Here is the Canon RF 100-400mm f/5.6-8 IS USM

LogicExtremist

Lux pictor
Sep 26, 2021
322
227
I've just made an interesting discovery, looking at the test results from the following review, this is essentially a RF 400mm f/11 lens when used at the long end, the shorter focal lengths and apertures are just a bonus.

In Canon's budget long telephoto range, we really have a zoom that operates optimally at 400m f/11 and primes that are fixed at 600mm f/11 and 800 f/11.

See extract from review below:


"Sharpness:

The sharpness scores in the two graphs below are produced by shooting a monochrome test chart covered in multiple sharp boundaries between black and white. This image is then assessed by specialist software, with the extent of blur on the contrast boundaries at the centre, mid and edges of the image converted into a spatial frequency value to determine how many line widths per picture height the lens is capable of resolving. A higher spatial frequency corresponds to a greater number of finer lines over a given distance that the lens can resolve - this number is the sharpness score.


nUK9MLVkXdS9h8qBNFGsuJ-970-80.png


KjFhxSDjF3VoYBVnm5YgzJ-970-80.png



Center sharpness is very good at all focal lengths and apertures, apart from a small blip wide open at 400mm. Corner sharpness is much less convincing, however, with the lens being noticeably softer regardless of focal length or aperture."


Using the TDP test tool, we can conform this when comparing the image at f/8 vs f/11

Use the larger, magnified view of the centre sharpness in the lower half of the test tool. The difference is discernible, but not drastically, probably because the test is done using the R5, which is well into diffraction territory at f/8, and more affected by diffraction at f/11, which iI suspect is offsetting a lot of the sharpness gains from stopping down the aperture.

 
Last edited:

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
I've just made an interesting discovery, looking at the test results from the following review, this is essentially a RF 400mm f/11 lens when used at the long end, the shorter focal lengths and apertures are just a bonus.

In Canon's budget long telephoto range, we really have a zoom that operates optimally at 400m f/11 and primes that are fixed at 600mm f/11 and 800 f/11.

See extract from review below:


"Sharpness:

The sharpness scores in the two graphs below are produced by shooting a monochrome test chart covered in multiple sharp boundaries between black and white. This image is then assessed by specialist software, with the extent of blur on the contrast boundaries at the centre, mid and edges of the image converted into a spatial frequency value to determine how many line widths per picture height the lens is capable of resolving. A higher spatial frequency corresponds to a greater number of finer lines over a given distance that the lens can resolve - this number is the sharpness score.


nUK9MLVkXdS9h8qBNFGsuJ-970-80.png


KjFhxSDjF3VoYBVnm5YgzJ-970-80.png



Center sharpness is very good at all focal lengths and apertures, apart from a small blip wide open at 400mm. Corner sharpness is much less convincing, however, with the lens being noticeably softer regardless of focal length or aperture."


Using the TDP test tool, we can conform this when comparing the image at f/8 vs f/11

Use the larger, magnified view of the centre sharpness in the lower half of the test tool. The difference is discernible, but not drastically, probably because the test is done using the R5, which is well into diffraction territory at f/8, and more affected by diffraction at f/11, which iI suspect is offsetting a lot of the sharpness gains from stopping down the aperture.

1. Don't waste your time making fine comparisons on the TDP site's image tool because some are no more than rough guides. There are glaring examples of where they are really misleading. Others are spot on, but you don't which.
2. What distances were the MTF measurements made at? Even if the measurements are done well, and there is not copy variation of the lenses, the MTF values can vary greatly with distance. Some lenses are optimised for infinity, some for medium distances, and others close up. I am willing to bet that the digital camera world's were not done at distances for which a 400mm lens is typically used at.

I test my wild-life lenses at close up distances of 1-3m, for where I use them for dragonflies and butterflies etc, which is shorter than typical for Imatest used as your example, and 20m, for birding etc, which is far further than used for Imatesting of MTFs. For example, the RF 100-400mm and RF 100-500mm, are excellent at very close up, and significantly better than the EF 100-400mm II. At 20m, they all perform very well. The EF 100-400mm II + 2xTC is OK at longer distances but much weaker close up whereas the 100-500mm II with RF 2x is outstanding close up and at 20m. The Nikon lenses like the 500PF and 200-500mm are optimised for shorter distances.

I'll check out the RF 100-400mm at f/8 and f/11 at 20m on the R5 and R6 to see how they compare and report back later.
 
  • Like
Reactions: 1 user

LogicExtremist

Lux pictor
Sep 26, 2021
322
227
1. Don't waste your time making fine comparisons on the TDP site's image tool because some are no more than rough guides. There are glaring examples of where they are really misleading. Others are spot on, but you don't which.
2. What distances were the MTF measurements made at? Even if the measurements are done well, and there is not copy variation of the lenses, the MTF values can vary greatly with distance. Some lenses are optimised for infinity, some for medium distances, and others close up. I am willing to bet that the digital camera world's were not done at distances for which a 400mm lens is typically used at.

I test my wild-life lenses at close up distances of 1-3m, for where I use them for dragonflies and butterflies etc, which is shorter than typical for Imatest used as your example, and 20m, for birding etc, which is far further than used for Imatesting of MTFs. For example, the RF 100-400mm and RF 100-500mm, are excellent at very close up, and significantly better than the EF 100-400mm II. At 20m, they all perform very well. The EF 100-400mm II + 2xTC is OK at longer distances but much weaker close up whereas the 100-500mm II with RF 2x is outstanding close up and at 20m. The Nikon lenses like the 500PF and 200-500mm are optimised for shorter distances.

I'll check out the RF 100-400mm at f/8 and f/11 at 20m on the R5 and R6 to see how they compare and report back later.

Thanks for sharing your observations that the RF 100-400mm and RF 100-500mm are excellent very close up, and much better than the EF 100-400mm II. That's really useful for those of us that also take close up shots of insects and flowers with long lenses when out photographing wildlife.

It would be great to see how the RF 100-400mm performs at real birding distances using apertures of f/8 and f/11. I'm curious to see if the R6 really is less affected by diffraction too. Looking forward to your results, thanks again Alan! :)
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
Thanks for sharing your observations that the RF 100-400mm and RF 100-500mm are excellent very close up, and much better than the EF 100-400mm II. That's really useful for those of us that also take close up shots of insects and flowers with long lenses when out photographing wildlife.

It would be great to see how the RF 100-400mm performs at real birding distances using apertures of f/8 and f/11. I'm curious to see if the R6 really is less affected by diffraction too. Looking forward to your results, thanks again Alan! :)
R5 vs R6 at 20m with RF 100-400mm at f/8 vs f/11, unsharpened files
Edge sharpness measured using Focal
R5 f/8 1970; f/11 1886
R6 f/8 2076; f/11 2043
Chart resolution (arbitrary units)
R5 f/8 2.2; f/11 2.0
R6 f/8 1.8; f/11 1.8

So, with the high resolution sensor of the R5 where f/11 moves further into diffraction limitation (DLA = f/7.1), f/11 is slightly worse for both acutance and resolution.
For the low resolution R6 (DLA =f/10.6); f/8 ~ f/11. So, for my copy of the lens, it is best wide open.
 
  • Like
Reactions: 1 user

Czardoom

EOS RP
Jan 27, 2020
503
1,110
I've just made an interesting discovery, looking at the test results from the following review, this is essentially a RF 400mm f/11 lens when used at the long end, the shorter focal lengths and apertures are just a bonus.

In Canon's budget long telephoto range, we really have a zoom that operates optimally at 400m f/11 and primes that are fixed at 600mm f/11 and 800 f/11.

See extract from review below:


"Sharpness:

The sharpness scores in the two graphs below are produced by shooting a monochrome test chart covered in multiple sharp boundaries between black and white. This image is then assessed by specialist software, with the extent of blur on the contrast boundaries at the centre, mid and edges of the image converted into a spatial frequency value to determine how many line widths per picture height the lens is capable of resolving. A higher spatial frequency corresponds to a greater number of finer lines over a given distance that the lens can resolve - this number is the sharpness score.


nUK9MLVkXdS9h8qBNFGsuJ-970-80.png


KjFhxSDjF3VoYBVnm5YgzJ-970-80.png



Center sharpness is very good at all focal lengths and apertures, apart from a small blip wide open at 400mm. Corner sharpness is much less convincing, however, with the lens being noticeably softer regardless of focal length or aperture."


Using the TDP test tool, we can conform this when comparing the image at f/8 vs f/11

Use the larger, magnified view of the centre sharpness in the lower half of the test tool. The difference is discernible, but not drastically, probably because the test is done using the R5, which is well into diffraction territory at f/8, and more affected by diffraction at f/11, which iI suspect is offsetting a lot of the sharpness gains from stopping down the aperture.

Here's a suggestion. Stop making "discoveries" (that aren't) and go take some pictures and stop analyzing test charts. I've had the lens for a few weeks now and have been shooting wide open at f/8 at the long end with excellent results.
 
  • Haha
Reactions: 1 user

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
Here's a suggestion. Stop making "discoveries" (that aren't) and go take some pictures and stop analyzing test charts. I've had the lens for a few weeks now and have been shooting wide open at f/8 at the long end with excellent results.
The RF 100-400 is the bargain of the year and, as I have written, everyone should have one! I have always used all of my telephotos wide open, especially now as none is now wider than f/7.1. The one exception was the original Tamron 150-600mm, which was noticeably soft at f/6.3 but sharpened up nicely at f/8. The TDP charts bug me because whereas the rest of the site is a mine of well-researched information, people take those charts as serious tools, but they are too variable. They misdirected some of my purchases until I learned by experience to ignore them.
 
Last edited:
  • Like
Reactions: 2 users

neuroanatomist

I post too Much on Here!!
CR Pro
Jul 21, 2010
27,151
6,611
Here's a suggestion. Stop making "discoveries" (that aren't)
Often the real ‘discovery’ is that the copy of the lens used by the testing site performs differently than the one you bought.

Copy variation occurs, even with expensive lenses. Some people buy several copies and keep the best. Personally, I buy one copy, test it properly and exchange it if needed. I’ve only had to do that once (with a Rokinon 14/2.8).

There are several blog posts on lens testing, none of which require fancy equipment. As has been discussed, sometimes than fancy equipment is misleading, because chart-based testers (DxOMark, TDP, and Imatest users like OpticalLimits) are using a close focus distance, and testing on an optical bench (LensRentals) has the lens focused at infinity.

Here is the advice from Uncle Rog:

 
  • Like
Reactions: 1 users

Dragon

EF 800L
May 29, 2019
642
659
R5 vs R6 at 20m with RF 100-400mm at f/8 vs f/11, unsharpened files
Edge sharpness measured using Focal
R5 f/8 1970; f/11 1886
R6 f/8 2076; f/11 2043
Chart resolution (arbitrary units)
R5 f/8 2.2; f/11 2.0
R6 f/8 1.8; f/11 1.8

So, with the high resolution sensor of the R5 where f/11 moves further into diffraction limitation (DLA = f/7.1), f/11 is slightly worse for both acutance and resolution.
For the low resolution R6 (DLA =f/10.6); f/8 ~ f/11. So, for my copy of the lens, it is best wide open.
Even the TDP shots (that I know you don't trust) show that the lens is softer in the vertical direction at f/11. From the TDP chart, my guess would be that f/9 is about the optimal aperture. The Canon MTF charts show that the lens has pretty strong astigmatism at 400mm, so that may explain why a little stopping down helps. Also, the Canon MTF charts show that the RF100-500 is the best of the three under discussion with the EF 100-400 II next and the RF 100-400 at the bottom, but certainly not bad. My copy of the EF 100-400 II is very good, particularly at distance and it holds up quite well with a 1.4x TC, but a little soft with a 2x TC (of course the comparison is an EF 800L f/5.6 and the test bodies are 5DSr and R5 , so "soft" is a relative word).
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
Even the TDP shots (that I know you don't trust) show that the lens is softer in the vertical direction at f/11. From the TDP chart, my guess would be that f/9 is about the optimal aperture. The Canon MTF charts show that the lens has pretty strong astigmatism at 400mm, so that may explain why a little stopping down helps. Also, the Canon MTF charts show that the RF100-500 is the best of the three under discussion with the EF 100-400 II next and the RF 100-400 at the bottom, but certainly not bad. My copy of the EF 100-400 II is very good, particularly at distance and it holds up quite well with a 1.4x TC, but a little soft with a 2x TC (of course the comparison is an EF 800L f/5.6 and the test bodies are 5DSr and R5 , so "soft" is a relative word).
The new Canon (theoretical) MTF charts that allow for diffraction seem pretty good to me. Just how much of the diffraction correction you will see in practice depends on the resolution of the sensor. I also suspect that they are calculated for infinity.
I have tested directly the RF 100-500mm, RF 100-400mm and EF 100mm-400mm against each other. The IQ at 400mm is 100-400mm II, the best; 100-500mm (@400mm, f/6.3) nearly as good; and RF 100-400mm last. The difference with the RF 100-400mm varies with the pixel density because f/8 is greater than the DLA of the R5 but less than that for the R6 so it is affected more by diffraction on the R5.
Whether or not you will see astigmatism depends on how far away you are from the centre of the frame. The RF 100-400mm is soft at the corners and you see it clearly as I have written somewhere in the thread I started on the lens. At the 20m distance for my shots above, the chart occupies just the centre of the frame (like a small bird would at that distance). Reiken Focal measures astigmatism, and there was minimal astigmatism in the size of image I used for the RF 100-400mm, in fact less than an EF 100-400mm II, much to my surprise (I have had 3 good copies). Both copies of the 400mm DO II I have had showed even more astigmatism.
My lens of choice is the RF 100-500mm. At 500mm it outresolves the EF 100-400mm II and has just about as good resolution as the EF 100-400mm II with a 1.4xTC at 560mm. It's hardly inferior to the Nikon 500mm f/5.6 PF at longer distances. The EF 100-400mm II is indeed at its best at far distances, where as I agree also it is OK with the 2xTC. The 100-500mm really pulls ahead with TCs, and is very good indeed at 1000mm.
For a very lightweight 400mm you hardly know you are carrying, the RF 100-400mm is in a league of its own.
 
  • Like
Reactions: 1 user

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
To show you how good the RF 100-500mm + RF 2x is, here are my charts comparing it with the RF 800 f/11 and the 400mm DO II + 2xTC at 20m. These charts are only 14cmx9cm at 20m away and just 600 dpi laser prints on paper, not proper high resolution ones so they are pretty rough but they are good enough to tell you whats going on.

309A5483-DxO__Chart_20m_1000mm_centre_marked.jpg 309A7403-DxO__Chart_20m_DO_800mm_centre_marked.jpg 309A5382-DxO__Chart_20m_800mm_centre_marked.jpg
 
  • Like
Reactions: 1 user

dcm

It's not the gear. But it helps.
CR Pro
Apr 18, 2013
1,054
733
Colorado, USA
To show you how good the RF 100-500mm + RF 2x is, here are my charts comparing it with the RF 800 f/11 and the 400mm DO II + 2xTC at 20m. These charts are only 14cmx9cm at 20m away and just 600 dpi laser prints on paper, not proper high resolution ones so they are pretty rough but they are good enough to tell you whats going on.

View attachment 201405 View attachment 201406 View attachment 201407

Thanks for this comparison. I had considered the EF 400mm DO II + extenders a while back but decided to wait. Then the RF 800 appeared and it seemed like something I should check out first. I've quite happy with the choice and these images suggest it was a good decision. Did you consider comparing the RF 100-500 at 1000mm to the RF 800 at 1120mm?

I'm also quite happy with the RF 100-400 for the weight difference from the EF 100-400 on extended carries.
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
Thanks for this comparison. I had considered the EF 400mm DO II + extenders a while back but decided to wait. Then the RF 800 appeared and it seemed like something I should check out first. I've quite happy with the choice and these images suggest it was a good decision. Did you consider comparing the RF 100-500 at 1000mm to the RF 800 at 1120mm?

I'm also quite happy with the RF 100-400 for the weight difference from the EF 100-400 on extended carries.
309A5375-DxO__Chart_20m_1120mm_centre_marked.jpg

I sold the 400mm DO II after these tests.
 
  • Like
Reactions: 1 users

Dragon

EF 800L
May 29, 2019
642
659
To show you how good the RF 100-500mm + RF 2x is, here are my charts comparing it with the RF 800 f/11 and the 400mm DO II + 2xTC at 20m. These charts are only 14cmx9cm at 20m away and just 600 dpi laser prints on paper, not proper high resolution ones so they are pretty rough but they are good enough to tell you whats going on.
Were the charts shot with an R5 or R6? If with an R5, I am very impressed with the RF800 f/11. It would be interesting to see how an EF 800 f/5.6 would look in that comparison. Other reports I have seen suggest that the RF800 is best at distance, but I have not seen any info as to what distance is required to get the best performance. Given the price, I would expect the RF 100-500 to be quite good, but it appears to be better than many expected given that the aperture was narrower than the unicorn they dreamed of. Certainly, the size/weight is favorable. I have just not been too tempted given that the EF100-400 II that I already have works so well and the EF800 is quite good for the really long stuff. I have been tempted by the RF800, though, thanks to its low cost and light weight.
 

AlanF

Stay at home
CR Pro
Aug 16, 2012
9,744
15,024
Were the charts shot with an R5 or R6? If with an R5, I am very impressed with the RF800 f/11. It would be interesting to see how an EF 800 f/5.6 would look in that comparison. Other reports I have seen suggest that the RF800 is best at distance, but I have not seen any info as to what distance is required to get the best performance. Given the price, I would expect the RF 100-500 to be quite good, but it appears to be better than many expected given that the aperture was narrower than the unicorn they dreamed of. Certainly, the size/weight is favorable. I have just not been too tempted given that the EF100-400 II that I already have works so well and the EF800 is quite good for the really long stuff. I have been tempted by the RF800, though, thanks to its low cost and light weight.
Sorry, I should have said R5. The EF 100-400 II is still a superb lens. I got the RF 800 f/11 for a bit of fun - WEX had one used 10+ at clearly a mistakenly low price and I snapped it up. However, now I have bought my wife an R6, we can travel, if covid ever permits, with her having the RF 100-400mm for hiking, and the 800mm when shooting from a vehicle, and I the 100-500mm + TCs. When you get down to f/11, the R5 has only about 20% more resolution than the R6,and the 800 on the R6 outresolves the 100-500mm on the R5 and gives very nice images.
 
  • Like
Reactions: 2 users

Jethro

EOS R
CR Pro
Jul 14, 2018
674
605
The RF 100-400 is the bargain of the year and, as I have written, everyone should have one! I
I've taken the plunge - Black Friday (15% off) discounts and continuing good reviews have proved too much for me! 3 - 7 days ...
 
  • Like
Reactions: 1 user

LogicExtremist

Lux pictor
Sep 26, 2021
322
227
Here's a suggestion. Stop making "discoveries" (that aren't) and go take some pictures and stop analyzing test charts. I've had the lens for a few weeks now and have been shooting wide open at f/8 at the long end with excellent results.
Well, whatever happened to the spirit of inquiry? :oops:

We did discover that distance affects test results!

Comparing Alan's test results vs lab tests published by photography sites performed at different distances, we get completely different outcomes! Bet you didn't know that the optimum sharpness of the RF 100-400mm varies, and that certain apertures may work more favourably at longer or shorter distances. Or that you're better off not stopping down this lens on the R5 because the diffraction limiting distortion is visible.

Here's a suggestion. I have the time, but not the money to buy lenses for testing. How about you send me a RF 100-400mm? Then I can take lots of real world photos and test charts, and share my results here like Alan does. Perhaps, also send me an RF 800mm f/11 too for a more thorough comparison... ;)
 
Last edited:
  • Like
Reactions: 1 user

neuroanatomist

I post too Much on Here!!
CR Pro
Jul 21, 2010
27,151
6,611
We did discover that distance affects test results!

Bet you didn't know that the optimum sharpness of the RF 100-400mm varies, and that certain apertures may work more favourably at longer or shorter distances. Or that you're better off not stopping down this lens on the R5 because the diffraction limiting are visible.
Some of ‘we’ have known that for years. What you write about the RF 100-400 is generally applicable to all lens designs, and some of ‘we’ have known that for years, as well.
 
  • Haha
Reactions: 1 user

LogicExtremist

Lux pictor
Sep 26, 2021
322
227
Some of ‘we’ have known that for years. What you write about the RF 100-400 is generally applicable to all lens designs, and some of ‘we’ have known that for years, as well.
Perhaps some context is necessary. Some of us are fairly new to the world of ultra-long telephoto lenses, and many photographers entertain the idea of dipping their toes into budget wildlife photography, or sometimes, more specifically, birding. Knowing what the different lenses available can and can't do really helps set realistic expectations for budget buyers curious to try long tele lenses, but not wanting to spend too much.

From what I can surmise, what's quite new is the idea of shooting at really high apertures where DSLR autofocus systems didn't really work, using high ISO values that would nave been disregarded, and concepts such as DO lenses in budget primes.

Seeing budget lenses such a the RF 100-400, and RF 800mm f/11 used with teleconverters to get crazy reach at apertures of f/16-f/22 surely must be new ground technologically? It looks like technology has pushed previous boundaries, but that raises many questions. How well do different setups and camera/lens combinations work, and what are the constraints in terms of light, shutter speeds, camera bodies, shooting distances, etc to get reasonable image quality.

Some lab tests show that the RF 800mm, f/11 is much sharper than its 600mm counterpart, but that not only raises the question of sample size and product variation as you've suggested, but how that compares with real-world test results, or tests carried out at MFD and longer distances more commonly encountered in this genre of photography.

Lots of questions to be answered for newbies looking for the optimum budget entry-level birding solution, or whatever other wildlife they might want to shoot. :)