[Edgar_in_Indy]

I've decided to go to a dual Pentax / m43 system, but since I'm not made of money, I've had to make some difficult decisions regarding my K-mount lenses. I sadly sold my Sigma 8-16mm a few days ago , and I'm trying to decide between the Olympus 9-18mm or the Panasonic 7-14mm lens to replace it.

In the meantime, I found out about the Olympus 9mm fisheye "body cap lens", and figured that for the price it was worth a try to hold me over until I re-acquire a proper UW lens. So how does an $88 body cap compare to the widest rectilinear lens available for Pentax cameras? For the price, better than I thought!

Here's a couple pictures of the lens...er, body cap.

And here's a couple pictures I took within minutes of getting the lens, as I was walking to meet my kids at the bus stop. The school bus picture had the top and bottom cropped, but it has not been resized; and the car picture is not cropped or resized. So if you click on the pictures, you can download the 100% picture to pixel peep it.

When I got home and loaded the picture onto my computer, the thing that amazed me is how sharp they were throughout most of the center part of the frame.

So the lens is usable, but how does it compare to the mighty Sigma 8-16mm? Well, in overall image quality, the Sigma definitely wins. The Sigma is sharp edge-to-edge, and has better contrast than the little fisheye. When shooting in RAW, you will probably want to add some zing to the fisheye pictures.

Here's an example from my Sigma 8-16mm showing how ridiculously sharp it can be across almost the entire frame:

So if you're wanting to shoot an ultra-wide angle just for fun, and/or you won't be printing large images, then the Olympus lens can provide a lot of utility. But if you're doing critical work, then you'll definitely want to use a real lens.

The other thing I was curious about is the comparative angle of view of the two lenses. I was under the impression that the fisheye would be a little wider, since the 8-16mm is supposed to have a 121 degree angle of view on APS-C, while the fisheye supposedly has 140 degree angle of view. But in practice, it seemed like the lenses were very evenly matched. In fact, it seemed like the 8-16mm captured a bit more at the left and right edges, but I don't know if that's really a fair comparison, since the APS-C sensor is wider side-to-side than the 4/3 sensor.

The big difference is the way the two lenses capture the wide angle. The Olympus fisheye shows a lot of barrel distortion in the middle, and curves the edges of the frame, while the Sigma gives you straight edges, but gives you kind of a pinched view, with objects at the edge stretched and objects in the center of the frame shrunk.

I've never used a fisheye lens before, but my understanding is that the Olympus is not quite a true fisheye, since it does not provide a 180 angle of view. As far as I'm concerned, this mild-fisheye makes the lens easier to use in more situations since it is not too exaggerated. Depending on the scene, it may not even be obvious that the lens is a fisheye.

Here's a couple comparison pictures, and again, you can click for a full-size image. (Sorry about the uninspiring and/or depressing photography, but I only had a few minutes to do a head-to-head comparison with the two lenses since I had to get the Sigma 8-16mm to the Post Office before it closed to ship it to the new owner.)

I'm not sure that I prefer one approach to the other. The fisheye is cool with some scenes, but in other cases it just looks strange. And kind of the same thing with the 8-16mm.

One thing I do appreciate from the fisheye is that the proportions of a scene are better preserved. These next two images demonstrate that.

In the bedroom picture, the fisheye picture does a pretty good job of representing the actual size of the room. But in the Sigma version, the room looks larger than it is, since the bed is shrunk and the floorspace in front of the bed is stretch. And in the Sigma picture, the little mirror in the corner looks huge.

I think that either picture looks okay, but if these pictures were used for real estate purposes, the Sigma picture could be considered a bit misleading since it seems to add square footage to the room. (BTW, that is something I have noticed with some real estate pictures, and it always annoys me.)

For the second picture in my home theater area, both lenses were kind of a fail. Again, the Olympus picture does a good job of showing the relative size of the speakers to the screen, etc, but it just looks bizarre seeing the screen and speakers curved like that.

And the Sigma does a good job of keep the lines straight, but the screen looks tiny and the center speaker look tiny, and the L/R speakers look huge. And since they are canted a bit to aim towards the center of the seating area, the rectilinear lens makes them look oblong.

And for the sake of comparison, here is a more normal shot showing pretty much how the home theater area appears in person. This picture was taken with my Olympus XZ-1 at 30mm equivalent:

So for everybody who has been dying for a comparison of those two lenses, there you go!

I'll finish up with a few pics comparing the physical dimensions of the Pentax K-30 & Sigma 8-16mm combo to the Panasonic GX7 & Olympus 9mm fisheye combo. (These pictures were courtesy of my old Olympus XZ-1 since that's the only other camera I have with a hotshoe!)

It still blows me away how little this thing is! There's been many times when I wanted to take my 8-16mm along, but I just couldn't bring myself bring add a 3rd heavy lens to my normal two-lens kit. Now I'll just be able to slip this little guy into a pocket or something, and forget I even have it with me until I want to use it.

---

Last edited by Edgar_in_Indy; 12-21-2015 at 12:12 PM.

[From1980]

Your last photo says a lot. I don`t have any experience with m43 and didn`t know of that tiny body cap lens. The Lumix beside the K30 is a difference like K30 to Pentax 67. On the other hand I had a similar decision back when I acquired the SMC 15mm while I had the SMC 17 fisheye for my 35mm Pentax cameras. 2mm difference in focal length but a world of difference in perspectives. Curvilinear vs Rectilinear. I kept them both and still have them.

[Edgar_in_Indy]

Your last photo says a lot. I don`t have any experience with m43 and didn`t know of that tiny body cap lens. The Lumix beside the K30 is a difference like K30 to Pentax 67.

The funny thing is, the K-30 is one of the more compact DSLR's, and the GX7 isn't exactly small for an m43 camera. This body cap lens on one of the really tiny bodies would be kind of humorous.

On the other hand I had a similar decision back when I acquired the SMC 15mm while I had the SMC 17 fisheye for my 35mm Pentax cameras. 2mm difference in focal length but a world of difference in perspectives. Curvilinear vs Rectilinear. I kept them both and still have them.

Yeah, I think you're right. Ideally it would be nice to have both options.

The picture of the stadium, for example, would have been completely different with a fisheye. The shape of the floor area would have been distorted, and the rafters and scaffolding in the ceiling would have obvious bowing. It would be nice to have the ability to choose which view you prefer for the project at hand.

[Rondec]

I definitely prefer the rectilinear projection. I owned the Pentax 10-17 which is a very nice lens and capable of 180 degrees of view, but I didn't enjoy shooting with it because of the fishiness of the output. At the same time, I don't really like ultra wides either because of the way they distort things. The stadium shot is cool and wouldn't be possible with a more narrow view, but in general the wider you go the more you make everything close up look big and everything far away look tiny.

[Edgar_in_Indy]

The stadium shot is cool and wouldn't be possible with a more narrow view, but in general the wider you go the more you make everything close up look big and everything far away look tiny.

That's the funny thing. The 8-16mm is considered rather extreme, and is in fact a good bit wider than all the other competing UWA zoom lenses, but I've had several occasions where 8mm was just wide enough to capture the entire scene/subject the way I wanted to in a single shot. I loved my old Sigma 10-20mm, but in each of those situations I was glad I made the decision to trade it for the 8-16mm.

[Rondec]

That's the funny thing. The 8-16mm is considered rather extreme, and is in fact a good bit wider than all the other competing UWA zoom lenses, but I've had several occasions where 8mm was just wide enough to capture the entire scene/subject the way I wanted to in a single shot. I loved my old Sigma 10-20mm, but in each of those situations I was glad I made the decision to trade it for the 8-16mm.

It's just the nature of the beast. Rectilinear lenses aren't as wide in terms of angle of view as fisheyes, due to their method of projection. From what I've seen the 8-16 is the best of the ultra wide lenses for APS-C.

[K-Three]

Interesting comparisons,very nice compact package in the Lumix/Oly set-up
For what it's worth,
Fish-eye only means that the lens has pronounced, uncorrected barrel distortion, in theory you could have a fish-eye at any focal length.
I have one of the old Pentax 17-28mm FEs. On a FF camera it is pretty fishy, but on APSc at 28, it just looks like something is off, nearly normal perspective & angle of view, but some odd curvature running through it. You can minimize the curves by carefully leveling the camera and aligning the important straight lines through the center vertically or horizontally, fun to play with.

If you took your Olympus FE photos and "de-fished" them in Photoshop or other editor, you would be nearly back to your Sigma field of view and perspective.

[Edgar_in_Indy]

If you took your Olympus FE photos and "de-fished" them in Photoshop or other editor, you would be nearly back to your Sigma field of view and perspective.

I actually already did something like that last night.

First, an uncorrected fisheye image of the bedroom:




And here I just used a high degree of manual barrel correction in Photoshop CS6 (nothing specifically tailored for this lens...just eyeballing it):




And finally, here is the corrected image with the blank areas cropped out:





As you said, I end up with an image that has most (but not all) of the original angle of view, but looks more natural. I much prefer this image to the 8-16mm image, and I've lost almost no angle compared to the 8-16mm. However, the final result does end up making the room look larger, kind of like the 8-16mm image, but to a lesser degree. And I believe there is still some fisheye distortion present when you look at the tops of the walls.


Here's the 8-16mm image again for comparison:

[Digitalis]

here is a more normal shot showing pretty much how the home theater area appears in person.

I know this has nothing to do with the lenses in question, or even photogaphy. But why do you have the Sub so close to the seating area?

[Edgar_in_Indy]

I know this has nothing to do with the lenses in question, or even photogaphy. But why do you have the Sub so close to the seating area?

Just slightly off topic. It's actually two subs (an SVS 20-39 cylindrical sub on top of a Velodyne DLS-5000R 15" sub), and that was just a convenient place to put them. As far as I know, there is not an issue with putting subs close to a seating area.

I'm now running a single Tuba HT subwoofer that's approximately the size of a washer or dryer, and it's to the left of the seating area.