What's the next EF-S Lens?

Image circle is not done by just increasing and decreasing the size of the front element. Go check out the Olympus 4/3 lenses - 150 f2, 300 f2.8 etc. These are 2x crop lenses but look and feel just like a 135 format lens.

On the other hand look at the Leica M 50mm. They are so tiny and yet a 135 format lens. The difference? The register distance for one.
Well, let's imagine a EF-S 70-300/2,8 IS. Wouldn't that be smaller
than a EF 70-300/2,8 IS because of the smaller image-circle?
Nope...same size (well, there would be a very tiny reduction).

300/2.8 = 107mm - entrance pupil size regardless of image circle size.



That entire front section on the left would still be the same, and
the overall length would be the same. Some of those tiny elements on
the right could get very slightly smaller. Overall, no real change
in size.

Notice that there was plenty of room for them to move the rear
element closer to the sensor, even with the EF mount, and they
didn't. That's because there's no real advantage to doing so on a
telephoto lens.
I seriously doubt that! The front element on the EF-S lens could have
a 1,6x smaller diameter, and still collect enough light for the 1,6x
smaller image-circle!
--

 
From a business stand point probably some 18-200 IS. Beginners love those super zooms. Or the "one lens for all my travels" people.
--

 
I hope Canon can produce both EFS300 f2.8 and EFS500 f4 IS at the size of EF200 f2.8 L II for my 30D.

That would be cooooool....

--

EOS 30D, EFS10-22 USM, EFS17-55 f2.8 USM IS, EF70-200 f2.8 USM L, EF50 f1.4, EF85 f1.2 USM L II & 550EX.
 
I thought the same thing, slept on it. The part of the image circle
that is outside the sensor corresponds to a field of view that is
wider than the 480 MM equiv. FOV. In short, you're throwing away the
portion that corresponds to the 300MM FF FOV.

Look at it another way - a perfect lens will have a FOV that is
inversely proportional to the sensor size. Taken to extremes, you
could have a 300 MM telephoto converted to a 3 MM wide angle with a
sensor that is 100 times larger than a FF sensor. So, an ideal lens
is always "throwing away" light outside the angle of view determined
by the sensor.

A pin hole is an example of such an ideal lens. Of course a real lens
always has some limitation of the ultimate FOV it's capable of of.
Might be due to optics, baffles, obstructions. I suspect it's
mechanical in most cases, and hence arbitrary. Better to block and
baffle the light not used that to let it through to create flair and
all.

So, every lens ever made is "throwing" away light outside it's
equivalent FOV. In most cases you don't even think about it because
this extra light is mechanically blocked before it reaches the end of
the lens. That's just good design practic.
Yes, you're right. It's the FOV or angle of view that's important. A 100x crop sensor with a 100mm lens would only collect the light from an angle 100x more shallow than the FF-sensor, and therefore the entrence pupil must have the same size.
 
They need to make an EF-S normal prime, one that is higher (build) quality than their 35/2. Sigma has the 30/1.4 which is very popular.

Also, the 17-55IS is too expensive to be a realistic kit lens for the 30D, so they need to upgrader the 17-85 to make it the premium kit lens it originally was. I don't know how you do that (add fancy elements to improve corner sharpness and reduce fringing etc.?)

So the next 3 EF-S lenses:

EF-S 30/1.4
EF-S 18-200 IS
EF-S 17-85 IS II
 
I hope Canon can produce both EFS300 f2.8 and EFS500 f4 IS at the
size of EF200 f2.8 L II for my 30D.

That would be cooooool....
See discussion above as to why that's impossible.

--
Lee Jay
(see profile for equipment)
 
I hope Canon can produce both EFS300 f2.8 and EFS500 f4 IS at the
size of EF200 f2.8 L II for my 30D.

That would be cooooool....
See discussion above as to why that's impossible.
Well, you could get fairly close using DO technology ..... and there would be a useful side effect - no one would complain about the price of the EF-S 17-55 any more! Say, in round figures, about US $6000 for the 500/4, and US$5000 for the 300/2.8? I bet they would sell like crazy to the 400D crowd. :)
 
Also, the 17-55IS is too expensive to be a realistic kit lens for the
30D, so they need to upgrader the 17-85 to make it the premium kit
lens it originally was. I don't know how you do that (add fancy
elements to improve corner sharpness and reduce fringing etc.?)
This is what i was getting at in earlier post:
EF-S 15-60 f/3.5-4.5 USM (with or w/o IS).
 
I hope Canon can produce both EFS300 f2.8 and EFS500 f4 IS at the
size of EF200 f2.8 L II for my 30D.

That would be cooooool....
See discussion above as to why that's impossible.
Well, you could get fairly close using DO technology ..... and there
would be a useful side effect - no one would complain about the price
of the EF-S 17-55 any more! Say, in round figures, about US $6000 for
the 500/4, and US$5000 for the 300/2.8? I bet they would sell like
crazy to the 400D crowd. :)
But they still would not be the size of the EF 200mm f2.8 L. Possibly the weight, and maybe a bit closer to the length, but still a bit "girthy".

And again, virtually no gain in making this an EF-S mount. Almost all of the weight/size savings would be because of the DO tech. So now (in this hypothetical world) you have one of Canon's most expensive lenses (in the top 3 anyway), and yet it only fits on a consumer and prosumer body. Sales might be a little stagnant.

T!
--

 
If Canon reduced the "kit lens tax" for the 17-85 IS to $299, this would be good enough for most 400D and some 30D/40D customers who would be unhappy with the 18-55 kit lens.

It is too bad that the 17-85 is not f/3.5-4.5 instead of f/4.5-5.6...

EF-S 15-60 IS would be a good lens, but I doubt it could be any cheaper than the 17-85 IS. An EF-S 24-105 IS would probably easier to design/manufacture.
 
"Sigma has the 30 f1.4. We (canon) can't put out a 30mm - no way! We have 28mm and 35mm marked on all our lenses and flashes. It must be either a 28 or 35. How about a 35 f1.4? Oh, that's right - we already have one of those. People can buy that one if they want a 35 f1.4. What about a 28 f1.4? Shoot - we have a 28 f1.8, and it's a pretty good lens, too - that's close enough.

Now that that's settled, lets put out a 18-200 zoom for the masses and make some real money!"
--



Bossier City, Louisiana
http://www.pbase.com/ericsorensen
 
Rumours say the next EF-S lens will be an updated kit lens 18-55IS accompanying the 40D.

IS is a selling feature these days. New buyers of DSLRs will have experienced the wonders of IS in point and shoots (I wouldn't dream of recommening a p'n's without IS to anyone these days). I think those rumours sounds plausible from a marketing point of view.
 
I really like the thought of the 17-85mm range on a 1.6x crop camera, along with IS, but the reviews have scared me off: optical quality doesn't seem good enough, especially wide open which is where I'd use it most (weddings and inside the home). So I'm looking hard at the 17-55 with a 2nd body for either my 100mm or the 70-200, or the 24-105 although that isn't quite wide enough on a 1.6 crop, nor fast enough even with IS. The 17-85 IS also seems like a perfect walkaround and travel lens, again except for the less-than-sterling optics.

So I vote for a 17-85/2.8 IS II.

What are the results of you 17-85 owners?
--
Mike D.
http://www.mikedphoto.smugmug.com
 
What are the results of you 17-85 owners?
It's a good lens with decent build and decent optical quality at all ranges. It's not an L-prime but it's a quality consumer zoom with effective IS, fast focusing, good sharpness, and relatively little CA and distortion. It does have some corner softness, some distortion and some CA but I've shot with it in all sorts of conditions and have never had any shots that showed any optical defects on prints up to 8x12.

--
Lee Jay
(see profile for equipment)
 
17-85 IS is much better for general purpose use than the impression you would get by reading these forms. 5x zoom range is very useful. It has a well documented set of problems, most of which are the result of the 5x zoom range:

1. Excessive barrel distortion at 17mm. Distortion is significantly reduced by 20mm and almost gone at 24mm. If you are going to be using this lens in 17mm range most of the time, you should probably buy a EF-S 10-22 instead.

2. Softness at extreme borders. I am not sure how many people care about the sharpness of extreme borders of their images for daily use. In many cases, those areas get cropped after recomposing the image during post processing.

3. CA, especially at the wide end of the zoom range. This is bad. It can be corrected to some degree in post processing, but still annoying. This is one thing Canon should fix if they decide to make a Mk II version of this lens.

I have gotten a good number of excellent shots with this lens in the 50-85mm range, and it does a decent job in the 24-50mm range. And it gives you OK shots in 17-24mm range without having to carry another wide angle lens.

Alternatives are:
  • EF-S 17-55 f/2.8 IS
Pros: faster, much less distortion at wide-end, sharper overall
Cons: limited focal range in tele end, heavier, larger, 2X price
  • EF 24-105 f/4L
Pros: faster, much less distortion, sharper overall, longer tele range
Cons: limited focal range in the wide end, heavier, larger, 2X price

For those who don't want to pay $1K for a big, heavy lens, EF-S 17-85 IS is a good compromise at half the price. It would have been really great if this lens had been f/3.5-4.5...
 
I agree with most of your statements but strongly disagree with your overall conclusion. IMO, the 17-85 is a good choice only for people who absolutely want IS and USM in this range and who shoot mainly at the tele end. - Yes, I had several copies of the 17-85 and I used one extensively. I really tried to like it but eventually sold it. Between 17-35mm my old kit lens was better (sharper, less vignetting, less CA) than the 17-85.
17-85 IS is much better for general purpose use than the impression
you would get by reading these forms. 5x zoom range is very useful.
It has a well documented set of problems, most of which are the
result of the 5x zoom range:
It cannot be only the zoom range: Have you noticed that the new Tamron 18-250
is sharper across the board than the 17-85 in tests on SLRgear and photozone?
1. Excessive barrel distortion at 17mm. Distortion is significantly
reduced by 20mm and almost gone at 24mm. If you are going to be using
this lens in 17mm range most of the time, you should probably buy a
EF-S 10-22 instead.
There are at least two outstanding choices if 17mm is important: Canon17-55/2.8 and Tamron 17-50/2.8. I would consider the 10-22 only if you really want to go wider than 17mm.
2. Softness at extreme borders. I am not sure how many people care
about the sharpness of extreme borders of their images for daily use.
In many cases, those areas get cropped after recomposing the image
during post processing.
It doesn't make much sense to shoot wide angle and then to zoom in later digitally in postprocessing if you want the best resolution. Sharp borders at wide angle are very important in landscape photography. The 17-85 unfortunately does not deliver, even stopped down.
3. CA, especially at the wide end of the zoom range. This is bad. It
can be corrected to some degree in post processing, but still
annoying. This is one thing Canon should fix if they decide to make a
Mk II version of this lens.
True, but this bothered me actually less than border softness and vignetting.
 
Okay, to answer the question.

These are just JPEGs from the Rebel XT and 17-85IS.

First, worst focal length, 17mm and f9:



Full size:
http://photos.imageevent.com/sipphoto/samplepictures/IMG_3131.JPG

Best focal length 85mm and f11:



Full size:
http://photos.imageevent.com/sipphoto/samplepictures/IMG_4639.JPG

Are they L-prime perfect? No. Are they as good as my 24-105L on the 5D? No. But print them 8x12 or 13x19 and see what you think. For a 5x consumer zoom with IS and ring USM, I think the 17-85IS is a great value and a lens that provides pretty decent images over the whole range. I even use it for 1:1 macros with tubes. We have several 13x19s on the wall here that came from that combo and they are quite good, and I examine them much more closely than anyone who actually views them.

--
Lee Jay
(see profile for equipment)
 
We have the 60mm macro, the 10-22, the 17-85IS, the 17-55IS. What's
next!?
What is IMO desperately needed is a very high quality and compact
EF-S 30/1.8 or 30/1.4.
Better than Sigma's good copies? Or just with less inter-copy variation? Isn't a good Sigma good enough?
More likely is probably a long range zoom,
e.g. EF-S 18-200 IS, because Nikon has already one.
Commercially, this is what they really should get pronto. They're missing numerous body sales to Nikon right now because they haven't one. Not that I would buy one. But the DSLR first-timers love such a lens. Very few of them come to the DSLR world primarily to enjoy the exhilarating joy of changing lenses. And a number of them come from dicgital superzoom cameras like Panasonic FZ50 or Canon Powershot S5 IS, and have come to expect a 10 x or 12 x zoom lens.
Will Canon go long for EF-S next time around? I think so!
It wouldn't make much sense because there are plenty of long EF
lenses and the smaller EF-S image circle affects mainly the design of
wide lenses. But something starting at 50mm like an EF-S 50-150/2.8
IS may be in the pipeline.
That's one that I'd really like as a complement to my 17-55. Sigma makes one without IS, er, OS, which is said to be quite good optically, but IS really starts to be a huge improvement at those focal lengths. I don't care if it's EF or EF-S, but I want one!
 

Keyboard shortcuts

Back
Top