User input on Canon 100-400/4.5-5.6L IS USM requested

[bobatkins]

If anyone has had time to <em>properly evaluate</em> the new Canon

100-400/4.5-5.6L IS USM, It would be good if they could post or send me their

comments. There is considerable interest in this lens from nature

photographers and a serious evaluation (perhaps with a direct

comparison with the 300/4L or 400/5.6L) would be useful to many.

I don't expect anyone has yet had time to do this, but it's

possible I guess. I've had a few questions about this lens and

I'd like to concentrate any <em>user</em> comments in this post.

 

 

<p>

 

I'd like to keep <em>speculation</em> on why this lens would be good

(or bad),

or

what people think it's performance <em>might</em> be, to a

<b>mimimum</b>. If you

don't own and haven't shot with with this lens, please don't

followup on this post. Try the <a href="http://www.greenspun.com/bboard/q-and-a-fetch-msg.tcl?msg_id=000Jvj">Camera Equipment Forum</a> if you

just want to chat about it. Thanks.

 

<p>

 

Before anyone asks the price it's $1700 and it's still hard to find.

If you want the specs, check out the <a href="http://www.usa.canon.com/camcambin/cameras/eflenses/ef100n400.html">Canon</a> or B&H web sites.

[jim_roetzel]

I bought the 100-400 the first weeks of December. I paid around

$1,600 USA (plus tax). I spoke with a B&H dealer the other day who

said they had some Gray Market version for $1,650. I was also told

that the US dealer cost is $1,500. It's one of those "new stuff =

prices and rumors of prices tend run a little weird time."

 

<p>

 

So far I like the lens- but I have only shot about 30 rolls of film

through it- all Velvia and Provia. 95% shots were with the lens wide

open and the camera on aperture priority. The reason for this- I used

to own the 300/IS lens and was not always happy with it wide open-

sharpness was inconsistent. Wide open the 100-400 is an excellent

lens. Its still a little too early to call it tack sharp. I want to

use it in more situations.

 

<p>

 

I am getting used to the one touch zoom of the lens- at 100 the lens

is about the size of the a straight 300mm but as it is zoomed it gets

a little awkward (but not impossible) to handle- at 400mm it is about

14 inches long. It should be noted that I shoot off of tripods or

monopods almost exclusively so I'm not the best judge of hand holding.

 

<p>

 

What's great about the lens- The ability to crop in the camera.

I shoot wildlife and spend a lot of time behind a 600mm. When doing

blind work I have always been envious of a close friend who owns the

150-600 FD lens that focuses to 10 feet! I now have the same (sorta)

ability to frame shots as him and the 100-400 focuses to just under 6

feet. The lens is of course silent and was quick and responsive in

auto focus.

 

<p>

 

The lens has a built in "smooth---tight" ring- I initially thought it

was somewhat useless but found it gave a little more control when

shooting. In the 'smooth'setting the lens zooms very comfortably. In

'tight' its locked on to a focal length. The utility of this feature

will be helpful when shooting down- no slip. I did use the lens with

a 500D close filter- and found the 'tight' feature useful.

 

<p>

 

The tripod collar is small- good for those who hand hold- I wish it

was a little larger or placed a little more forward on the lens.

Working off a ball head with RRS arca plates attached the camera binds

up on the quick release clamp. But this wouldn't be the first

lens/camera combo to do that.

 

<p>

 

I'm from the old school so converters are something I use sparingly.

I did use the 1.4x with the lens- lost AF on my EOS-1n. The manual

focus ring is a little skinny but you get used to it and it is

comfortably placed. The lens was sharp wide open with the 1.4x

(F/8).

 

<p>

 

I have owned both the 400/5.6 and 300/4 (IS and non). Of course both

of those were lighter than the 100-400, but the new lens seemed

slightly faster in AF and sharper than the 300/4 IS lens. I could cut

the weight of the lens by removing the collar for "hip" or flight

shooting. One little hint for flight shooters- manually preset the

focus toward infinity. The times I forgot to do this, the lens did

have a tendency to go toward close focus and "hunt" for something

close.

 

<p>

 

So far I like the lens- but 30 exposed rolls may not be enough to

measure a lens.

 

<p>

 

Sorry to go on so long.

 

<p>

 

Jim

[bobatkins]

Thanks Jim. Make sure to post a followup later if you have more

thoughts about the lens.

 

<p>

My only comment on what you posted is that I'm a little surprised

you haven't been totally happy with the sharpness of the 300/4 IS

wide open. I shoot with the 300/4 (not IS) quite a lot and I've

been pretty impressed with it's wide open sharpness, even when

used wide open with a 1.4x TC. I'd rate it right up there with

the 300/2.8, maybe a hair less sharp at any given aperture, but

you'd have to look pretty close to see any difference at all. Maybe

it's an IS vs non-IS issue, though previous posters have claimed to

see no difference between them.

 

<p>

 

Here is the first test data I've seen. It was sent to me by Don

Gauger, and I'd like to thank him for sending it. It's from

the UK magazine "Amateur Photographer", Dec 12th 1998, page 42.

 

<p>

 

Let me preface this data by cautioning readers not to try to

compare these lpm (lines per mm) numbers with numbers from any

other source (published or personal). The issues of AP I have do

not describe their test method, whether it's film based or not or

even if lpm is lines/mm or line-pairs/mm (they are normally the

same, but I have seen reviews where they differed by a factor of

2!). It's only worthwhile comparing AP results with AP results,

the usual rider being given that single sampling tests are prone

to possible errors, even assuming good testing methodology.

 

If anyone <b>knows</b> the AP testing method, that would be useful

info to add to this discussion.

<p>

 

 

<pre>

Their Summary Rating are:

Optical Quality 27/30

Ease of Use 27/30

Construction 19/20

Value 16/20

 

Total 89/100

 

color shift neutral

distortion below reporting levels

vignetting not significant

performance excellent

 

the below lpm are read off a very small graph and hence are

approximate:

 

hi contrast low contrast

FL F edge/center Lines/mm Lines/mm

100 4.5 C 80 60

100 4.5 E 70 55

100 8.0 C 110 85

100 8.0 E 105 82

100 11 C 120 90

100 11 E 115 90

100 16 C 110 90

100 16 E 110 90

 

400 5.6 C 75 50

400 5.6 E 65 45

400 8.0 C 90 70

400 8.0 E 90 70

400 11 C 115 90

400 11 E 115 90

400 16 C 105 90

400 16 E 105 90

 

</pre>

Compared to all other lens they have tested (I think this includes

fixed focal lengths). Lpmm performance wide open at 400mm is close

to

some of the worst lens they have measured. Performance becomes mid

range at f8. Above average at f11, and about as good as any lens

they

have measured at f16.

<p>

<em>Moderator's note: By the time you get to f16, most lenses are

diffraction limited, so most lenses show pretty much the same

resolution</em>

<p>

The only other test I have is one on the Sigma 35-135/4-5.6. This

is only useful to give some context to the numbers they report on

the 100-400/4.5-5.6L. The data is as follows:

 

<pre>

Sigma 35-135/4-5.6

35C = 35mm center

35E = 35mm edge

135C = 135mm center

135E = 135mm edge

 

High Contrast

f-stop 35C 35E 135C 135E

 

4 80 70

5.6 88 80 75 69

8 102 100 88 81

11 111 108 110 102

16 104 103 105 101

 

Low Contrast

f-stop 35C 35E 135C 135E

 

4 50 42

5.6 72 60 60 55

8 81 79 78 71

11 88 87 82 81

16 88 87 87 87

 

</pre>

Again, let's try to keep speculation to a minimum here and keep to

a discussion of facts and user experiences. Remember that it's how

a lens performs in the field that's important, not how well it does

in the lab. Whatever the bench numbers, if you have to handhold the lens

(which is what IS is all about), this lens will certainly give you sharper

images than anything else out there most of the time!

[danny_weber]

Bob, the cover shot on the Feb 99 issue of Outdoor Photographer was

taken by George Lepp with this lens. I know you don't get the

magazine, but I'll assume you don't mind looking at the cover.

[ShunCheung]

The Arctic Fox image on the cover of the February

issue of Outdoor Photographer is clearly very sharp, but of course

George Lepp and the editor wouldn't pick a poor picture for the cover

of OP. I have little doubt that this Canon L lens is a good lens, but

what remains to be answered is how well it works at the long end with

the aperture wide open and hand held, and what percentage of images

are sharp, etc. Jim says he is using it mainly on either a tripod or a

monopod, so the very important IS feature doesn't get fully tested in

his case.

[bobatkins]

(Trying not to initiate a long thread of opinions on IS, which we have

covered before).

 

<p>

 

I'd disagree with Shun. I think we all know IS works great, and if

you are going to hand hold the lens, it's going to give you much

sharper images than you could get without it. The IS performance

is pretty much a given.

 

<p>

 

It's the non-IS performance that matters. All serious nature shooters

will be using the lens most of the time on a tripod, hand holding it

only when that's impossible (flight shots, on a boat etc.). The

concern then is whether the additional IS optics result in a loss

of sharpness (already tricky to get in a long 4x zoom).

 

<p>

 

The matter of a cover shot by George Lepp isn't really much of a test.

I think Galen Rowell shot a cover for Audubon with a Nikon P&S one

time. It's not what a lens is ultimately capable of that counts, it's

what sort of results it gives on a day-in-day-out basis. Having a

somewhat cynical opion of OP, I wouldn't be surprised if the cover

shot was part of a promotional stratagy for the lens.

 

<p>

 

The real question is can the 100-400L replace say, a 300/4 in your

bag. Do the extra shots you gain make up for the few shots you lose

due to slower speed and differences in optical quality. That remains

to be answered by those shooting with both lenses. How many people

shooting with both will want to (as opposed to have to!)sell their

300/4?

[ShunCheung]

Obviously different people purchase lenses for different reasons, but

if I were to buy a 100-400 IS zoom, I would be looking into using it

hand held most of the time for bird in flight shots, etc. I wouldn't

pay the extra price and possible performance penalty for IS if I am

going to use the lens primarily on a tripod. Again, your mileage may

vary. Moreover, so far the longest IS lens is 300mm, therefore I still

would like to find out how well IS works on a 400mm.

 

<p>

 

I do agree that one cover shot from OP doesn't really say much. George

Lepp's strong tie to Canon aside, he doesn't specify the aperture and

shutter speed used. The subject is a well lit white fox; it could have

been shot at f11 and 1/250 sec. As I pointed out earlier, how well the

lens works wide open at 400mm/f5.6 is the big question.

 

<p>

 

And yes, inside the magazine, there is a double-spread Canon ad for

the 100-400 IS. I certainly wouldn't be all that surprised if the

cover shot is part of a promotional deal.

[dick_ginkowski]

Bob,

 

<p>

 

When I was checking EOS 3 availability a couple of weeks back, I did

find quite a few dealers who had the new 100-400. It's not as scarce

as the EOS 3. Local or regional dealers may be the best source.

 

<p>

 

I didn't buy one because my existing stable of lenses is adequate.

[eddie_barnes]

Bob,

 

<p>

 

A few comments on the Amateur Photographer lens testing procedures.

 

<p>

 

You are quite right to comment that test results are not comparable

between different sources. Those published in AP are carried out by

Dr Stewart Bell, who has worked on various UK photographic titles for

a few decades now, using equipment of his own devising, having

graduated up from proprietary MTF test equipment, as far as I

can remember.

 

<p>

 

The tests are carried out on an optical bench (i.e. independent of

film/processing variables). What the tests do is to provide a relative

measure (versus best and worst) rather than an absolute (which is

effectively meaningless unless you always shoot under controlled

conditions with film x, processed at exactly y minutes and z:C.

 

<p>

 

Best and worst are taken from a tested population of both fixed focal

length and zoom lenses.

 

<p>

 

For what it's worth, the lens was given a glowing report in AP. At

first sight, lenses of this specification look attractive (IS aside),

until you realise that the performance wide-open at the business end

of the zoom is pretty average (where I guess it will be used most),

while the best performance is obtained when the lens is operated at

the other end. Who really wants a heavy/long 100mm/f4.5?

 

<p>

 

Regards,

 

<p>

 

Eddie.

[ming_kuo]

I met a man up at Bosque Del Apache yesterday using the new IS zoom

along with his 400 2.8. I asked his opinion on it, and he loves it.

He also owns the 300 IS and I asked him how the two compared. He

preferred the zoom for most cases, simply because of the flexibility

a zoom gives you. When I pressed him for more specifics, he wouldn't

say much, aside from saying that they're both sharp enough for him.

He uses the 300 on boats for shooting whales, and says it's the ideal

lens for those situations.

[glen_johnson]

This lens is turning out to be an excellent choice for photographers in sports where 100 might be good if you can get close enough, but 400 might be necessary sometimes - horse racing for example. These guys were often using the 70-200 f/2.8L with a 1.4x or 2x converter. The new lens is definitely sharper than the old zoom and tc combo, and the IS mode is a plus too. This lens will log a lot of time at the track this season.

 

<p>

 

It doesn't seem all that hard to find, and $1700 isn't much in the grand scheme of things, especially compared to the cost of the hotel, transportation, food, etc. associated with a trip to shoot at a major event.

[bobatkins]

Well, our friends at Pop Photog have tested the 100-400/4.5-5.6L now.

Here's my take on their results, based on my experience of their

earlier tests, general optics and my experience of Canon lenses:

 

<p>

 

The numbers they report at 100 to 300mm don't seem too far out

of line. They show peformance being pretty good at the short end, and

starting to drop as you zoom out. Sharpness peaks a stop or two down

from maximum as you'd expect. Looking back to tests on the Sigma and

Tokina 300/4 lenses, you see the new zoom has similar performance, if

you stop it down an extra stop (i.e. to get the same performance as

the 3rd party primes at f5.6, you need to be at f8 with the zoom). The

zoom appears to be a little soft wide open (f5.6) at 300mm, softer

than the 3rd party f4 primes wide open. All pretty reasonable.

 

<p>

 

Now we get to 400mm. PP can't do their SQF tests on this lens at 400mm

for some reason, so they switch to lp/mm. That's the first problem.

You now have no real way to compare 400mm performance with 300mm

performance. At least they could have given lp/mm results for 300mm

as well, so then you might have a "rosetta stone" to translate between

SQF and lp/mm, but they don't of course. <em>[in another part of the same issue

PP tests 50/1.4 lenses and does give both SQF and lp/mm. Problem is

that there isn't a correlation between SQF and lp/mm. On one lens,

72 lp/mm (center) and 36 lp/mm (edge) gives an SQF of 96.3 at 11x14.

On another lens 75 lp/mm (center) and 36 lp/mm (edge) gives an SQF

of 84.1 at 11x14. Go Figure!]</em> The second problem is in the

numbers. They show a big <em>drop</em> in sharpness going from wide

open (f5.6) to one stop down (f8). They go from 72 lp/mm to 57 lp/mm.

Ignoring the magnitude of the numbers (since we have no basis for

comparision, they don't mean much at this point), the drop on stopping

down is very odd. Very unexpected, very unusual and at odds with the

Amateur Photographer (UK) number trend, which shows performance

peaking 2 stops down (f11), just as you would expect and just as about

every other telephoto zoom does. The fact that this lens seems to do

what you'd expect at 300mm (peak at f11) from the SQF data, then do

something really odd at 400mm thows up a bunch of red flags to me. Not

to Pop Photog though. PP make a big deal about low resolution at f22,

f32 and f40 - also a bit odd since almost all lenses are diffraction

limited by such small apertures - where the lens will rarely, if ever

be used, but don't seem to find anything odd in the "wide open"

numbers. Perhaps this is because the wide open numbers seem good, and

so questioning them wouldn't be wise.

 

<p>

 

The overall conclusion (which you could probably reach with no test

data!) is still that if you need a ZOOM and you need IS, this is

<em>the</em> lens to get. If you need a telephoto that's sharp wide open, the jury is

still out, but the likely verdict is that the 300/4 (even + 1.4xTC) or

400/5.6 may still be the better (and cheaper) lens.

[bobatkins]

Another small addition to the database. Note that these are "facts"

about the Pop Photog lens tests, not "facts" about the lenses!!

 

<p>

 

I compared the PP lens test of the 75-300/4-5.6 IS (Feb 1996) to

their recent (Feb 1999) test of the 100-400/4.5-5.6L IS USM. The

numbers in [parentheses] are SQF values at 11x14

 

<p>

 

At the short end (75mm or 100mm), the 75-300 has a higher SQF wide

open [96.9/92.6] and stopped down [f8 97.6/96.0]. Plus it's 1/2 (or

1/3) stop faster. The 75-300 seems to be the winner here.

 

<p>

 

At 200mm the 75-300 has a higher SQF wide open [93.1/92.6] and a

slightly lower SQF stopped down [f8 97.5/98.1]. This looks pretty

much like a tie.

 

<p>

 

At 300mm the 75-300 is significantly better wide open [88.9/77.0]

and very slightly worse stopped down [f8 89.3/90.1]. Again the

75-300 looks like the "winner".

 

<p>

 

At 400mm of course, there is no comparison possible. No contest. The

100-400 wins!

 

<p>

 

 

 

<p>

 

In case PP says, well, you have to look at the comments not just the

numerical data, they call

both lenses "average" at 300mm. The 75-300 is described as "excellent"

at 75mm, the 100-400 as "above average" at 100 and 200mm. The field

curvature of the 100-400 was described as "high" at 300mm, while that

of the 75-300 was effectively zero.

("we could detect no field curvature at longer focal lengths").

 

<p>

 

What does this show? Either the new lens is no better than the

much cheaper, smaller, ligher 75-300IS in the range up to 300mm,

or you can't tell anything from Popular Photography lens tests.

You decide.

[tiago silva]

Last issue of Chasseur d'Images includes a test on this lens.

As I don't have a copy with me right now, I will only summarise the

results. Later I will be able to post a more exact report on the test

results.

 

<p>

 

Very good quality in the lower and mid focal range, from f/4.5 to

f/11. Good quality at 400 f/5.6 slightly increases at f/8.0.

Reasonably small vigneting at all focal lengths and also small

geometric distortions.

 

<p>

 

The results at 400/5.6 seem comparable to the Sigma 300 APO macro at

f/4 and maybe just a bit lower than the Sigma 400 APO macro at f/5.6.

 

<p>

 

The optical quality was rated as ****. The results seem impressive for

a zoom lens.

[sean_hester]

i saw and looked through this lens for the first time, first hand

yesterday. it looks and feels almost exactly like the 35-350. which

prings me to a question.

 

<p>

 

why is (or just is) the 100-400 a better range then the 35-350. the

extra 50 reach hardly makes a difference. and if you need 400 instead

of 350 you probably really need 500 or 600 or 800. on the other hand,

the range from 35-100 is often times useful.

 

<p>

 

i can't think of a shot that's "untakable" because i'm at 350 and need

400. but i can think of lots of shots that are "untakable" at 100

that would be "takable" at 50 or 35.

 

<p>

 

the IS is of course a big benefit. but i'd rather have an IS 35-350

then an IS 100-400.

[bobatkins]

The 35-350 doesn't take the Canon TCs, the 100-400 does. Now given

it's a zoom, and given it's already a bit slow, adding TCs isn't

exactly desirable, but in a push it's a useful ability.

 

<p>

 

The 35-350 never had a sparking reputation for sharpness at 350

either. I presume everyone's expecting the 100-400 to be better.

[ShunCheung]

The 35-350mm is a 10x zoom. There has to be a lot of major compromises

in the design to make a zoom with that kind of range, especially from

wide angle to long telephoto.

 

<p>

 

However, we can expect very high quality lenses from 70-200mm type

zooms that have a smaller range and is all telephoto in the whole

range. The same expectation applies to a 100-400 zoom.

[scott_macbutch]

Was wondering if any one had any comments on the speed and accuracy of

the focus when in servo mode. It looks like the new zoom would be

killer for stills, especially with the flexibility allowed with adding

a 1.4x converter and IS, but I never have been impressed with any of

the Canon IS lenses when it comes to quick focus acquisition such as

in shooting flying birds.

[sean_hester]

i haven't done formal tests, but when i've used the 75-300 and 300 f4

IS lenses, i've noticed that initial AF lockon seemed to be slowed a

bit. this seems a pretty obvious side effect of IS. if you need the

fastest possible AF you can turn the IS off.

 

<p>

 

for flying birds you're probably not getting much use from IS anyway

because you're moving the lens around. i guess if you're birds are

going in a straight line (parallel to the orinetation you want to

frame the shot) you can use "mode 2" but then you only get one axis

of IS.

 

<p>

 

for some subjects, IS is useless. in those cases you can turn off

the IS and reclaim the AF speed. with the 300 f4, and the IS turned

off, it focuses EXACTLY the same as the 300 f4 non IS (as far as i

cold tell in my 5 minute comparison test of the two.

[birdie_chew]

Just wondering if anyone has compared the optical performance of the Canon 100-400 with the Tokina 80-400? Thanks.

[robin_chok]

I recently bought a canon 100-400 IS L, it's great and rather easy to use.

It was difficult to choose between the 70-200L vs. 100-400L and I am not certain if I had made the right decision for my shooting style.

Anyway, first few rolls of film shows very sharp results with crisp, and contrasty pictures magnified on a 60 inch screen. AF is very fast with EOS 3 and EOS 50. Push pull zoom takes a bit of getting use to but it was fun when you get the hang of it.

In response to 100-400L vs Tokina 80-400 => no contest in sharpness and colour except for the price. But I think the price of a Canon 100-400L with USM and IS is reasonable.

 

Anyone out there has a flaw with the filter thread on the front of the lens? Mine looks as if the thread on the lens cap has bent the thread. Is the thread metal or plastic? feels like plastic to me.

 

Will be testing the lens further in terms of the handholdability with IS. Correction seems to be 1 F-stop at 400mm.

[birdie_chew]

Thanks Robin. Could you elaborate a little more on the comparison. If I may draw a parallel, is it like a Russian Mig 21 (Tokina 80-400) compared with an American F16 (Canon 100-400)?

 

As a Nikon user (longer than I remember), I am frustrated. They seem so slow in coming up with Silent Wave motor lenses. I am waiting impatiently for the Nikon 100-420mm with SW!!! I had thought of changing to a Canon system many times but the cost just didn't permit me. On, Nikon, get up!

[les_gyug]

I went down to the local photo dealer and (surprise!) they actually had one in stock, and at the same price as I would have had to mail order it from New York!, and without the hassle, customs charges, and duties (7% if its over $1500 CDN). Living in a small city (100,000 people) in Canada, I had no right to expect anything of the sort.

 

Anyway, I shot one roll (Sensia 100) in the parking lot, had it developed, and then bought the lens the next day. I already have a 28-135 IS so I knew what to expect more or less from IS. I've spent half a year slowly switching from Contax and still hadn't got any Canon long lens. I've used 300's (Nikons, Tokinas) and found them too short as a single length lens, so didn't really want the 300IS.

 

I took a sharp shot (under 20x loupe) of the cameras in the display case at 1/3 second at 100mm! At 400mm out in the lot, handholding at a distance only got really sharp at over 1/90 sec, but handheld at closest focus was sharp at around 1/30. I decided that was good enough for me. The quality is also head and shoulders above the old Tokina 400/5.6 (pre-ATX) that I had been using (and had given up on and already sold).

 

What I would really like is a Canon (or whoever) 400/5.6 IS, but I may have a really long wait for that one, in the meantime I'm hoping this one will do. I am still field testing this lens, but am kind of slow on this, it being winter here, evenings being very dark etc.

[robin_chok]

My parallel would be kinda like the tokina is the F-16 and the canon is the F-22. ie. more advancement because of time and also higher cost. Technology in the canon is state of the art when it come to zoom lens design. It's the only zoom lens in the world with Fluorite and Super UD glass and even Nikon hasn't got one. Plus the USM and IS it's a killer.

 

My advice to you is , maximise your Nikon gear, get a Nikkor 80-200mm 4.5-5.6 ( very good , too good for the price) or the Nikkor 70-210 4.5-5.6. Or look into the Sigma 70-200mm 2.8 EX lens with HSM. Rivals the Wave motors but minus the Canon and Nikon price ( A$1300 ). a truly excellent lens. Get a 2x converter for A$340 and you are at 400mm.

[bobatkins]

It's a fine lens but there are factors you shouldn't lose sight of.

For example, at $1700, it's almost twice the price of a Canon 300/4, which is not only sharper, but also a full stop faster. If you are a hopeless zoom addict who hates tripods, the 100-400L IS is your dream

come true I guess. Just try to think clearly if it's the best lens

for your particular needs.