5D2 + 17-40mm or 500D + 10-22mm?

[jabbott]

Neuro brings up a good point...  the 17-40 is indeed quite a bit worse in terms of distortion than the 10-22.  If you are using Lightroom you can apply a lens correction profile which does a good job of cleaning it up.  I first noticed the distortion when taking a photo of a door near the edge of the frame, and noticing that the door was noticeably bowed.

[neuroanatomist]

Neuro brings up a good point...  the 17-40 is indeed quite a bit worse in terms of distortion than the 10-22.  If you are using Lightroom you can apply a lens correction profile which does a good job of cleaning it up.  I first noticed the distortion when taking a photo of a door near the edge of the frame, and noticing that the door was noticeably bowed.

Yep - and of course, while that distortion can be corrected, that correction further reduces corner sharpness (in this case, maybe 'further increases corner softness' is the better phrasing).

[sagittariansrock]

Now, another question, is it 17-40 or 16-35?  I'm taking mostly landscape photos using a tripod and long shutter speed.  Is the 16-35 a lot better than 17-40 @ F/8 and above for example?  I don't worry about low-light.  I've got primes to take care of that.

In that case, no - the 17-40mm would be the way to go.  I got the 16-35L II because I did want to use it in low light

While I agree that the 16-35mm may not be $800 worth of IQ compared to the 17-40, there are non-Canon UWA zooms that I like far more than the underwhelming 17-40. I would particularly look at the excellent Tokina 16-28mm f/2.8. The fast aperture will give you additional flexibility for DoF and low lights if you need it some day.

[wayno]

Gosh. Am I the only champion for this "underwhelming" lens? Where I am it's an $800 L lens. It's not the sexiest beast by any stretch but as I said I have found it slightly better than the 10-22. I agree it has more distortion than the former- marginally - but anyone who is concerned about correction of this affecting corner performance should take a reality check. I'm all for pixel peeping - believe me I'm obsessed - but I know enough to know that correcting in LR makes no discernable difference at all in reality. We can choose to ignore that fact for the same of spending $$$$ more for no 'real' difference or we can just roll with it. I spend a lot of time mulling over the former but generally settle on the latter: reality.

I too will be keen to look at the new alleged UWA zoom in the pipeline but for now the 17-40 is doing a perfectly good job for me. I also appreciate the extra length relative to the 10-22 (or by comparison the 16-35).

Interestingly I used my 24-70l quite a bit on a recent trip and I found I was getting just as much distortion and in some instances more out of it than my 17-40 for certain shots. Which surprised me.

[robbinzo]

This means that on a crop sensor at 10mm, you will get slightly more distortion of verticals than with 16mm on a FF camera.

That's ony true if you're talking about the exact same lens, i.e. a 10mm lens with a FF-sized image circle.  When you compare different lenses...

EF-S 10-22mm @ 10mm on APS-C


EF 16-35mm f/2.8L II @ 16mm on FF


1.25% distortion on APS-C vs. 3.26% distortion on FF.  Note that that's with the 16-35 II - the distortion on the 17-40L @ 17mm is even worse.

Now, another question, is it 17-40 or 16-35?  I'm taking mostly landscape photos using a tripod and long shutter speed.  Is the 16-35 a lot better than 17-40 @ F/8 and above for example?  I don't worry about low-light.  I've got primes to take care of that.

In that case, no - the 17-40mm would be the way to go.  I got the 16-35L II because I did want to use it in low light, e.g.:


EOS 5D Mark II, EF 16-35mm f/2.8L II USM @ 27mm, 1/25 s, f/2.8, ISO 3200

Good info. Thanks!

[Mt Spokane Photography]

Just recently, I'm thinking of buying a 17-40mm L lens.  I intend to upgrade to either 5D2 or 6D after sometime.  My question is, is it really better than just having an APS-C + 10-22mm lens?  This will be used for landscape photography.  I'm also into macro photography but perfectly happy with a 500D + 100mm F2.8 USM lens.  Should I just forget going FF and go for APS-C?  I'm confused because I really love both macro and landscape photography.  I'm planning to complete the set by next year March since summer starts in my place during that month.

Just hold off.  The 17-40mm L is no better than your 10-22.  Bank the money and wait to see what is announnced in January next year.  There is no use buying lenses for a FF body, particularly since prices are dropping due to the world economy.
I did not fall in love with a 17-40mmL on my 5D MK II, it was good but did not really inspire me.  I recently bought a 16-35mmL and it is sharper, but that could be a sample variation thing. 

[verysimplejason]

Neuro brings up a good point...  the 17-40 is indeed quite a bit worse in terms of distortion than the 10-22.  If you are using Lightroom you can apply a lens correction profile which does a good job of cleaning it up.  I first noticed the distortion when taking a photo of a door near the edge of the frame, and noticing that the door was noticeably bowed.

Yep - and of course, while that distortion can be corrected, that correction further reduces corner sharpness (in this case, maybe 'further increases corner softness' is the better phrasing).

Will the DLO in DPP help?  I still have not used that feature.

[neuroanatomist]

Neuro brings up a good point...  the 17-40 is indeed quite a bit worse in terms of distortion than the 10-22.  If you are using Lightroom you can apply a lens correction profile which does a good job of cleaning it up.  I first noticed the distortion when taking a photo of a door near the edge of the frame, and noticing that the door was noticeably bowed.

Yep - and of course, while that distortion can be corrected, that correction further reduces corner sharpness (in this case, maybe 'further increases corner softness' is the better phrasing).

Will the DLO in DPP help?  I still have not used that feature.

It will correct the distortion, yes.

[sagittariansrock]

Gosh. Am I the only champion for this "underwhelming" lens? Where I am it's an $800 L lens. It's not the sexiest beast by any stretch but as I said I have found it slightly better than the 10-22. I agree it has more distortion than the former- marginally - but anyone who is concerned about correction of this affecting corner performance should take a reality check. I'm all for pixel peeping - believe me I'm obsessed - but I know enough to know that correcting in LR makes no discernable difference at all in reality. We can choose to ignore that fact for the same of spending $$$$ more for no 'real' difference or we can just roll with it. I spend a lot of time mulling over the former but generally settle on the latter: reality.

I too will be keen to look at the new alleged UWA zoom in the pipeline but for now the 17-40 is doing a perfectly good job for me. I also appreciate the extra length relative to the 10-22 (or by comparison the 16-35).

Interestingly I used my 24-70l quite a bit on a recent trip and I found I was getting just as much distortion and in some instances more out of it than my 17-40 for certain shots. Which surprised me.

When I say underwhelming I compare it to the EF-S 17-85mm (yes, the $ 350 lens) and 10-22 I had at the time on 7D as well as the 24-105mm on the 5DII.
I am not talking about pixel peeping- but the colors and the sharpness were just not great. This is, of course, subjective. YMMV, so I am sure there are hundreds of people who are perfectly happy with their 17-40.
Additionally, on a crop sensor Rebel with limited ISO capabilities, IS is often necessary for the 17-40mm focal length with f/4 or smaller.

[Zv]

I found the 10-22 to be a bit too soft all round, especially compared to my 17-55 f2.8. Also @ 10mm it wasn't that great. Plus CA and flaring were issues. However, it's like the only UWA option from Canon for crop sensors. Now the 17-40 might not improve spectacularly on this because it is just the beginner level of UWA on FF. Fine for most peope (who are content with an L lens under $1000! ).  For those who need the larger aperture and better performance they will go for the 16-35 II. If all you ever do is landscape then primes would be the way to go. Get a TS if your really serious. But at the end of the day it's really FF where all the landscape action is. 

I swapped my 10-22 for 17-40 cos I bought a 5D mkII and needed wide coverage for cheap. Have only had it for a few weeks but so far it's doin well.  Quite impressed with the IQ.

[verysimplejason]

Now, another question, is it 17-40 or 16-35?  I'm taking mostly landscape photos using a tripod and long shutter speed.  Is the 16-35 a lot better than 17-40 @ F/8 and above for example?  I don't worry about low-light.  I've got primes to take care of that.

In that case, no - the 17-40mm would be the way to go.  I got the 16-35L II because I did want to use it in low light

While I agree that the 16-35mm may not be $800 worth of IQ compared to the 17-40, there are non-Canon UWA zooms that I like far more than the underwhelming 17-40. I would particularly look at the excellent Tokina 16-28mm f/2.8. The fast aperture will give you additional flexibility for DoF and low lights if you need it some day.

There's only one thing I don't like with Tokina 16-28mm is that it can't take any filters.  I think for landscape photographers especially those with Canon (DR) requires filters.  Are there any UWA that takes filters but as cheap as this Tokina and with the same IQ?  That's the reason why I'm looking hard at 17-40mm...

[verysimplejason]

I found the 10-22 to be a bit too soft all round, especially compared to my 17-55 f2.8. Also @ 10mm it wasn't that great. Plus CA and flaring were issues. However, it's like the only UWA option from Canon for crop sensors. Now the 17-40 might not improve spectacularly on this because it is just the beginner level of UWA on FF. Fine for most peope (who are content with an L lens under $1000! ).  For those who need the larger aperture and better performance they will go for the 16-35 II. If all you ever do is landscape then primes would be the way to go. Get a TS if your really serious. But at the end of the day it's really FF where all the landscape action is. 

I swapped my 10-22 for 17-40 cos I bought a 5D mkII and needed wide coverage for cheap. Have only had it for a few weeks but so far it's doin well.  Quite impressed with the IQ.

I love TS especially the 24mm but it's way too pricey for me.  Maybe if I'll be able to earn something from photography then I'll buy the 24mm or 17mm TS.

[wayno]

Gosh. Am I the only champion for this "underwhelming" lens? Where I am it's an $800 L lens. It's not the sexiest beast by any stretch but as I said I have found it slightly better than the 10-22. I agree it has more distortion than the former- marginally - but anyone who is concerned about correction of this affecting corner performance should take a reality check. I'm all for pixel peeping - believe me I'm obsessed - but I know enough to know that correcting in LR makes no discernable difference at all in reality. We can choose to ignore that fact for the same of spending $$$$ more for no 'real' difference or we can just roll with it. I spend a lot of time mulling over the former but generally settle on the latter: reality.

I too will be keen to look at the new alleged UWA zoom in the pipeline but for now the 17-40 is doing a perfectly good job for me. I also appreciate the extra length relative to the 10-22 (or by comparison the 16-35).

Interestingly I used my 24-70l quite a bit on a recent trip and I found I was getting just as much distortion and in some instances more out of it than my 17-40 for certain shots. Which surprised me.

When I say underwhelming I compare it to the EF-S 17-85mm (yes, the $ 350 lens) and 10-22 I had at the time on 7D as well as the 24-105mm on the 5DII.
I am not talking about pixel peeping- but the colors and the sharpness were just not great. This is, of course, subjective. YMMV, so I am sure there are hundreds of people who are perfectly happy with their 17-40.
Additionally, on a crop sensor Rebel with limited ISO capabilities, IS is often necessary for the 17-40mm focal length with f/4 or smaller.

Yeah, fair enough, it's a personal view and there's no way I can argue with that.

My experience is certainly quite different. I must have a good copy because I can't be more impressed with mine to be honest. It was probably the most unexcited I've been in buying a lens (at the time) as I was effectively replacing my old 10-22 but I was quite surprised at a number of things with regards to how it improved the IQ of my night photography: much cleaner and punchier starbursts, better and less lens flare (but slightly odd flare from time to time), and stronger and more impressive colours. I do believe the 5D2 sensor has a bit to play with the difference but it cuts to the nub of the original poster's comparison query.

I agree the 17-40L has a fair whack of distortion but I don't see it as anything terribly unusual.

For me (personally), I enjoy the zoom aspect at night as it means I can get a shot quickly without pacing around. I did consider using the Zeiss 21 for awhile but decided that the 17-40 was better suited to the more quickfire guerilla style photography of night stuff. So it really suits me.

As for comparisons with the 16-35L, several other posters have nailed it appropriately. As I understand it (and I cannot speak from personal experience, granted), the IQ comparison between both is not significant and most crucially, the 16-35 is advantageous in a more photo-journalistic sense as opposed to the more traditional tripod landscape usage given the 1 stop faster aperture. I have READ that the 16-35 exhibits equivalent (if not worse) corner sharpness. Again, I stress I haven't tested it for myself. I think both would fit well into a cash-flushed (or pro) camera bag. The 17-40 is also far handier to lug around at night (or up hills) than the 16-35 on account of it's comparative weight. But it would be a dog in a late afternoon or evening.

The final advantage of the 17-40 is the slightly improved reach. Might not seem like much but I'm happy to have that extra zoom when compared to the 16-35ish equivalent focal length of the 10-22.

There's no doubt to me that at a pixel level, both lenses are not stellar. The alleged new Canon UWA zoom, the Nikon equivalent and the Zeiss ZE 21 have the advantage but for the relative cost of the 17-40 I reckon it's a good alrounder that I'm happy to settle on for the moment and it hasn't harmed my IQ/photography in the slightest (and for whatever it's worth, I pack a bit of L glass so I know what good glass is).

For reference, mostly everything on the page below has been taken using the 5D2 and the 17-40L. Processing has just been colours/contrast etc and the usual RAW fix-ups. No HDR silliness. The lens is a little champion, in my opinion.

www.facebook.com/waynegrivellartandphotography

[sagittariansrock]

Now, another question, is it 17-40 or 16-35?  I'm taking mostly landscape photos using a tripod and long shutter speed.  Is the 16-35 a lot better than 17-40 @ F/8 and above for example?  I don't worry about low-light.  I've got primes to take care of that.

In that case, no - the 17-40mm would be the way to go.  I got the 16-35L II because I did want to use it in low light

While I agree that the 16-35mm may not be $800 worth of IQ compared to the 17-40, there are non-Canon UWA zooms that I like far more than the underwhelming 17-40. I would particularly look at the excellent Tokina 16-28mm f/2.8. The fast aperture will give you additional flexibility for DoF and low lights if you need it some day.

There's only one thing I don't like with Tokina 16-28mm is that it can't take any filters.  I think for landscape photographers especially those with Canon (DR) requires filters.  Are there any UWA that takes filters but as cheap as this Tokina and with the same IQ?  That's the reason why I'm looking hard at 17-40mm...

Exactly the reason I didn't buy it though I was getting a really good deal. I really hope someone can respond to this question knowledgeably.
By the way, I think the general consensus is that getting a FF like the 5DII will far outweigh any shortcomings from having the 17-40 and I can't agree more.

[extremeinstability]

I'm not sure you can find a better flare resistant lens than the 10-22.  It's hard to get it to flare even pointed at the sun.  The 17-40 isn't the same story at all. 

I prefered my 10-22 on a T2i than I did when I had a 5D II and a 17-40L.  I had actually sold the T2i and 10-22 to buy the 5D II back then and already had the 17-40, which was fine on a crop.  It wasn't up to snuff enough full frame and I didn't have enough extra money for a full frame worthy lens, so I sold the 5D II and 17-40L and went back to T2i 10-22.  Unless you have some magical 17-40 copy I'd count on bad enough corners that make having the full frame out there useless/pointless.  And toss in big vignetting too. 

But I finally recently went 5D II again and have been trying to get worthy lenses for it.  14mm Samyang is a great cheap choice, with higher resolution than the Canon 14mm F2.8.  Sharp corners.  Only $400.  The vignetting is bad though(big whoop though really) and I'd see the lens as more of a F5.6 lens full frame for vignetting on any higher dynamic range scene that one's shadows don't need any vignetting "help" on.  I'd rather have that on there with F2.8 if I ever needed it, than a 17-40L again. 

But for crop the 10-22 is a great lens.  Thing is it isn't parafocal, so if you live view set the focus while at 10mm, if you zoom in it's going to be off again by enough to matter.  I could see that alone being a reason some think it's soft.  It's not.  THe other thing is using autofocus on such wide lenses is often an utter joke.  Simply get a nicely lit subject with contrast, then repeatedly autofocus on one spot, watching where the lens indicator winds up.  I was amazed how much it varied.  3 feet to past infinity marker was pretty much the norm.  Just useless to autofocus.