An Experiment - T Zoom Vs Canon 50mm

[gbealnz]

Advertisement (gone after registration)

I have a wedding coming up in a few months and am looking to take a few informal shots.

 

Assuming the slow old T 18-56 will be less than ideal I had figured I should use the likes of the Canon 50mm f1.4 I have.

 

But since I have a perfect model staying with us for the school holidays I thought I should try them both side by side. Nothing scientific, simply did my best to keep the subject in one place.

 

Both with the T, on a tripod. Both at f 5.6, which means the T zoom is wide open. Aperture priority used, and auto ISO. Both Superfine JPEG, and the two here are straight out of camera, nothing done except reduce from 300 dpi to 72 dpi, which makes them fit the forum sizing better.

 

Such a huge difference though, unreal. If I've done it right the Canon shot should be on top.

 

Is it possible the camera "processes" the zoom shot differently, on account of it "knowing" what the lens is? The Canon is on a after-market 6-bit adaptor, and is recognized as a 50mm. Jury is out at this end.

Gary

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[jdlaing]

#2. 

[jdlaing]

I did that unfairly without explaining why.

I look at three things in the images and this is why I like #2:

 

The sunlight on the hair at her forehead.

The sunlight on her nose.

The blacks in the background.

 

The first image, lacking these three things appears flat............to me.

[jdlaing]

And........

 

Let us not forget that zoom is tuned to the camera and is a native lens.

[gbealnz]

Thanks JD,

it sort of goes with what I had guessed then, camera and lens combo. I'm certain that as a DNG, then the difference "may" not be as significant, and any difference could be adjusted perhaps.
 

I wish I had tried the older Summicron M 50mm collapsible which I usually leave bolted to the M6. I might just do this all again.

Gary

[jankap]

How is this one? I changed your Canon picture somewhat. 

More black, the a and b values of LAB modified (more plus), but the white of the eye kept white. 

Jan

 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[jankap]

Advertisement (gone after registration)

You can influence the jpgs, that are made by the camera. There should be a option in the firmware to make the colors vivid, natural, etc. 

And of course a non coded lens is handled standard (what ever that may be).

 

The wb is something different. 

Jan

[Guest]

How is this one? I changed your Canon picture somewhat. 

More black, the a and b values of LAB modified (more plus), but the white of the eye kept white. 

Jan

 

 

post-3.jpg

 

I would be also happy with this shot.

[gbealnz]

I would be also happy with this shot.

 

Yes, me too, thank you Jan.

Looking at all of them, I could use and fiddle with them to produce similar results.

Just surprising (to me) that a native lens, and a non-native lens produced such differing results.

Gary

[bencoyote]

From my tests which don't use non-leica lenses overall the Leica cameras are a tiny bit quirky.

If you just directly import the DNG files into LR and use the "Adobe Standard" camera profile it looks quite a lot like that. However, if you flip to the "Embedded Profile" which is embedded into the DNG the colors shift to be much more like the shot with the 18-56. If I take the Color Checker Passport and calibrate the camera I find that the profiles that X-Rite's software generates are very similar to the embedded profile. I have to spend quite a while flipping between the two profiles to notice any difference. It suggests to me that Leica spent a considerable amount of effort getting their "Camera profile" pretty darn accurate.

 

Then in the DNG there is also an embedded lens profile which fixes optical distortion in the lens. This is automatically applied when you bring the photo into LR.

 

Here is my guess about what is happening: I bet when the Leica T does not know what lens is on the camera due to the fact that you were using an unencoded adapter, when it generates the the JPG from the data it gets from the sensor, it not only doesn't apply the lens profile, it doesn't apply the camera profile.

 

It would be interesting to see if the DNG file included an "Embedded profile" for the camera or if Leica also forgets to put it in the DNG profile. If it doesn't include an "Embedded profile" then I think we have a reportable bug to Leica. It should always generate DNG files which include camera profiles.

 

If it does include an embedded profile, it would be interesting to see if when you apply that profile in LR if the shots are more similar. If that happens, then I think that may also be a bug that you could report to Leica. It should apply the camera profile when generating JPGs even if the lens and lens profile are unknown.