A Classic EF Lens Reaches the End of Production

[neuroanatomist]

...but my wife and I witnessed the murder of a mole (trigger warning: if you are too sensitive to the more brutal facts of life, don't scroll down).

Oh, that just looks like they're playing. This is also not with the EF 100-400L, but with its grandchild the RF 100-500L. Mr. Cooper doesn't play with his food.

"Hawk’s Prey"

EOS R3, RF 100-500mm f/4.5-7.1L IS USM @ 500mm, 1/250 s, f/7.1, ISO 12800

...and neither does his cousin, Red.


EOS 1D X, EF 70-200mm f/2.8L IS II USM @ 200mm, 1/80 s, f/2.8, ISO 320

Both of those were taken at my house.

 

[HeavyPiper]

One from my old files with the EF 100-400mm L. Cropped.

 

[justaCanonuser]

Oh, that just looks like they're playing. This is also not with the EF 100-400L, but with its grandchild the RF 100-500L. Mr. Cooper doesn't play with his food.

"Hawk’s Prey"
View attachment 228907
EOS R3, RF 100-500mm f/4.5-7.1L IS USM @ 500mm, 1/250 s, f/7.1, ISO 12800

...and neither does his cousin, Red.

View attachment 228908
EOS 1D X, EF 70-200mm f/2.8L IS II USM @ 200mm, 1/80 s, f/2.8, ISO 320

Both of those were taken at my house.

great images! You have nice neighbors (as long as you aren't commenting from a prey's viewpoint).

In fact, the stork finally swallowed the mole, after he stabbed the poor guy several times with and afterwards flattened him with his beak.

 

[GMCPhotographics]

The EF 400mm f/5.6 L had quite a following, who thought it very sharp. Then along came the EF 100-400mm ii that was sharper than the prime, had excellent IS, not much heavier and was a zoom. A splendid lens, fully worth its price. Fortunately, it morphed into the RF 100-500 so I don't miss my two 100-400s.

I had the EF 400mm f5.6 L for a long while, which I replaced with the EF 100-400mm f5.6 II LIS. The Prime was amazing for locking on and tracking Birds in flight "coming towards you" shots and was pretty amazing at that one trick. The zoom was way more versatile and superior in every area except when birds came at you head on. I still use this lens with an adapter as my main long tele / short wilde life lens. It takes a 1.4x TC really well, but not the 2x TC so much.

In terms of sharpness, the prime was a legend, however our cameras got more sophiticated and our sensors more densly populated with pixels. In time our camreas became sharper than some of our lenses. The EF 100-400mm f5.6 II LIS was an answer from Canon to higher resolving sensors. Many of the earlier EF lenses were refereshed with sharper optics, the EF 35mm f1.4 II L for example. optically, there's not much between it and the current VCM RF versions (optically).

I'd like to trade it up for the RF 100-500, however, that's a lot of cash for a mild upgrade and I'm still getting great results from my EF zoom. Sure, the RF lens is lighter, doesn't need an adapter, has a removable collar (not just the foot), has slightly better AF and IS and had a bit more reach. That's a lot of £££ for mild upgrades across the board. Perhapse I should flip my EF 100-400mm f5.6 LIS II before it gets too battered and looses too much of it's resale value. It's probably more an economics choice more than the features / specs of the newer lens.

 

[GMCPhotographics]

Here's a few images from my EF 100-400mm f5.6 LIS II, both with a 1.4x TC on a R6mkII.

 

[AlanF]

I had the EF 400mm f5.6 L for a long while, which I replaced with the EF 100-400mm f5.6 II LIS. The Prime was amazing for locking on and tracking Birds in flight "coming towards you" shots and was pretty amazing at that one trick. The zoom was way more versatile and superior in every area except when birds came at you head on. I still use this lens with an adapter as my main long tele / short wilde life lens. It takes a 1.4x TC really well, but not the 2x TC so much.

In terms of sharpness, the prime was a legend, however our cameras got more sophiticated and our sensors more densly populated with pixels. In time our camreas became sharper than some of our lenses. The EF 100-400mm f5.6 II LIS was an answer from Canon to higher resolving sensors. Many of the earlier EF lenses were refereshed with sharper optics, the EF 35mm f1.4 II L for example. optically, there's not much between it and the current VCM RF versions (optically).

I'd like to trade it up for the RF 100-500, however, that's a lot of cash for a mild upgrade and I'm still getting great results from my EF zoom. Sure, the RF lens is lighter, doesn't need an adapter, has a removable collar (not just the foot), has slightly better AF and IS and had a bit more reach. That's a lot of £££ for mild upgrades across the board. Perhapse I should flip my EF 100-400mm f5.6 LIS II before it gets too battered and looses too much of it's resale value. It's probably more an economics choice more than the features / specs of the newer lens.

The EF 100-400mm II was and still is a superb lens. Posted below are the real MTF values measured by Uncle Roger, averaged over several copies of it and the EF 400mm f/5.6. Apart from the extreme corners, the zoom beats it on contrast and increasingly so for higher resolution. Nowadays with the R5ii, AF acquisition seems instantaneous for BIF with my telephotos, and it was certainly damn fast with the EF 100-400mm II on DSLRs using a 9 point centre setting. I really love the RF 100-500mm, and it works well with the 2xTC at 1000mm. I was out yesterday and the 2x extender on when a tiny Longtailed Tit suddenly appeared 10m away, I took a quick shot at 1000mm, a crop from the centre posted below. The shutter speed was too low at only 1/100s but it was still sharp enough (I don't like going below 1/500s as these small birds twitch rapidly and IS in any case is a bit dodgy when cropping such long focal lengths).

 

[GMCPhotographics]

According to optyczne's sensor testing, the R5II scores a value (on his MFT-50 test) of .76. So in a reasonable approximation, any lens that scores a wide open value of over .76 on the Canon MFT charts with keep up with or out resolve the sesnor and render a sharp image. Any lens under this value might render a slightly soft image (which might need a lot more sharpening in post). Although Canon's resoluton and contrast MFT's are slightly different to optyczne's charts, they are both a common and give a similar and relatable score.
When I look at the files from my R6ii and R5, I generally find that I need slightly sharper lenses on my R5 than my R6ii and a lot sharper lenses than I needed on my 5DIII. When I look at optyczne's results for these three cameras, I broadly agree with his results.
When I migrated from my 5DIII to an R8 and R6ii, I broadly found that I need a lens that resolved over .6 lpm on Canon's MFT charts. I was finding that my images were slightly softer with my EF 35m f1.4, EF 24-70mm f2.8 L and 85mm f1.2 II L on my newer cameras. The Af on those cameras were giving those old lenses a new lease of life, but the optics were not as sharp as they once were on the older cameras. If I stopped down a bit, there sharpness was there again.
With my R5, I find the same is broadly true for lenses over .8 lpm on Canon's MFT charts. Thankfully, this puts my EF 100-400mm f5.6 II L and 1.4x TC comfortably in the category, along with my EF 400mm f2.8 LIS II with a 2X TC mkIII. The images I get are super sharp on either camera, wide open.

Here are the results for the canon cameras that he's tested that are of interest to me personally (he has a lot more cameras tested on his site)

R6 60 https://www.optyczne.pl/465.4-Test_aparatu-Canon_EOS_R6_Rozdzielczość.html
R6II 64 https://www.optyczne.pl/485.4-Test_aparatu-Canon_EOS_R6_Mark_II_Rozdzielczość.html
R6III 71 https://www.optyczne.pl/521.4-Test_aparatu-Canon_EOS_R6_Mark_III_Rozdzielczość.html
R5II 76 https://www.optyczne.pl/510.4-Test_aparatu-Canon_EOS_R5_Mark_II_Rozdzielczość.html
R5 82 https://www.optyczne.pl/457.4-Test_aparatu-Canon_EOS_R5_Rozdzielczość.html
R7 93 https://www.optyczne.pl/483.4-Test_aparatu-Canon_EOS_R7_Rozdzielczość.html
EOS-R 54 https://www.optyczne.pl/413.4-Test_aparatu-Canon_EOS_R_Rozdzielczość.html
R8 65 https://www.optyczne.pl/490.4-Test_aparatu-Canon_EOS_R8_Rozdzielczość.html
5DII 42 https://www.optyczne.pl/92.4-Test_aparatu-Canon_EOS_5D_Mark_II_Rozdzielczość.html
5DIII 46 https://www.optyczne.pl/179.4-Test_aparatu-Canon_EOS_5D_Mark_III_Rozdzielczość.html
5DIV 55 https://www.optyczne.pl/351.4-Test_aparatu-Canon_EOS_5D_Mark_IV_Rozdzielczość.html
5Ds 67 https://www.optyczne.pl/324.4-Test_aparatu-Canon_EOS_5Ds_Rozdzielczość.html
5DSR 79 https://www.optyczne.pl/324.4-Test_aparatu-Canon_EOS_5Ds_Rozdzielczość.html

You might find some fault in his methodology or even my attempt to corelate optical sharpness to sensor resolution. But for me and what i've seeing, I would strongly suggest that these figures work and that the two sytems are roughly comparible. Maybe a slight adjustment between the two systems is in order, but as a generalisation they seem to work.

However...in real world use, I'm sure we can take a slightly optically soft image and apply some post processing sharpness routines to eek out more resolution that what was appraent in the original source image. Image Sharpensing ahs come a long way over the last 3-4 years.

 

[GMCPhotographics]

The EF 100-400mm II was and still is a superb lens. Posted below are the real MTF values measured by Uncle Roger, averaged over several copies of it and the EF 400mm f/5.6. Apart from the extreme corners, the zoom beats it on contrast and increasingly so for higher resolution. Nowadays with the R5ii, AF acquisition seems instantaneous for BIF with my telephotos, and it was certainly damn fast with the EF 100-400mm II on DSLRs using a 9 point centre setting. I really love the RF 100-500mm, and it works well with the 2xTC at 1000mm. I was out yesterday and the 2x extender on when a tiny Longtailed Tit suddenly appeared 10m away, I took a quick shot at 1000mm, a crop from the centre posted below. The shutter speed was too low at only 1/100s but it was still sharp enough (I don't like going below 1/500s as these small birds twitch rapidly and IS in any case is a bit dodgy when cropping such long focal lengths).

View attachment 229324View attachment 229325View attachment 229326

Hi Frank, using this data and assuming https://www.optyczne.pl/testy_aparatów.html site's tesing is accurate / correct then any lens that scores over .75 lpm will render optically sharp images on your R5ii. So the resolution difference between these two optics is clear in the MFT charts but won't necessarily carry over to anything observable on a camera sensor that has a max optical sharpness of around .75 lpm.
the only time we really get to see anything observable is when the optical resoluton drops to below the sensor's required value, usually when we use tele coverters. I've found that my EF 100-400mm f5.6 II L isn't very sharp with my R6ii when using a 2x TC. This correlates with the MFT data from Canon.
Your RF 100-500mm LIS seems to fare a bit better than my EF 100-400 with a 2X Tc and it is quite possible that is in this area we see (and prove) that the RF 100-500 LIS is a slightly optically shaper lens.

 

[AlanF]

Hi Frank, using this data and assuming https://www.optyczne.pl/testy_aparatów.html site's tesing is accurate / correct then any lens that scores over .75 lpm will render optically sharp images on your R5ii. So the resolution difference between these two optics is clear in the MFT charts but won't necessarily carry over to anything observable on a camera sensor that has a max optical sharpness of around .75 lpm.
the only time we really get to see anything observable is when the optical resoluton drops to below the sensor's required value, usually when we use tele coverters. I've found that my EF 100-400mm f5.6 II L isn't very sharp with my R6ii when using a 2x TC. This correlates with the MFT data from Canon.
Your RF 100-500mm LIS seems to fare a bit better than my EF 100-400 with a 2X Tc and it is quite possible that is in this area we see (and prove) that the RF 100-500 LIS is a slightly optically shaper lens.

To be frank as I am not Frank, I can do better than optyczne and put some approximate numbers on things and try and give some meaning to the MTF in simple terms that is relevant to this discussion. A crude analogy is one with human vision, in a way pertinent to telephoto lenses. If you have 20/20 vision you are deemed normal. 20/30 is you resolve at 20' what normal vision would resolve at 30' and so are worse than normal, and 20/15 means you are better than average and resolve at 20' what normal would resolve at 15' (or 6/6, 6/9, 6/4.5 in m for metric). Similarly, a lens with MTF 0.75 has to be at 15' away to resolve detail as well as one with an MTF of 1.0 at 20'. The MTF at 50 lp/mm, the highest detail measured in those charts for the 100-400 zoom is 0.61 and for the 400/5.6 prime is 0.55. That is telling you the zoom resolves roughly as well at say 65m as the prime at does at 55mm for high detail. (On top of all of this you acutance differences etc.)

I think the difference between the EF and RF 2x extenders on the lenses is not necessary the sharpness of the two lenses but in the way the extenders are tailored to the lenses.

 

[GMCPhotographics]

To be frank as I am not Frank, I can do better than optyczne and put some approximate numbers on things and try and give some meaning to the MTF in simple terms that is relevant to this discussion. A crude analogy is one with human vision, in a way pertinent to telephoto lenses. If you have 20/20 vision you are deemed normal. 20/30 is you resolve at 20' what normal vision would resolve at 30' and so are worse than normal, and 20/15 means you are better than average and resolve at 20' what normal would resolve at 15' (or 6/6, 6/9, 6/4.5 in m for metric). Similarly, a lens with MTF 0.75 has to be at 15' away to resolve detail as well as one with an MTF of 1.0 at 20'. The MTF at 50 lp/mm, the highest detail measured in those charts for the 100-400 zoom is 0.61 and for the 400/5.6 prime is 0.55. That is telling you the zoom resolves roughly as well at say 65m as the prime at does at 55mm for high detail. (On top of all of this you acutance differences etc.)

I think the difference between the EF and RF 2x extenders on the lenses is not necessary the sharpness of the two lenses but in the way the extenders are tailored to the lenses.

Sorry Alan....I was messaging another chap, over on Whatsapp called Frank at the same time as i was writing my forum reply here and for some reason my brain got the two names mixed, I am so sorry.
Canon's lens MFT charts are measuring lens optical resolution at 30 lpm and contrast at 10 lpm. Optyczne's sensor tests are measuring 50 lpm, which is similar to what some other lens brands have used in the past to determin their Lens optical MFT charts. However, the two approaches do align to some degree and hold merit, but there needs to be adjustments between the two scales. As they currently stand, they are a reasonable approximation between sensor and lens sharpness testing.

 

[AlanF]

Sorry Alan....I was messaging another chap, over on Whatsapp called Frank at the same time as i was writing my forum reply here and for some reason my brain got the two names mixed, I am so sorry.
Canon's lens MFT charts are measuring lens optical resolution at 30 lpm and contrast at 10 lpm. Optyczne's sensor tests are measuring 50 lpm, which is similar to what some other lens brands have used in the past to determin their Lens optical MFT charts. However, the two approaches do align to some degree and hold merit, but there needs to be adjustments between the two scales. As they currently stand, they are a reasonable approximation between sensor and lens sharpness testing.

What I was getting at is that a lens of MTF 0.75 will give a sharp image under certain conditions, but a lens of MTF 1.0 will give you a sharp image under more extreme conditions. If it's a telephoto lens, it will resolve as well 30% further away.

The overall MTF of the sensor/lens system, MTF(L/S) = MTF(L)xMTF(S), that is the product of the MTF of the Lens times that of Sensor. A lens of MTF 0.75 will lower the overall MTF(L/S) by 25%, irrespective of the resolution of the sensor.