Leica CL vs TL2 dynamic range

[lct]

Advertisement (gone after registration)

It is simply not credible (to me) than any Leica APS-C camera does better than MF or FF Leica cameras like the S, SL or M10. YMMV.

[jaapv]

Sensor size has little to do with DR.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[lct]

I've never seen a better DR out of APS-C compared to MF or FF so far but i may be wrong. Any example of such an APS-C superiority among current sensors? Just curious.

[jaapv]

Well, aren't we discussing just that here in the case of the TL2?

Anyway, the use of the word superiority might be out of place. The DR of negative film was always twice that of slide film, yet nobody considered one superior to the other.

 

[lct]

Not sure what film has to do with this topic but anyway let me ask my question another way if you don't mind. Do you have any experience or knowledge of a current APS-C camera having better DR results than current MF or FF ones? I have none personally but i may be wrong. If i'm not wrong, the TL2 would be the only exception according to the questionable (to me) chart above. 

[jaapv]

I’d have to dig through PhotonstoPhotons for that. Given the low priority of the question to me I won’t The DR of the present-day cameras is so much higher than the DR of prints and screens that it appears to serve mostly as cover-up for exposure errors. 

[frithjof.b]

Advertisement (gone after registration)

On 12/4/2018 at 9:55 AM, jaapv said:

I've just sent this mail to Bill: Let's see what his take on this puzzle is.

Looking at the curves, it is surprising that there are some non-linearities in the TL2 curve. This suggests that Leica is using a technique called dual conversion gain, which they also use on the Q and SL.

Dual conversion  gain is standard in many modern CMOS sensor and kicks in about ISO 400 or 800..  But just add the D500 or the Sony alpha 6500 curves in Bill' website. You find that they are all behaving nearly identical to the CL. You are correct, the sensors in a lot of cameras are produced by the Sony Semiconductor Cooperation, which differs from Sony Electronics who make the Sony cameras.  Usually companies such as Nikon design tailored sensor for there needs combining Nikon patents with Sony   Semiconductor patents. Hence there are many different Sony sensor depending on the design of integrated after pixel electronics. 

However, for a certain base technology, the BSI-CMOS sensor for 2016/2017, we expect a dynamical range which is very similar. This is in fact the case for most APS-C cameras. I have no problem that the CL might have a slighly different sensor than the TL2 , But I have a problem if there is a claim that the TL2 APS-C Sensor used by Leica as nitch optics company shows a larger (!) dynamical range as the Nikon Z6, with the same pixel count but 2.25 times the area. An Nikon really puts man power onto optimizing the sensors since the advent of the D3.  The published number would imply that Leica would  have found a magnificent way of havesting information on this sensor type which is unprecedented in other cameras even Sonys flagship APS-C. Does that make sense?  Sorry not to me.

[jaapv]

Well, don't forget that portraying Leica as an electronically inept company is not consistent with the facts. Not only are they well-embedded in the Western European industrial complex, which boasts companies like ASML, Philips and Jenoptik, to name a few,  they have full access to technology by Panasonic as well - and Panasonic is one of the few companies in the world that is on par with Sony technologically. Fujitsu is involved as well.

Anyway, confirmation of these figures is needed, I agree.

 

BTW, Bill Claff suggests in an email to me that Leica is varying the black point, with the effect that noise is "hidden", but with the drawback that the deepest shadows are crushed irretrievably.

[ramarren]

6 hours ago, lct said:

I've never seen a better DR out of APS-C compared to MF or FF so far but i may be wrong. Any example of such an APS-C superiority among current sensors? Just curious.

Go to Bill Claff's page, listed in the first post of this thread, and leave the TL2 enabled while you click on other Leica models. The measured TL2 dynamic range is bested at some points only by the Leica S (007), and measures better than the M10 at most points by a small amount. All the others are below the TL2 curve. Same holds for many other cameras ... including what is supposed to be one of the industry leaders (Nikon D850). 

Not that I think this is a useful way of comparing photographic results, mind you.

[jaapv]

Well, the D850 with ~50 MP will have smaller pixels than quite a few other cameras. That will be visible in things like DR and noise. 

[hoppyman]

14 hours ago, lct said:

It is simply not credible (to me) than any Leica APS-C camera does better than MF or FF Leica cameras like the S, SL or M10. YMMV.

I have no idea on the significance or relevance of the test  shown as a practical matter. As an owner of the S (Typ 007), SL and CL I can comment that the S has obviously superior dynamic range where that is a factor in my images.

I sometimes use the CL in room light typically up to 1600 ISO and am very impressed with its noise performance for example. Control and handling wise and most especially due to the finder, for me it’s a superior camera to use than the T’s ( I have an now unused T and its bulky add on EVF)

[SrMi]

9 minutes ago, hoppyman said:

I have no idea on the significance or relevance of the test  shown as a practical matter. As an owner of the S (Typ 007), SL and CL I can comment that the S has obviously superior dynamic range where that is a factor in my images.

I sometimes use the CL in room light typically up to 1600 ISO and am very impressed with its noise performance for example. Control and handling wise and most especially due to the finder, for me it’s a superior camera to use than the T’s ( I have an now unused T and its bulky add on EVF)

I assume you are talking about S 007, not 006.

Your experience with the superior dynamic range of S corresponds to the measurements: http://www.photonstophotos.net/Charts/PDR.htm#Leica S (Typ 007),Leica SL (Typ 601),Leica T

While TL2 measures surprisingly well, it measures worse than S (007): http://www.photonstophotos.net/Charts/PDR.htm#Leica S (Typ 007),Leica TL2

I am not qualified to say how much the measured data is transferable to practice.

[IkarusJohn]

What an interesting toy to play with (not that it really means anything to me in practice) - the TL2 and M10 seem to track together reasonably well.  The Monochrom (v1) doesn't fare as well as I thought it might, nor the SL.  I would have expected the SL and M10 to have been reasonably close, as their development was over a similar period.

http://www.photonstophotos.net/Charts/PDR.htm#Leica S (Typ 007),Leica SL (Typ 601),Leica T

[hoppyman]

14 hours ago, SrMi said:

I assume you are talking about S 007, not 006.

Your experience with the superior dynamic range of S corresponds to the measurements: http://www.photonstophotos.net/Charts/PDR.htm#Leica S (Typ 007),Leica SL (Typ 601),Leica T

While TL2 measures surprisingly well, it measures worse than S (007): http://www.photonstophotos.net/Charts/PDR.htm#Leica S (Typ 007),Leica TL2

I am not qualified to say how much the measured data is transferable to practice.

Yes, as mentioned I am using the (Typ 007) I have shot with a loan S (Typ 06) and I still have my S2.
 

Edited December 6, 2018 by hoppyman

[jankap]

Am 5.12.2018 um 07:42 schrieb jaapv:

Sensor size has little to do with DR.

Is that generally true? The effective area of a pixel is a main parameter still.

Jan

[jaapv]

That is right. Effective area is geometrically, related to well, yes, sensor size, but equally to pixel count and pixel technology. I guess I should insert the words "on its own"  

[tom0511]

I found the cl to give a slightly more contrasty output with a little more pop, and the TL2 a little smoother. But I cant believe there is that much difference after playing a bit in post processing.

[jaapv]

Nor I. It happens too often that it really takes some effort to get the histogram free of clipping at both ends.

[tom0511]

For sure one can do much more post to images of the S and even the SL over dx sensors.

So its more than just larger size and more pixels.

I also enjoy the dx sized sensors, but believe one has to expose more accurate in the first step to get good final image, and its harder to bring back shaddows or midtones compared to the large sensor.

 

[Advertisement]

Hallo,

Look here: Leica L-Mount

[jaapv]

Not that I experience any difficulty managing midtones and shadows on the CL, on the contrary, it is considerable easier than my M240 was - I have no comparison to SL, S  or M10-, but I have always maintained that the beginning of any technically good photograph is correct exposure.