Olympus OM on M9

[plewislambert]

Advertisement (gone after registration)

I sometimes use a cheap Tokina f3.5 17mm on my M9 which gives better results than a 15mm Voigtlander, provided the lens is stopped down to f5.6 or f8 (which also applied if using it on an OM film camera). I also have had excellent results on my M9 with a Nikkor 28mm PC lens shifted to get in the top of a building without tilting the camera back. This lens is very sharp when stopped down, as it was on an F100 film slr. The images are not as sharp as those I get from a 28mm f2.8 Elmarit but few lenses are (and those are probably all by Leitz!)

The only lenses I have reservations about are my Cosina Voigtlander 21mm which needs Cornerfix and the 12mm CV, which does too, and has some very strange stretching effects on nearby objects like cars situated at the edge of the field of view. The need for Cornerfix relates to the close proximity of the rear element to the sensor (quite unlike slr lenses); the stretching of images might be caused by the perspective of such a short focal length. I have never had another make of 12mm lens to compare.

Is there an expert in the house?

Philip:)

[jdlaing]

Oh I see, I'm so sorry, you were demanding a personal answer. Its a shame that in those fifty years you didn't learn any manners at the same time as learning about zone focusing. I was only trying to help. Over to the OP.....

 

Steve

 

I didn't demand anything. If you'd had read the question you'd have understood that. I asked a really simple question of the OP. You saw fit to stick your nose in and then tell me I have no manners.

 

To the OP: Thank you for your answer.

[250swb]

You saw fit to stick your nose in and then tell me I have no manners.

 

.

 

Well perhaps you do have manners, in the style of the DPR forums perhaps? I mistakenly answered the wrong question, but didn't give a wrong answer, and what was it you said "Now give the OP a chance to answer my question unless you are his official spokesperson." Charming. I doubt you'd say that in real life to somebody.

 

Steve

[jdlaing]

Well perhaps you do have manners, in the style of the DPR forums perhaps? I mistakenly answered the wrong question, but didn't give a wrong answer, and what was it you said "Now give the OP a chance to answer my question unless you are his official spokesperson." Charming. I doubt you'd say that in real life to somebody.

 

Steve

 

Try me.

[sinjun]

A good comparison, and although it has already been naively dismissed earlier in this thread this is an example of smearing caused by the micro lenses of the sensor.

 

Just because we mostly hear about RF lenses doing it on other bodies that are in the news at the moment, it does not mean it can't happen with non-RF lenses on a RF body. The micro prisms on the M9 sensor are designed to point at the light coming from the rear element of a RF lens. Add an SLR lens with an adapter and with the increased distance the prisms, and especially the corner prisms, are no longer gathering light from the direction they were designed for. And as usual it is the wide angle lenses that are worst affected. It is easy to draw out on a piece of paper, but if the prisms are at an acute angle to deal with wide angle RF lenses, as the alternative SLR lens is moved further away from the sensor that angle increases and the prisms stop working properly.

 

Another way to think of it is this. The lens to sensor distance of an m43 body is small, and the prisms are designed to match the angles of light coming in at a very acute angle. But put a wide angle RF lens on the body with an adapter and everybody acknowledges the corner smearing is down to the angle of the prisms not dealing with a lens that is further away. It is a well known phenomenon. So scale this up, the m43 camera is now the M9 body, still a relatively small lens to sensor distance compared with a DSLR/SLR lens. The same thing happens with an SLR lens, the prisms are no longer at the optimum angle when a lens from a camera system with a greater lens/film/sensor distance is used. It is exactly the same effect as we see with a wide angle RF lens causing smearing on the A7R.

 

Steve

 

Thanks Steve. I thought that the sensor sites could always gather light falling at a large angle to the plane of the sensor, and that it was just that falling too obliquely which caused a problem. But if, as you imply, they all have a limited cone of angles then your answer makes sense.

 

However, it seems to me that this reasoning would imply that an SLR lens of any other focal length would cause problems, e.g. 50 mm or 90 mm, and that does not seem to be the case.

[sinjun]

I sometimes use a cheap Tokina f3.5 17mm on my M9 which gives better results than a 15mm Voigtlander, provided the lens is stopped down to f5.6 or f8 (which also applied if using it on an OM film camera). I also have had excellent results on my M9 with a Nikkor 28mm PC lens shifted to get in the top of a building without tilting the camera back. This lens is very sharp when stopped down, as it was on an F100 film slr. The images are not as sharp as those I get from a 28mm f2.8 Elmarit but few lenses are (and those are probably all by Leitz!)

The only lenses I have reservations about are my Cosina Voigtlander 21mm which needs Cornerfix and the 12mm CV, which does too, and has some very strange stretching effects on nearby objects like cars situated at the edge of the field of view. The need for Cornerfix relates to the close proximity of the rear element to the sensor (quite unlike slr lenses); the stretching of images might be caused by the perspective of such a short focal length. I have never had another make of 12mm lens to compare.

Is there an expert in the house?

Philip:)

 

Interesting about your Tokina - implies my OM ought to be OK.

 

Regarding stretching at the edge of your 12 mm CV, I think this would happen with all rectilinear lenses of that focal length - if you want to preserve straight lines, you've got to distort in other ways.

[01af]

Advertisement (gone after registration)

... here are two pictures from the Zuiko 24 mm 1:2.8 and a Voigtländer Snapshot Skopar 25 mm 1:4, both taken at f/5.6.

... this is an example of smearing caused by the micro lenses of the sensor.

No, it most definitely isn't.

 

This Zuiko 24 mm lens clearly is damaged. Either an element has come loose inside, maybe due to a strike or sustained vibration. Or the lens has been disassembled, and then re-assembled with one element missing or flipped upside-down.

[LucisPictor]

Interested that you have used them with success on a DSLR. Can you think of any reason why they should come out differently on the M?

 

Well, the sensor in the 5Dclassic has 12MPix and a different design.

Then in a DSLR you havea mirror box etc.

[250swb]

 

However, it seems to me that this reasoning would imply that an SLR lens of any other focal length would cause problems, e.g. 50 mm or 90 mm, and that does not seem to be the case.

 

And that is the case with nearly every other mixed make combination where the wide angle lenses smear, anything longer than 35mm tends to work perfectly. The issue of a Tokina 17mm working is possibly the same issue as found in the long discussions of the Sony A7R. It works fine with the Leica WATE at each focal length, but not a Leica 21mm or indeed anything up to 35mm. It is down to the design of the lens, not the focal length but the general rule applies, more often than not its the wides at fault because many are designed in a similar way.

 

By the way, have you dropped it or had it repaired and re-assembled, just to clear things up?

 

Steve

[sinjun]

No, it most definitely isn't.

 

This Zuiko 24 mm lens clearly is damaged. Either an element has come loose inside, maybe due to a strike or sustained vibration. Or the lens has been disassembled, and then re-assembled with one element missing or flipped upside-down.

 

And that is the case with nearly every other mixed make combination where the wide angle lenses smear, anything longer than 35mm tends to work perfectly. The issue of a Tokina 17mm working is possibly the same issue as found in the long discussions of the Sony A7R. It works fine with the Leica WATE at each focal length, but not a Leica 21mm or indeed anything up to 35mm. It is down to the design of the lens, not the focal length but the general rule applies, more often than not its the wides at fault because many are designed in a similar way.

 

By the way, have you dropped it or had it repaired and re-assembled, just to clear things up?

 

Steve

 

I always looked after then lens well, but it has been lying around unused for about 15 years. I've just gone and checked it. When I shake it hard there is a bit of a rattle, suggesting perhaps a loose element. I think that must be the answer to the problem.

 

I still don't understand in Steve's explanation why light from an SLR wide-angle lens arriving at a large angle of incidence would be any more likely to cause smearing than light from an SLR 50 mm lens arriving at a similar angle.

[250swb]

All auto diaphragm (aperture) SLR lenses rattle, it is the mechanism. Conversely you shouldn't hear a rattle from an RF lens because it doesn't have an auto diaphragm.

 

As the focal length of the lens increases it becomes more telecentric, so the light hits the sensor at a squarer angle. Think of it like point for point, a single light ray hits a single pixel square rather than a wider angle light ray skimming one or two pixels in the corners (obviously more than one light ray will hit a pixel, this is just for illustrative purposes). The design of the sensor is a compromise based on getting the best performance from the lenses designed for it, this isn't so with film which is why smearing (as opposed to a poor lens design) has never been a factor in lens or sensor design before.

 

Steve

[01af]

I still don't understand in Steve's explanation ...

Don't waste too much time or effort trying to understand those half-baked "explanations"

 

Here's a few sample shots I took today with my Olympus OM Zuiko MC Auto-W 24 mm 1:2 lens via Novoflex LEM/OM adapter, at f/2.8 and f/5.6, on the Leica M9 and M (Typ 240) cameras. Lens detection was off; no lens profiles manually selected. No red-corner correction or vignetting elimination in post-processing. The M9 has shifted micro-lenses, the M has them straight. According to Steve's theory, this should make an overwhelming difference ... but in fact, there is none, not even at 100 % view on the original files.

 

At f/2.8, the tiny twigs in the frame's center are sharp down to the pixel level but the corners leave a lot to be desired, with both cameras. At f/5.6, the corners sharpen up quite nicely, again with both cameras—still not perfect but fairly good for a 35-year-old wide-angle lens which is no Leitz, Schneider, or Zeiss. If sensor smearing was the cause for the poor corners at f/2.8 then the corners (1) would look different with the two cameras and (2) would remain poor at any aperture.

 

Bottom line—on the M9 as well as on the M (Typ 240), sensor smearing is no issue with retrofocus wide-angle lenses originally made for 35-mm-format SLR cameras (I also tried 21 mm, 17 mm, and fish-eye lenses—no issues at all). Furthermore, the Zuiko 24 mm 1:2 is a decent wide-angle lens but cannot hold the water to a modern Elmar-M or Elmarit-M 24 mm Asph. Finally, the Zuiko 24 mm 1:2.8 should perform very much like the 24 mm 1:2 ... if it doesn't, as in sinjun's sample shot, then there's something terribly wrong with it.

 

 

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

 

 

 

 

 

[sinjun]

 

Bottom line—on the M9 as well as on the M (Typ 240), sensor smearing is no issue with retrofocus wide-angle lenses originally made for 35-mm-format SLR cameras (I also tried 21 mm, 17 mm, and fish-eye lenses—no issues at all). Furthermore, the Zuiko 24 mm 1:2 is a decent wide-angle lens but cannot hold the water to a modern Elmar-M or Elmarit-M 24 mm Asph. Finally, the Zuiko 24 mm 1:2.8 should perform very much like the 24 mm 1:2 ... if it doesn't, as in sinjun's sample shot, then there's something terribly wrong with it.

 

 

Thanks for going to the trouble 01af - for me this is an interesting test. I hope you did not put your life in too much danger setting a tripod up on that road.

 

I'd love an Elmar but can't complain about my CV Snapshot Skopar which does a very good job without much colour shift using one of the profiles. I'm still interested in an SLR wide angle for closeups. As a matter of interest, what is the close focus on the Zuiko 24mm f/2?

[01af]

I'm still interested in an SLR wide-angle for closeups. As a matter of interest, what is the close focus on the Zuiko 24 mm 1:2?

Same as on the Zuiko 24 mm 1:2.8—that is, 0.25 m. But as mentioned before, it has floating elements which improve sharpness at close range (the 24/2.8 hasn't).

 

If I remember correctly then there was a manual-focus Sigma 24 mm 1:2.8 lens made for all kinds of SLR lens mounts, including Olympus OM, which had a good reputation and an extremely short minimum focus distance ... 0.18 m or thereabouts. Maybe you can find one on eBay some day. You can also always augment your existing lens with a close-up diopter lens. For a wide-angle, you'd need the stronger ones ... at least +3 dpt, or better yet, +4 or +5 dpt. Achromatic double-element close-up lenses are visibly better (and more costly) than simple single-element close-up lenses.

[250swb]

Don't waste too much time or effort trying to understand those half-baked "explanations"

 

 

At f/2.8, the tiny twigs in the frame's center are sharp down to the pixel level but the corners leave a lot to be desired, with both cameras. At f/5.6, the corners sharpen up quite nicely, again with both cameras—still not perfect but fairly good for a 35-year-old wide-angle lens which is no Leitz, Schneider, or Zeiss. If sensor smearing was the cause for the poor corners at f/2.8 then the corners (1) would look different with the two cameras and (2) would remain poor at any aperture.

 

You are as ever making wide generalisations and presenting them as facts. Your images show corner smearing, it's there, produced by your 'perfect' lens, and you said it couldn't happen!

 

But your 'test' photographs are in no way a comparison with the OP's examples, the focus point is further away and there are no nearby details (branches) to exaggerate the smearing. So it's funny that somebody who thinks of themselves as the Einstein of the photography world should sidestep such critical details. Perhaps 'proving' a theory was more important than truth? Branches yards away up in the canopy and focusing on the middle distance are not going to exaggerate smearing like the OP encountered in his example, but I guess you knew that.

 

Steve

[tgm]

 

By the way, here are two pictures from the Zuiko 24 mm f/2.8 and a 25 mm CV Snapshot Skopar f/4, both taken at f/5.6.

 

[ATTACH][ATTACH]410683[/ATTACH][/ATTACH]

 

The smearing is indeed extrem. Is the adapter planar? If not, the focal plane is tilted with respect to the sensor. I would check if you get similar smearing with a telephoto lens.

 

Thomas