EF 100-400mm f/4.5-5.6 USM L IS

[Loswr]

Get ridd of the push pull. Get a new 100-400 L is with inner focus and weather sealed. Then this lens would be perfect.

What does push-pull have to do with inner focus? AF on the current 100-400mm is rear focusing (meaning the focusing elements are behind the iris diaphragm, as opposed to inner focusing, where the focus elements are in front of the iris diaphragm). Both rear focus and inner focus occur entirely within the lens, as opposed to front focusing where the front element extends outward with focusing (e.g. 85mm f/1.2L II and many consumer-grade lenses).

Presumably, you mean internal zoom, not internal focus. I'd still prefer the push-pull - one big advantage of the 100-400mm is it's (relatively) compact size when retracted. Would you really want the lens to be 11" long all the time (which it would be with an internal zoom)? I wouldn't.

 

[Edwin Herdman]

200-400 isn't a "short zoom," it's a very useful focal range - just as 100-300 is (or 70-300, or 120-300, to name a few other focal length ranges found in real lenses).

A short zoom is a short zoom. 2x is nothing for a zoom and we only really see that sort a range on more exotic lenses like wide angle lenses nowadays. But I didn't say it couldn't be useful. If it couldn't, then prime users would need serious help! Of course, by short I'm not referring to the focal length, but the zoom ratio.

You do realize that "zoom ratio" is marketing fluff and has really no bearing on the final image quality, right? A better predictor of quality would probably be percentage change between the respective angles of view or perhaps even between utilized surface of the glass surfaces. Why do you suppose the 70-200mm is "only" a bit over a "2X zoom"? It's because all the elements of making a good lens, as opposed to a merely okay one that tries to be the jack of all trades, force compromises in making professional zooms. The 70-200mm already sees criticisms for being relatively heavy, but a 70-300mm f/2.8 IS would probably be at least 50% heavier (and more likely nearly twice as heavy).

As you say a bit further on, in this segment optical quality is important - but even more so for people buying 200-400 lenses (and the Canon 200-400 won't be a highly restricted lens, either, due to the built-in TC - 200-600 is a very useful focal length range). Many people buying 100-400 lenses, or 28-300mm lenses for that matter, are placing versatility more highly (although many are in search of the best-quality yet affordable wildlife or sports lens, and many of these people are ultimately disappointed).

To reframe what I meant by Canon serving the 400mm segment poorly: The f/2.8 IS lenses arereally the only options that are universally well-respected. The 400mm f/5.6 is old and outclassed yet still is being manufactured. Again, a "bargain" option that's likely to be bought by the more desperate. The DO lens is interesting although some people don't like the DO effects (mainly a lack of sharpness or slight blurriness), but the main issue is price. Currently the DO seems to sell around the same price point as the original 400mm f/2.8 IS (the new one, of course, is over $10K). Meanwhile, the 100-400mm isn't in the same league (scroll down for direct comparison shots) as either other lens. I suspect that if Canon had released a regular EF 400mm f/4 that it would have been significantly cheaper (although also larger as well). Unfortunately it seems that a primary consideration for Canon was market-testing the technology and possibly the marketing potential of the smaller lens to meet with acclaim (not sure it's any lighter than a "regular" f/4 would have been, though).

For the 400mm segment, Canon essentially has relatively cheap yet long-outclassed options at one end, and super expensive options at the other. I suppose it doesn't hurt that there really aren't many serious contenders in the 100-400mm area, but nevertheless even cheaper options show up some of the shortcomings of the cheap 400mm options.

Get ridd of the push pull. Get a new 100-400 L is with inner focus and weather sealed. Then this lens would be perfect.

Presumably, you mean internal zoom, not internal focus. I'd still prefer the push-pull - one big advantage of the 100-400mm is it's (relatively) compact size when retracted. Would you really want the lens to be 11" long all the time (which it would be with an internal zoom)? I wouldn't.

Push-pull designs themselves don't confer any particular advantage in size - Sigma's design has a zoom ring and yet also changes length when zoomed out to 400mm. Of course, the Canon 70-300mm shows that this may be possible with a weather-sealed lens - though I'd expect the lens to also be a bit slower to zoom than the "dust pump" or even the unsealed Sigma. The main thing would be the likely additional friction from the front zooming segment being even larger (and thus slower / more finicky).

On the other hand, I never really felt that the internal zooming design of the 120-300mm Sigma was a big drawback - it's a somewhat large lens but even when immediately switching over it never really felt much longer than the 100-400mm, which I always had to worry about changing its focal length from zoom creep, needing to hold the zoom ring when pointing the lens up or down (I recall I once said it didn't seem to zoom creep much, but this was wrong). The 120-300mm mainly feels heavier. Of course, the 120-300mm isn't a 100-400mm, and its minimum focus distance suffers a bit in comparison.

 

[Loswr]

Push-pull designs themselves don't confer any particular advantage in size

True, and as you point out the 70-300mm L (like many zoom lenses) has an extending design with a rotating zoom ring vs. push-pull. But given the presumption that Heidrun was talking about an internal zoom, as opposed to a rotating ring-driven extending zoom, I think it's fair to say that an internally zooming 100-400mm would be substantially longer than the current version (or a new version with a rotating zoom ring that still extended).

Of course, the Canon 70-300mm shows that this may be possible with a weather-sealed lens - though I'd expect the lens to also be a bit slower to zoom than the "dust pump" or even the unsealed Sigma.

FWIW, the Canon 100-400mm has weather seals for the zoom/focus ring, and the switches. It lacks only the mount gasket seal. The similarly-designed (also push-pull) 28-300mm L is a weather-sealed lens.

I love that people still call the 100-400 a dust pump, when in fact, that part of the lens is sealed (although a filter is required on the front to complete the seal). Roger Cicala (lensrentals.com) recently blogged that, "People assume that zoom lenses with extending barrels will “pump dust.” They do to some extent, but the dustiest lenses we see are actually some primes and short range zooms."

 

[lol]

Edwin, I think you've got the wrong end of the stick. My comment on a 200-400 range being a short zoom range remains. Yes, the 200-400 extender would mitigate that by the built in extender, so overall we're about a 3x zoom then.

I never said the zoom ratio was a significant indicator for image quality. It may be a compact camera marketing spec dream, but even in DSLR-land it has some use. As a generalisation, for a given general quality level, if you want faster, you get less zoom range. Primes get the biggest apertures but no zoom. f/2.8 zooms have a short range. f/4 zooms get a bit longer. At f/5.6 you can get pretty much as much as you want. That was my point in looking at the zoom ratio. For a f/4, the 200-400 is a relatively short range. Maybe they wanted to make it extra special? Or it was required to make the extender work better.

On the 100-400L vs the 400mm fast primes, that's not exactly comparing like with like is it? Does a zoom really have much of a chance keeping up with something of fixed focal length and big aperture with price tag to match? If you want much better you have to pay for it. The 100-400L is still very competitive for the focal length range and cost.

My interest in these longer lenses is for wildlife. Sure, many times I wish I had something longer than a 400, but the 100-400L does the job well. The two sticking points for me with the more exotic teles is not the price, but a combination of them being prime and the weight. The weight can't be helped, but the prime doesn't sell it for me. I had a 300/2.8 for a while, but didn't find myself using it as it lacked flexibility. Even the short zoom of the 200-400 extender would help a lot there I think, even if it wont help on the weight side, and I'm not looking forward to finding out how much Canon want for it!

On the "dust pump" I've never had a dust problem with the lens or sensor and I do use the 100-400L a lot. Probably over 80% of my lifetime photos are taken with it. You can tell the front isn't sealed, as the pressure when pumping it does change with a filter fitted compared to without. I found that air vent handy actually as it helps dries out the lens elements faster when I get condensation inside. Stick a cap on the back end and pump away!

Hypothetically, if they remade a future 100-400L in the style of the 70-300L I would be disappointed. The 70-300L just isn't as good to zoom with, although it has its charms like not needing me to dry out its insides like I do on the 100-400L...

 

[Edwin Herdman]

We're on a tangent somewhat, but I don't see where I am misreading you all that badly. And I would appreciate it if you didn't use language like "short end of the stick."

A 200-400mm f/5.6 should still be an interesting lens, but I don't see any particular reason for the limited constant aperture - there is a reason the 100-400mm lenses and most consumer zooms allow wider apertures at their longer focal lengths. From the utility and marketing perspectives, a constant aperture lens is most sensible when it's a relatively fast aperture (f/4 or faster). Of course, with the hunt for good 400mm options, any improvement would be welcome, even in a "limited" 200-400mm.

With the 100-400mm L versus primes situation - yes, with modern materials there is every reason to believe a new 100-400mm would be much better. One thing I mentioned before is worth expanding on - not all lenses treat the long end as well as others. Overall, the Sigma 120-400mm seems a better design than the Canon 100-400mm, and when you look at the (not exactly compatible, due to them being different manufacturers) MTF graphs (Sigma, Canon) a couple things appear clear. One, the Canon's quality drops off rather rapidly past the frame center on the 400mm end, whereas the Sigma is more consistent across. Second, Canon optimized the 400mm setting where it seems reasonable to expect people will be getting effectively closer than is usual with the less-used 100mm setting. The Sigma is the reverse of this - the 400mm setting is noticeably weaker than the 100mm setting.

A big point to be made in all this is that the 400mm focal length is very useful at many different distances, even very close up, and so on these lenses people generally are much more interested in the far end than the 100mm - otherwise why not go for a dedicated macro or even a standard prime? I wouldn't necessarily sacrifice the 100mm range to go with a 200-400mm, and with new materials that shouldn't be as necessary to get a big boost in quality. I rather strongly believe that a lot of people are simply getting the 100-400mm as the best, cheapest option. The zoom range, IS, and minimum focus distance counteract any slight image quality benefit from the 400mm f/5.6.

@ Neuro:

Incidentally, and I don't think it's a big point either way (being such an old design), I see in Thom Hogan's review of the Nikon 200-400mm f/4 ED that the original 1984 version was a "much bigger push-pull design" than the current one. I don't know which dimension he is referring to though. Can't seem to find reliable dimensions though it looks like a big lens.

I love that people still call the 100-400 a dust pump, when in fact, that part of the lens is sealed (although a filter is required on the front to complete the seal). Roger Cicala (lensrentals.com) recently blogged that, "People assume that zoom lenses with extending barrels will “pump dust.” They do to some extent, but the dustiest lenses we see are actually some primes and short range zooms."

And a second thought: The 100-400mm may not be all that dusty itself, but it is moving more air around and more vigorously than most primes and short-range zooms. I believe what Roger says - that the other lenses get dustier - but this doesn't necessarily mean they are pushing as much air into the camera internals as the others. Especially if a lens is usually mounted to the camera (as is likely to be the case with a privately owned one), as opposed to lenses that are being put on and taken off of all sorts of different cameras as part of a business, it should be fairly dust-free. But what our man Ken calls "eyeblow" appears to be pretty well-documented happening with the 100-400mm. No, I don't think it's blowing all sorts of dust around, but I do see where the term comes from. Anyhow, I'll avoid using the term even in jest in the future.

 

[lol]

Edwin, you are definitely seeing things I did not write. I think what I wrote on each previous occasion was very clear, particularly taken in context of the talk at the time.

For example, the constant f/5.6 zoom was only mentioned in passing as something that would be dead easy to do, since some people keep asking for constant aperture zooms. f/5.6 is constant too. I'm not suggesting that I'd want such a thing.

 

[Edwin Herdman]

For example, the constant f/5.6 zoom was only mentioned in passing as something that would be dead easy to do, since some people keep asking for constant aperture zooms. f/5.6 is constant too. I'm not suggesting that I'd want such a thing.

I can't help but come back to this: My question is why you would suggest such a thing, seeing how you don't want it? Would anybody ask for replacing (or even adding to) the 100-400mm with a zoom that is only f/5.6 throughout the range? I think that part of the reason my replies to you have gotten so long-winded and seemingly tangential is that I'm trying to suss out what use these things are, and what you are suggesting might be useful and why. I'm trying to find your actual position.

I also think you may be wrong on the "constant aperture is easy" point - what I've seen indicates that the constant aperture zoom, while not really requiring significant tradeoffs, still does require some extra design components. Well, a quote is in order:

"An f-stop is the ratio between the focal length of the lens and the *apparent* size of the lens opening as viewed through the front. It must take into account the magnification factor of all lens elements in front of the diaphragm, because it is the size of the opening that the light "sees" as it passes through the lens, not the actual physical diameter of the diaphragm opening.
It is this fact that allows companies to make constant aperture zoom lenses which maintain a constant f-stop when the focal length changes, because such lenses are designed so that the magnification factor (diopter value) of all elements in front of the diaphragm changes as focal length is changed to hold the aperture value constant." - Bob Shell (January 2003, qtd. at Digital Grin forums)

Naturally, variable aperture zooms do not need to change the magnification factor of the front elements. It may or may not come down to lens grinding, not even fewer elements perhaps (but likely a couple elements fewer), but conceptually at least it is a simpler design.

 

[lol]

Look back at the original statement in context. Someone said they wanted a constant aperture zoom. Fixed f/5.6 was constant too, as an example, and returning the question if that was also desired. Even if you don't want it yourself, is it not interesting to see what is possible?

And my statement on constant f/5.6 aperture being easier (not any faster constant aperture, that was never mentioned) is based on the specific case you would only need to make a lens slower than needed on the wide end to achieve it. For example, take the existing 100-400L and do nothing other than fix it to f/5.6 max in the entire range. There's your constant aperture, and easy to do too. Now would you prefer that or the existing variable aperture? Of course that is a question to those asking for a constant aperture without specifying what value that would be.

Personally I don't care if a lens is variable aperture or not. I'll use whatever I have. I think the goal of more affordable long focal length zooms is not to provide a big aperture, but to provide a long focal length at a reasonable cost and quality. f/5.6 on the long end is more than acceptable to me there, since you get a long lens if you need a long lens, and long end performance is often defining the reason for it's existence. If the wide end is faster, that's a bonus. There is also a place for more premium models, as we're seeing with the upcoming 200-400 extender, but that's another story...

 

[Edwin Herdman]

And my statement on constant f/5.6 aperture being easier (not any faster constant aperture, that was never mentioned) is based on the specific case you would only need to make a lens slower than needed on the wide end to achieve it. For example, take the existing 100-400L and do nothing other than fix it to f/5.6 max in the entire range. There's your constant aperture, and easy to do too. Now would you prefer that or the existing variable aperture? Of course that is a question to those asking for a constant aperture without specifying what value that would be.

Personally, I think that's a question not worth having. A choice between a slow lens and a slower one?

But your reasoning seems sound when you consider even longer focal lengths, like for instance a (just pulling a range from the mirror lens thread) possible 500-800mm f/5.6 (which would be a big, big lens of course). A 200-500mm f/5.6 would probably be a variable aperture zoom able to start out wider, and I think that if a lens can be made wider reasonably it probably should be.

Personally I don't care if a lens is variable aperture or not. I'll use whatever I have. I think the goal of more affordable long focal length zooms is not to provide a big aperture, but to provide a long focal length at a reasonable cost and quality. f/5.6 on the long end is more than acceptable to me there, since you get a long lens if you need a long lens, and long end performance is often defining the reason for it's existence. If the wide end is faster, that's a bonus. There is also a place for more premium models, as we're seeing with the upcoming 200-400 extender, but that's another story...

I agree with you completely here. f/5.6 is the "minimum" acceptable for autofocus functions, but as long as that minimum is met, it's going to meet many people's needs for a long range lens.

I personally find that, even as an amateur, f/5.6 isn't usually enough, but when I only had two choices (the Canon 100-400mm and the Sigma 120-400) I made do, and made some decent enough pictures with it. And when I put the 2X teleconverter on, it functions essentially like a 240-600mm f/5.6. I don't expect the dimensions and weight of a "native" 240-600mm f/5.6 would be much different from the lens I have - perhaps the front element could be a bit smaller.

 

[afira]

Originally I thought Canon's new 200-400mm announcement was going to do a "soft replace" this lens, but I'm not so certain now. Either way, I don't see a 200-400 and a 100-400 announcement coming in the same year. I would highly doubt the marketing genius behind it. If a 100-400 replacement was coming, we'd be looking at a minimum of a year, possibly two out.

Which reminded me, I played with a 100-400mm this weekend. I won't discount my initial impression that the lens was probably stolen and/or very badly mistreated for awhile - lacking a rear cap, didn't have a case but had a very expensive Pro Hoya filter and the lens hood; the guy who was selling it didn't have a Canon camera and had about zero clue on the lens itself. The guy wanted $885 AUD (normally 1550+ AUD used here), which should have been my first clue, but I went to check it out anyway. The push-pull mechanism was really stiff and I felt like I was fighting with it the whole time. I also had extreme issues with getting it past the 350mm mark, I think I gave up finally. Images were soft, edges were fuzzy and indistinct - a close up of a palm tree leaf looked like it had a ghosted image behind it. While on manual focus images objects in the centre were blurry and slightly ghosted, with a small barrel distortion in the image - shooting the roof line and balcony of the house across the street. This was quite possibly the worst L lens I've ever handled and seen. I felt dirty just looking at it. Nothing obvious through the lens, no marks or scratches and even removing the filter, it was just bad. I'm guessing it had been dropped a few times off a four story building, but the guy said he thought it was the calibration of my camera to the lens. Needless to say, I didn't buy it, but that experience just put me off buying a used lens for awhile.

 

[Heidrun]

Originally I thought Canon's new 200-400mm announcement was going to do a "soft replace" this lens, but I'm not so certain now. Either way, I don't see a 200-400 and a 100-400 announcement coming in the same year. I would highly doubt the marketing genius behind it. If a 100-400 replacement was coming, we'd be looking at a minimum of a year, possibly two out.

Which reminded me, I played with a 100-400mm this weekend. I won't discount my initial impression that the lens was probably stolen and/or very badly mistreated for awhile - lacking a rear cap, didn't have a case but had a very expensive Pro Hoya filter and the lens hood; the guy who was selling it didn't have a Canon camera and had about zero clue on the lens itself. The guy wanted $885 AUD (normally 1550+ AUD used here), which should have been my first clue, but I went to check it out anyway. The push-pull mechanism was really stiff and I felt like I was fighting with it the whole time. I also had extreme issues with getting it past the 350mm mark, I think I gave up finally. Images were soft, edges were fuzzy and indistinct - a close up of a palm tree leaf looked like it had a ghosted image behind it. While on manual focus images objects in the centre were blurry and slightly ghosted, with a small barrel distortion in the image - shooting the roof line and balcony of the house across the street. This was quite possibly the worst L lens I've ever handled and seen. I felt dirty just looking at it. Nothing obvious through the lens, no marks or scratches and even removing the filter, it was just bad. I'm guessing it had been dropped a few times off a four story building, but the guy said he thought it was the calibration of my camera to the lens. Needless to say, I didn't buy it, but that experience just put me off buying a used lens for awhile.

You are maybe right in the prediction of a year or two. But that brings me to the other discussion i had. The ef 400 f. 5,6 L. Will this be relpaced with a new one with is. Because this one now is going out on date

 

[afira]

Originally I thought Canon's new 200-400mm announcement was going to do a "soft replace" this lens, but I'm not so certain now. Either way, I don't see a 200-400 and a 100-400 announcement coming in the same year. I would highly doubt the marketing genius behind it. If a 100-400 replacement was coming, we'd be looking at a minimum of a year, possibly two out.

You are maybe right in the prediction of a year or two. But that brings me to the other discussion i had. The ef 400 f. 5,6 L. Will this be relpaced with a new one with is. Because this one now is going out on date

Doubt it, at least you won't see this redesign in the next year. With the full redesigns of the prime telephotos in the f/2.8 range, I wouldn't see them rushing to the plates to put this one in production.

I'd doubt Canon would bother with IS on this lens anyway, as it is slow and pretty much designed for tripod use. Keeping the IS from the lens in this case keeps the manufacturing cost down (read cheaper for users) and the lens light(er) and... portable. Although, I will note that it seems every Canon lens is getting the IS treatment in the II versions, anything over f/2.8 anyway, so you may get your wish.

I only ever see the 400mm f/5.6Ls stuck to a massive tripod with hood fully extended and a guy in a baseball cap glued to the eyepiece.

 

[Loswr]

I only ever see the 400mm f/5.6Ls stuck to a massive tripod with hood fully extended and a guy in a baseball cap glued to the eyepiece.

You must not go where there are a lot of flying birds. The 400/5.6L is commonly used to capture birds in flight - long focal length, fast AF, easy to handhold, reasonable cost, and flying birds demand shutter speeds that render IS pretty ineffective. Still, I'm firmly in the camp of wanting an IS version of this lens - then, it would be useful for stationary subjects, too.

 

[TexPhoto]

I think the 70-300 L and the 200-400 L will replace the 300mm f4 IS, and the 100-400mm.

IS is BTW awsome both on and off a tripod for longer lenses if you know how to use it.

 

[afira]

I only ever see the 400mm f/5.6Ls stuck to a massive tripod with hood fully extended and a guy in a baseball cap glued to the eyepiece.

You must not go where there are a lot of flying birds. The 400/5.6L is commonly used to capture birds in flight - long focal length, fast AF, easy to handhold, reasonable cost, and flying birds demand shutter speeds that render IS pretty ineffective. Still, I'm firmly in the camp of wanting an IS version of this lens - then, it would be useful for stationary subjects, too.

When I was back home in Texas, the only people I saw near flying birds had shotguns in their hands, not cameras.

Here in Australia, I find that the lenses of choice are typically much smaller, as the birds are quite tame and the white and black cockatoos and other parrots will let you get pretty close, and anything under 300mm is suitable for bird watching near the beach. Usually I find it's just a long waiting game for some kid to run up and spook them into flying away if you want to catch them in flight. Even in the bush, I believe you'd be hard pressed to need something with a longer range. Even the galas will let you get within 5-20 feet, and give you a massive show of pink and white feathers up close and personal.

 

[Loswr]

I only ever see the 400mm f/5.6Ls stuck to a massive tripod with hood fully extended and a guy in a baseball cap glued to the eyepiece.

You must not go where there are a lot of flying birds. The 400/5.6L is commonly used to capture birds in flight - long focal length, fast AF, easy to handhold, reasonable cost, and flying birds demand shutter speeds that render IS pretty ineffective. Still, I'm firmly in the camp of wanting an IS version of this lens - then, it would be useful for stationary subjects, too.

When I was back home in Texas, the only people I saw near flying birds had shotguns in their hands, not cameras.

Here in Australia, I find that the lenses of choice are typically much smaller, as the birds are quite tame and the white and black cockatoos and other parrots will let you get pretty close, and anything under 300mm is suitable for bird watching near the beach. Usually I find it's just a long waiting game for some kid to run up and spook them into flying away if you want to catch them in flight. Even in the bush, I believe you'd be hard pressed to need something with a longer range. Even the galas will let you get within 5-20 feet, and give you a massive show of pink and white feathers up close and personal.

Once, I managed to get close enough to a great blue heron to fill the frame (a 1.6x crop frame, mind you) at 400mm. That was after about 20 minutes of careful approach.


EOS 7D, EF 100-400mm f/4.5-5.6L IS USM @ 400mm, 1/640 s, f/5.6, ISO 640

...but the majority of the time, even 400mm on a crop body is not long enough.

I think the 70-300 L and the 200-400 L will replace the 300mm f4 IS, and the 100-400mm.

The 70-300mm is quite possibly the replacement for the aging 100-400mm. A zoom cannot really replace a prime, though, RE the 200-400 being a replacement for anything in the sub-$2K range, that's like saying the 1DsIV is the replacement for the 7D - preposterous.