6
\$\begingroup\$

I've been considering getting an L-series zoom lens, and the two that I am looking at are the Canon 24-105mm f/4L IS, and the 24-70mm f/2.8L.

Aside from the extra reach of the 24-105, the main difference is obviously the f/2.8 aperture versus the active Image Stabilisation.

Canon state that the IS on this lens "permits the use of shutter speeds up to three stops slower than normal". It's the "up to" in this statement that got me thinking about how image stabilisation is rated.

There is a general rule of thumb that your shots should be good (with a steady hand) at shutter speeds of up to 1/focal length, so clearly the focal length is a factor of image stability. Now, I can't imagine that Canon would publish anything but the best numbers for their spec sheets, so this leads me to assume that the "three stops" of IS they quote are at a focal length of 24mm. So my next question is, what does this mean at the 105mm end? I've done a few calculations, and I invite you to check my workings.

3 stops is a linear factor of 8 (2^3). I think this equates to a linear IS factor of 1.83 @105mm (8*24/105). Or, 0.87 stops. Less than 1 stop!

Using the same calculation but for a focal length of 70mm, the IS of the 24-105mm comes out at 1.46 stops.

Do I have my calculations correct?

PS: I know that the increased aperture of the f/2.8 will give greater background blur, better capture of moving subjects, blah, blah... so please don't feel that you have to lecture me on that. All I'm asking about is IS.

Improve this question
\$\endgroup\$
5
  • 1
    \$\begingroup\$ I generally agree with your argument BUT suspect that the claim is somewhat more subjective than measured. I have or have had a number of Sony & Minolta in body stabilised cameras* but the general principle is the same. I'd say the actual performance tends to be not as good as the most enthusiastic claim but more than 1 stop at "usefully long" focal lengths. (*A200, 7D, 5D, A700, ...). I'd say "1 to 2" in many circumstances - treat it as a bonus, don'r forget the old tricks re bracing and breathing, remember that IS doesn't help with subject movement. . \$\endgroup\$
    – Russell McMahon
    Aug 11, 2011 at 11:46
  • \$\begingroup\$ I'm pretty sure that Canon would have measured the IS capability. It would be fairly easy to do with a mount that oscillated left-right at a speed that was just quick enough so that the IS motors were working at their maximum, but not hitting the end stops. \$\endgroup\$
    – ltn100
    Aug 11, 2011 at 12:27
  • 5
    \$\begingroup\$ @ltn100 I have no idea what you're trying to do with your calculations or how you've calculated that a three stop stabiliser is equates to 0.87 stops! Three stops simply means you can leave the shutter open 8 times as long. \$\endgroup\$
    – Matt Grum
    Aug 11, 2011 at 13:40
  • 1
    \$\begingroup\$ @Matt Grum: I think the assumption made is that the stabilization mechanism provides the same amount of reduction of movement regardless of zoom, and that the actual effective on sharpness will be proportional to focal length. The basis for this is the traditional min-shutter-speed rule, which makes plenty of sense. The question is, is the other part of the assumption true? \$\endgroup\$
    – mattdm
    Aug 11, 2011 at 14:02
  • 2
    \$\begingroup\$ Your math is totally flawed. At 105mm, the 1/FL rule of thumb is 1/100. At 70mm, 1/FL=1/60 or 1/80. At 24mm, 1/FL = 1/25. Three stops slower than 1/25 is a lot longer exposure than three stops slower than 1/100! You are still expecting three stops slower than 1/25 at 100mm instead of three stops slower than 1/100! \$\endgroup\$
    – Michael C
    Mar 18, 2019 at 22:01

5 Answers 5

Reset to default
11
\$\begingroup\$

Reverting to some empirical data, the 70-200 f/2.8L II lens has a supposed 4 stops of IS at all focal length. DPreview tested it at 70mm and at 200mm and revealed it had just under 4 stops and 70mm and over 4 stops at 200mm!

http://www.dpreview.com/lensreviews/canon_70-200_2p8_is_usm_ii_c16/page5.asp

From the review:

We're used to seeing Canon's latest IS systems do well in our tests, and the 70-200mm F2.8 II doesn't buck that trend. Under these controlled conditions, the IS system is delivering something pretty close to the claimed 4 stops of stabilization (for example, at 70mm we get similar results at 1/5 sec with IS on as at 1/80 sec with it turned off), which is about as good as things get at the time of writing.

Unfortunately they haven't tested the 24-104 IS yet. But both their review and my experience with the 100 f/2.8L macro indicates the actual performance is pretty much in line with the official specs (again, four stops). So I'd be inclined to believe them and expect about three stops throughout the range on the 24-105.

Improve this answer
\$\endgroup\$
3
  • \$\begingroup\$ This exact review of the new 70-200mm 2.8IS II is what I thought of as well. It performs excellent on this lens, as spec'd, and I would expect the 24-104 to be the same. \$\endgroup\$
    – dpollitt
    Aug 11, 2011 at 15:20
  • \$\begingroup\$ @Matt: thanks, this is just the sort of objective results that I was after. I think in my calculations I didn't take into account the reduction of shutter speed that is accounted for using the 1/FL rule. I was simply thinking about lateral movement of the camera and how much counteractive movement the IS lens would have to make to compensate for this before hitting its limits. However, if the shutter time is quicker then it won't need to move as far. \$\endgroup\$
    – ltn100
    Aug 11, 2011 at 15:49
  • \$\begingroup\$ I know out of experience that the EF-S 17-55mm F/2.8 IS is pretty bloody fantastic and that is rated 3 stops (it's a $1,000 lens). \$\endgroup\$
    – Nick Bedford
    Aug 11, 2011 at 23:08
7
\$\begingroup\$

From first-hand experience, I can shoot my 24-105L at 105mm at 1/10 second on a full frame camera and expect reliably sharp shots. (If the image is blurry, it is because I yanked the shutter button or did something similarly stupid!)

1/10 is just about where a three-stop IS ought to be (1/100-1/50-1/25-1/12.5 sec) according to the 1/focal length rule of thumb. Which is only a rule of thumb of course, your shooting technique will affect the results, as will your caffeine intake, but that has nothing to do with the IS as such :)

I am more sceptical about IS being quite as effective on very short focal lenghts actually. If you can handhold at 1/25 second at 25mm without IS, a three-stop IS ought to give you about one third of a second shutter speed. At this point body sway starts to come into play, both up-down and left-right... a different, slower, larger kind of movement than the fast hand-shake vibration that IS normally handles.

Improve this answer
\$\endgroup\$
4
  • \$\begingroup\$ Good answer, although I don't own any IS lenses, I'd also say that the IS will probably work better at longer focal lenses than at a wide angle! \$\endgroup\$
    – eflorico
    Aug 11, 2011 at 13:25
  • 4
    \$\begingroup\$ IS does not work BETTER at longer focal lengths, rather it is more useful at longer focal lengths. \$\endgroup\$
    – dpollitt
    Aug 11, 2011 at 15:19
  • \$\begingroup\$ Thanks Staale, it's good to hear from someone with the actual lens that it seems to live up to its specs. Are you happy with the lens in other respects? Any regrets in getting it? \$\endgroup\$
    – ltn100
    Aug 11, 2011 at 15:53
  • \$\begingroup\$ Yes, I am happy with it. It has quite a bit of distortion at the wide end but this is easily fixable in Lightroom. Colour and contrast are good, the IS is a godsend, the zoom range is nice, build quality is good L standard, size and weight are reasonable. The lens-hood is far less effective than the 24-70 one, but the lens is well-behaved with regards to flare so this is not a showstopper. f/4 makes for a rather dark viewfinder indoors and does nothing good for autofocus performance, but the 1DsII AF system performs well even then. A consumer body may struggle a bit though. \$\endgroup\$
    – Staale S
    Aug 11, 2011 at 16:39
4
\$\begingroup\$

Three stops should be three stops regardless of the focal length. If you can do 1/100 @ 105mm, then with IS it would be 1/15 or so.

* However, keep in mind that this will only compensate for camera shake and not for subject motion. If your main interest is people or dynamic scenery, then the f/2.8 may be more handy. This subject is discussed in this question.

UPDATE: OK, I think I have found a supporting evidence at the dpreview review of the EF70-200/2.8LIS here. Looking at the results achieved at 70 mm and at 200mm, it seems that at 200mm the IS keeps up with the stated 4 stops of compensation.

Improve this answer
\$\endgroup\$
12
  • \$\begingroup\$ IS relies on keeping the image where it belongs and this depends on acceleration and position control under motion. As focal length increases the acceleration/ deceleration and required accuracy of positioning of the lens elements (or sensor in an in body system) increase linearly with focal length, so the "3 stops" ability becomes harder to achieve with increasing focal length. Yes? \$\endgroup\$
    – Russell McMahon
    Aug 11, 2011 at 11:33
  • \$\begingroup\$ @Russel MacMahon - I don't think so. To my understanding, the compensation is relative. At 105mm the effect of shake is indeed more noticeable, but the compensation works harder as well - assuming there is some inertial sensor and a control circuitry that takes the FL into effect. It might not be as good as at the wide end, but I don't think that it is linearly bad. \$\endgroup\$
    – ysap
    Aug 11, 2011 at 11:40
  • \$\begingroup\$ Thanks for the answer, but I'm afraid I don't buy it at all -- I'm the sort of person that needs hard numbers to believe anything ;-) Also, thanks @Russel for pointing out that it is al about shifting the image to keep it in the centre. Perhaps its not as linear as I first thought. \$\endgroup\$
    – ltn100
    Aug 11, 2011 at 12:28
  • 1
    \$\begingroup\$ I actually tried to look for some data, I assume you did as well before posting the question. I could not find any evidence for the reduction in performance when using zoom lenses. In Canon's available publications, they give few focal lengths and appropriate shutter speeds as examples. However, they don's state whether they refer to zooms or different primes, but for all of them they use the 1/FL as a guide. \$\endgroup\$
    – ysap
    Aug 11, 2011 at 12:43
  • 1
    \$\begingroup\$ IS used to be a hallmark and unique selling point of the huge 3-4-5-600, even 800mm, Canon tele lenses. Primes, of course. So I think we can safely assume that they have had the problem of several-stop stabilization at long focal lengths licked for quite some time now. \$\endgroup\$
    – Staale S
    Aug 11, 2011 at 13:04
1
\$\begingroup\$

Do I have my calculations correct?

Your math is totally flawed. At 105mm, the 1/FL rule of thumb is 1/100. At 70mm, 1/FL=1/60 or 1/80. At 24mm, 1/FL = 1/25. Three stops slower than 1/25 is a lot longer exposure than three stops slower than 1/100!

You are still expecting three stops slower than 1/25 at 100mm instead of three stops slower than 1/100!

Improve this answer
\$\endgroup\$
0
\$\begingroup\$

I posted a similar question without the flawed math: Can I trust in advertised 4 stops of image stabilization in a cheap kit lens? ...that was marked as a duplicate, so I'll repeat the main findings of me here, so that the information in the duplicate question doesn't get lost. Note: this answer doesn't even attempt to give an answer to the flawed math.

Firstly, I have found that IS doesn't help much at 1/5 s on a cheap 18-55mm kit lens (see later for 24-240mm $1000 lens results). You need 1/10 s or shorter exposure for IS to help. It seems that if I really try, I can hold the camera steadily at approximately 1 / (0.5 * 35mm-equivalent-focal-length), being somewhat better than the 1 / (35mm-equivalent-focal-length) rule. This was based on 10 test images (55mm focal length, 1.6x crop sensor) taken at 1/5 s with IS (mostly shaken), 1/10 s with IS (mostly fine), 1/20 s without IS (mostly shaken), or 1/40 s without IS (mostly fine).

So, at 55 mm focal length, for my test conditions, the IS helped only for two stops (advertised as four-stop IS). The 55 mm is the longest focal length on this 18-55mm zoom, so clearly, you (or well, I) can't trust the image stabilization ratings.

I also tested a 55-250mm lens @ 250mm and took some non-shaken pictures at 1/30 s shutter speed. If we believe the 1 / (35mm-equivalent-focal-length) rule, the shutter speed should be 1/400 s, and if we believe the slightly modified 1 / (0.5 * 35mm-equivalent-focal-length) rule, the shutter speed should be 1/200 s. Based on whichever rule we believe, the IS helps for at least 2.7 stops or 3.7 stops. I say "at least" because I didn't test longer exposures than 1/30 s.

I also took ten pictures using 24mm lens at 1/20 s exposure. They were mostly non-shaken, although one had very slight camera shake. If IS can't help at longer exposures than 1/10 s, on a 24mm crop lens, IS helps for one stop only. A Canon 24mm f/2.8 has twice the light collecting ability of Canon 18-55mm f/3.5-5.6 zoom which @24mm opens to f/4. So, on a 18-55mm lens used at 24mm, it's a much better idea to remove that lens and attach a 24mm f/2.8 prime.

Main findings, applicable for cheap crop sensor zoom lenses:

  • Short focal length on a 18-55mm f/3.5-5.6 zoom: use a 24mm f/2.8 prime instead, no need for IS as the prime is equivalent unless you really need lots of depth of field, and the prime is better for fast-moving subjects, too.
  • Long focal length on a 18-55mm f/3.5-5.6 zoom: use a nifty fifty (50mm f/1.8) instead, no need for IS as the nifty fifty collects 2x the amount of light at longest acceptable exposure time compared to the zoom, and it's much better for fast-moving subjects. However, if you really need lots of depth of field, a kit zoom with IS may be better.
  • Short focal length on a 55-250mm f/4-5.6 zoom: use a nifty fifty instead, again.
  • Long focal length on a 55-250mm f/4-5.6 zoom: now image stabilization helps a lot!

However, the above apparently applies only to cheap kit lenses. I bought a $1000 24-240mm Canon RF lens for EOS RP. It is using a much better "dynamic IS" technology that combines sensors in the camera and in the lens. It is advertised as 5 stops.

I took 10 pictures at 1.3 second exposure times @ 24mm. None of the pictures can be classified as genuinely shaken, although very slightest hints of camera shake can be seen in about 2 or 3 images. The very slight hints of shake are visible only if you view the image at 1:1 ratio on a 96 dpi computer screen, which most people don't.

If we believe the 1 / (35mm-equivalent-focal-length) rule, we get from 1/24 s to at least 1.3 s, which is at least 5 stops of improvement. If we believe the slightly modified 1 / (0.5 * 35mm-equivalent-focal-length) rule, we get from 1/12 s to at least 1.3 s, which is at least 4 stops of improvement.

So, it appears that on the cheapest kit lenses, the IS ratings are deceptive. But, on the more expensive lenses, the IS ratings can actually be true.

Improve this answer
\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.