Geek Articles

Just MTF Charts: Canon Prime Lenses

By
Published March 25, 2019
Canon 85mm f/1.4L IS MTF Testing

This is the first post of a series of posts publishing all of our MTF results so that methodology is consistent, easy to find, and up-to-date. (For some of the older lenses there are graphs done with older software floating around. For a couple of lenses there are incorrect graphs done before we worked out all of the details for sensor glass in the testing pathway. These are all current.)

Otherwise, there are no comparisons, no commentary, just the test results for you to use and abuse as you see fit. The major purpose is just to get these MTF charts organized and in one place where you can find them.

Just to avoid 4,200 comments: these are all average MTFs from multiple copies. They are all measured at the widest aperture. Yes, I know you’d like several apertures. I haven’t got the time or money to do that. Sorry.

A Quick How to on Reading MTF Charts

If you’re new here, you’ll see we have a scientific methodology to our approach, and use MTF charts to measure lens resolution and sharpness. All of our MTF charts test ten of the same lenses, and then we average out the results. MTF (or (or Modulation Transfer Function) Charts measure the optical potential of a lens by plotting the contrast and resolution of the lens from the center to the outer corners of the frame. An MTF chart has two axis, the y-axis (vertical) and the x-axis (horizontal).

The y-axis (vertical) measures how accurately the lens reproduces the object (sharpness), where 1.0 would be the theoretical “perfect lens”. The x-axis (horizontal) measures the distance from the center of a lens to the edges (measured in millimeters where 0mm represents the center, and 20mm represents the corner point). Generally, a lens has the greatest theoretical sharpness in the center, with the sharpness being reduced in the corners.

Tangential & Sagittal Lines

The graph then plots two sets of five different ranges. These sets are broken down into Tangential lines (solid lines on our graphs) and Sagittal (dotted lines on our graphs). Sagittal lines are a pattern where the lines are oriented parallel to a line through the center of the image. Tangential (or Meridonial)  lines are tested where the lines are aligned perpendicular to a line through the center of the image.

From there, the Sagittal and Tangential tests are done in 5 sets, started at 10 lines per millimeter (lp/mm), all the way up to 50 lines per millimeter (lp/mm). To put this in layman’s terms, the higher lp/mm measure how well the lens resolves fine detail. So, higher MTF is better than lower, and less separation of the sagittal and tangential lines are better than a lot of separation. Please keep in mind this is a simple introduction to MTF charts, for a more scientific explanation, feel free to read this article.

 

Canon Prime Lenses

Canon 14mm f2.8 L MkII

Lensrentals.com, 2019

 

Canon 20mm f2.8 USM

Lensrentals.com, 2019

Canon 24mm f1.4 L Mk II

Lensrentals.com, 2019

Canon 24mm f2.8 IS

Lensrentals.com, 2019

Canon 28mm f1.8

Lensrentals.com, 2019

Canon 28mm f2.8 IS

Lensrentals.com, 2019

Canon 35mm f1.4 L

Note – this is a 5 sample chart because we didn’t have many left when we retested for proper cover glass thickness. 

Lensrentals.com, 2019

Canon 35mm f1.4 L Mk II

Lensrentals.com, 2019

Canon 35mm f2 IS

Lensrentals.com, 2019

Canon 40mm f2.8  STM

Lensrentals.com, 2019

 

Canon 50mm f1.2 L

Lensrentals.com, 2019

Canon 50mm f1.4

Lensrentals.com, 2019

Canon 50mm f1.8 STM

Lensrentals.com, 2019

 

Canon 85mm f1.2 L Mk II

 

Lensrentals.com, 2019

Canon 85mm f1.4 IS L

Lensrentals.com, 2019

Canon 85mn f1.8

 

Lensrentals.com, 2019

Canon 100mm f/2

 

Lensrentals.com, 2019

Canon 100mm 2.8 Macro USM

 

Lensrentals.com, 2019

Canon 100mm f2.8 Macro IS L

 

Lensrentals.com, 2019

Canon 135mm f2.0 L

Lensrentals.com, 2019

 

Canon 200mm f2.0L II

Lensrentals.com, 2019

 

 

 

That’s all the Canon primes we have done except tilt-shifts, which will come out separately.

For a look at all the Just MTF Articles we’ve done so far, be sure to check them out here.

 

Roger Cicala, Aaron Closz, and Brandon Dube

Lensrentals.com

March, 2019

Author: Roger Cicala

I’m Roger and I am the founder of Lensrentals.com. Hailed as one of the optic nerds here, I enjoy shooting collimated light through 30X microscope objectives in my spare time. When I do take real pictures I like using something different: a Medium format, or Pentax K1, or a Sony RX1R.

Tweet
Posted in Geek Articles
← 
Newer Comments  →

  • Just want to say “thank you” for this and the other similar work you share.

    Dan

  • To be honest, with the 85m f1.8 and 100mm f/2, the sample variation is large, so given one copy of each, I’d hesitate to guess which would be sharper. Remember also the 100 is being tested at a slightly smaller aperture, which helps it a bit.

  • Donga

    Between 85mm f/1.8 and 100mm f/2, do we notice these MTF chart differences in real use? I have 100mm f/2. I like this focal length over 85mm. I have 50mm stm lens. 85mm is too close. I do not have 135mm. I am using 100mm for portraits.

  • Frank, a good question and one I’ll try to answer here. I’d also suggest asking Klaus to do the same. Optical Limits is, in my opinion, the best and most consistent of the Imatest review sites. They also do a very thorough job of analyzing distortion, vignetting, and presenting stopped down data, which I don’t.

    So, to your point, the differences are as follows.

    We’re testing at infinity using an optical bench; no camera. They test at closer distances (varies with focal lengths but maybe 5-6 feet on a wide angle, 30 feet on a telephoto) using a camera.

    The MTF bench gives a lot more data and presents it as curves. Imatest gives a subset of data, usually the MTF50. One nice thing is you can estimate the MTF50 at different frequencies from the MTF graph, but not to the degree they measure it. In the graph below, the yellow dots are basically the MT50 at center, mid, and corner.

    https://uploads.disquscdn.com/images/7d40dbd72c170577638ad5ae4ba4e4c56301ef490d7a5e25d83edcd85acaa9e1.png https://uploads.disquscdn.com/images/7d40dbd72c170577638ad5ae4ba4e4c56301ef490d7a5e25d83edcd85acaa9e1.png

    In a well-controlled Imatest setup like optical limits has, you can make some comparisons between what the lens alone looks like at infinity, and what the lens-on-camera looks like close up. Some lenses do better closer, some worse, most not too different. Some cameras apply in-camera sharpening and corrections (OpticalLimits.com turns everything off they can, but in a few cases, not Canon that we know off, there’s some correction applied to the RAW files.)

    And finally, as you point out, it’s easier to read 3 or 4 points than 80 points.

    Roger

  • Actually tested way over 100. That’s just the subset I grabbed for charting.

  • Most are too big to test on the optical bench. It has size constraints. We can work around them, but since we’re technically overloading the bench the results are questionable.

  • Alan Fersht

    Your measurements on 6 copies of the 400mm f/2.8 II, 8 of the 400mm f/5.6 and, admittedly, just 2 of the DO II on an adapted optical bench may not be up to your usual standards but are far, far more credible than the one-off measurements of single copies using charts under ill-defined conditions that is the stock-in-trade of virtually every other lens-testing site. And you did state the caveats.

  • JJason Lee

    Nice work here. Wonder if you would be doing the same for super-tele photo lenses?

  • Andreas Werle

    Thanks for this Roger! Did you realy test 40 Samples of the 50/1,2, perhaps a writing error? Greetings Andy

  • Hunter45

    And the 85 1.8 is listed as: “Canon 85mn f1.8”.

  • Hunter45

    And thankfully I have one…….

  • They are, although we’re still waiting on parts for those. Making the mounts is pretty complex: we need a camera mount, the lens bayonet, the electronics for both. Then the camera mount has to have an optical bench plate machined to receive it which can be a bit complex and is to really high tolerances. After all that is done we have to shim the mount to the plate so it’s optically flat and wire everything together in a way that still lets the plate rotate 360 degrees.

    That requires sacrificing a lens and at least an adapter; we’d obviously rather sacrifice a well used one of each.

  • Woops. I’ll add a text to show that. Thx.

  • Someone

    What about Canon and Nikon mirrorless mounts? Are these in the works too?

  • Michael

    Are you sure you are measuring the 85mm 1.2 L instead of the 85mm f1.2 L Mk II? I ask because all your other lenses are the latest versions.

  • dadohead

    “Fanpeople are why we can’t have nice tests.” LOL. A most excellent Rogerism.

  • It’s all about getting the mount made, which I know I’ve been saying for a year. But it’s actually at the machinist now.

  • In the center, yes almost all can. Most primes do easily at f/2.8. Towards the edge many primes can reach that at f/4, and some zooms at f/5.6.

  • Thanks for the hard work, Roger et al. Any chance we will see some Fuji charts as well?

  • Ernest Green

    Thank you! Great list. I understand you can’t test all apertures on all lenses, but for a point of reference, could you take a middle of the road lens such as say, the 100 F2, and test it at f/5.6? I am curious what the 50 line pair would be? Over 5? In other words, can a number of lenses, if not most, achieve a score of over 5 with the 50 line pair test? Cause for me, I use a lot of these lenses for landscape stuff where i wouldn’t shoot them wide open. So it would be nice to see how fine detail is resolved at 5.6 (the most common sharpest aperture for full frame glass)

  • xeppelin

    hehe, fully confirms my opinion, that EF 40/2.8 STM and 50/1.8 STM deliver max. bang for the buck.

  • Brandon Dube

    All at full aperture. The “chart” is infinitely far away.

  • Max Manzan

    All tests wide open at infinity.

  • xeppelin

    sorry, if i overlooked it, but don’t see at what f-stop … all charts for lens wide open?

  • Alan, if you recall that article, I made people swear a solemn oath before reading it not to take it out of context: the bench was over its supposed limits and the number of copies were low – it was an experiment. But, of course, people made all kinds of silly statements that all started with “Roger Cicala showed”. So I’m not going to put experimental results in here with things that I feel the results are scientific.

    It would just give more fodder to the Fanpeople. Fanpeople are why we can’t have nice tests.

  • Ed, we’ll have all the primes out by brand and mount within a couple of weeks. Then I’ll start the zooms.

  • Joe Blow

    To my untrained eyes, the 135mm f2 looks pretty darn good. Maybe this is why Canon hasn’t upgraded this lens for years.

  • Brenda

    Most recent poll shows that more than 75% people are occupied into online jobs. Online world-wide is becoming bigger and better and bringing an ample amount of money making opportunities. Working at home on line tasks are trending and transforming individual’s everyday lives. Exactly why it really is extremely popular? Mainly because it allows you to work from anywhere and any time. You receive more time to spend with all your family and can plan out tours for holidays. Persons are generating great income of $22000 each week by utilizing the efficient and smart approaches. Carrying out right work in a right direction will always lead us towards success. You can start to earn from the 1st day after you browse through our web-site. >>>>> PLU.SH/njexd

  • Ed Hassell

    Thank you for all your hard work. A few surprised me. Others were pretty much as I expected them to be.

    I’m mainly a Nikon shooter (hint for a similar post of Nikkor optics); however, I have used Canon equipment belonging to friends often over the years. I figure the two companies are more similar than different and leap-frog each other with their technologies from time to time. I’m way too invested in glass that I have no desire to replace to switch systems just because someone has a better widget this year.

  • Samuel H

    It’s official: I’ve been spoiled by Sigma

    (Thanks for these!)

←  
Newer Comments  →
Follow on Feedly