature of any camera.
EVF design is probably quiet a bit about firmware. You need to present an image that is good enough on a limited budget of power.
Thanks to Jim, I have tons of information on the GFX, I don't have that kind of input for the X1D. Rumor here in Sweden says that Hasselblad was in a very bad shape a few years ago, but I hope they are doing well now.
I am much interested in what I would call the new medium format, or small medium format. It makes a lot of sense to select a format for a sensor and build a system for it, instead of building a system around a cropped version of old film formats. So, both the GFX and the X1D makes a lot of sense for me.
By incident, I plan to visit a place selling Hasselblad next Monday, so I hope I can have a look at the X1D.
As things are now, I don't see myself as a potential buyer of MFD, right now. The Sony A7rII I have fulfills my needs. The same goes for the A7rIII. I think it is much improved, but I don't think it will yield significantly better images. So I put my money elsewhere, like travel.
Just to say, I am an amateur photographer, but I am also interested in imaging science.
Best regards
Erik
Now, I did neither use the GFX or the X1D. But, I communicate frequently with Jim Kasson who is very happy with the GFX. He uses magnified live view for focusing, combined with peaking. Jim also shoots Sony A7rII/A9/A7rIII, Leica M, Hasselblad H3D and some other stuff. Great scientist.
Personally, I feel that magnified live view is about the most important feature of any camera.
EVF design is probably quiet a bit about firmware. You need to present an image that is good enough on a limited budget of power.
Thanks to Jim, I have tons of information on the GFX, I don't have that kind of input for the X1D. Rumor here in Sweden says that Hasselblad was in a very bad shape a few years ago, but I hope they are doing well now.
I am much interested in what I would call the new medium format, or small medium format. It makes a lot of sense to select a format for a sensor and build a system for it, instead of building a system around a cropped version of old film formats. So, both the GFX and the X1D makes a lot of sense for me.
By incident, I plan to visit a place selling Hasselblad next Monday, so I hope I can have a look at the X1D.
As things are now, I don't see myself as a potential buyer of MFD, right now. The Sony A7rII I have fulfills my needs. The same goes for the A7rIII. I think it is much improved, but I don't think it will yield significantly better images. So I put my money elsewhere, like travel.
Just to say, I am an amateur photographer, but I am also interested in imaging science.
Best regards
Erik
I am torn between both-EVF and OVF.
OVF needs a reliable AF system. If it works phase AF has an advantage as soon as things move.
I have very good experience with the S AF accurancy. However I see another major advantage of the Fuji (or Hassy) that one can move the focus point around.
EVF has also increased my first image keeper rate in regards of exposure (at least with the SL, I have not yet enough experience with the x1d to judge).
On the other side looking in an EVF also means to not see the real light and to not see fine details. Examples: I saw a bird in long distance on a electric wire in fog. looked good and I wanted to take an image what I did. Looking through the EVF there was not enough detail to realize the bird was gone when I took the image. The image was useless, what I found out at home.
Things like this have happed more than once to me. Also in regards of light. Special light, it takes me some time to frame, and through the EVF I can not see if an dhow the light changes.
I think it is a different style of shooting. With the OVF you can see more through the viewfinder while taking the image, with EVF you need to see and imagine the image before taking the camera to the eye, and then use the EVF just for framing and timing the right moment. In an OVF you see the original light real time, with an EVF you see the image, but in low resolution and with low dynamic range and reduced color depth.
I must say that since the SL I am not prefering one over the other in general. I can get along with both types.
I can imaginge for people putting the camera on a tripod and taking long time for an image an EVF/live view might be the better option.
OVF needs a reliable AF system. If it works phase AF has an advantage as soon as things move.
I have very good experience with the S AF accurancy. However I see another major advantage of the Fuji (or Hassy) that one can move the focus point around.
EVF has also increased my first image keeper rate in regards of exposure (at least with the SL, I have not yet enough experience with the x1d to judge).
On the other side looking in an EVF also means to not see the real light and to not see fine details. Examples: I saw a bird in long distance on a electric wire in fog. looked good and I wanted to take an image what I did. Looking through the EVF there was not enough detail to realize the bird was gone when I took the image. The image was useless, what I found out at home.
Things like this have happed more than once to me. Also in regards of light. Special light, it takes me some time to frame, and through the EVF I can not see if an dhow the light changes.
I think it is a different style of shooting. With the OVF you can see more through the viewfinder while taking the image, with EVF you need to see and imagine the image before taking the camera to the eye, and then use the EVF just for framing and timing the right moment. In an OVF you see the original light real time, with an EVF you see the image, but in low resolution and with low dynamic range and reduced color depth.
I must say that since the SL I am not prefering one over the other in general. I can get along with both types.
I can imaginge for people putting the camera on a tripod and taking long time for an image an EVF/live view might be the better option.
It is certainly a very key feature, and that's why I was so surprised by how poorly implemented it is in the GFX, which has an EVF that on paper is superior to the EVF in the X1D. When you look through the GFX viewfinder with magnified live view and compare it to the X1D or the Sony A7rII, the difference is not subtle. It is dramatic. I remember trying to focus the GFX and struggling to see the optimum focus point. I do recall Jim Kasson stating that he "solved" the issue with live view in the GFX by turning on peaking at the same time. I haven't tried it myself, but that strikes me as a kludge that may not work well in many situations. Personally, I would not like having peaking on all the time in my viewfinder. (I assume you can't set it up so that it only comes on when you go into magnified live view; with the X1D, if you have peaking on, it goes OFF when you go into live view.) I find it compromises my ability/enjoyment of the composition process as I look through a viewfinder. Peaking also does not seem to work well in low light or low contrast conditions, and you would probably have to adjust the sensitivity in certain situations.
When you visit the Hasselblad dealer, take a look at the magnified live view with the X1D. I think it works very well and is quite comparable to the A7RII.
ErikKaffehr
Well-known member
Hi,
Yes, it is my way of shooting. One advantage with EVF is that you can actually see what you are shooting in dark places.
The other thing is that EVF is improving. A few months ago, Sony has released the A9 that offers continuous view at 20 FPS without viewfinder black out. It was conceived as a major improvement over OVF.
The clue is that the A9 has an "front side ASIC" that takes care of the signals from the sensor combined with a multi layered sensor that does readout in parallell.
The single lens reflex concept has been around for over 100 years and gone trough a lot of refinement in that period. EVF has been around for say 20 years but made rapid development in the recent few years. In all probability EVF will be better in the coming years.
Just as an example, the Sony A7rII is known for eating batteries. The A9 has a battery with twice the capacity, but users say they can shoot all day on a single battery.
Douglas Herr is a very fine photographer of birds who used to shoot Leica R9 with Leica's digital back. It seems that he is a quite happy camper with Sony A7rII (?), but he feels that the A9 lacks the resolution to handle aliasing on his Leica 280 APO. He is well known for his photographs of birds in flight, I don't know how it works with the EVF of the Sony systems.
Anyway, EVF is getting better while OVF may actually going into the opposite direction. It seems that modern OVF systems prioritize brightness over focusing accuracy. That said, my Hasselblad 555/ELD is not an up to date system, but it feels like a dark hole. I also have a Sony A900, which happens to have a very bright viewfinder, but there is no way I can switch to 11X magnification on either.
So, I am pretty sure that EVF is where the future is, but that future may not be here and now.
Best regards
Erik
Yes, it is my way of shooting. One advantage with EVF is that you can actually see what you are shooting in dark places.
The other thing is that EVF is improving. A few months ago, Sony has released the A9 that offers continuous view at 20 FPS without viewfinder black out. It was conceived as a major improvement over OVF.
The clue is that the A9 has an "front side ASIC" that takes care of the signals from the sensor combined with a multi layered sensor that does readout in parallell.
The single lens reflex concept has been around for over 100 years and gone trough a lot of refinement in that period. EVF has been around for say 20 years but made rapid development in the recent few years. In all probability EVF will be better in the coming years.
Just as an example, the Sony A7rII is known for eating batteries. The A9 has a battery with twice the capacity, but users say they can shoot all day on a single battery.
Douglas Herr is a very fine photographer of birds who used to shoot Leica R9 with Leica's digital back. It seems that he is a quite happy camper with Sony A7rII (?), but he feels that the A9 lacks the resolution to handle aliasing on his Leica 280 APO. He is well known for his photographs of birds in flight, I don't know how it works with the EVF of the Sony systems.
Anyway, EVF is getting better while OVF may actually going into the opposite direction. It seems that modern OVF systems prioritize brightness over focusing accuracy. That said, my Hasselblad 555/ELD is not an up to date system, but it feels like a dark hole. I also have a Sony A900, which happens to have a very bright viewfinder, but there is no way I can switch to 11X magnification on either.
So, I am pretty sure that EVF is where the future is, but that future may not be here and now.
Best regards
Erik