24mm lens comparison: Canon FD SSC, Nikkor, Tamron

[cooltouch]

I have three manual focus 24mm prime lenses and three more zooms that have 24mm within their focal range, two of which are AF. For this test, I'm comparing the three primes and, as a benchmark, I'm comparing them to my 18-55mm IS kit lens. They are a Canon FD 24mm f/2.8 SSC, a Nikkor AIs 24mm f/2.8, and a Tamron 24mm f/2.5.

The camera is a 10.1mp EOS XS (1000D). To test the Canon FD 24mm, it requires that I use an FD-EOS adapter. Given this restriction and so that all lenses have a level playing field, I have tested all three using a Fotodiox FD-EOS adapter. Previously I have done numerous tests with the Fotodiox and I have found that, working within its limitations, that there is no perceptable image degradation that occurs from using that adapter with its corrective element. At apertures of f/3.5 and faster, however, the adapter produces a sort of ghosting flare. The Fotodiox is much better at controlling this aberration than another one I have made by Bower. But as you can see from the following photos, at f/2.8, some ghosting flare is evident. But using the Fotodiox also required that I use a glassless Nikon lens to Canon FD adapter, which I happen to own, so I could mount the Nikkor to the Fotodiox.

First thing I noticed was the Fotodiox does not behave as a 1.2x teleconverter. It behaves as a 1.25x teleconverter. So that means a 24mm lens has an effective focal length of 30.0mm. But with my camera's 1.6x APS-C sensor, this duplicates the angle of view of a 48mm lens on a 35mm camera.

In order for me to put together a test such that it is convenient to evaluate the lenses' performance, it ends up being a lot of work. I have to crop and combine 100% shots, label them, assemble them, post them, and then write about them. I hope it's worth it for you folks. For me, the results were quite interesting, and I think you might find them to be interesting as well.

About the images: the camera was mounted to a stout tripod and Live View was used, both to confirm focus and to pre-release the mirror. The images were recorded raw and converted to .jpgs. Each one is a 100% crop, unless mentioned otherwise, and I've done no post processing of any kind. Those images exhibiting the blooming flare caused by the FD-EOS adapter could have been greatly improved with a bit of Curves adjustment, but I though it better just to show them as they came out of the camera.

First is a cast iron plaque, I guess you'd call it, hanging from an exterior wall of my house. I was set up about 2.5 meters from the plaque when I took the shots. The ghosting flare I mentioned above is quite evident at f/2.8. By f/4 it is mostly gone.






Now, for the Canon EF-S 18-55 IS, which I've used as sort of a reference standard:


Next time people try to convince you that the EF-S 18-55 IS lens is a piece of crap (which happens to me fairly often), just show them this test. That should shut them up.

I have more. Continued in my next post.

[fermy]

Well, sorry to say, but this is a flawed test and you know it. All it shows is that FD-EOS converter is a piece of crap, which is hardly a revelation. Frankly the wide open IQ that you show might be (perhaps even a lot) worse than uncoated 1937 Leitz Summar with a scratch on front element that I just got. Get the NEX and you will see fundamentally different results. Does your Nikkor with glassless adapter really look like that?

[Lightshow]

Sorry, this test just doesn't show what these lenses can do, my SSC 24/2.8 is sharp and clear wide open, if you can take the glass out of the adapter you can at least shoot at a close distance without a coke bottle in the way.

[cooltouch]

Have you actually bothered to use the Fotodiox FD-EOS adapter? Have you done comparison tests to determine what sort of image degradation, if any, it is responsible for? I have. If I had found it to degrade IQ other than the flare I mentioned, why would I bother going to all the trouble of these tests?

To convert a Nikon lens to FD mount, no glass is necessary to achieve infinity focus. I mentioned "glassless" in case folks weren't aware of this.

Lend me a NEX and I'll redo the test. No I tell you what -- I'll remove the glass from the Fotodiox, which means macro only, and shoot some more photos. Best I can do with the tools I have at my disposal.

[fermy]

Michael,

we simply have/had FD 24mm/f2.8 and what you are showing is not even close to it in IQ. If it were only FD showing poor results, then one could think that your copy is not healthy, but since two other decent primes look awful, the only logical conclusion is that Fotodiox coke bottles them.

[iangreenhalgh1]

Interesting the IS version of the 18-55 is a hell of a lot better than the first version without IS, which is a piece of crap.

I have the Nikkor 2.8/24 and it's a really great lens, I use it on my EOS 450D quite often.

I'd be interested to see the same comparison without glass, what is the max focusing distance for the Canon without it?

[cooltouch]

[cooltouch]

[visualopsins]

Even flawed test does show some differences, however.

Lens comparisons fail (imo) mostly because identical focus is difficult to achieve -- these are very close, except for the first FD@f/2.8. Interesting, the degree of oof of the plant branch in each image gives an additional focus-point check. The branch is mysterious missing from Tamron@f/2.8.


Loss of contrast is evident from use of 1.2X TC.


10.1 effective megapixels, 22.2 x 14.8 mm, 3,888 x 2,592 pixels.
3888 / 22.2 = 175 pixels/mm = ~ 87.5 lpm MAXIMUM (probably closer to 44 lpm in real world use)
Is that enough resolution to show lens resolution differences?


+1 Canon 18-55 VERSION II lens has MUCH better IQ than version I.


The Vivitar 24/2.8 by Tokina was mentioned. I'm liking the Soligor 24/2.8 made by Tokina on FF Canon 5D. I don't know if it is the same (T4 mount) lens...

[iangreenhalgh1]

I had the Hoya HMC 2.8/24, which is the Tokina RMC 2.8/24 rebadged. I thought it was a very good lens, I sold it because I then found a Konica Hexanon 2.8/24 and that is a really stunning lens.

[fermy]

[Bille]

[Bille]

[kyonthinh]

Thanks for your hard work but I also think that FD is more better.

[Attila]

I had all three also, Nikon and Canon was fine really, my Tamron lens was pretty crap at infinity distance.