Canon EF70-200 f2.8 L vs Nikkor 80-200 f2.8 ED

I was getting so tired about hearing of continuous rants between canon vs nikon lenses and how inferior canon lenses were that I decided to do my own tests. (Yes, I've read most of those lens tests, from those by pop photo to Klaus' postings to those from Photodo.com). But it's always better to convince yourself right? Working as a journalist, I often hear pros make sweeping statements that Nikon lenses are better than Canon's. Nikon has better contrast and sharpness, they say. As EOS users, I'm sure you must have sometimes wondered if such statements are true.

Anyway, I have a good friend, a pro who uses the Nikon system so we paired up and decided to have a lens test. Our first test was between the new Tokina 20-35 f/2.8 vs Canon 17-35L f/2.8, since he was planning to buy the tokina for his F5. I won't go into the details of that test now, but suffice to say, my friend won't be buying the Tokina after comparing the results with the 17-35L. Wide open, the Tokina is no good while Canon at f/2.8 (at all focal lengths, including that "dreaded" 17mm) can still rival the Tokina stopped down to f/8. No flaming pls, these were both our observations.

We then ran another test on his new Nikon 80-200D f/2.8 ED (with tripod collar) vs my one year old EF 70-200L f/2.8. Setup is as follows:- Cameras EOS1n and F5 triggered with cable release, using mirror lock. Fuji Velvia rated at ISO 40 (we agreed that's truer to its real speed). Exposure taken using a Minolta autometer IVF ambient light setting. Cameras mounted on the heavy Manfrotto 055C tripod with 029Mk2 pan and tilt head. Test target was a "home made" computer designed "high resolution" chart with words ranging from large fonts to fonts only 2mm large. The target measured approx 24 inches x 16 inches. It also had grids drawn by computer on it to check for distortion, and a colour chart with various colours and patterns including gradations plus a few items glued on like bar codes and some pictures to check for color reproduction. Target mounted on a flat wall. All lense protection filters were removed.

Cameras precisely framed with the targets almost completely filling up the viewfinder. Composition was reconfirmed by both of us to double check it was properly aligned. Lenses tested at: (70/80mm at f/2.8, f/4, f/8), (135mm at f/2.8, f/4, f/8), (200mm at f/2.8, f/4, f/8). Results were examined with a schneider 4x loupe and two 8x loupes on a light box. We tried to be as neutral as possible. and remember, I was cross examining the results with a Nikon die hard.

At 70/80mm setting:
Wide open, very little but noticeable light falloff and vignetting at the edges of pix taken by both lenses but was slightly more pronounced on the Nikon. Both lenses showed very light barrel distortion. By f/4, Canon lost all traces of vignetting and had very even illumination,
while a tad of vignetting could be spotted on the Nikon's pix. Sharpness was really neck and neck, both are so damn good from centre to edges even wide open, but we both agreed that Canon had the slight edge in terms of sharpness and slightly higher contrast. (do I hear cheers already? :) Stopping down to f/4 for both lenses improved sharpness marginally. I'd confidently use both lenses wide open, but if I wanted it that extra bit sharper, closing down one stop would do the trick. At f/8, frankly, we couldn't discern any improvements in sharpness...but perhaps psychologically, we thought it was ever so very very slightly sharper.

At 135mm setting:
Wide open, Canon had more severe vignetting and light fall off at edges compared to Nikon, but again, it was only marginal and still within reason. Nikon however, had more pronounced pincushion distortion than Canon. Interesting to note, the distortion of both lenses have now turned from barrel to pincushion. Even so, distortion on both lenses was not disturbing.
If it were not for the grid lines, it would hardly be noticeable. Again, we agreed that canon's shots looked marginally sharper as it had very slightly more contrast. Both lenses had good sharpness from centre to corners.

Stopping down to f/4, Canon still had a hint of light fall off at the edges while Nikon had good even illumination across the frame. As expected, stopping down one stop improved sharpness quite a bit, although wide open, both lenses were already very good. At the centre, there was no difference in lens sharpness between both lenses but Canon had the advantage at the frame's edges...a very small advantage I must point out. By f/8 illumination on both lenses were on par. Absolutely no vignetting detected. Edge definition of both lenses improved marginally. Frankly, both are equally good here, stopped down to F/8.At this setting, centre to edge sharpness was outstanding for both lenses.

At 200mm setting:
Good news, this is where the Canon shines above the Nikon. The Nikon at 200 displays obvious pincushion distortion and while the Canon also has some pincushion, it is only very minor. Distortion for the Nikon is at its most severe at this tested focal length but still within reason. Wide open, the Nikon has pretty heavy vignetting and light fall off which crept almost to the centre of the frame compared to the Canon which fared quite a bit better. By f/4, Canon had very even illumination, no vignetting while the Nikon still had some. By f/8, traces or vignetting or fall off for both lenses were virtually gone. In comparison however, Canon still provided more even illumination across the frame at all settings even up to f/8.

In terms of sharpness and contrast, again it was so very close, but my pro-Nikon friend grudgingly admitted that the Canon still had the slight advantage as the image looked slightly more crisp especially at bigger apertures. Wide open, both lenses were already very good, but edge sharpness of both improved quite a bit when stopped down to f/4. By f/8- there was no difference in both lenses.

General observations:
Both lenses are very closely matched in terms of optical and mechanical performance. Both lenses focus very fast on their respective bodies, although the Nikon seems to have a silkier and smoother zooming feel. However, without USM, the Nikon "jumps" everytime you focus. I don't like that. I think in focus tracking, that jump may actually cause some shake. However, the Nikon F5 beats the Eos1n hands down in terms of AF sensitivity. On many occasions, the F5 could lock on a man's patch of black hair without hesitation while the 1n hunted. Perhaps this is where the Nikon's EV-1 focusing sensitivity has it's edge.

AF speed on both were comparable...heck, if there's any difference, you need instruments to measure it, but I noticed the canon with its USM lens could focus marginally faster in bright light and could drive it's lens from nearest focus to infinity faster than the F5.

It is also interesting to note that the black edges separating each frame of exposure on the emulsion was sharper and more defined in those shot by the Nikon. Each exposure had a solid defined black square around it while that of the Canon was soft and not so well defined. I wonder if this means the Nikon pressure plate holds the film more securely compared to the Canon? Both lenses's colour rendition is also very neutral, but I think the Canon is a little warmer.

All in all, I must admit that this lens test echoed the results of the MTF lens test posted in www.photodo.com/ To those who feel that Nikon is always better than Canon, I'd say...use your Canon L lens with confidence and pride. As far as telephoto zooms are concerned, I'd go for the Canon 70-200L any day - less distortion, better illumination, slightly sharper in the edges and slightly better contrast than our rival's equivalent lens. Add the smooth USM focusing and ability to use converters and we have a lens that is very hard to equal and surpass. Both lenses managed to reproduce even the finest resolution targets sufficiently for us to read while all gradations and colours were accurately rendered. Amazing!

Sorry for the long posting. If this is too long, just tell me, I won't post such results again. BTW...will be doing more tests on my 28-70L and perhaps the 300f/2.8L. and also more tests with the 70-200 attached with 2x converter. Anyone interested to know the results? I know this test is subjective, since numbers are not attached. Go check out the tests for these lenses at photodo. I find their's quite accurate at least for these two lenses I've tested. Now both my friend and I need to go for a break, our eyes are tired after countless hours of peering through loupes. So, my EOS comrades....you have good reason to hold your head up high with your white piece of Luxury USM glass. :)

A practical EOS user,
Jonathan Kwok.