Test: Tokina 300mm f6.3 Mirror vs m.Zuiko 75-300mm @ 300mm

SXR_Mark · Joined: 12 Jun 2010 Posts: 506 Location: England

This is a quick resolution comparison of the Tokina micro-4/3 300mm mirror lens against the Olympus micro-4/3 75-300mm AF zoom lens. The zoom is the first version of the lens, which has now been replaced by a Mk II version. However, the optical scheme is the same for the two versions, so the comparison should be valid for the newer version.

I used my OMD E-M5 for the tests, securely tripod mounted. I took two test shots with each lens of my standard test target, allowing me to probe performance in the centre and at the edge / corner of each lens. The first shot has the 48 lpi marks on the test chart in the centre of the frame. The second shot placed the marks towards the top left hand corner. The target was at 7.8m from the camera. The chart was in bright sunlight for all the shots and there was no change in lighting over the (short) duration of the test.

The zoom lens was set to its longest focal length and its largest aperture (F6.7). The mirror lens is of course a fixed focal length and aperture (F6.3)

For simplicity, I will use screen shots of the recorded images as displayed in Apple's Aperture to illustrate the results. The Tokina image is always on the left.

The first screen shot shows the two full pictures with 48 lpi marks in the centre.

You can see that the m.Zuiko zoom is a shorter focal length (and hence wider field of view) than the Tokina. Scaling from the imaged size of the chart, the m.Zuiko is only 94% of the focal length of the Tokina. If the Tokina is exactly 300mm, this means the m.Zuiko has a focal length of 282mm. Other tests I have done confirm the m.Zuiko is indeed "short", at least at this focus distance. (Internal focusing lenses do show a change in FL length with focus distance).

You can also see that the colour is not quite the same. This is a genuine effect of the lenses as the camera was set to a fixed colour temperature. The Tokina gives a cooler rendering. Also of note is that the Tokina shot was taken with 1/1600s shutter and the m.Zuiko 1/2000s, even though the exposure is similar and the lighting did not change. Thus, although the Tokina is f6.3 and the m.Zuiko f6.7, the m.Zuiko has the higher transmission. This is surely due to the central obstruction in the mirror lens. The Tokina is probably something like T7.5.

Now lets look at the detail of these two pictures. Below is the screen shot of the centre of the above images viewed at 100%.

We see that both lenses easily resolve the 48 lpi marks, but the m.Zuiko subjectively looks to have higher contrast over the lines. This would suggest it would have a better MTF. There is no evidence of colour fringing from either lens.

Moving to a screen shot of the bottom of the two "centre" shots, we see that m.Zuiko is sharper at the text "Printed in England", but it also exhibits a small amount of purple fringing.

The screen shot below shows the two off-set shots in full, illustrating how the 48 lpi marks are positioned towards the top left corner. The most striking difference is the classic mirror-lens bokeh. Love it or hate it!

Below is the 100% view of the top left corner of the pictures.

Both the lenses still resolve the 48 lpi marks. However, we see that the m.Zuiko shows a complex pattern of slight purple and green fringing and the 48 lpi marks are clearly not as well resolved as in the centre. The Tokina shows no such fringing (as expected) and subjectively has the higher resolution.

The last screen shot below is of the middle of the left edge. The m.Zuiko still exhibits green and purple fringing. The Tokina again looks the better in terms of resolution.

My summary and conclusions are as follows.

� Both the lenses are actually good/very good performers. I have done many similar tests with MF lenses of this focal length, and the results here are better than most and not hugely worse than the best.
� The Tokina mirror lens is actually the sharpest mirror lens that I have tested (and I have tested a lot!)
� The Tokina is also beautifully made. It looks and feels fabulous.The only thing that lets it down is that the focus is very sensitive i.e. a tiny twist of the focus ring makes a big change. This is because the lens has a very short min. focus distance (0.8m), so most of the angular range of the focus ring is used up in going below 5m. This makes the handling at longer distances a real pain.
� The m.Zuiko zoom is not the last word optically, but is is still capable of very good results, even at its (admittedly very slow) maximum aperture. Centre performance is actually very good and it is only the fringing at the edges that mar the performance.
� It is important to note that the fringing is not very big and does not increase if the lens is not accurately focussed. In the centre, fringing doesn't appear even with significant defocus. This is not always the case with MF 35mm lens. For example, I have a Nikon 100-300mm f5.6 zoom which is really excellent and has well controlled fringing when accurately focussed, but shows vivid purple or green fringing with even a small amount of defocus.
� The m.Zuiko is the only autofocus lens I have with a 300mm focal length. I bought it because it was heavily discounted (being superseded) and because I was curious to see how a modern zoom 300mm lens would stack up against the many older MF 300mm lens (zoom and prime) that I have. In general, I would say it performs very well, but does not render prime lenses like the Tair-3 photo sniper redundant, mainly because of the relatively weak edge performance and slow max aperture.
� Comparison with the previously mentioned Nikon 100-300mm lens is more difficult. The Nikon has better edge performance (as expected being a cropped 35mm lens) and is first rate when accurately focussed. But accurate focus is very critical and handling is poor as it is a large and heavy device with a one-touch mechanism and no possibility of adding a tripod collar.