Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
next PZ lens test report: Fujinon XF 50-140mm f/2.8 R LM OIS WR
#11
How about stop talking bullshit and prove with showing, linking or writing a formula in which a given area does matter and changes ISO or sensitivity from a larger to a smaller sensor. And YES a exposure meter does tell me this sort of stuff because normal use of it is to choose a shutter speed or an aperture and an ISO number. You're always trying the same boring trick, using a distraction and questioning common sense. I don't know any exposure meter where I have to put in a film size or sensor size into the equation.

 

And the fact that a sensor is no film but is dealing with ISO, shutter speed and aperture as if it were a film doesn't matter. Basics is, two different sensors will need different ISO but a single sensor will be at the same parameters if you use it full frame or half frame. Any math acrobatics is caused by different sensor technologies, but not by sensor size. By comparing sensors, we need to stay in the same type, not compare X-Trans with Bayer or Foveon. And it's the topic of "area" which is discussion base, not the sensor type.

 

Anyway is this whole ISO thing metaphorical as sensors behave differently and their output of data is influenced by tons of other parameters. One stop faster remains one stop faster.

#12
Sorry, but the "bullshit talking one" is not me.

 

There you go, you figured out that the exposure meter does NOT tell you that an ISO number is fixed. That is half a step to the right direction.

 

Your film ideas are about.... film with a set sensitivity. This may for instance be ISO 60. ISO 60 film has a quite fine grain. Rather comparable with a very high resolution sensor. If you a more sensitive film, like ISO 1600, you will soon notice two things: the grain is more coarse grain. And the resolution of that film is lower.  

 

Now expose those films the same time and with a similar aperture. You will soon notice that that "exposure per square unit" is a rather nonsensical thought experiment: the different films will show very different results now. Apparently exposure is not about "area" that much. Even in the film times it was rather a meaningless thing. 

 

It is not clear to me why you feel you can't/should not compare for instance X-Trans with Bayer sensors... Both use similar silicon sensors, both use CFAs of the R, G and B type. 

 

But you are free to focus on a meaningless things.

 

But lets then just use the Nikon D810. At FF it has about 36mp, when used in APS-C crop mode about 15mp.

When we look at images printed the same size, we should be able to notice that the APS-C crop image has less resolution, but that aside.

 

We shoot the FF image with 100mm f4. 

To get a similar FOV, we shoot the APS-C crop image with 100 / 1.5 = 66.7mm.

To get a similar DOF, we simply use an f-value of 4 / 1.5 = f2.67.

 

Now, the exposure times will be different, because of the different f-values. But fear not, we can set cameras easily to other amplification factors in the digital age. We call this "equivalent ISO settings". Suppose we shot the D810 in FF mode with ISO 800. We then can set the D810 in APS-C mode to 800 / 1.5 / 1.5 = ISO 355.

 Of course, focal lengths, aperture settings and ISO settings are not THAT fine grained, but you can set these to the next close values and you get pretty close to similar results.

 

Now we will look at the images we printed again. We notice that the images look pretty similar still, even though the ISO settings were different for both shots. Why is that? Because the higher ISO FF 36mp shot is much finer grained than the 15mp APS-C shot.

 

Darnet, even in this odd experiment with just one sensor used in different sizes we get similar results when using equivalent settings.

 

Maybe you want to argue against that the result is what counts, in photography, and across different formats?

Like for instance that the only thing that matters is is lumens per square foot?

#13
Quote:Not perfect but nice:

http://www.opticallimits.com/fuji_x/969-fuji50140f28ois
Interesting that the center resolution remains perfect at longer focal lengths but the border resolution drops quite a bit compared to 50 mm.
#14
Quote:Interesting that the center resolution remains perfect at longer focal lengths but the border resolution drops quite a bit compared to 50 mm.
I think I would have substituted  "slightly disappointing" for "interesting".......the best end was  shown in the guessing game.....it might have been tougher call if we'd been shown the 90mm chart!.......

 

 .........I wonder how the resolution charts of my Sigma 50-150mm F2.8 EX DC HSM would look on a modern sensor?   Tested here it did very well..... (I guess on a Nikon 12mp )........... I say that because it is Sigma's older/cheaper generation , albeit without the sealing and IS.

Dave's clichés
#15
Quote:How about stop talking bullshit and prove with showing, linking or writing a formula in which a given area does matter and changes ISO or sensitivity from a larger to a smaller sensor. And YES a exposure meter does tell me this sort of stuff because normal use of it is to choose a shutter speed or an aperture and an ISO number. You're always trying the same boring trick, using a distraction and questioning common sense. I don't know any exposure meter where I have to put in a film size or sensor size into the equation.

 

And the fact that a sensor is no film but is dealing with ISO, shutter speed and aperture as if it were a film doesn't matter. Basics is, two different sensors will need different ISO but a single sensor will be at the same parameters if you use it full frame or half frame. Any math acrobatics is caused by different sensor technologies, but not by sensor size. By comparing sensors, we need to stay in the same type, not compare X-Trans with Bayer or Foveon. And it's the topic of "area" which is discussion base, not the sensor type.

 

Anyway is this whole ISO thing metaphorical as sensors behave differently and their output of data is influenced by tons of other parameters. One stop faster remains one stop faster.
 

Example:

Take the EOS 5Ds (50mp) and the 7D II (20mp).

They have roughly the same pixel density thus roughly the same noise level per square-millimeter - give or take a little.

 

However, when comparing these cameras you have to do that on identical grounds - at the SAME megapixels.

 

Thus you have to downsize the 50mp to 20mp. As a consequence, the noise level decreases substantially - by roughly "1 stop" (a little more). Vice versa this also means that the 5Ds (@ 20mp) is about as good at ISO 800 as the 7D II (@ 20mp) at ISO 400.

So combined with a lens the 5Ds (@20 mp) is as fast with a f/4 lens (using ISO 800) as the 7D II (@ 20mp) with a f/2.8 lens (using ISO 400).

 

Is that so hard to understand ? You don't take picture with a lens. You take pictures with a camera plus a lens.

 

Feel free to point to the flaw in this logic. 

#16
Or alternatively, upsample the 20mp image. This then shows the noise up more coarse.

#17
We will not get on the same line. Two different cameras remain different, no matter what and how you up- or downsample images. In my opinion this is the worst idea to compare systems. Comparing a Minox 8×11mm with a Hasselblad 56×56 mm: What do you gain by sampling up or down until something fits to something else by just putting one of those systems out of it's proportions? You loose the advantages of one system for the benefit of another. There are and always were projects for which one system suits better than another - what good is a tripod based large format camera, when you want to do a reportage and want to remain mobile? How succesful one could blow up a DX-picture, no matter what MP or ISO to get a highly detailed view inside of a church?

 

Shrinking an oceanliner to the size of a canoo makes no sense at all to me. Different cameras for different tasks, but feel free to go on with equivalenting pears to bananas. I consider that useless and also totally ignoring the idea of creating pictures, it's just a tool. I would not take a watchmakers little hammer to demolish a wall, although it might be possible.

#18
Quote:We will not get on the same line. Two different cameras remain different, no matter what and how you up- or downsample images. In my opinion this is the worst idea to compare systems. Comparing a Minox 8×11mm with a Hasselblad 56×56 mm: What do you gain by sampling up or down until something fits to something else by just putting one of those systems out of it's proportions? You loose the advantages of one system for the benefit of another. There are and always were projects for which one system suits better than another - what good is a tripod based large format camera, when you want to do a reportage and want to remain mobile? How succesful one could blow up a DX-picture, no matter what MP or ISO to get a highly detailed view inside of a church?

 

Shrinking an oceanliner to the size of a canoo makes no sense at all to me. Different cameras for different tasks, but feel free to go on with equivalenting pears to bananas. I consider that useless and also totally ignoring the idea of creating pictures, it's just a tool. I would not take a watchmakers little hammer to demolish a wall, although it might be possible.
Understanding lens equivalence, what do you gain from it? The simple understanding of the simple facts that enable you to get a certain result with different equipment. 

 

There are only three variables that determine the image. Field of view. Depth of field. Exposure time.

And you want to argue that it makes no sense to understand these three simple variables across different formats? Ok...

 

And of course, a small camera is easier to put into a pocket. A FF camera with the appropriate lens is capable of very shallow DOF. A large format camera may have the edge for extremely high resolution requirements. 

 

It still makes sense to see what are equivalents across formats. ESPECIALLY when we are looking at an APS-C mirrorless camera from Fuji with an expensive, relatively large f2.8 zoom.

Handy to know that that is equivalent to a 70-200mm f4 L IS USM on a Canon EOS 6D, or a Sony 70-200mm f4 OSS on a Sony A7 II.

 

Also makes sense when someone has like a Canon EOS 100D and wonder if there is any way to get similar looks to images when they look at someone else's work, someone with deeper pockets and more professional equipment, like for instance a portrait or wedding photographer who is using a Canon EOS 5D mk III with a Canon EF 135mm f2 L USM lens. It is nice for that Canon EOS 100D owner to understand that he can get similar image parameters by understanding that for instance a Samyang 85mm f1.4 will get him close to the same FOV and DOF, even with his smaller camera.

 

Or, indeed, it is nice to know for Fuji X MILC owners that they can get similar DOF and FOV from their Fuji X-T1 with Fuji 56mm f1.2 lens as for instance a Sony A7R II user gets with a Sony Zeiss Batis 85mm f1.8. So that Fuji photographer does not have to run out and empty his pocketbook on that Sony + Zeiss, for similar images.

In short: I have no idea what you have against lens equivalence. It is simply theory based on facts, science and optics, does not pose confusion, and makes things very simple to understand. 

 

And  (real) photographers have always used this equivalence understanding when dealing with different film formats. They have understood equivalent focal lengths for similar FOVs, they have understood equivalent f-values for similar aperture sizes (DOF), they have understood the relation to film sensitivity and exposure time and the relation to film sensitivity and formulation to grain size and film resolution.

They could both pick the best tool for the job, AND make do with the tools at hand, to get the results they were after.

#19
Quote:We will not get on the same line. Two different cameras remain different, no matter what and how you up- or downsample images. In my opinion this is the worst idea to compare systems. Comparing a Minox 8×11mm with a Hasselblad 56×56 mm: What do you gain by sampling up or down until something fits to something else by just putting one of those systems out of it's proportions? You loose the advantages of one system for the benefit of another. There are and always were projects for which one system suits better than another - what good is a tripod based large format camera, when you want to do a reportage and want to remain mobile? How succesful one could blow up a DX-picture, no matter what MP or ISO to get a highly detailed view inside of a church?

 

Shrinking an oceanliner to the size of a canoo makes no sense at all to me. Different cameras for different tasks, but feel free to go on with equivalenting pears to bananas. I consider that useless and also totally ignoring the idea of creating pictures, it's just a tool. I would not take a watchmakers little hammer to demolish a wall, although it might be possible.
 

Sorry, the discussion was about equivalence that ----> YOU <----- started.  Big Grin

 

YOU questioned my comparison with the Canon 70-200mm f/4 & Nikkor 70-200mm f/4 - which I referenced in the review to show the weight issue and high pricing of the Fujinon.

 

The only thing that I did was to show the validity of my arguments.

 

The reasons for choosing a mirrorless system go beyond the "equivalence" discussion, of course.

I don't touch the 5Ds R except for doing reviews. That should be clear by now.  Tongue

 

PS: If you don't like the 5Ds example - just take the Canon EOS 5D III and compare it to the 7D II. Both have ca. 20mp. The noise advantage (-> speed advantage) of the 5D III should be crystal clear. 

And if even that feels odd - just take the Sony A7 II with Sony 70-200mm f/4 G. The Sony lens is more light-weight and cheaper than the Fujinon.

#20
Thanks for the review. Vignetting seems a little high for such a lens @ f2.8 as well as corner resolution at the long end but maybe these are good trade offs given that the lens appears to be nearly apo. Seems an interesting contrast to the m4/3 version but it is a bit on the heavy side.

  


Forum Jump:


Users browsing this thread:
1 Guest(s)