1. Reminder: Please use our affiliate links for holiday shopping!

Extreme dynamic range test of G5 vs. E-M5

Discussion in 'This or That? (MFT only)' started by DHart, Aug 30, 2012.

  1. DHart

    DHart Mu-43 Hall of Famer

    Jan 7, 2010
    Scottsdale, Arizona
    Don
    This scene is an extremely high contrast scene ranging from direct sunlight reflecting off of polished steel to dimly lit black fabric falling into shadows.

    This scene represents an extreme setting for a camera to capture, far more extreme than the typical scene one might wish to capture - but as such, makes a good test to see how much highlight and shadow data can be recovered from the RAW files generated by the G5 and the E-M5.

    All variables were held constant except the camera bodies. RAW captures, shot with Pana-Leica Summilux 25mm f/2.4 lens set at f/8, ISO 200, with a tripod.

    First up, two images with simple processing in LR 4.2 RC.

    E-M5
    Z8190055.

    G5
    P1000200.

    And now, with the highlight recovery at -100 and lifting shadows at +100

    E-M5
    Z8190055-1.

    G5
    P1000200-1.

    Impressions are that the G5 and E-M5 are remarkably close in their ability to capture data in the extremes of dynamic range, which is recoverable in LR 4.2RC. The most obvious difference between them being the color shift in the recovered deep shadow areas with the G5. In this aspect the E-M5 sensor makes a better showing. And there is a bit more data recovered in the extreme highlight area with the E-M5. All in all, though, the DR of the two sensors are not far apart.

    The demands made by such a test are not likely to be needed with the vast majority of image capture generally done, but it's nice to know that both cameras are capable of recovering a remarkable amount of data in the extremes!
     
    • Like Like x 11
  2. jyc860923

    jyc860923 Mu-43 Hall of Famer

    Feb 28, 2012
    Shenyang, China
    贾一川
    Thanks Don

    that's really very informative, and gives a good idea to people like me who's been hesitating if DR is concerned. I hope I'll be getting the G5 next year.

    plus, I don't know if there is anything to do with lumix's colour shift in LR, but I find it necessary to get the correct colour by applying +25 to the red channel in hue, it makes a difference.
     
  3. 71taa

    71taa New to Mu-43

    6
    Jan 28, 2012
    Russian Federation
    Alexander Tretyakevich
    Great comparison - thanks a lot!
    E-M5 still capable of recovering some more details from highlited area (see the center of round metal thing, texture of stain is restored on E-M5), yet not a dramatic difference.

    On color shift - I could see some blueish casts near the cloth (on the table), could the reflected light hit the cloth as well so some traces of color components from it were restored while pushing up shadows in G5 shot? Only a speculation, though.
     
  4. DHart

    DHart Mu-43 Hall of Famer

    Jan 7, 2010
    Scottsdale, Arizona
    Don
    Thanks for your comments, guys. Having both cameras, if I were to have to choose between them and only keep one, DR capability would NOT be a factor in my decision... it comes down to feature sets, ergonomics, feel-in-the-hand. I like both cameras, but will say that at present, the G5 is the one I'm shooting, for the most part. :wink:
     
  5. arad85

    arad85 Mu-43 Veteran

    477
    Aug 16, 2012
    Thanks Don,

    I'm glad you repeated it with LR - just shows what a cruddy RAW processor SilkyPix is - especially the highlight area.

    I think you are likely to get different results if you tried this at higher ISO - the impression that I get from all the reading I have done on this is the GH2 sensor is great at low ISOs, but falls rapidly in DR and colo(u)r range as you up the ISO when compared with the EM5.

    I'm saying that as someone who does NOT have an EM5 but shoots Panny only. I agree, the handling of the G5 is great!
     
  6. DHart

    DHart Mu-43 Hall of Famer

    Jan 7, 2010
    Scottsdale, Arizona
    Don
    Arad... Yes, high ISO results may not be quite so comparable. And thats just fine, as well. I won't likely do another test, however, as I am plenty happy with both cameras and just not compelled to push them to extremes for the purpose of crowning a winner. Clearly, they're both winners! :smile:
     
  7. arad85

    arad85 Mu-43 Veteran

    477
    Aug 16, 2012
    Have to say that I try and keep to low ISOs anyway and compensate for needing high ISOs with aperture (hence the 25/45 primes) or fast zooms and OIS. The impression is that for a well exposed image (i.e. one that isn't pushing ISO to get the light) they are pretty much equivalent. The further you push ISO, the more the gap widens.

    I've written elsewhere that I view bodies at this sort of price as effectively disposable (I managed to get the Lumix deal in the UK so bought mine for £423 delivered). In 9-12 months when the G6 comes out (or even if the GH3 turns out to be a game changer), the most I will lose is £200 on resale or I keep it as a second (or third!!) body. That is peanuts compared to if I'd have bought a 5D3 (not to mention the 5D3 isn't as user friendly when taking pics so I wouldn't have got the same pics (e.g. crop one I posted elsewhere on here) with a 5D3.

    Now the quality lenses are available (or will be with the 35-100), we're set. Give it 12-24 months - or perhaps 3 if the GH3 is a breakthrough - and we'll all be wondering how we managed with only the performance of todays sensors.
     
  8. sLorenzi

    sLorenzi Mu-43 Top Veteran

    586
    Sep 15, 2010
    Brazil
    Sidinei
    Impressive results from both cameras. What is interesting is the noticeable the difference between the G5 cooked in LR and in Silkypix, from the first test you did. Damn, I just got a copy of Paintshop Pro, but now that I'm gonna get a G5, I'll "need" Lightroom 4 too. :redface::redface:
     
  9. mattia

    mattia Mu-43 Hall of Famer

    May 3, 2012
    The Netherlands
    Sorry, but if there's one thing I miss about my Canon gear when shooting M43 (the 5DII in my case), it's the significantly better ergonomics in every respect but weight. Setting all key parameters on a Canon DSLR is very easy to do quickly. There's a learning curve with the more complex AF modes on the 5DIII and 7D, but in every other way I still love shooting the Canon. I just hate carrying it around.
     
  10. arad85

    arad85 Mu-43 Veteran

    477
    Aug 16, 2012
    My 5D2 left here this morning as I have sold it on. In fact I have have sold all my Canon gear in the last week.

    I have owned and used GX1, G3 and now a G5. It wasn't until I used the G5 that I felt the ergonomics were better. I can put the camera where I want to without having to look through the viewfinder (a few inches from the floor in the case of the photo I mentioned). I can move the focus points anywhere on the sensor both whilst looking at the LCD and whilst looking through the viewfinder simply by using the touch screen. The combination of lever, thumbwheel and sensible placement of the Q menu button means I can change everything quickly without looking away from the viewfinder. I also have an AEL/AFL button right under my thumb.

    IMHO, the G5 is the gamechanger ergonomically for a photographer coming from a DSLR. Everything I found constraining about the G3 has gone with the G5. It is simply that good IMHO.
     
  11. meyerweb

    meyerweb Mu-43 Hall of Famer

    Sep 5, 2011
    Just shows how we all have different preferences. I've shot Canon AF gear since the very first Canon AF camera, the EOS 650. I certainly know and understand the Canon interface. But when I picked up a Panasonic GH2 for the first time, I found it largely superior. It suffers as a consequence of the small size and resulting lack of real estate for back panel controls and a top LCD, but over all I find the mechanical settings for things like focus modes, drive mode and bracketing to be superior to Canon's button and dial approach.
     
  12. jloden

    jloden Mu-43 Hall of Famer

    May 15, 2012
    Hunterdon County, NJ
    Jay
    Good to see real world results show a minor margin between the two, just means more great choices for m4/3 users.

    It'll be interesting to see how the GH3's new "wide dynamic range" sensor performs in a similar comparison. I'm curious to see how much of that is marketing terms versus actual increase in D.R. on the new sensor purported to be coming in the GH3.