Can we judge a lens by its edited pictures?

Discussion in 'Image Processing' started by Julia, Jan 19, 2017.

  1. Julia

    Julia Mu-43 Regular

    190
    Mar 9, 2013
    Dresden, Germany
    I think this might be a bit of a philosophical question: When we look at pictures produced with a particular lens, are we actually judging the lens or are we judging the editing skills of the photographer?

    I'm not saying it's bad to edit images, I do it myself quite heavily. But do those images then give us a real representation of what the lens is capable of?

    I like sharp images. Razor sharp. Lots of detail, little noise. Yeah, I know, mFT isn't exactly kind when it comes to noise and that was a big hurdle when I switched from DSLRs back in 2011/2012.

    I also like to go through the picture threads and browse Flickr to get a feeling for how my lenses stack up against other copies of the same model, and of course to help me decide which lens to buy next.

    Just last week(end) I had a rather frustrating experience when two different copies of the Panasonic 12-32 I tried produced mediocre images at best. It was not even close to what I have seen on Flickr or around the web. After editing the images, I could improve the sharpness a bit, but I was still really disappointed (and returned both).

    I just edited a few shots taken with my cheap Panasonic 45-150 on the EM5. The lens is ok. It doesn't produce razor sharp images out of the box, but with some tweaking in LR, I can make the images look pretty sharp.

    Here's a crop of a recent image:

    Screen Shot 2017-01-19 at 17.34.39.

    As you can see, the Candidate (right) is the image without Clarity, Sharpening, Unsharp Mask, and Noise reduction. So the right picture is actually more of what comes out of the camera and more representative of what the lens is capable of?

    Or, here's another example (crop again):
    Screen Shot 2017-01-19 at 17.36.11.

    Once more, the Candidate on the right is the one without anything changed in the Detail section (sharpening etc).

    So, when judging a lens, should we judge was comes out of the camera, without any changes applied (this would mean that OOC Jpegs shouldn't have any sharpening, contrast etc. changes applied either, of course), or should we judge what we can potentially make of those images in post?

    As I said initially, this isn't about "faking" how good a lens is, or implying that image editing isn't part of the photographic process (at least it is for me). But I'm having trouble to figure out how to judge a lens. And, of course, when I look at my unedited pictures, I think constantly that my lenses or my camera must be bad because they don't look perfect like the many pictures I find online.

    Is that just a misconception in my mind, something I need to get rid of and judge my image after I ran it through my editing workflow? After all, all we see online is a finished product and we don't know how heavily an image was edited and what the original "look" was. We just what the photographer wants us to see, but if we (I) are not careful, it's easy to fall into the trap of comparing non-edited pictures to edited ones and despair :)

    I'd love to hear your thoughts!
     
    • Like Like x 1
    • Agree Agree x 1
  2. Turbofrog

    Turbofrog Mu-43 Hall of Famer

    Mar 21, 2014
    If you can sharpen an image and add contrast and vibrance without getting halo'ing, artifacts, or detail degradation, then it's totally fair game to me.

    Oversharpening an image until it's crunchy and overcooked is often used to disguise soft (or out of focus) images, but that doesn't seem to the be case here.
     
    • Agree Agree x 1
  3. Giiba

    Giiba Something to someone somewhere Subscribing Member

    505
    Aug 19, 2016
    Burnaby, BC
    I'd suspect that starting from a sharper, more contrasty image allows a sharper more contrasty final output. Or needing less sharpening might preserve details better and avoid enhancing noise.

    I doubt you could make an image from a kit zoom look as good as one from a f/2.8 zoom... but I have never tried so I'm just continuing to philosophize.
     
  4. Drdave944

    Drdave944 Mu-43 Top Veteran Subscribing Member

    818
    Feb 2, 2012
    My philosophy is that "What is in the soup is known only to the boss." So you can't really isolate how good the lens is performing.
     
    • Funny Funny x 1
  5. PeeBee

    PeeBee Mu-43 Top Veteran Subscribing Member

    725
    Sep 17, 2012
    UK
    Editing is just one of the many variables that can effect the final image. Sensor generation, camera settings, lighting, composure, camera shake etc etc. Even with PP ruled out, where is the consistent reference point?

    I look at online images to see what is possible from a certain lens, not what to expect. The more images that impress me, the higher the chances that I'll be happy with it.
     
    • Agree Agree x 5
    • Like Like x 2
  6. ijm5012

    ijm5012 Mu-43 Hall of Famer

    Oct 2, 2013
    Pittsburgh, PA
    Ian
    Based on the exif shown in the first image, you were shooting the kit telephoto a near max reach, wide open, correct?

    If so, you're expectations are a bit unrealistic IMO. There's a reason the 45-150 is only a couple hundred euro, compared to a lens like the Olympus 40-150 PRO. If you want "razor sharp" images, be prepared to either stop down, or pony up the money for one of the better performing lenses.
     
  7. Julia

    Julia Mu-43 Regular

    190
    Mar 9, 2013
    Dresden, Germany
    This is a very good approach. I think my "flaw" was always assuming that the images I saw online would pop out of the camera just like that. I somehow rarely ever stop to think: oh, here was sharpening, gradation, noise reduction applied.

    But you are right: looking at a wide variety of images gives one at least an overall impression of the possibilities. Then it's up to the individual photographer to make the most of it.

    Yes, you are right. And of course, I know that I cannot expect this lens (which goes for about 200 Euros nowadays) to be razor sharp, otherwise what would be the point of more expensive lenses.

    I am currently saving up for an Oly 12-40 or Pana 12-35, after having tried the 12-32 (two copies) because of the praise it had received and some amazing samples I saw online. I thought, going by what some people said, that using it stepped down might give me great image quality, or at least good enough quality. Sadly, with the copies I had, that wasn't the case. So now I am hoping that once I invest into one of the PRO lenses the images that come out of the camera will be sharper than what I get from my cheaper lenses at this point. My assumption/hope is that what is OOC will already be sharper, more detailed and that I will not have to add so much PP to make the image appealing.
     
  8. Chris5107

    Chris5107 Mu-43 Top Veteran Subscribing Member

    910
    Jan 28, 2011
    USA
    Chris
    Could your problems with the 12-32 be shutter shock? I find mine to be sharp on the EM5-2.

    I ended up selling of my copy of the Oly60 before I understood the shutter shock issue.
     
  9. greenboy

    greenboy Mu-43 Top Veteran

    One of the biggest improvements I got for my typical landscapes shots was to use the zooms that are sharper to begin with and get a good polarizer, which affects at least some angles -- especially since much of my favorite shooting season is affected by wildfire smoke/haze. Then it just comes down to me.
     
  10. gryphon1911

    gryphon1911 Mu-43 Hall of Famer

    Mar 13, 2014
    Central Ohio, USA
    Andrew
    My take:
    Images from the camera, JPG or RAW are never "finished" until I work on them in post - same as the image from a film negative is not finished until it is printed and processed.

    Even when shooting JPG files, I have a specific way they are captured in camera that provides me more latitude in reducing the low ISO noise and widening the shadow areas a bit. If I showed those images as is from the card - they would look flat and unsharp. Also remember, that ever camera/lens combo has a slightly different need in post processing. I have a base level preset for my m43 cameras, but sharpness is one of the variables that gets tweaked in every image.

    The P12-32 is a great lens. No, it is not going to be the same level as an f/2.8 lens - however, after using it some, I realized that the lens suffers from shutter shock on my EM1.1/EM1.2 and PEN-F cameras. I resolved this problem by using anti-shock mode or e-shutter. As others have mentioned, perhaps you fell victim to the shutter shock issue with the lens.
     
    • Informative Informative x 1
  11. Julia

    Julia Mu-43 Regular

    190
    Mar 9, 2013
    Dresden, Germany
    I was hoping that would have been the case. I had actually never heard of shutter shock before I started to research the 12-32, and I had set both the EM5 and the EPL5 to Anti-Shock (1/8s). I also tried it on a GF7 (that came with one of the lenses) but I didn't get one image that looked pleasing to me. I could keep trying and order more copies (local stores don't have any others in stock) but the cost of shipping them back, and having to wait for refunds, is just not worth it to me.

    Yes, that is my hope too. That the images OOC will already be sharper and won't have my guess if I made a mistake, at least until I've finished editing them and can then tell if I like them or not.
     
  12. Photozone has this issue too, when they process in ways to avoid the camera's autocorrections. It's more work to get the unprocessed image in this case, but it does point to how much the autocorrection process is doing for distortion and 'chromatism'. It's fair to do this in my opinion since corrections often soften corners or other areas of the image.. but it's a quandary. I have a 12-32 and 12-60 in hand, some day I'll check them over to see whether portability or versatility exacts the higher price.
     
  13. fredlong

    fredlong Just this guy...

    Apr 18, 2011
    Massachusetts USA
    Fred
    To me the answer is that a lens must be judged by edited images. There's really no unedited image to see. If an image comes straight out of the camera as a jpeg, which set of settings is "natural". If it's a raw image which raw converter do you use to get an unedited image. They all do some processing by default. What counts as unedited.

    An image has to be processed or edited to get as close to the photographers vision to judge whether the lens will do the job.

    Fred
     
    Last edited: Jan 19, 2017
    • Agree Agree x 3
    • Like Like x 1
  14. wjiang

    wjiang Mu-43 Hall of Famer Subscribing Member

    You can't easily separate the lens from the camera when evaluating images. SOOC JPEGs depend heavily on camera processing capability (even minimum noise reduction is still some on Olympus JPEGs) and even RAWs depend on sensor characterisitics and how Adobe treat the RAWs from different models. I've always found my GMs/GF comparatively a little softer with the same lenses compared to my E-M1, I put this down to the lack of AA filter on the E-M1 but who knows? If you look at DXOMark the same lens somehow scores very different results on different camera models. It's only when you measure with an optical bench like LensRentals do you get 'fair' results, but that doesn't give a easily digestible representation of real world image quality.

    Comparing against online images is going to be a limited exercise - it is doubtful you'll be able to replicate the same camera settings, lighting, subject distance and bsckground. I suggest rather than comparing what you get with what other people are getting, use your own benchmarks. I know the two copies of the 12-32 I've had were both decent, and definitely better than the 12-50 with the same cameras and PP treatment. In the same way, I know the 12-40 is even better. If a lens is truly meh, there's no way you can rescue it with PP.
     
    • Agree Agree x 2
  15. Hazza

    Hazza Just Clicking Subscribing Member

    644
    Aug 10, 2013
    Huddersfield UK
    Harry
    I process all my images to a set of parameters in Photoshop CS6 camera raw, then I tweek the settings to give me a look that I like. I have had 3 Olympus Pro zooms (7-14, 12-40 and the 40-150) all are superb and very sharp but I wouldn't say that the images straight out of camera are the images that I end up with. They all get the same treatment and tweeked to taste depending on the subject matter, ie portrait, landscape, cityscape etc. Portraits tend to have less sharpening than say a cityscape which would have the sharpening turned up more.
    I use an EM5 with the anti shock set to zero and it works for me pretty well.
    I no longer have the 40-150 as I tend to prefer a wider FoV so I sold it and bought the 7-14.
     
    • Agree Agree x 1
  16. norman

    norman Mu-43 Rookie

    13
    Jan 22, 2014
    Corby, Northants, England
    Norman Shearer
    I think you can judge a lens by looking it up in say a Flickr group and going through lots of images taken with it. In ways it is more accurate than testing a single copy of the lens yourself because you may have a bad copy. Sure there is post processing going on but there's a saying about polishing turds that springs to mind!

    Regarding the 12-32mm zoom. I had a copy in the past and I loved it. Sharp across the frame, good contrast, fast AF, lovely and small. It was a bit flare prone and the build quality leaves much to be desired. Also lack of controls on the lens is missed at times. Great value though. I sold my 14mm pancake because the 12-32mm made it reduntant. I kept the 20mm pancake for low light shooting otherwise that would have been sold too. Must be copy variation rather than shutter shock. I was using my copy on GX7 and GF7..
     
  17. ionian

    ionian Mu-43 Veteran Subscribing Member

    486
    May 20, 2016
    Kent, UK
    Simon
    I don't think you can judge a lens based on someone else's photos- there are too many variables. All you can do is know what you want a lens to excel at and then get a range of opinions from photographers with similar aims to you.

    Also remember your output - when is a lens sharp enough? For me it's usually when I can post it online and occasionally make a modest sized print. Any sharpness after that is mustard, and will actually have no effect on how my image is perceived.
     
    Last edited: Jan 19, 2017
    • Agree Agree x 2
    • Like Like x 1
    • Winner Winner x 1
  18. wjiang

    wjiang Mu-43 Hall of Famer Subscribing Member

    I've found that print is very kind. I've printed 12 inch long and up prints that looked fine from the mediocre 12-50 because you just can't pixel peep a print!
     
    • Agree Agree x 3
  19. Wisertime

    Wisertime Mu-43 All-Pro

    Aug 6, 2013
    Philly
    Steve
    Agree...and I've printed quite big with the 12-50 and 40-150 kit lens and sold prints from them. It's more about the picture than technicalities, not that there was anything to complain about in that regard. I've also found exported photos viewed on a big screen (full screen) with any given lens make me appreciate them more than viewing them in "normal" reduced screen sizes.
     
    • Like Like x 1
  20. Machi

    Machi Mu-43 Veteran

    242
    May 23, 2015
    Both.
    Or more specifically:
    1) Lens + sensor in case of raw with basic processing (deBayer + Gamma + white balance).
    2) Lens + sensor + in-camera processing in case of JPG directly from camera.
    3) Lens + sensor + editing skills in case of processed raws (or JPGs).

    You can always do comparison of different lenses with basic processing of raws but it's difficult to find such images on the net.
    Edited images or images from camera are much easier to find so they are often the only way to "compare" different lenses.