What is the maximum aperture at 400mm on the Canon RF 100-500mm f/4.5-7.1L IS USM?

AlanF

Stay alert, control the camera, save photos
Aug 16, 2012
6,880
5,948
I am trying to make sense of why there is a difference when using 1/3rd rather half stops. The first is to note that f/4 to f/7.1 can be divided nicely into five 1/3rd stops: 7.1 - 6.3 - 5.6 - 5.0 - 4.5, with values we are all used to seeing. The progression in 0.5 stops doesn't fit, e.g. 7.1 - 6.0 - 5.0 - 4.2 is the progressive increase in 1/2s and doesn't use f-numbers that are usually seen. Or, 4.5 - 5.3 - 6.3 - 7.4 is a similar bad fit. A plot of f-number against focal length based on when the f-number starts on the Cameralabs chart gives a nice straight line. All this suggests to me that the 1/3rd stops are more realistic, and the 1/2 stops are just numbers spewed out by Canon to be ones we are used to seeing, like 5.6, and aren't real.

100-500mm_f-number_vs_f.jpg
 
  • Like
Reactions: Eclipsed and Joules

Act444

EOS R
May 4, 2011
1,112
183
So basically, beyond 250mm you’re giving up 1/3 stop compared to the 100-400. F5.6 at 300mm (vs f5) and f6.3 at 400mm (vs 5.6). But in exchange, you get to reach 500mm. Ok. About what I expected.

I’d like to see if there’ll be an 80-400 4 or 4.5 to 5.6 type lens in this class on the horizon for the RF system. There is space for one, methinks...
 
  • Like
Reactions: YuengLinger

Sharlin

EOS R
Dec 26, 2015
1,314
1,076
Turku, Finland
All this suggests to me that the 1/3rd stops are more realistic, and the 1/2 stops are just numbers spewed out by Canon to be ones we are used to seeing, like 5.6, and aren't real.
Both the 1/3 and the 1/2 stop scale are approximations. The lens is almost certainly physically not exactly 100–500mm f/4.5–7.1, those are just familiar figures that the actual physical values are rounded to. At least at some focal lengths the 1/2 scale may actually be a better approximation, but 1/3 happens to be the one that most people use and is, of course, the more fine-grained of the two.
 

AlanF

Stay alert, control the camera, save photos
Aug 16, 2012
6,880
5,948
Both the 1/3 and the 1/2 stop scale are approximations. The lens is almost certainly physically not exactly 100–500mm f/4.5–7.1, those are just familiar figures that the actual physical values are rounded to. At least at some focal lengths the 1/2 scale may actually be a better approximation, but 1/3 happens to be the one that most people use and is, of course, the more fine-grained of the two.
Where would a coarser-grained approximation be better for some focal lengths? A genuine question, I realise that a clock that is stopped is absolutely correct twice a day but one that is 1 minute fast never is never right.
 
  • Like
Reactions: scyrene

BillB

EOS R
May 11, 2017
1,393
659
It would be nice if Canon finally released the manuals of the R5 and R6. Do they expect people to order those camaras without having read the manual? Then we would know how easy it is to change from 1/2 to 1/3 increments and back.
If it works the same way on the R5 and the R6 as on current Canon cameras it won‘t be hard to change From 1/2 to 1/3 increments if you read the manual.
 
  • Like
Reactions: Sharlin

BillB

EOS R
May 11, 2017
1,393
659
Where would a coarser-grained approximation be better for some focal lengths? A genuine question, I realise that a clock that is stopped is absolutely correct twice a day but one that is 1 minute fast never is never right.
The difference between 1/3 stop and 1/2 stop is 1/6 stop, which isn‘t all that much.
 

Sharlin

EOS R
Dec 26, 2015
1,314
1,076
Turku, Finland
Where would a coarser-grained approximation be better for some focal lengths? A genuine question, I realise that a clock that is stopped is absolutely correct twice a day but one that is 1 minute fast never is never right.
Given that the apparent aperture size varies continuously and smoothly as a function of the focal length, there’s necessarily a point where, say, f/6.7 (the half step between 5.6 and 8) is a better approximation of the real aperture ratio than either f/6.3 or f/7.1. But of course differences of just 1/6 stops are mostly academic anyway.
 

subtraho

Birds and Macro
Nov 20, 2018
24
33
Maryland
www.flickr.com
This whole discussion makes me wish cameras let you adjust their internal rounding in higher-granularity increments like 1/10 (or finer) of a stop, we'd probably learn a lot about our lenses if they did. Likely not practical for real-world usage, however - plus it would probably expose quite a few creative marketing shenanigans - so it'll never happen. I do have an older Pentax 55-300mm f/4-5.8 that was a bit of a head-scratcher at the time, but in retrospect was just using a higher-granularity f/stop scale.
 
Last edited:

AlanF

Stay alert, control the camera, save photos
Aug 16, 2012
6,880
5,948
The difference between 1/3 stop and 1/2 stop is 1/6 stop, which isn‘t all that much.
The difference between reported values of f/5.6 and f/6.3 at 400mm is a 1/3 stop so how does a 1/6 stop account for that?
 

dcm

Good or bad - it's not the gear.
Apr 18, 2013
828
184
Here it is again as it is most pertinent.
View attachment 191338
Something similar happens with the 100-400L on my 1DXII. If you set the 1DX II exposure level increment to half stops with the 100-400L II, the f/4.5 (not a half stop) range extends from 100-312mm and jumps to f/5.6 for 312-400mm. In this scenario f/5 never occurs. It's somewhat interesting that it never chooses the half stop at f/4.8, even when you manually control the aperture.


Changing the exposure level increment simply changes the values that can be selected and gives you less granularity in exposure choices.
 

BillB

EOS R
May 11, 2017
1,393
659
The difference between reported values of f/5.6 and f/6.3 at 400mm is a 1/3 stop so how does a 1/6 stop account for that?
If the lens is reporting f6.3 using 1/3 stop increments the actual measurement has to be closer to f6.3 than to f5.6. If same lens is reporting f5.6 using 1/2 stop increments, the actual measurement has to be closer to f5.6 than to f6.7, or whatever the half stop value is. So, for the lens to read F5.6 using half stop increments and read f6.3 using 1/3 stop increments, the actual measurement has to be between (half the difference between f5.6 and f6.3) and (half the difference between f5.6 and f6.7). Dividing a stop into twelfths the actual difference has to between (1/2 (4/12)) stops and (1/2( 6/12)) stops or between 2/12 and 3/12 stops above f5.6. So, the max discrepancy resulting from the lens reading f5.6 with half stop increments and reading f6.3 with third stop increments would be a a twelfth of a stop. (Not a sixth of a stop I said previously.)
 

juststeve

EOS M50
Nov 29, 2018
37
56
For those who were sleeping through third grade arithmetic, 1/2 = 3/6. 1/3 = 2/6. 3/6 - 2/6 = 1/6.

And for what it is worth, memory is Canon calculates f stops in camera in tenths, since early EF days. Those wishing for the info in print may want to check specs for Cinema EOS and EF lenses.

So what you are seeing is simply the rounding off of numbers to the nearest whatever. The diameter of the aperture at any given focal length is going to be a constant, but one determined by a bunch of variables. After-all, there are going to be anywhere from six to eight groups of elements moving around in that lens at it is zoomed. The actual physical diameter of the aperture may be changing as well as the lens is zoomed. Its position within the lens relative to other groups of elements almost certainly is changing.

A lot of things moving and a grooving in that lens.
.
 
  • Like
Reactions: Jack Douglas

LSXPhotog

EOS RP
Apr 2, 2015
429
319
www.diossiphotography.com
If any of you are seriously considering buying this lens and switching your cameras on 1/2-stop increments just so your camera displays 5.6 at 400mm...you should reconsider the importance of that arbitrary difference. LOL

Great to hear the lens isn't 7.1 at 400mm, that's for sure. I really like this lens a lot but not that price tag! If I sold my 70-200 and 100-400 it STILL wouldn't fund the purchase!