GEAR: Canon EF-S 10-22 f/3.5-4.5
Tutorials and Reviews
Registered Users, Retired Mod Posts: 138
Canon EF-S 10-22 f/3.5-4.5
Review by Scott_Quier
I'm posting this review for a couple of reasons:
I don't have any "professional grade" glass. I have read that some hold the opinion that this lens is, in fact, "L" grade glass but that Canon didn't put the red ring on the lens because it is an EF-S lens - something about "L" glass should support full-frame sensors. Looking in from the outside, I really can't talk to that, just thought I would throw that in the mix.
All the above having been said, let me say this is the sharpest lens I have ever used and has convinced me that all future purchases will be "L" glass (I'll just have to save my pennies a little longer than I'm used to). I sure wish I had read and paid heed to some of Andy's (and others') comments before I did the "get glass, sell glass, replace glass, repeat" process. The advice I've read goes like something this:
"DSLR bodies come and go, but lens - they can last a very long time. So, put your $$ into the glass. Get the best you can, even if you have to wait a bit. You will not regret your decision."
Some rules of the game:
OK, onto the (maybe) useful information.
Because of all the technical difficulties in engineering a wide lens (I'm taking that on faith, I don't have the math background to support this contention) these lenses typically have problems in three areas: flare, contrast loss (under certain circumstances), and Chromatic Aberration (CA). I look at each of these problems, one at a time.
This is my copy of the lens. I thought I would include a picture of it, JUST to make sure that everyone is on the same page. Yeah, I should have purchased the hood. I thought it came with the lens. It doesn't - go figure.
Flare: This isn't a work of art. It was intended to demonstrate this lens' propensity to flare.
Point the camera/lens at a point source of light that is brighter than the rest of the image, and you typically get a ghost of that light source at some point on a line defined by the bright light and the center of the lens, but on the "other side" of the center.
I think you will agree, there is little, if any, flare to be seen.
OK, that didn't work. Let's try a much more gruelling test. This was taken at about 10:50AM with the sun just out of the frame. Looking in the upper left corner, you can see some evidence of flare. I have other examples under similar harsh conditions, but this is good enough to demonstrate that this lens is not perfect.
But, when you shade the lens (like you're supposed to), all those problems seem to go away!
Contrast: Point your camera/lens such that your objective lens is not shaded. What do you usually get, aside from flare? All that unwanted light enters the lens in such a way that it bounces around and around. This results, for reasons I don't completely understand, in a lowering of the contrast of your images. You can see it here when you compare the above image to this one. The contrast in the second image is just *that* much sharper than it is above.
Chromatic Aberration: (also known as "Purple Fringe") - you can observe this best in images where you have a very bright region next to a very dark region. CA is also, usually, most pronounced at the corners (because these areas are further from the axis of the lens where all the math seems to work so much better). The images below are examples of what is possible under very extreme conditions.
This image is a "whole picture" example of the scene used to for this purpose. There is some image distortion here, but, hey, give the lens a break. This is at 10mm with the camera tilted at about 30 degrees above level. There is no way to avoid the distortion given the situation. It's the nature of the beast. This image was taken at f/3.5
This, and all the following images are 100% crops (400px X 400px). This crop is from near the center of the above image and demonstrates that my copy of the lens exhibits little, if any, CA near the center. This is about what one would expect from a lens of this reputed caliber.
I don't include center crops below because things only get better with decreased aperature and, to my mind, things here are good enough for my needs.
This, and all the following images, are 100% crops (400 x 400 pixels) of the upper left corner of the scene. Each were captured using a different aperature.
This was captured at f/3.5. Here, you can see significant CA along the right edges of the branches.
At f/5.6, the CA is significantly less pronounced.
By f/8.0, the CA is well controlled for a lens of this focal length.
At f/16, the CA observed in this crop is about the same as that seen at f/8.0
But, stopping down to f/16 shows us something else. It appears, from this small data set, that the sweet spot for this lens, aperature wise, to be in the 5.6 -> 8.0 range.
Conclusions:
Do I like this lens? Would I buy it if I knew then what I know now? Oh, yeah, I really love the lens and am looking forward to the time when my JOB allows me to get out and do camera type things. (I'm not out photographing now because I kind of committed to Andy (and others) that I would get this together and I like to meet my obligations).
Review by Scott_Quier
I'm posting this review for a couple of reasons:
- I just got the Canon EF-S 10-22 lens and WOW!
- I've read a lot of posts here and elsewhere saying how wonderful this lens is, but only in one did I find any images to support that position. I thought adding the detail that follows might help someone.
I don't have any "professional grade" glass. I have read that some hold the opinion that this lens is, in fact, "L" grade glass but that Canon didn't put the red ring on the lens because it is an EF-S lens - something about "L" glass should support full-frame sensors. Looking in from the outside, I really can't talk to that, just thought I would throw that in the mix.
All the above having been said, let me say this is the sharpest lens I have ever used and has convinced me that all future purchases will be "L" glass (I'll just have to save my pennies a little longer than I'm used to). I sure wish I had read and paid heed to some of Andy's (and others') comments before I did the "get glass, sell glass, replace glass, repeat" process. The advice I've read goes like something this:
"DSLR bodies come and go, but lens - they can last a very long time. So, put your $$ into the glass. Get the best you can, even if you have to wait a bit. You will not regret your decision."
Some rules of the game:
- All images were captured using my Canon 20D, tripod mounted. All but the second image was captured without a shutter release cable. The first was with flash at 1/125 and used a different lens (I've only got one copy of the 10-22). For the rest, I figured, at these shutter speeds, the cable release would not be needed.
- All images are as "straight from the camera" as possible. I shot the images in RAW, imported them into PS CS2, via ACR, changed mode to 8-bit, and saved them as maximum resolution/quality JPG files.
- All images were shot at 10mm. Results were only better at long focal lengths, so I chose to shorten the review by concentrating only on "worst case scenarios".
- Where it makes sense, a link to the "original" JPG is also provided.
OK, onto the (maybe) useful information.
Because of all the technical difficulties in engineering a wide lens (I'm taking that on faith, I don't have the math background to support this contention) these lenses typically have problems in three areas: flare, contrast loss (under certain circumstances), and Chromatic Aberration (CA). I look at each of these problems, one at a time.
Point the camera/lens at a point source of light that is brighter than the rest of the image, and you typically get a ghost of that light source at some point on a line defined by the bright light and the center of the lens, but on the "other side" of the center.
I think you will agree, there is little, if any, flare to be seen.
Contrast: Point your camera/lens such that your objective lens is not shaded. What do you usually get, aside from flare? All that unwanted light enters the lens in such a way that it bounces around and around. This results, for reasons I don't completely understand, in a lowering of the contrast of your images. You can see it here when you compare the above image to this one. The contrast in the second image is just *that* much sharper than it is above.
This image is a "whole picture" example of the scene used to for this purpose. There is some image distortion here, but, hey, give the lens a break. This is at 10mm with the camera tilted at about 30 degrees above level. There is no way to avoid the distortion given the situation. It's the nature of the beast. This image was taken at f/3.5
I don't include center crops below because things only get better with decreased aperature and, to my mind, things here are good enough for my needs.
This was captured at f/3.5. Here, you can see significant CA along the right edges of the branches.
But, stopping down to f/16 shows us something else. It appears, from this small data set, that the sweet spot for this lens, aperature wise, to be in the 5.6 -> 8.0 range.
Conclusions:
- Flare - Is very well controlled, especially for a lens as wide as this. If you practice appropriate technique to avoid this problem (like use a lens hood and/or shade the lens) you will probably never see this "problem".
- Contrast - Loss of contrast due to light bouncing around inside the lens in an uncontrolled manner is nearly unnoticable. Again, good technique will virtually eliminate contrast issues.
- CA - yeah it's there. You can easily find it if you pixel-peep a bit (or enlarge unprocessed images for printing). If, on the other hand, if you shoot RAW and adjust for the CA in ACR (or other similar RAW converter) you can greatly reduce or eliminate this artifact from your images.
- As an aside, the last series of image would seem to support that the sweet spot, aperature wise, is somewhere in the range of 5.6 to 8. By 16, things in the corners start to get a little fuzzy. The "fuzzy-ness" could have been caused by camera shake as the shutter speed at f/16 was only 1/25 sec.
Do I like this lens? Would I buy it if I knew then what I know now? Oh, yeah, I really love the lens and am looking forward to the time when my JOB allows me to get out and do camera type things. (I'm not out photographing now because I kind of committed to Andy (and others) that I would get this together and I like to meet my obligations).
0