Olympus 75 mm: sharp from f/1.8?

Discussion in 'Native Lenses' started by SojiOkita, Jan 24, 2016.

  1. SojiOkita

    SojiOkita Mu-43 All-Pro

    Feb 23, 2014
    Hello everyone,

    I just got an Olympus 75 mm yesterday.
    I just had time to make some tests (to see if I keep the lens or not) and I'm a bit disappointed by the sharpness I get at f/1.8.

    The lens has a great reputation for being sharp wide open.
    I would not call my lens "sharp" at f/1.8 (I would call it "usable").
    It's sharp at f/2.8, and very sharp at f/4.0, but it's not impressive in any way at f/1.8.

    Here are some of my test shots, center crops at the point of focus.
    All shots on tripod, no IS, MF with focus peaking and zoom.
    (I tried AF too, with the same results).

    For all examples, f/1.8 is first, f/2.8 second, f/4 last.
    you can click on images to compare them in different tabs.

    * test 1 *

    * test 2 *

    * test 3 *

    I wonder: did I make something wrong? Am I pixel peeping? or is my 75 sample not as good as it should be?
    • Informative Informative x 1
  2. 50orsohours

    50orsohours Mu-43 Hall of Famer Subscribing Member

    Oct 13, 2013
    Portland Oregon
    I am not a techy, but when I had my 75mm, at 1.8 I'd get tack sharp pictures handheld. So there is something going on. Especially the second set, that first shot is way off. What camera and what were the shutter speeds?
  3. heli-mech

    heli-mech Mu-43 All-Pro

    Mar 9, 2012
    Vancouver Island, Canada
    • Agree Agree x 2
  4. MRM

    MRM Mu-43 Regular

    Jun 9, 2013
    Seattle, WA
    I found that stopping down to 2.0 improved the sharpness. Beyond that there was further improvement but not enough to stop down unless increased dof was needed. Sharpness is a relative term and wide open the 75 is plenty sharp for portraits. Also in my lens testing I have found that the 40-150 pro improved significantly from a 1/3 stop down to 3.2 at 150. At 40 and 70 it is sharp as it can be wide open. And the 12-40 doesn't improve from stopping down until 40 mm.
  5. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    Sharpness is subjective and depends a lot on processing. Here are 4 shots I took this this lens wide open which I consider to be sharp. Both JPEG and RAW files included in the download: 75mm.zip
    • Agree Agree x 1
  6. SojiOkita

    SojiOkita Mu-43 All-Pro

    Feb 23, 2014
    1/4000s for the first one, 1/800s for the second one, 1/640s for the third one.
    All with my E-M10.

    Yes, sharpness is subjective, that's why I wonder if my copy is bad or if was hoping for too much.
    I made first some quick portrait tests and wasn't impressed. I was hoping to take portraits with very sharp subject and great bokeh.
    I plan to use it mostly wide open, so it has to be good at f/1.8.
    (didn't have the time to study the bokeh, it's hard to judge when not in real conditions)

    Thanks for the samples, I will study them tomorrow (RAW is nice, because that eliminates the post-processing differences).
  7. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    I do think it gets a bit sharper at f/2.0 with bokeh at that setting almost indistinguishable from f/1.8.
  8. heli-mech

    heli-mech Mu-43 All-Pro

    Mar 9, 2012
    Vancouver Island, Canada
    One other thought, what camera and what shutter speeds are these at? I ask because the other possiblity is shutter shock interfering with your tests.

    Edit- nevermind missed your post with shutter speeds
  9. David A

    David A Mu-43 All-Pro

    Sep 30, 2011
    Brisbane, Australia
    Some observations:

    1- the first 2 sets of examples are shot at an angle to the door/ground so there is a difference in the distance to the left/right and top/bottom of the frames respectively. At F/1.8 you have shallower depth of field with DOF increasing as you stop down. I would expect some difference in edge to edge sharpness across those sets of examples simply because of the fact that the subject plane is not parallel to the sensor plane and you are varying aperture which changes your DOF. I am not saying that this is the only issue, just that this will make a contribution to the overall appearance of sharpness. That factor is not an issue with the third set of examples which appears to be straight on with subject and sensor planes parallel. Since the subject in this set is a flat plane, DOF is not going to make a difference if the lens is accurately focussed. The difference in sharpness between the 3 images in the 3rd set does not appear to me to be as great as it does in the first 2 sets which tends to confirm my suspicion that part of the issue in the first 2 sets is simply the difference in DOF coming into play. I will add that I'm not certain that the first frame of the 3rd set, the F/1.8 example, is quite as accurately focussed as the other 2 frames. I also wish you had shot the door straight on like you shot the text so that we could exclude DOF as a factor in that set of images.

    2- the first 2 sets of examples are outdoor and I'd guess that the 3rd set was taken indoors. I have no idea of the angle of the light source in the outdoor shots or whether you were using a lens hood but it is possible that there is some slight loss of contrast from flare in the first 2 sets and one would expect that to be less of an issue as you stop down. Once again, I'm not saying that this is the only issue, just that it may be be making a contribution.

    3- In all 3 comparisons I think the F/1.8 frame seems to have slightly lower contrast than the other frames and increasing contrast increases the appearance of sharpening. In fact increasing contrast at edges is what we do when we sharpen an image in processing. I think a difference in contrast is a factor in all 3 of these comparisons but once again that will only be part of the picture. It doesn't explain everything.

    It's really hard to design a test for sharpness which isn't affected by other factors and I do think there are factors in all 3 of these comparisons which are affecting the sharpness evident in the results. Those factors will not account for all of the difference because there will be a difference anyway. No lens performs at it's best wide open and there will be an improvement as you stop down. I just don't think the performance at f/1.8 that you're getting in these examples is as good as you should be seeing with the lens and I think your choice of comparison subject and setup are adversely affecting the result you are getting at f/1.8.
    • Informative Informative x 1
  10. Replytoken

    Replytoken Mu-43 Hall of Famer

    May 7, 2012
    Puget Sound
    Is it difficult for you to switch copies?

  11. b_rubenstein

    b_rubenstein Mu-43 All-Pro

    Mar 20, 2012
    Melbourne, FL
    Asking people if the lens is "sharp" wide open is heading down a crowd sourced rabbit hole, because "sharp" not only is subjective, but depends on the technical shooting skills of the responder. There are probably over a dozen web sites that do very good lens measurements, or highly experienced photographers who know how to shoot and know what they're looking at. Now, if you're going to read an actual lens test (actual tests generate numerical quantifying data), then you have to understand what the numbers mean and how they correlate to visible differences. This requires more work than asking "Ralph", but is more useful than many internet "Ralphs" who have never touched the lens in question, let alone taken a picture with it.

    FWIW, I'm not Ralph, I owned one of these lenses for years and it's pretty sharp wide open, but progressively sharper stopping down to f2, f2.8 and f4. So I agree with all the lens tests I've read for the lens.
  12. hazwing

    hazwing Mu-43 Hall of Famer

    Nov 25, 2012
    I haven't done any official tests with my 75mm, but I agree with others... wide open is pretty sharp, sharpens further as it's stopped down. I don't think your findings are out of the normal.

    It's not just the sharpness that I like about the 75mm... more so I like the subject compression, shallow DOF and bokeh it gives.
    • Agree Agree x 1
  13. SojiOkita

    SojiOkita Mu-43 All-Pro

    Feb 23, 2014
    Hello David.
    I agree with pretty much all your comments;) 
    On your 1st comment: yes, the door is shot with an angle. In fact I think pretty much all of my test shots are not very good "test shots" ;) 
    I didn't find something without any angle, with a flat plane, and where it was possible to assess sharpness.
    Except on shot (3) that was taken inside, at a lower distance.
    The difference in sharpness is indeed lower on this shot (it's mostly contrast) so I don't know if due to focus distance or better focusing.

    On test (1) though, I can't find anything perfectly sharp on the f/1.8 picture (but DOF is quite thin).

    Concerning flare, it's possible. I didn't use a hood (considering the price of the lens, it would have been a good idea to include it ;) 
    The shot is taken from inside, so the window frame is a sort of big hood;) 

    There is a noticeable difference in terms of contrast between f/1.8 and f/2.8
    I'll try to increase contrast on the f/1.8 to match the others.

    Yes it's very hard.
    I always have a sort of "new lens paranoia" ;)  and I don't know the lens well as I just received. But there's very little time to decide to keep a lens or not.
  14. SojiOkita

    SojiOkita Mu-43 All-Pro

    Feb 23, 2014
    It's quite difficult.
    There's no other option than ordering online, and there's a big olympus promotion ending on January 31st (150€ cashback on the lens).
    If I try another copy, I would prefer from another source (to avoid getting a lens produced nearly at the same time which could be almost identical).
    I can order another one on amazon and compare, but it's more expensive (or in silver, and I don't want the silver version ;)  ).

    That's probably what I will do, just to be sure...

    From what I saw, it's not a "bad lens", but considering the price and the reputation, I'm disappointed.
    The performance wide open is not different from what I saw in other wide aperture lenses (Olympus 45 f/1.8 and Canon 85 f/1.8 for example) which are 2-3 times cheaper.
    Less sharpness, less contrast, more AC.
    I think, for example, that a lens like 20 f/1.7 is much more impressive wide open (but the focal length is different, the DOF is bigger).
  15. David A

    David A Mu-43 All-Pro

    Sep 30, 2011
    Brisbane, Australia
    You could have shot the door straight on rather than from an angle.

    If you repeat the 3rd example or have a go at the door again from straight on, what I would try is this. Shoot RAW and set the contrast setting for JPEGs, which affects your viewfinder/screen display to maximum contrast. That may make manual focussing a little easier and it may also make autofocus a little more precise given that the main focussing method for m43 bodies is contrast autofocus. Try both autofocus and manual focus. I don't know it that will affect your results but it may.

    I didn't try any testing with my 75, just put it on the camera and started using it. I've always been impressed with the sharpness and resolution when I've used it but I also rarely use it wide open. I take most of my shots at around F/4-F/5.6 regardless of the focal length. I rarely chase shallow depth of field and I usually like a reasonable amount of depth of field so sharpness at F/1.8 isn't usually an issue for me.
  16. listers_nz

    listers_nz Mu-43 Veteran

    Nov 22, 2013
    Christchurch, New Zealand
    +1 for trying it with autofocus, taking note of where the focus point is. Not suggesting there is anything wrong with your focussing, but at f/1.8 it isn't always that easy to nail it.

    Also, I'm assuming you don't have any kind of filter on the lens - if you do, test the lens with it off. I recall another thread about a lens not being sharp, and it improved out of this world once the filter was removed, despite it beings a reputible brand - I guess you can get a dud of anything.
  17. cptobvious

    cptobvious Mu-43 Veteran

    Jan 8, 2013
    The two copies of the 75 I had were sharp wide open, although I had to return the 2nd for another issue (defective coating). Should be noticeably sharper than the Canon 85 1.8 wide open though. As far as the 45 1.8, there is pretty significant sample variation in the 5 or so copies I've tried/owned. A couple I tried were soft wide open, but the best were about on par with the 75 in the center, although the 75's sweet spot extends further out from the center. I would try another copy.
  18. stripedrex

    stripedrex Do or do not. There is no try.

    Jun 8, 2012
    Long Island, NY
  19. SojiOkita

    SojiOkita Mu-43 All-Pro

    Feb 23, 2014
    I tried another copy from another store, I was hoping for better... but my second copy is a worse at f/1.8.
    I made new tests trying to do a little focus bracketting (manually, as I have an E-M10) to get the more "in focus" shots.
    Each time the second lens was worse than the first.

    And in both cases there was a big difference in sharpness between f/1.8 and f/4.0.
    I compared with my 45-150 at 75 mm (f/5.1 and f/6.3), the 75 mm is sharper (without a doubt) at f/4.0, but softer at f/1.8.
    (it's very difficult to compare sharpness between a f/6.3 shot and a f/1.8 shot, though)

    I made comparisons with the 85 but as my Canon 40D definition is only 10 Mpix, and the lens is a little wide, it's hard to conclude (except that there's more detail on the E-M10 shots, and more AC on the 85).

    At first I was only doing sharpness tests to spot a defective lens (as I always do).
    I think I can eliminate this possibility now (or else I'm very very unlucky, or Olympus quality checks are very very bad).
    So I'll try the first lens in normal conditions (portrait shots) to see if I want to keep it, and I'll return the second.
  20. Klorenzo

    Klorenzo Mu-43 All-Pro

    Mar 10, 2014
    This is about the norm for all lenses. There are very few exceptions where the lens is at maximum wide open (only in the center) or where it improves only marginally (even Nocticron peaks at f/4).
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.