DJL329 wrote:
How could *anything* that *anyone* posts here be taken as *fact* without an official press release from Canon?!? It's *all* speculation!
*That's my point*. Don't just toss speculation around like you've got a solid source (again, the first poster presented it as fact). The real fact is NO ONE knows whether or not it's a possibility, even if CR, some random source, Chuck Westfall or any other Canon employee says so. Stuff like this is kept under lock and key.
M Vers wrote:
*That's my point*. Don't just toss speculation around like you've got a solid source (again, the first poster presented it as fact). The real fact is NO ONE knows whether or not it's a possibility, even if CR, some random source, Chuck Westfall or any other Canon employee says so. Stuff like this is kept under lock and key.
Wait, if we both agree, then what are we arguing about? Let's grab a beer...
S Dilworth wrote:
The heart of this matter is that 24-70 mm f/2.8 lenses are not yet a solved problem. The f/2.8 tele-zooms have been around for decades, and they already achieved very high optical quality in the nineties. Since then, Canon and Nikon have added optical image stabilisation and refined performance further, partly because they could and partly because stabilisation is particularly useful for longer lenses: with a narrow angle of view, the shutter speed is more often limited by handshake than subject movement.
Trans-standard zooms are harder to design and manufacture, and even the widely praised Nikkor from 2007 has significant optical defects (and mechanical defects). With cameras getting more pixels, and photographers using zooms for studio and landscape work, the manufacturers probably feel it's more important to fix the optics than add stabilisation. It's hard to disagree, especially since stabilisation is less important for a trans-standard zoom than a tele-zoom (though still a desirable feature).
Here's a to-scale comparison of some trans-standard zooms with and without stabilisation:
Note how much larger the internal elements are in the f/2.8 lenses. Stabilisation is easiest, cheapest, and lightest to add when the stabilising group can be small. And it gets worse: adding stabilisation would mean making all the elements in front of the stabilising group even larger. It may also be more difficult to implement stabilisation on a negative-lead type lens than on a positive-lead type; it's probably not a coincidence that all trans-standard zooms with stabilisation are of the positive-lead type (positive lens group at the front).
The main complaints about 24-70 mm f/2.8 lenses are:
- their excessive size and weight: users frequently report giving up on these lenses after the novelty wears off
- their high purchase price, considered in tandem with their "boring" focal length and speed
- their less than flawless optical quality, e.g. serious field curvature at 24 mm, chromatic aberrations, heavy vignetting, etc.
Adding image stabilisation would only exacerbate these key problems.
For these reasons and others, I don't expect the next Canon to have stabilisation. I do expect Canon to release a replacement lens soon, and I'd expect it to come with better optics, more refined mechanics, possibly a slight weight reduction, and definitely a large price increase....Show more →
Thank you. So it will be the 24-105 again, I guess.
timbop wrote:
Boy does this get rehashed a lot. It also brings out the narrow minded "everyone should shoot the way I do and I do/don't need it so everyone else does/doesn't need it" crowd.
Amen to that, Tim.
I shoot APS-C cameras, and I find the IS on my 17-55 to be very useful. (I hope that if/when I move to full-frame cameras (if I stick with Canon) there will be a 24-70 IS.) But I wouldn't say that someone who doesn't find it neccesary is wrong.
I shoot APS-C cameras, and I find the IS on my 17-55 to be very useful. (I hope that if/when I move to full-frame cameras (if I stick with Canon) there will be a 24-70 IS.) But I wouldn't say that someone who doesn't find it neccesary is wrong.
Different courses for different horses.
The point is that some folks, when they do not think IS is useful for certain focal length(s) for their particular type of shooting, claim that IS is not useful at all for these focal length(s). (I am not saying anyone on this particular thread is making such claims; I found the arguments here mostly civil and well reasoned).
The truth is that such claims are extremely easy to falsify: even if we find one person who finds IS useful for these focal length(s), then IS is useful, at least for some people.
I, for one, believe IS is useful for all focal lengths. The only type of lens that would not benefit from IS is perhaps wide-angle T/S lenses -- these are really specialty lenses that are almost impossible to use without a tripod.
timbop wrote:
Boy does this get rehashed a lot. It also brings out the narrow minded "everyone should shoot the way I do and I do/don't need it so everyone else does/doesn't need it" crowd.
Do you really need f/2.8? Do you really need macro? Do you really need ring USM?
The answer to those is the same as "do you really need IS in this focal length"? The answer of course, is sometimes yes and sometimes no. The 17-55IS sells extremely well, with third party manufacturers following suit. There's even a Nikon 16-35VR, for those that pay attention. The reason canon hasn't done it yet is because they haven't felt the need to - nothing else.
As for the "added weight" argument, compare the weights of the 70-200 variants before you parrot that statement....Show more →
Yup.
Consider the above mentioned EFS 17-55 f/2.8 compared to the EF 24-105 which is larger, heavier, and slower (great set of comparative attributes eh?). Incidentally, I have both.
If Canon's 17-55 is so popular with croppers (at what many consider to be an outrageous price), why not an equivalent for FF? It's going to cost more to get to L series build quality, but I think it would sell well. That's why I think it may well be in the design phase already.
Glenn NK wrote:
If Canon's 17-55 is so popular with croppers (at what many consider to be an outrageous price), why not an equivalent for FF?
I know what you mean by this (i.e. f/2.8), but I'd argue the EF 24-105mm f/4L is the equivalent full-frame lens to the EF-S 17-55mm f/2.8. It has a similar total light throughput (or etendue) and angle-of-view range, so it provides images with approximately the same field of view, depth of field, shutter speed, and noise characteristics; in other words, equivalent images.
But we use full-frame cameras to get better images, not equivalent ones, so I do understand your point.
Unsurprisingly, the optical layout of the 17-55 mm f/2.8 and 24-105 mm f/4 is quite similar (positive-lead types, etc.), and they are very similar in size (including the internal elements, a couple of which form the stabilising group). The full-frame lens is heavier primarily because it's built to a higher mechanical standard.
It's much harder to make a full-frame 24-70 mm f/2.8 lens with image stabilisation than a stabilised 17-55 mm f/2.8 for APS-C or 24-105 mm f/4 for full-frame. The optical elements must be larger, a higher image quality is expected, the mechanics must be robust enough to withstand years of professional use, and in general the manufacturer can take fewer risks with the design.
Third-party manufacturers can do things a bit differently, and Sigma did in 2008 with the 24-70 mm f/2.8 IF EX DG HSM. This 790 g lens is a positive-lead type, and thus shorter (94.7 mm) at minimum focal length than the Canon and Nikon lenses in my diagram earlier in the thread. However, it's even larger in diameter, takes 82 mm filters, and has pretty lacklustre image quality. It doesn't have stabilisation.
Mike V wrote:
Canon will release this when people stop buying the non IS version of this lens.
As long as this lens still has good sales, they won't do it.
Not true. The 70-200s (both f/2.8 and f/4) non-IS versions were selling well; in fact, Canon sold them along side the IS versions for a while. Just different price point. I can totally see Canon having 3 full-frame L "standard" zooms in the lineup:
24-105 f/4 IS: ~$1,000 -- "budget" zoom with IS
24-70 f/2.8 ~ $1,500 - large aperture zoom that is not too expensive
24-70 f/2.8 IS ~$2,500 initially, maybe settling for $2,000 - large aperture zoom with IS for people with fat wallets
S Dilworth wrote:
...I'd argue the EF 24-105mm f/4L is the equivalent full-frame lens to the EF-S 17-55mm f/2.8. ...it provides images with approximately the same field of view, depth of field, shutter speed, and noise characteristics; in other words, equivalent images.
I'll have to disagree about providing the same shutter speed. For a given exposure level, f/2.8 will yield a faster shutter speed than f/4, regardless of the sensor size.
My Sekonic L-758DR light meter has a lot of modes and settings, but "What sensor size?" isn't one of them.
BrianO wrote:
I'll have to disagree about providing the same shutter speed. For a given exposure level, an f/2.8 lens will yield a faster shutter speed than an f/4 lens, regardless of the sensor size.
My Sekonic L-758DR light meter has a lot of modes and settings, but "What sensor size?" isn't one of them.
He's using the quasi-theory of "equivalence" (no offense intended), which makes very simplified (and not very realistic) assumptions about noise characteristics of FF and crop sensors, and what people want to achieve in photography.
Yes, I agree with you, f/2.8 is f/2.8 no matter what sensor size.
Well, f/2.8 is certainly f/2.8 for any sensor size, but it provides a different depth of field and signal-to-noise ratio for sensors of different sizes. For equivalent images, the depth of field and shutter speed must stay the same, and the ISO setting can be changed to allow that.
Anyway, I understand that full-frame shooters want better images than they'd get from a crop-sensor camera, not equivalent images.
The point here, though, is just that the EF-S 17-55 mm f/2.8 and EF 24-105 mm f/4L, both stabilised, are similarly difficult to design and make. A stabilised 24-70 mm f/2.8 for full-frame is much harder. That, more than any other reason, is why Canon hasn't made one yet. It's quite hard enough to make one that isn't stabilised, which is why nobody did until 2002, and why nobody has yet made one that pleases all photographers with its image quality.
S Dilworth wrote:
The point here, though, is just that the EF-S 17-55 mm f/2.8 and EF 24-105 mm f/4L, both stabilised, are similarly difficult to design and make. A stabilised 24-70 mm f/2.8 for full-frame is much harder. That, more than any other reason, is why Canon hasn't made one yet. It's quite hard enough to make one that isn't stabilised, which is why nobody did until 2002, and why nobody has yet made one that pleases all photographers with its image quality.
Canon is a huge company. If they decide to make a great 24-70/2.8 IS (superb IQ on FF, internal zoom, IS etc.) I have zero doubt that they can. The only question is if they want or, more precisely, if their marketing managers think there's a market for one.
howard wrote:
Not true. The 70-200s (both f/2.8 and f/4) non-IS versions were selling well; in fact, Canon sold them along side the IS versions for a while. Just different price point. I can totally see Canon having 3 full-frame L "standard" zooms in the lineup:
24-105 f/4 IS: ~$1,000 -- "budget" zoom with IS
24-70 f/2.8 ~ $1,500 - large aperture zoom that is not too expensive
24-70 f/2.8 IS ~$2,500 initially, maybe settling for $2,000 - large aperture zoom with IS for people with fat wallets
+1
Make the IQ as good as the Nikon or better and not only people with fat wallets will buy it.
The 24-70L has been great EOSfun for the professionals with their 1.3 crop bodies. Most of them prefer opening speed over image stabilization. The enthusiasts though with their full frame 5D's mostly preferred the 24-105L IS. Now that Canon's new full frame strategy for the pro is there with the demise of the 1.3 crop 1D and the professional version of "one for all 1DX" things are gonna change. There are no technical hurdles, it's just a marketing thing
The 24-70/2.8 was released with the 11MP 1Ds and the 24-105/4 was around the time of the 12MP 5D as a kit lens. The 1Ds II pushed both lenses to the breaking point and the 21MP bodies made the weaknesses of both lenses even more glaring. I'm not sure that an 18MP (step backward body) will push Canon to making a better 24-70, but if there are future high-res bodies new lens designs will be mandatory.
Yakim Peled wrote:
Canon is a huge company. If they decide to make a great 24-70/2.8 IS (superb IQ on FF, internal zoom, IS etc.) I have zero doubt that they can. The only question is if they want or, more precisely, if their marketing managers think there's a market for one.
eosfun wrote:
There are no technical hurdles, it's just a marketing thing
These statements all say the same thing. The difficulty in making a stabilised 24-70 mm f/2.8 lens lies in making it marketable, which is the same as saying the difficulty lies in making it good, small, light, and cheap. These are the usual concerns of the engineer. Nobody at Canon cares what could be achieved for a billion dollars, in some kind of moon-walking effort to prove what's technically possible in 2012.
I think Canon will make an EF 24-70 mm f/2.8L II, with a big price increase and no stabilisation. It will establish a new benchmark of image quality in this class of lens, and sell pretty well despite its price. I would be very surprised if it had stabilisation, for all the reasons I've mentioned in this thread.
It's *all* speculation!
