Tuesday, 9 June 2009

comparing legacy 35mm adapted fifties

One of the great things about the micro 4/3rds cameras (and as of this moment there is only the G1) is that not only do they make using older manual 35mm lenses possible, they make it easy. On the 4/3rds sensor camera good old 50mm lenses become nice medium portrait lenses, with these being as cheap as chips (and often you'll have one floating around) its really quite attractive to add one to your camera.

However, some decades ago I changed over from using more or less manual cameras to the modern Autofocus cameras like the Canon EOS, for one reason or another I just don't have any manual focus gear anymore. So if you're like me you might not have any manual 50mm lenses lying around.

But never fear eBay is here, and there's a whole wealth of stuff to choose from.

So, where to start?


Well, back in the 70's and early 80's when 35mm SLR cameras like the Pentax MX or the Olympus OM-1 were the cameras to have I used Olympus. I liked the OM-1 and I personally thought that the 50mm f1.8 was quite a good standard lens. Years later I bought into the Canon EOS system and have been using it since 1989 with (at first) film bodies and (later) digital bodies.

When Canon changed from FD series to EOS they changed a few things which meant that FD lenses couldn't be used on EOS bodies without an adaptor which required a lens in it to correct for the different distance between the lens flange and the camera body mount (with the EF mount being some 2mm longer). This seemed to be a show stopper for many and FD lenses just dropped away in price with eventually the entire series becoming worth peanuts.

It wasn't just because of the change to digital either, as many people keep using Pentax and Nikon lenses on their digital bodies even though they do not have autofocus (just see eBay for some market price comparisons in the same focal length).

Being a Canon user (and hearing many glowing stories from some professional Canon users I knew pre EOS) I thought "why not pick up some bargains" for putting onto my G1.

Remember that micro 4/3rds is significantly smaller than a 35mm film frame, and so the "crop factor" means that the angle of view of a lens is essentially equal to a lens with double the focal length.

What this means is that if I want to see what 50mm lens looks like in a 35mm camera viewfinder I need to put a 25mm lens on it.

Well Canon don't make and FD 25mm, but they do make a 24 and a 28. Since the 28 is much cheaper than a 24 I thought I'd go for a 28mm and a 50mm pair of lenses and an adapter to fit them onto the G1.

For one reason or another (a future blog article on this one) I have decided to start re-acquiring Olympus lenses, the humble old Olympus 50 f1.8 is among my first. Naturally I wanted to compare the Oly to the FD and see what was what.


Unlike comparison back in the film daze this was going to be easier, cheaper and more accurate. To even the comparison field even more (we're already using the same sensor on the same camera) I decided to do the following:
  • mount the camera on a tripod
  • use RAW capture
  • use dcraw to decode the files in a uniform way
  • focus on the same point using the "zoom tool" in manual focus
  • take images at f1.8, f4 f8
It seems that the 'crowd' using lenses by adapter seem particularly fond of shooting wide open. I can understand that for trying to obtain shallow DoF more in line with what one expected on 35mm (well so few of us remain these days, and full frame is too exotic for most folks) but it is often the worst to show what any lens can do ...

Since these lenses typically only work in stop down mode (meaning that the aperture is closed down just like when the camera takes the picture) it is more difficult to obtain your focus precisely on what you are wanting to focus on when focusing at f8.

While the G1's wonderful EVF certainly makes that easier by compensating for brightness you still end up focusing when using the greater depth of field, making critical focus more difficult (more on that later). So for these tests I focused at f1.8 using the manual assist magnifier in the EVF for very accurate placement of the focus plane.


So, first an overview of what I photographed with the Olympus OM 50mm @ f1.8

Then the Canon FD 50mm @ f1.8

even at this overview size its clear that the Olympus lens is much sharper than the FD lens and has better contrast too.

Now for all images I focused on this patch of yellow flowers. Please click on any of the images to load a full 100% pixel screen snapshot that I've loaded.

The image above is at f1.8 and in my opinion the Oly is just streets ahead. Lets look at some other apertures.



At f4 the gap is closing, and by f8 its nearly gone, but you know, the little Oly 50 is still just that bit better. The funny thing is that the FD 50 is the same design as the "Nifty Fifty" prized by EOS shooters presently while the Zuiko 50 f1.8 was occasionally known as the standard body cap. Anyway, before we get to any conclusions, I'll show you the rest of what I found.

While we're at f1.8, lets look at the upper middle left of the image.

then at f4

then at f8

the gap is certainly closing, and if you weren't looking at a digital capture I'd say that any differences would well be lost in printing. But if you have to make a call, the Olympus 50 is still ahead to me.

The Olympus is so far better than Canon away from the center, so lets look right at the center (where both lenses should be at their best)




No matter where you look in the image the olympus lens performs better at every f-stop (the one below is at f8).

Now this brings back memories of the times (I was in my 20's then) and I remember many people saying that Olympus (Zuiko) lenses were by and far the best lenses short of Leica lenses.

It really leaves me asking just why it is that Canon developed such a market lead. Was it:
  • the better automation in their bodies?
  • price?
  • professional perception? (all those Olympic games and Footy matches)
  • ignorance of people in being able to compare lenses?

I can't be sure of that, but what I am sure is that FD lenses are not a match for Zuiko lenses

Since I've got a couple of other duplicate focal lengths I'll be making comparisons with other Oly and FD lenses in future pages. As well I'll publish soon a comparison of the Olympus 50mm on my EOS body compared to the EF50 f1.8

meantime, I hope that this has made your legacy 50mm lens purchase decision easier.


This was essentially the first comparison I did with adapted lenses on micro43 (and look at the date in the URL). I have subsequent to this put together some other comparisons of 50mm lenses which can fit onto micro43 cameras with an adapter here. That article focuses on contrast and flare, but also shows the differences in contrast. If you are interested in adapted lenses on micro43 you can find some other articles of interest here on:
 (although there will be some overlap in those tags)



Administrator said...

Interesting comparison. A lot of Zuiko lenses were (and remain) stellar performers.

The 24/2.8 was recently rated against modern glass - and was second only to Canon's 1.4 - their current (very expensive) current top of the range lens.

I think Olympus fell behind in the 20th century because the OM system was never developed. Canon's AE series were brilliant and made OMs look very old fashioned. That, and Canikon have built up solid importer and dealer networks. Interesting question though.

snatre said...

Very neat job. I have always thought about the question you have asked, Canon vs. Olympus. And somehow I have always been away from Canon since I have always felt some sort of bias...

Although there is no Canon competitor in my test the link is here and here

Anonymous said...

Hhhmmmmm...nice work on your test but I don't think you can make a blanket statement about Zuiko lenses being superior to Canon FD, or anything else. I spent many hours studying "Chasseur d'Image" lens tests in the late 1990s, I came away with the impression that every manufacturer makes some good ones, and some bad ones. Even Leica had a couple that were rated poorly.

obakesan said...


I agree that blanket statements are dangerous generalisations, its like saying all Australians are anglo-saxon decedents. However as far as sampling goes my test is essentially random. I would be very happy to compare another FD 50 with mine to see how much variation exists within the batches. Please let me know if you would like to progress with that.

email me at the reverse of:



Anonymous said...

thank you for posting the results.
I have only two Oly primes (50 1.8 and 35mm shift lens) the rest are Minolta MD and and Canon FD primes from 24 through to 200)

I wanted to do similar test to yours all the time (just for fun, not that it will make any difference to my pictures as I hand hold and scan them). If you have other lenses please keep posting the results.

I would very much like to know about Minolta lenses if you have any as well any zooms.

thank you again

Admin said...

Would be good to know what versions of each lenses you tested. Also, with legacy lenses, it makes sense to check more than one copy. You never know who a particular lens was treated in the last, say, 20 years...

Personally I am a Canon FD system user and collector. There are great lenses in the system and not so great ones.

If you can, try one of the early SSC ASPHERICALs. Would love to see them perform on digital.

obakesan said...


as stated in the text the lenses were both 50 f1.8 lenses. The FD lens is a New FD 50 f1.8 with the A setting available on the aperture ring, the OM is the only OM 50 f1.8 which I know.

I realize that the test is neither statistically significant in terms of samples or through in terms of resolution testing. It is what it is. Should it inspire anyone to do more meaningful testing or even compare their own pair of the same lenses I would be quite keen to see that and would be happy to add links to that.