fredmiranda.com
Login

Moderated by: Fred Miranda
  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Leica & Alternative Gear | Join Upload & Sell

1       2       3              78              80              85       86       end
  

Archive 2017 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)

  
 
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #1 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
But that's gimping the 5DsR unnecessarily, how you test has to match match how you shoot. If you need to use f/5.6 on the 5DsR to get sufficient DOF then you need to use f/8 on the GFX, however that doesn't mean you need to use a faster shutter speed on the 5DsR. And if you're not DOF limited you'll find the maximum aperture on the 5DsR to be at least a stop faster than the GFX.


I am totally with you on this, Matt, that would be a better "low light efficiency" test. However there are two issues at hand here:

Actually I was only addressing the ISO normalisation. Since the differences between cameras are so large, often two thirds to one ful stop, this cannot be overlooked. Many things can be said about the DPR tests, but they have made two giant leaps towards fair ISO noise tests. That is the provision of normalized images sizes and that they compensate for the manufacturer's "cheating" on the ISO.

The other issue is that the world is not ready for presentation of results that correspond to "taking the same picture". Even when we have walked through the theory and demontsrated with samples, and people agree that it is correct, they don't accept the premises. They say:
"But that is not the way I shoot".
"When I take a smaller camera, I will get more noise, but I don't care".
"When I take a larger camera, I will get shallower DOF. That's the whole point of taking a larger camera"
"I will take lens X and shoot it wide open on both cameras anyway, because I love the rendering. I don't care if I have to change the perspective to get similar framing of the foreground"
"I shoot mostly focal length limited anyway"
And so on.




Mar 14, 2017 at 02:58 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #2 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


charles.K wrote:
It always great to see comparisons from DPReview, but we are now entering into the realm of diminishing returns with high MP sensors and often the selection of lens, lens alignment, technique is so crucial and may explain the wider variances in bench tests. I even compared the D810 at 64 ISO, and it is surprisingly sharp when we compare at base ISO's.



When it comes to resolving power or acuity of a system, the dominant parameter is still the megapixel count of the camera. 24 MP APS-C cameras outresolve 20 MP FF cameras under ideal conditions like this, with a lens and aperture that is supposed maximize the potential of the camera. I expect that when Canon releases a 120 MP camera, the resolution will be comparable to the current 100 MP digital MF sensors.

When it comes to lenses, I do expect the Fujinon 32-64 mm to be a gem and easily outshine the best 24-70 mm offerings for edge to edge performance at equivalent FOV and DOF. Actually that lens may turn out to be my reason for buing into the GFX at all



Mar 14, 2017 at 03:13 AM
tunisia
Online
• •
Upload & Sell: Off
p.79 #3 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Quick question: I've read some reviews - just read a translated French review focus-numerique, for eg - that sort of trashes the evf of the GFX. Is it really that bad?

Anyone who's handled/shot with it: How does it compare to the Leica SL, Q, and the Sony a7rII. From what I've read too date the belief was - I thought - that the SL was the best and Sony the worst. I thought the Fuji was closer to the Q which is said to be much better than Sony.

While I find Sony acceptable I was hoping for a more meaningful improvement in the Fuji's evf. This is vitally important to me as I shoot manually primarily and will be using the GFX this way with some exception with adapted lenses.

Finally, on a more personal note, I've been the route with the 5ds r using live-view and a zacuto and given it up for the Sony's evf system. Love the 5ds r for auto when I do auto which is not very often, but that's about it. The personal note is just that: vision problems that demand the best or better than present with sony manual focusing system.

As I've got the GFX pre-ordered, any help anyone can offer would be greatly appreciated.
By the way, I'm really impressed with the Fuji and what it offers for my kind of shooting; what I don't want is to be seriously disappointed by the manual shooting experience.
Thanks.






Mar 14, 2017 at 03:57 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #4 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
But that's gimping the 5DsR unnecessarily, how you test has to match match how you shoot. If you need to use f/5.6 on the 5DsR to get sufficient DOF then you need to use f/8 on the GFX, however that doesn't mean you need to use a faster shutter speed on the 5DsR. And if you're not DOF limited you'll find the maximum aperture on the 5DsR to be at least a stop faster than the GFX.


Well, I don't think it is quite that simple. If I want more depth of field on the GFX I might stop down more, but I might also put the camera on the Cambo Actus that I plan to get and use movements to get the depth of field. Neither is it always true that the maximum aperture of FF 35mm is always at least a stop faster. That won't be true in the absolute sense of lenses available and it certainly won't be true in the particular cases of what lenses a given individual has. The comparison you point to is an interesting test, but I don't see it as the only test that makes senses and anything else is gimping the 5DsR. For your uses the comparison you suggest may be better, but I don't think it is the only sensible test and I think the one they did is sensible for at least some people too.



Mar 14, 2017 at 07:26 AM
Matt Grum
Offline
• • •
Upload & Sell: Off
p.79 #5 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Steve Spencer wrote:
Well, I don't think it is quite that simple. If I want more depth of field on the GFX I might stop down more, but I might also put the camera on the Cambo Actus that I plan to get and use movements to get the depth of field.



But you can use the Canon 5Ds with a tilt-shift lens to increase DOF and then open the aperture and you end up back to parity etc. etc. Of course you could construct some scenario where you needed to use an f/2.0 lens on the Cambo, but anyway that is slightly irrelevant, the point I'd make is this:

1. Actual real world performance will always come down to lens availability (I could go into business and produce a digital 4x5, but it will be no use in low light if I decide to bundle it with an f/64 lens).

2. With that in mind, comparing at the same entrance pupil diameter is a better basis or starting point for comparisons than comparing at the same f-stop (or changing f-stop and the shutter speed in dpreview's case). This is because the entrance pupil diameter is a better invariant between FF and MF lenses (i.e. it varies a lot less than the maximum aperture does) and it has the nice side effect that DOF and diffraction also stay the same.

Steve Spencer wrote:
Neither is it always true that the maximum aperture of FF 35mm is always at least a stop faster. That won't be true in the absolute sense of lenses available and it certainly won't be true in the particular cases of what lenses a given individual has.


Yes there will always be edge cases. But I'd rather look at equivalent exposures first and then discuss special cases (such as not caring about DOF and being able to use a fast lens designed for 35mm on the GFX), than look at comparisons which don't match real world usage very well on the whole, as then almost everything becomes a special case for most people.



Mar 14, 2017 at 08:04 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #6 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
But you can use the Canon 5Ds with a tilt-shift lens to increase DOF and then open the aperture and you end up back to parity etc. etc. Of course you could construct some scenario where you needed to use an f/2.0 lens on the Cambo, but anyway that is slightly irrelevant, the point I'd make is this:

1. Actual real world performance will always come down to lens availability (I could go into business and produce a digital 4x5, but it will be no use in low light if I decide to bundle it with an f/64 lens).

2. With
...Show more

I guess my point is that I don't think any one way of comparing across formats makes senses in a general way. I think you have to think of a particular use and then think about the most sensible way to compare for that use and what would be needed in it. Again I think that comparing at the same entrance pupil makes sense in some cases, but I don't think it is necessarily the best way or even the first. Yes, you have your preferences about what you would like to do first, but those very likely differ from mine. I think arguments about the "best" way or the way that should be "first," are a bit silly and point to particular points of view as if they are normative. Instead, I think it is better to talk about ways to do comparisons without the value laden "best," "first," and "edge" characterization. What is one person's "edge" use might be another person's primary use. What is "best" depends on typical usage. What should be done "first," depends on a lot of factors.



Mar 14, 2017 at 08:25 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #7 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
2. With that in mind, comparing at the same entrance pupil diameter is a better basis or starting point for comparisons than comparing at the same f-stop (or changing f-stop and the shutter speed in dpreview's case). This is because the entrance pupil diameter is a better invariant between FF and MF lenses (i.e. it varies a lot less than the maximum aperture does) and it has the nice side effect that DOF and diffraction also stay the same.



Actually DPR try to use about the same entrance pupil to get the same DOF and diffraction, that is the reason they use f/5.6 for FF36x24 and f/8 for 44x33 and get different shutter speeds at the same ISO. They are not consequent about it though. They use 32 mm f/4 for 1", 45 mm f/4 for m4/3, 50 mm f/5.6 for APS-C, 85 mm f/5.6 for FF and 90 mm f/8 for the Pentax 645Z. If they had the 120 macro for the Fuji, I am sure they would have used it, and it would make sense at f/8.

It is not easy to make one entrance pupil diameter work optimally for such a borad range of formats. f/2.8 on 1" and f/11 on MF would cause problems in both ends. f/2.8 on 1" would make it difficult to get sharp corners. f/11 on the XF 100 could be seen as limiting the full resolution.



Mar 14, 2017 at 08:43 AM
Matt Grum
Offline
• • •
Upload & Sell: Off
p.79 #8 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Steve Spencer wrote:
I guess my point is that I don't think any one way of comparing across formats makes senses in a general way. I think you have to think of a particular use and then think about the most sensible way to compare for that use and what would be needed in it.


I'm not suggesting people should think a certain way - of course not, I'm suggesting websites present the comparison a certain way by default (I mean, they already do this, I just think they made the wrong choice for the default).

Steve Spencer wrote:
What is one person's "edge" use might be another person's primary use.


Of course there is no single right answer between f-number and entrance pupil invariance, but there is one answer which applies to more people than the other* - yet the more niche one is presented, often with no indication that this could be misleading.

*unless you want to argue that using the camera as a digital back on a view camera, or adapting lenses designed for a smaller format is going to apply to more people than using native lenses.

If you're using native lenses that at any given moment you are either DOF limited or you are not. If you are, then you can open the aperture more on the 35mm system and you end up capturing the same total light. If you are not DOF limited then you can also open the aperture more on the 35mm system because the native lenses go wider. This is a simplification but it applies to the vast majority of shooting scenarios.



alundeb wrote:
Actually DPR try to use about the same entrance pupil to get the same DOF and diffraction, that is the reason they use f/5.6 for FF36x24 and f/8 for 44x33 and get different shutter speeds


Yeah they almost do it right, but then change the shutter speed for no reason (yes I know the reason they do it - Richard Butler has claimed doing the comparison at the same total light is too hard to explain to people ).

alundeb wrote:
It is not easy to make one entrance pupil diameter work optimally for such a broad range of formats. f/2.8 on 1" and f/11 on MF would cause problems in both ends. f/2.8 on 1" would make it difficult to get sharp corners. f/11 on the XF 100 could be seen as limiting the full resolution.


Corners are already a major problem, case in point there is a lot more contrast in the 5DsR corners than the GFX. The solution is to only show the centre of the frame, that's more than enough to compare detail and noise, and centre performance is much more stable between lenses. The solution to the entrance pupil is to choose the best setting for the larger format and work your way down, when you reach the limit keep the shutter speed but increase the ISO setting after all the lack of an equivalent aperture setting is a deficiency of the smaller format. It's imperfect but it's in the spirit of making meaningful comparisons (again people generally choose settings according to the needs of the scene not according to the size of the sensor they are using).

Anyway I think I've made my point I'm not going to labour it any more, it's a criticism of gear websites rather than anyone here (all of whom are intelligent enough to make their own comparisons).




Edited on Mar 16, 2017 at 05:53 AM · View previous versions



Mar 14, 2017 at 09:55 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #9 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Hi,

I was able to play with both the Fuji GFX and the Hasselblad XTI this morning at Midwest Photo in Columbus, OH. Neither are available for sale, but they have the demo cameras and you can even rent them. They only had the 32-64 zoom for the Fuji and the Hassy they only had the 45 f/3.5 lens. Here are my impressions:

Appearance: The Hassy is beautiful and much smaller. It is a lot easier to hold and the operation is wonderfully simple. The Fuji is a lot bigger especially in thickness and really quite ugly. It has a huge number of knobs and dials. On first picking it up, it took me quite a bit of time to get my head around how to control the camera. As they say, however, appearance is only skin deep and I think that is very much true in this case.

Operation: The Hassy is very slow in operation. It takes several seconds to turn on and the delay between shots is noticeably long. The EVF is fine and a bit better than my Sony A7rII. The lens worked very well in manual focus with a nice feel and was pretty quick to AF as well. The shutter made to me a very annoying high pitched tinny sound. It was loud and I suppose I would get used to it, but at least during the demo I found it distracting. After using the Fuji for about 5 minutes I finally understood how to control most functions on the camera. I found the menus pretty easy to navigate and once I shut of picture review the camera was pretty responsive. A short blackout between shots, but it was pretty ok. I found the EVF to be much nicer than the Hassy or my Sony A7rII. It is big bright and easy to use. Magnification worked well in conjunction with the joystick. It automatically had zebras on, which I would almost surely shut off. The AF was quite quick and reminded me a lot of the AF on my Sony A7s. All in all, it was heavy with the zoom, but very nice to operate. I should add that I loved the tilt/swivel for the EVF. I think I will end up often shooting the camera holding it to my chest with the EVF pointed up. I found that an excellent way to stabilize the camera and comfortably frame shots. The shutter was a little louder than the Hassy, but much less noticeable as it was to me a more typical shutter sound. I think I would actually prefer the Hassy to shoot if they ever get the firmware to a state that it isn't painfully slow and if the shutter wasn't such an odd annoying sound. On the other hand, I am fairly confident that when I use the Fuji for awhile it would be quite easy to use. I do like the shutter speed dial, the aperture ring on the lens, and the ISO dial a lot. I also found the joystick for focus placement quite helpful. That the LCD was tillable in both direction is a nice touch as well, but my favorite feature was the tillable EVF.

So, I still don't have my adapters in yet, but when I do I hope to test a few F mount and Mamiya 645 mount lenses. I should be able to do that within a week. or so.



Mar 14, 2017 at 12:23 PM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #10 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
I'm not suggesting people should think a certain way - of course not, I'm suggesting websites present the comparison a certain way by default (I mean, they already do this, I just think they made the wrong choice for the default).

Of course there is no single right answer between f-number and entrance pupil invariance, but there is one answer which applies to more people than the other* - yet the more niche one is presented, often with no indication that this could be misleading.

*unless you want to argue that using the camera as a digital back on a view
...Show more

Matt,

I think the most common use case for the Fuji GFX might well be as a studio camera. If that is true, then a test at ISO 6400 is a pretty edge use for that most common use of the camera. We also know next to nothing about what the fastest native mount lenses will be for this system. When Fuji started out with their APS-C system they didn't announce their fastest lenses right away. It took a couple of years to announce the 56 f/1.2. Would it be all that surprising if they made a 150mm f/2.4 lens or a 75 or 50 or 40 or 35 f/1.4 lens? I don't think it would be. So deciding this early in the game that there won't be any narrower depth of field capabilities for native Fuji lenses is a pretty big assumption. Suffice it to say that I am not convinced your assumption that you can always go to a wider aperture with FF 35mm lenses. I am not at all sure that will be the case. It isn't the case already with adapted lenses and I suspect in time it won't be with native lenses either.



Mar 14, 2017 at 12:39 PM
Matt Grum
Offline
• • •
Upload & Sell: Off
p.79 #11 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Steve Spencer wrote:
I think the most common use case for the Fuji GFX might well be as a studio camera. If that is true, then a test at ISO 6400 is a pretty edge use for that most common use of the camera.


Fuji will sell far more GFXs to amateurs than professionals, purely on the basis that there are orders of magnitude more users in that category (this is true of a lot of "professional" equipment).

Also if you are working predominantly in the studio you'll either be using a DSLR for the speed they offer or you will have a medium format system that you're already tied into. I can't really see the appeal of the GFX to studio photographers who aren't really concerned about weight or portability, for the same money you can get a PhaseOne P65+ outfit with body, back and lens.

The GFX is geared towards and will be used mostly by landscape/cityscape photographers.

Steve Spencer wrote:
When Fuji started out with their APS-C system they didn't announce their fastest lenses right away. It took a couple of years to announce the 56 f/1.2


Their 56mm f/1.2 seems ambitious but considering correcting for lower f-number is equivalent in difficultly to designing for the larger image circle of the 35mm format, it's akin to Canon announcing an f/1.8 lens (not saying it's a bad lens at all, quite the opposite, it's just not pushing the envelope in terms of f-stop considering the image circle, which is also fine shallow depth of field is not everything).

Steve Spencer wrote:
Would it be all that surprising if they made a 150mm f/2.4 lens or a 75 or 50 or 40 or 35 f/1.4 lens? I don't think it would be. So deciding this early in the game that there won't be any narrower depth of field capabilities for native Fuji lenses is a pretty big assumption


I'd be frankly astonished if Fuji came out with an f/1.4 lens for the GFX, especially one wider than 50mm. I don't think it's a big assumption given the way the industry is at the moment, the major manufacturers are being cautious and making very predictable releases. I will back it up to say that if they do release an f/1.4 I will come on here an eat a hat (and photograph it with the GFX), so bookmark this post now!



Mar 14, 2017 at 05:18 PM
Lee Saxon
Offline
• • • •
Upload & Sell: On
p.79 #12 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Seems like more 35mm lenses than we thought are covering this format, and the G lenses aren't that huge. I'm thinking the 54.7mm circle - smaller than a FF35 T/S - isn't as challenging as we tend to think.

I think they can do f/1.4 equivalent, at least. The 110/2 is already essentially there. If they can do an 85/1.6 equivalent I wouldn't be shocked by a 50/1.4 equivalent (63/1.8). If the system meets sales expectations, I can see them adding some flexibility.

But if we're taking bets I say the next lens announced after these six is a 32/33/34/35 f/2.8.



Mar 15, 2017 at 06:28 AM
Matt Grum
Offline
• • •
Upload & Sell: Off
p.79 #13 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Lee Saxon wrote:
I think they can do f/1.4 equivalent, at least.


Sure, just to clarify, my hat eating is contingent on Fuji releasing an actual f/1.4 lens!



Mar 15, 2017 at 07:54 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #14 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:
Sure, just to clarify, my hat eating is contingent on Fuji releasing an actual f/1.4 lens!


I still owe the whole community free pizza because I thought the 28 mm Otus wouldn't be f/1.4. Steve will remember that! Your bet is a much lower risk



Mar 15, 2017 at 08:16 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #15 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Matt Grum wrote:


Fuji will sell far more GFXs to amateurs than professionals, purely on the basis that there are orders of magnitude more users in that category (this is true of a lot of "professional" equipment).

Also if you are working predominantly in the studio you'll either be using a DSLR for the speed they offer or you will have a medium format system that you're already tied into. I can't really see the appeal of the GFX to studio photographers who aren't really concerned about weight or portability, for the same money you can get a PhaseOne P65+ outfit
...Show more

First, I think that you and I just disagree about who is going to buy this camera. It is $6,500. You may be right that it will be primarily amateurs, but I think Fuji would like to tap the professional market with this camera. I also think that the amateur/professional distinction is not always so easy to make. What do you call a stay at home dad that I know who shoots primarily senior pictures and a couple of weddings a year and makes about $30,000 a year? Perhaps you would see him as an amateur, but the distinction isn't always clear. Anyway, I think the camera will be sold to a lot of different people and exactly who is going to buy the camera is hard to know at this point. You seem very sure in your opinions about everything regarding this camera and I wonder where this certainty comes from and whether it isn't just a tad misplaced.

Second, I want to correct a couple of factual errors in this point, or what I think are factual errors. You say, "for the same money you can get a PhaseOne P65+ outfit with body, back and lens." You're joking right. Just that back, used, can go for $10,000. For example at this link:

http://www.digitalback.com/product-category/0-pre-owned-digital-backs/

It is a pretty big exaggeration it seems to me to suggest that you can get not only the back, but a body and a lens for $6,500.

You also claim, that, "correcting for lower f-number is equivalent in difficultly to designing for the larger image circle of the 35mm format." I believe that is just totally wrong. I have seen several lens designers comment that lowering the f-number increases the aberrations exponentially. Increasing the image circle is focal length depend and is harder at shorter focal lengths than longer focal lengths, but I have never seen any indication that increasing the image circle increases aberrations exponentially. I believe at longer than normal the increase is just linear and if that reasoning is correct it is a huge misstatement to equate an exponential increase with a linear increase. Simply put it is much harder to decrease the aperture than to increase the image circle for a longer than normal lens There is no question it is much harder to make an 56 f/1.2 lens for APS-C than an 85 f/1.8 lens for FF 35mm, that you equate the difficulty seems to betray a fundamental misunderstanding of lens design or just an attempt to over argue your case. I suspect it is the latter as in the past we have agreed that it is harder to increase the size of the aperture than it is to increase the size of the image circle.

You also claim, "the major manufacturers are being cautious and making very predictable releases." I think you may have missed some of the recent releases. Let's see within the last two years Zeiss has released a 28mm f/1.4 APO, lens; Nikon has released a 100 f/1.4 lens; Canon has released a much enhanced 35 f/1.4 lens with a new type of glass and an 11-24mm f/2.8 zoom, Leica has released a very small 28 f/1.4 lens, Cosina Voigtlander has announced a 40 f/1.2 and a 65 f/2 APO macro, Sigma has released a whole suite of Art lenses included an f/2 zoom and a 135 f/1.8 prime. I am sorry but although there have certainly been cautious and predictable releases, I also see a number of releases that have pushed the boundaries..

So, if you dismiss the factually inaccurate statements in your post can see that more professional might be interested in the camera than you expect because it is much cheaper than Medium format digital backs. It is not nearly as hard to make f/1.4 lens as you suspect because increasing the image circle of already existing designs for 35mm FF is not nearly as hard an increasing the maximum aperture of these lenses. For example, we already know that one 85mm f/1.2 design (the quite old by now nFD 85L) covers the mini MF image circle at f/1.2. So, I think you greatly overstate you case that such fast lenses won't be coming. I won't hold you to eating a hat, but unless you really like doing that sort of thing I would be a bit more cautious in your proclamations.

We of course do not know what Fuji will make beyond the six lenses they have announced. I agree with Lee that a 35 or maybe a 30 f/2.8 is likely the next lens, but I would not be surprised at all if they make a 70 f/1.4. I think a 50 f/1.4 would not be surprising at all either. I agree it would be pushing the boundaries pretty far for a 40 or 35 f/1.4, but I think that could happen and I think it is more likely than not that we could get at least a 40 f/2.



Mar 15, 2017 at 08:46 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #16 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Steve Spencer wrote:
but I have never seen any indication that increasing the image circle increases aberrations exponentially.


There are indications on this. The CA profiles measured by DXO show something like an exponential curve, on most lenses.

Here I picked a random one, clik on "Chromatic aberrations" and "Profile"

https://www.dxomark.com/Lenses/Canon/Canon-EF-35mm-F2-mounted-on-Canon-EOS-5DS-R---Measurements__1009

My tests of lenses for 35 mm format with a shift adapter, indicate the same, a dramatic increase of CA in the extended image circle.



Mar 15, 2017 at 09:21 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #17 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


alundeb wrote:
There are indications on this. The CA profiles measured by DXO show something like an exponential curve, on most lenses.

Here I picked a random one, clik on "Chromatic aberrations" and "Profile"

https://www.dxomark.com/Lenses/Canon/Canon-EF-35mm-F2-mounted-on-Canon-EOS-5DS-R---Measurements__1009

My tests of lenses for 35 mm format with a shift adapter, indicate the same, a dramatic increase of CA in the extended image circle.


There is a huge difference between a dramatic increase and an exponential increase. That DXO chart looks nothing like an exponential increase. The mathematical model for an exponential would be much much steeper than in that graph. This is also at 35mm where increasing the image circle is at its hardest. At 50mm or 70mm it would be much easier. Also given that CA is hard to control it is not surprising that it gets bad outside of the intended image circle for a lens. Why would they bother to correct it at all out there? That it isn't corrected well outside of the intended image circle tells us nothing about how hard it would be to correct it. I don't see anything in either point that suggests anything close to an exponential increase in aberrations as the image circle gets bigger.



Mar 15, 2017 at 09:36 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #18 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


Steve Spencer wrote:
That DXO chart looks nothing like an exponential increase. The mathematical model for an exponential would be much much steeper than in that graph.


I would disagree, or even say that you are wrong here. It looks much closer to exponential that to a polynomial, for example. One of the characteristics is that it looks almost flat for a long time. The visual steepness at the end is just accoring to scale.
Maybe you can find another mathemathical funtion it correspnds even better to than exponential, but we don't have to take it literally. "Dramatic" and "exponential" are close enough regarding the difficulty of extending the image circle.



Mar 15, 2017 at 09:42 AM
Steve Spencer
Offline
• • • • • • •
Upload & Sell: On
p.79 #19 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


alundeb wrote:
I would disagree, or even say that you are wrong here. It looks much closer to exponential that to a polynomial, for example. One of the characteristics is that it looks almost flat for a long time. The visual steepness at the end is just accoring to scale.
Maybe you can find another mathemathical funtion it correspnds even better to than exponential, but we don't have to take it literally. "Dramatic" and "exponential" are close enough regarding the difficulty of extending the image circle.


No, exponential means something specific and dramatic is a much more general term. Lots of things are dramatic without being exponential. Very few things are more dramatic than exponential. I don't even find that graph as being dramatic. It certainly isn't exponential. We can do the math if you want and plot the exponential function, but I am quite sure that is much much flatter than exponential, but it is a matter of math so there is a right and wrong answer. I don't have the time or inclination to do the math, but if you want to I would welcome being proven wrong.



Mar 15, 2017 at 10:31 AM
alundeb
Offline
• • • • •
Upload & Sell: Off
p.79 #20 · Pre-order: Fujifilm GFX 50S Medium Format body ($6,499)


It is only a curve approximation to some measurements, so no point in proving what it exactly looks like.
The point is that CA extends with the image circle in a way that makes the cost of extending a 43 mm IC design to 55 mm IC, while keeping the aberrations down, substantial.

You can find similar patterns for field curvature, so there clearly are "indications".



Mar 15, 2017 at 11:02 AM
1       2       3              78              80              85       86       end




FM Forums | Leica & Alternative Gear | Join Upload & Sell

1       2       3              78              80              85       86       end
    
 

Welcome back
Log in to your account