[Lak]

This test was done on a 300D on tripod, with manual setting. Shutterspeed was changed accordingly. This test did not harm any animals.
All done at 50mm. The 18-55mm was slightly off (at 49mm, yeah sorry my bad ).
Original pic with the prime lens at f-5.6



All these are at 100% crop focused on the "R" in "HERE"

Canon EF 50mm f/1.8 @ f-5.6 and @ f-8



Canon EF 28-135mm f/3.5-5.6 @ f-5.6 and @ f-8



Canon EF-S 18-55mm (300D kit lens) @ f- 5.6 and @ f-8



(adding in right, off-centre, bottom edge at f/5.6)

prime 50mm lens, 28-135mm lens, and 18-55mm kit lens


(adding in left, off-centre, bottom edge at f/8)

[Lak]

As you can see the prime 50mm is tack sharp both at f/5.6 and f/8. No complaints. Its Canon's miracle/bargain lens for a reason. Using the prime as the reference, the 28-135mm has very similar performace and I found them very hard to tell at f/8. I was fairly suprised with its performance at f/5.6, reading many reviews that it isn't much of a performer at the wider apertures. From what I can tell, its still very usable!

The kit lens compared with the 28-135mm, its a step lower in performance. Make that a significant step lower. Doesn't seem to be as contrasty, noticeable chromatic aberrations, and definetly not as sharp as the 28-135mm. Stepping the aperture to f1.8 seemed to reduce some of these. It's more sharper (but still not as sharp as the others), CA is less noticeable.

I'll post a few more comparisons up later. All in all, I'm very happy that i bought the 28-135mm to replace the 300d kit lens.

[adam]

Mr fifty-em-em is one of the reasons why I want to get primes instead of zooms.
Then I take into account my shooting-style, and the fact that I don't usually print at full size or zoom right in, but I still want haha

How about some corner crops! *pixel peeps*

[Irma]

Thanks Lak for your pictures and comment...

I have a 50mm, and I am very happy with the pictures I have taken till now... I have still very much to learn about how to use it, but I find it very useful...

Congrats for your new lens!!

[Lak]

Added in the centre edges, sorry adam, didn't really have anything in focus on the corner edges. Maybe when I have time (and when my parent's come back from holiday ) for another shoot, I'll get to some corner shots, for those pixel peepers out there.

Thanks Irma for you comments. Yeah the 50mm is really a gem for its price. I try to use it where possible! I definately love this lens for the low light shoots.

[slejhamer]

No real surprises, but a useful comparison anyway. Thanks for doing it.

[Kombisaurus]

Very interesting little comparison Lak...
Even more interesting because I just got the EF-S 17-85 IS, which is generally considered to be the "1.6 crop factor equivalent" of the 28-135 you have.

I've been curious about comparing the images of the 17-85 to the 28-135, as I expect the good and bad points to be a little different for each lens. It seems a lot of the people who's opinions I've read in various places (not here) think that just because the 28-135 displays a certain characteristic then the 17-85 will have the same one. I would be very suprised if this were true. For example, I'd be suprised if the 28-135 has anywhere near as much barrel distortion as the 17-85 when used at its shortest focal length.

Anyway, seeing your comparison with the 18-55 kit lens and the 50mm f/1.8 prime prompted me to do my own little comparison... seeing as I have both of these lenses to compare against too. I was a bit suprised at how soft your kit lens was at f/5.6.. so I wanted to replicate your test as closely as I could to see the difference.

I know that you would have used "version I" of the kit lens while I used "version II", but supposedly they are optically identical. I also used a 350D instead of a 300D, but I don't expect that to have much impact on the test (which isn't that scientific to begin with).

The first results here are trying to replicate your test.. high contrast red/white text smack in the middle of the image at 50mm focal length and a few feet away. I even used my desk with monitor behind it ( ) but my desk is messier than yours Camera was mounted on a tripod and I used a remote shutter release (but didn't bother with mirror lockup).



Here are the results... note that I took all shots in RAW and then converted them using Adobe Camera RAW with the sharpness and noise reduction set at zero... so they have no in-camera sharpening or PC sharpening applied to them at all.

First up, 50mm prime lens at f/5.6 and then f/8.0:


Next, 17-85mm at f/5.6 and f/8.0:


And finally the 18-55 kit lens at f/5.6 and f/8.0:


As you can see, all images are a bit softer than yours (I suspect you didn't turn all sharpening off, because when sharpening was set to default levels they came out very similar to yours).... but.. yeah I think its fair to say my kit lens is just as bad as yours! And while the 50mm prime wins convincingly, the 17-85 appears closer to the 50mm prime quality than the kit lens quality perhaps.

But like Adam, I was curious about the results in the corners too, which is where the kit lens really falls apart I find.
So I set up another test with contrasty detail in the corners and everything on the same focal plane (using the bottom middle focus point to focus on some black text on the grey box), and then used the same lenses to basically do the same test as before.
Here is the overall shot:


And here is the bottom-left corner using the 50mm prime at f/5.6 and f/8:


The same crop with the 17-85 lens at f/5.6 and f/8:


And the 18-55 kit lens:


A huge difference. While the 50mm prime and 17-85 hold together nicely at the edges (the 50mm still way better), the kit lens turns to slush. Strangely, the f/8 result seems even blurrier than the f/5.6 result. I quick comparison of 100% crops of the part of the image I was focussing on shows that the problem seems to be that the kit lens simply didn't focus correctly. If this is the case, it simply highlights the better focussing of the other lenses, or perhaps that the camera struggles to focus through the edges of the kit lens (which I suspect to be the case - often the kit lens seems to only lock focus when using the centre focus point).
Here are 100% crops of the area where the camera was (supposed to be) focussing on:

50mm prime at f/5.6 and f/8:


17-85 at f/5.6 and f/8:


18-55 kit lens at f/5.6 and f/8:


Sorry Lak, I didn't mean to hiijack this thread.. but I thought your results said as much about the kit lens and 50mm f/1.8 as they did about your 28-135... and I also thought my 17-85 was a very relevent comparison to the 18-135 because it is marketed as a "digital equivalent" to your lens.

Either way, you must be very happy with your lens as it obviously performs like it should... I saw it advertised in the Quokka, and know you got it at an absolute steal.

ps: I notice you just included edge crops.. yaay.

[Lak]

wow! great write up! No worries Kombi, I thought it was very relevent and supported my comparison on it. Yeah definately no real suprises slej, I just had to do the test for myself, just to give me peace of mind that i didn't get a dud lens and to shrug off those negative reviews on the 28-135mm.

[slejhamer]

I just had to do the test for myself, just to give me peace of mind that i didn't get a dud lens and to shrug off those negative reviews on the 28-135mm.

LOL; I always test my lenses this way too. We 'R' Geeks!

Those negative reviews of the 28-135 are usually not well founded. You did good work! I really liked that lens; it just didn't suit my needs anymore.

[slejhamer]

I've been curious about comparing the images of the 17-85 to the 28-135, as I expect the good and bad points to be a little different for each lens.

Nice work Kombi.

The main difference I would expect is that the 28-135 should have better performance in the corners when used on a 1.6x camera (at comparable focal lengths.) This is because you are only using the center "sweet spot" of the lens, where it naturally performs better (check the MTF chart for any lens), whereas with the 17-85 you are using the full glass.

This may or may not be true; I don't have the lenses to compare, but it's simply what I would expect.

:promises no more geekiness for the rest of the day:

[StudioJ]

Great idea guys, we should get together and do even more lenses

[Lak]

LOL; I always test my lenses this way too. We 'R' Geeks!

Haha too true!

Great idea guys, we should get together and do even more lenses

Yeah! I would love to do a comparison at 135mm, with the Canon 135mm f/2.0 L, Canon 70-200mm f/2.8L, along with the 28-135mm! Well when adam get's his 135mm and when my 70-200mm arrives, maybe we can do a joint comparison adam! Or if whoever has a sigma 70-200mm f/2.8. I would love to do my own comparison with the canon 70-200mm.

*lak, who wonders if L series is what its cracked up to be*

[slejhamer]

Yeah! I would love to do a comparison at 135mm, with the Canon 135mm f/2.0 L ...

I warn you: Do NOT compare anything to the 135mm f/2L. You will hate your other lenses after using that one, especially zooms. It's that good.
:o

[Lak]

whoa damm didn't realise it was that good! I think I might hafta reconsider lenses! Then again, I really do need the versatility of the zoom. Slej do you have one?? Was wondering if you could review it for us shuttertalkers?

[slejhamer]

whoa damm didn't realise it was that good! I think I might hafta reconsider lenses! Then again, I really do need the versatility of the zoom. Slej do you have one?? Was wondering if you could review it for us shuttertalkers?

Alas, no. Way out of my price league. But I've played with one briefly.

The speed ... the sharpness ... the bokeh ...

:shivers:


Most people would consider the 70-200 f/2.8 lens to be "pretty darned good." Here is a comparison that makes that zoom look "okay" :

http://www.outbackphoto.com/the_bag/uwe_...essay.html

We find that the 135mm f/2 lens delivers more detail and also more contrast. Interesting how much people discuss the softness of the images from the cameras and leave the lenses out of the equation.

For us the quality difference is reason enough to use in the future the 135mm f/2 as much as the scene allows.

[shuttertalk]

Haha, enter the era of collaborative Shuttertalk lens reviews! Awesome stuff guys!

Wow, these shots really show the weakness of the kit lens when compared with real glass.

[EnglishBob]

I knew my 50mm was sharper than my kit lens... didn't realise the difference was this significant! Nice work.

[Lak]

Yeah same, it was an apples and oranges comparison. I just didn't realise that the kit lens was that much of a rotten apple.

[adam]

gahahhaa, and here I sit, still waiting until december.
135 weeee

[Kombisaurus]

We could start the shuttertalk measurebators club!

Seriously though, I think a couple of these lens comparisons could be really useful. I know when I was shopping around for lenses that there are soooooo many subjective and biased opinions floating around the 'net for and against just about any lens, that any objective test like this I could find was invaluable.
Sometimes the objective tests disagreed with each other, sometimes they went against the general opinion, and they often couldn't determine particular qualitative characteristics of a lens (such as the appeal of its bokeh), but taken in context and with the understanding that some copies of lenses are better than others, I think they great.

And I also found that actually performing the tests myself helps me learn more about my lenses and their strong and weak points, and essentially gives me a deeper understanding of photography and helps develop that seamless 2nd-nature approach to using my gear which only comes from practice.

I guess it is all a matter of balance though... as soon as we become more worried about gear than about our photos is when we become measurebators. One of the great things about shuttertalk is the focus on photographs, which at the end of the day is the only thing that matters to me.

But I'm happy to perform some tests and contribute. I'd actually suggest picking some focal lengths to try with and then getting some reproducable scenes and setups so we can post our results.... in much the same way as I tried to replicate Lak's experiment as best I could @ 50mm.
Perhaps something like 18mm, 35mm, 50mm, 80mm, 135mm and 200mm might cover a lot of combinations? I'm just trying to think of good places where common zooms overlap and might coincide with a popular prime focal length.

What do you guys think?

[adam]

A measurebation section/thread where we can run standardised testing on the lenses so that they can be compared. Yes, sounds good

With so many lenses and differing opinions out there, measurebation and pixel-peeping would be one way to settle the sharpness/softness/bokeh/colour/pictures using the lens decision, which is hard to make based on the name of the lens - walk into the shop and have so many choices.

I think comparisons and reviews useful, so much that I could even buy / do buy some things based on those :o