Jump to content

Leica T or APO-Extender-R 2x?


Recommended Posts

Advertisement (gone after registration)

I have a M10 with a APO-Telyl 3,4/180mm and Visoflex 020. I would like to get some more reach with the 180mm. One way to do that is with an APO-Expender 2X. Another way is with aT because it has more pixels pr area than the M10.

I think the Leica T have 1,25 more pixels pr area than the M10 (I could be wrong). The Extender obviously degrade the optical quality, but is it better or worse than an upscaling of a 16MP picture from a Leica T?

 

I ask this because a used model T 16MP and APO-Extender 2X cost about the same. 4-600,- Euro.

Link to post
Share on other sites

Actually the image degradation of the APO-Extender is minimal and can be disregarded for normal use. However, the T as such will give a 1.5x extender effect and improve the rendering of the lens. In the end, I doubt whether you will see much difference between the two.

Link to post
Share on other sites

Are you sure about the 1,5x extender effect on a 16MP model T vs a 24MP M10?

Anyway. I would like to see some pictures that show the difference between a APO-Telyt and a APO-Telyt with APO-Extender-R 2X resized to the same size. It do not need to be 3,4/180. Any APO R lens from 70-80-90 will do.

Link to post
Share on other sites

The "extender effect" has nothing to do at all with the number of MP, but with the sensor size.

The resolution when comparing cropping (AKA smaller sensor) versus extender is obviously another matter, but then, image quality is not really determined by resolution. The resolution will determine the amount of small detail contained in the file -only visible at large magnification-, but not the way that detail is rendered.

 

 

 

 

0

Link to post
Share on other sites

Extender eats light for breakfast, especially 2x ones. So it should be better to mount it on the T. 

The T is like a 1.4x teleconverter. You lose around 1 stop of light and get 1.5x crop

a 2x extender will be like using m4/3 camera You loose around 2 stops and get 2x crop. 

Edited by nicci78
Link to post
Share on other sites

Advertisement (gone after registration)

You don't lose any light when you use a smaller format, you just lose subject area. The amount of light energy falling per unit area of the sensor is the same.

A teleconverter loses light because it expands the image circle of the lens: the total light energy coming through the lens is invariant, so if you expand the image circle, the total energy per unit area falling on the sensor is reduced. 

 

50 minutes ago, nicci78 said:

Extender eats light for breakfast, especially 2x ones. So it should be better to mount it on the T. 

The T is like a 1.4x teleconverter. You lose around 1 stop of light and get 1.5x crop

a 2x extender will be like using m4/3 camera You loose around 2 stops and get 2x crop. 

 

  • Thanks 1
Link to post
Share on other sites

Whatever, so you get your answer.

T is a better option than light eating teleconverter, if we follow your logic. 

Anyway full frame sensor has more unit of area as you say. So more light in total.

 

For a month I am using a Q and CL+TL18 alongside, for comparison sake.  

Almost same angle of view. ISO quality match each other. Ex : Q ISO 1600 match exactly CL ISO 1600 with same amount of noise and sharpness. 

Anyway I consistently found that CL receive two thirds of a stop less light than the Q with the same exposure values.

That’s why I am pretty sure that full frame sensor received more light than APS-C in total. And it perfectly makes sense. 

Edited by nicci78
Link to post
Share on other sites

Nonsense/bollocks - same aperture+shutterspeed = same exposure and image density, irrespective of sensor size.

A converter loses you aperture stops, so you need use a longer shutter speed.

Godfrey is 100% right.

  • Like 1
Link to post
Share on other sites

4 hours ago, nicci78 said:

Whatever, so you get your answer.

T is a better option than light eating teleconverter, if we follow your logic. 

Anyway full frame sensor has more unit of area as you say. So more light in total.

 

For a month I am using a Q and CL+TL18 alongside, for comparison sake.  

Almost same angle of view. ISO quality match each other. Ex : Q ISO 1600 match exactly CL ISO 1600 with same amount of noise and sharpness. 

Anyway I consistently found that CL receive two thirds of a stop less light than the Q with the same exposure values.

That’s why I am pretty sure that full frame sensor received more light than APS-C in total. And it perfectly makes sense. 

Um, no. :) 

Different lenses have different T-values and different sensors have different sensitivities. Never mind different image processing/data handling systems in the cameras, different metering systems (with different calibrations). Are we speaking of in-camera JPEGs or raw captures? ... And all that. I would not leap to the notion that the difference in sensor size is an immediate given as to why your best exposure might be slightly different on one camera vs another ... It takes a MUCH more rigorous set of specifications and tests to isolate and determine that as the critical factor. 

Speaking to the original post: 

The 1.5x magnification gain with an APS-C format camera is generally painless, but carrying a camera with different controls, different batteries, etc, poses a much larger burden on you than carrying a 2x teleconverter and having to accommodate the differences in exposure, imaging, etc. Whether it's worth it or not depends on what you're trying to do. Cropping FF to APS-C format using a FF 25 Mpixel sensor nets a 10 Mpixel image, which is often enough for long telephoto work in my experience; I don't know whether it's sensible to buy a T just to do that, since you're only gaining back about half of the pixel losses from doing a simple crop, and less than that of the resolution (which is a square root function on the area difference). 

I used very long lenses fairly infrequently so for me the sensible thing to do was to buy a 2x Extender-R to use with my Elmar-R 180mm f/4 lens for the occasional need. While not the ultimate in lens performance, it produces very nice results still and the telextender cost me less than $60, so it was worth it even if only for the experimentation it enabled. That was when I had the SL. Now I can achieve a 540mm f/8 lens's eqFoV with the same equipment on the CL body ... I did some testing of the setup and was not disappointed with the results. An example at fairly close up distance ...

Wide open: 

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!

Two stops down: 

Edited by ramarren
  • Like 1
  • Thanks 1
Link to post
Share on other sites

On 1/12/2019 at 2:16 AM, jaapv said:

Actually the image degradation of the APO-Extender is minimal and can be disregarded for normal use. However, the T as such will give a 1.5x extender effect and improve the rendering of the lens. In the end, I doubt whether you will see much difference between the two.

How are you so sure?  After one year of CL. I notice that it uses higher ISO value than sunny 16 required. This phenomena makes me sell my Super-TL 11-23mm and APO-TL 55-135mm. Because there are too slow for indoor use. 

In my experience in low light environment, f/2.8 lens is the absolute minimum for CL. Otherwise I had to use 25K ISO or more. I never needed 25K ISO with M10.

 

So where is the truth ? 

In my experience there is a 2/3 stop difference in exposure between Q and CL. The difference between 24x36 and APS-C should be 5/4 stop of difference. But only 2/3 in my case. Maybe better transmission from the tiny Elmarit-TL 18mm (8 lenses in 6 groups) than the Summilux-Q 28mm (11 lenses in 9 groups) and more efficient 24MP APS-C sensor technology. But all theses differences should factor in favour of CL. While there are still 2/3 stop less light hitting the CL than the Q. Thus I have to adjust for the 2/3 difference to get same file density and same exposure. 

Exactly same difference between Q and CL + Super-Elmar-M 21mm asph. But field of view is a little bit different : 28 vs 31mm. So distance was also adjusted. 

 

However there is zero exposure difference between Q, M (typ 262) and M10. 

I also check that same ISO value between Q and CL yields to same results (noise and sharpness) at every ISO value except at 50 000

 

So if you have a better explanation, with proof, than just plain faith in your theory, I am open to read it.

 

 

Edited by nicci78
Link to post
Share on other sites

Unfortunately, your observation has nothing to do with sensor size.

It is a well-established fact that there can be considerable differences between the rated ISO values of different digital camera types - up to two stops-, due to the way the ISO standard is defined. It has been discussed often in this forum, no need to fall into repetition.

Secondly, focal length( rather, field of view) differences between wide-angle, normal and tele render different exposure values.

Thirdly, different internal exposure meters give different results, this was also the case in the film era.

Fourth, different lenses have different T-stop values, which can vary over half a stop compared to the f-stop

On top of that, in my experience, the CL has a tendency to expose to the left by the histogram by up to one stop, depending on circumstances, probably to protect the highlights

Exposure has nothing to do with sensor size, as it is defined as the amount of light transmitted onto (say) a square mm of sensor area. The number of square millimeters does not change it. It is a property of the lens. The sensitivity of the sensor itself plus the in-camera processing determine the values that you see.

If your idea were correct, there would be different handheld exposure meters for different film sizes, and a larger mirror would give a brighter reflection than a smaller one...

 

I wish that the Internet Idiot Bloggers hadn't started this particular red herring.

  • Like 2
Link to post
Share on other sites

I will rephrase my discovery. CL has a 2/3 stops disadvantage in term of image quality against the Q for the same exposure value. 

Is it better ? 

So 2/3 of a stop worst noise. Instead of 5/4 of a stop, if the sensor were made from the same manufacturer, with the very same technology. 

So the CL is very good for an APS-C sensor. Or the Q is very bad for a Full frame sensor. How to you see it ? 

Unfortunately, for this topic, we have to measure how the T performs against the M10. Anyway a 2x teleconverter will sure yield to 2 stops worst noise level. 

Edited by nicci78
Link to post
Share on other sites

Well, the CL is a generation ahead of the Q in sensor technology, so the terms "better" or "worse" are not really to the point. Nor is sensor quality determined by noise performance alone. There are many other aspects. There is only one way to find out. Print. Large.

But yes, in the end, per pixel capacity to absorb photons is related to the sensitive surface of the pixel, A higher pixel density will result in smaller pixels. Thus reducing sensitivity and dynamic range - given identical sensor technology - and that's the rub. Complicated, of course, by the subsequent in-camera processing.

However, this has nothing to do with the physical size of the sensor. You should compare a 24 MP APS-C sensor with a 50 MP ff one - and still notice little  if any difference in the end  Unless you print as large as a 100% crop on your screen - like 0.001 % of photographers. ;)

Link to post
Share on other sites

You incited some curiosity on my part. I set up an exposure target with lighting and used the same lens on my CL, M-D 262, and Olympus E-M1 (a Pentax-L 43mm) to make manual exposures at the same exposure setting as determined by my Sekonic L358 incident meter. I did the M-D last, and the only hitch was that the M-D didn't have the nominal 1/20 second that the manual exposure on the CL and E-M1 called for. But eh? Between 1/15 and 1/20 sec is a minuscule change ... I just bracketed all the E-M1 and M-D shots by half stops from f/4 to f/8. 

Then I took the raw files into Lightroom, cropped the M-D and CL frames to show the exposure target exactly the same as the E-M1 (with the smallest sensor). And measured the illumination on the tri-tone target's white, black, and gray sections at the nominal exposure. The overall histograms at the nominal exposure in LR are almost identical, and the difference in readings between the RGB settings at the white patch amount to less than 4% in R, G, and B channels. Compared to the M-D, the CL recorded 0.1EV darker, and the E-M1 recorded 0.2EV brighter. 

Since the Xrite color checker was in the frame for these tests, I then generated a custom color profile for each of the cameras, with the Xrite Passport software, and checked the default processing. Now the whites, blacks, and greys all match to within 0.5%. 

This proves beyond a shadow of a doubt that the illumination captures by three different formats using the same lens and same lighting target is absolutely identical, all three of my cameras have shutter calibrations that are well within tolerance, and that the largest influence on the default rendering output is the camera calibration profile rather than anything else. The format has NO influence on the exposure per unit area, and as I look through the patches on the color checker and compare the RGB values, the differences in the sensors' color response is so small as to be inconsequential for all but work requiring the most critical color accuracy. 

Fun stuff. It didn't prove anything about format vs exposure that I didn't already know, but it confirmed that all three of these cameras are working well. It also showed that the M-D's camera calibration profile produces the warmest, "softest" color tones of the three standard CCPs ... which is useful to know since I really like the M-D's default palette. I can build a custom profile for the CL which will allow it to be exactly the same if I feel so moved now. :D 

If you do a similar test with your Q and M10, and find there is still a largish difference, the most probable cause is shutter calibration differing between the two cameras. Maybe enough that I'd have both checked and adjusted by Leica. 

  • Thanks 2
Link to post
Share on other sites

21 hours ago, jaapv said:

 

However, this has nothing to do with the physical size of the sensor. You should compare a 24 MP APS-C sensor with a 50 MP ff one - and still notice little  if any difference in the end  Unless you print as large as a 100% crop on your screen - like 0.001 % of photographers. ;)

Funny, but I do only print large. I see no reason to print smaller than 30x45 size photographs. I am already looking at them on an iMac 27"

If I do not print them larger, what's the point ? 

If I need smaller photographs I look at them on my iPhone. Otherwise I print one to two photo books per year. 

 

By the way, I really shoot everyday in dim environment, so f/2,8 is really my absolute minimum for CL. And of course I only shoot with auto ISO on 3x per focal length, to be sure to get zero motion blur. I can get away with it at 2x, but I still get some blurry pictures. Even my Q is set to 1/120th minimum, because 1/60th with OIS is not reliable enough. 

When I used the 11-23mm, I only shoot with 1/60 minimum. 

About 55-135mm, I preferred 4x per focal length. 

So now you see why I need fast optics. And please don't tell me to hold my breath to get a perfect 1/10th shot handheld. My toddler do not stop when I am taking a photograph. 

Edited by nicci78
Link to post
Share on other sites

To go back to Q vs CL differences. I just think that ISO real value is different between the two cameras. 

Anyway, as expected the CL has a little bit worst ISO performance than the Q.

However the CL has much better dynamic range at 100-400 ISO range than the Q. 

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...