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!

NEX-5 vs. NEX-5N pt. 2: ZM 35/2

douglasf13

New member
Hi. After comparing the CV 15 on the NEX-5 and NEX-5N in this thread,http://forum.getdpi.com/forum/showpost.php?p=364857&postcount=1, I decided to give it a go with my ZM 35/2. This test was difficult, because, despite using the same exposure for each, the major difference in illumination across the frame made it tough to get the two files to look similar. These were processed in LR3 with default settings, except for exposure comp. to even out the exposures, as well as using the WB tool in the same place on each photo. The color and illumination differences are still quite noticeable, as you can see, but I didn't want to process the files too much. I had to raise the exp. comp of the NEX-5 by 1.71 EV so that the centers of both images were exposed similarly.

NEX-5, ISO 200, 1/3200, f2:


NEX-5N, ISO 200, 1/3200, f2:



NEX-5 top right crop:



NEX-5N top right crop:



NEX-5 top left crop:



NEX-5N top left crop:



NEX-5 bottom left crop:



NEX-5N bottom left crop:



It seems to me that the sensor differences between the NEX-5 and NEX-5N extend to beyond only the wider angle rangefinder lenses. What do you think?



note* With these shots, I made sure that the corners were as in-focus as they could be, and infinity was the best focus for all corners. Also, the contrast of the NEX-5 shot can be deceiving, especially in the top right crop. If I add around .75 EV of vignetting to the NEX-5N photo, it results in similar contrast. The evenness of the illumination on the NEX-5N is MUCH better, it seems.
 

Rawfa

Active member
Were these shots taken exactly at the same time? The reason I ask is because at first glance the 5N seems to have a slightly better dynamic range.
On a pixel level the 5N seems to be a bit cleaner too.

I don't mean to threadjack or anything, but how much faster was the AF on 5N when compared to the regular 5?
 

Paratom

Well-known member
I believe if you had to raise the exp of the Nex-5 by so much we should be very carefull comparing the result regarding noise/DR.

Did you by any chance also compare the kit lens on those 2 cameras or maybe the kit with the 35mm?
 

douglasf13

New member
That's what is so outrageous. These shots were taken close together in midday at the same exposure, and the light was relatively constant. I have some other shots that show the same differences. The evenness of illumination of the scene between the two sensors is so drastically different that I would have guessed that they were taken at different times with different lenses. While the difference in color shift and smearing is relatively subtle, the difference in vignetting between the two sensors was surprising. Still, I wouldn't read into noise and DR too much in this test.

I've not used autofocus or the kit zoom, so I can't help you there, yet.
 
Last edited:

Jonas

Active member
After having downloaded the images and equalized the exposure (well, not really, but adjusted the images to reach the same colour and brightness values on the white wall under the roof, in the center) I'm puzzled.

The colours are very different, perhaps one or both of them are developed with no custom colour profile?

The plain 5 seem to be much darker in the upper right, the 5N seem to be a little darker in the lower left. Strange.

If you had to raise the EV comp in the plain 5 image by a value of 1.7 there is something strange going on. Do you see these exposure differences all the time between the cameras? I can't recall having seen anything about the 5N being a stop and a half "faster" than the old 5.

/Jonas
 

douglasf13

New member
Hi, Jonas. So I dug out another set of files that I took within seconds apart yesterday, as well as a set of pics with a tissue over the lens, as if I'm making a CornerFix profile. The ISOs of each camera are said by DxO Mark to be the same, so I think what we're seeing is a drastic difference in the way the sensor receives light with the near-symmetrical 35mm Biogon, which is significant, because we're not just talking really wide M lenses anymore. I can repeat the difference over and over again. The following files are raws from LR3 with identical settings and exposures:

NEX-5:


NEX-5N:


NEX-5:


NEX-5N:
 

douglasf13

New member
I guess another reason that they are so wildly off is that LR3 could have different base line exposure compensations for the two cameras, or one of my cameras has a problem. Interesting.
 

Jonas

Active member
Hi,

As expected the 5N tissue looks better. IMA-test agrees:


and


That is measuring luminance only, and for fun only. It does however make me think a little about different lens reviews we have seen through the years...
 

Jonas

Active member
I guess another reason that they are so wildly off is that LR3 could have different base line exposure compensations for the two cameras, or one of my cameras has a problem. Interesting.
Indeed. I hope somebody can help with insights on that.
 

douglasf13

New member
Very interesting, Jonas. Thanks.

For the heck of it, I ran the last two house shots above through RPP with default settings, and the difference in exposure is still there. Both shots are 1/2500 sec, f2, ISO 200.

NEX-5:


NEX-5N:



p.s. Jonas, are those imatest images from my 35/2 tissue shots?
 
Last edited:

douglasf13

New member
Shoot, I forgot to make sure that I did my own white balancing from the same spot. I changed the photos above with appropriately white balanced versions.
 
C

curious80

Guest
The difference in the brightness between NEX5 versus NEX5N might in part be due to a different tone curve used by each camera. In LR3 if you use the brightness control instead of exposure then you should be able to bring the NEX5 result closer to NEX5N without changing the exposure. It might be worth passing these through dcraw to get the linear RAW output and see how much the illumination varies before applying the tone curve.

DxoMark's results wont be impacted by the tone curve as they define the ISOs in terms of linear RAW output
 

Godfrey

Well-known member
Douglas,

You should use a tri-band calibration target to establish the proper exposure with each camera, then make your comparison photos. E.g.:


This is a view of such an exposure using the Ricoh GXR, displayed in Lightroom. Note the three peaks in the Histogram ... they are spaced equidistantly and the center is aligned with the center of the histogram. That's the ideal exposure as a starting reference.

If you compare it with a known-accurate incident light meter, it can also inform you as to how accurate the camera's metering calibration and ISO values are.
 

douglasf13

New member
I think you're right, thanks. Here are the two files from LR3, but the NEX-5 file has an extra +38 in brightness. I think we should ignore my first files, as it seems there is a problem with those. I'm going to delete them to avoid confusion.

NEX-5:

NEX-5N:
 

uhoh7

New member
And another thing :)

New rumour says SLR glass like nikon 28/2 etc may be sharper on 5 than 5n.

Whoever still has both, like me, needs to test :)
 

Jonas

Active member
One thing I don't understand really is the vignetting (using a tissue) test. Here is the result from my plain Nex-5 with the CV 35/1.2 v2 mounted to it:



It's not a perfectly aligned system but pretty good and as expected the shape of the light distribution is circular.

What can the reason be for Douglas' Nex-5 more oval result? A leaning sensor?!! Hardly. Did you use the same adapter when taking those images?
 

douglasf13

New member
Yeah, it is strange. My shots were taken seconds apart with the same adapter/lens combination. The Biogon 35 is a much more symmetrical design than the 35/1.2, so I wonder if somehow the sensor toppings of the 5 react differently to the angle light rays?

I have seen some talk in regards to the 5's AA filter being stronger in one direction, but I'm not sure about that. Maybe that could cause the phenomenon? Hmmm.
 
Top