1. Welcome to Mu-43.com—a friendly Micro 4/3 camera & photography discussion forum!

    If you are thinking of buying a camera or need help with your photos, you will find our forum members full of advice! Click here to join for free!

Analysis of Panasonic G5 RAW Files from Focus Numerique

Discussion in 'Micro 4/3 News and Rumors' started by Amin Sabet, Aug 8, 2012.

  1. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    French site Focus Numérique published their studio tests of the Panasonic G5 complete with RAW files for download. They have previously granted permission for us to publish processed crops of their files with analysis, and I set out to determine how the new 16MP sensor in the G5 compares with that of the Panasonic GX1 (most recent 16MP sensor from Panasonic) and Olympus OM-D E-M5 (reportedly uses a Sony sensor).

    I processed all RAW files using Raw Therapee. Although RAW Therapee will open G5 files, the black point is way off, so I followed these directions to get better support for the files for the sake of comparison.

    Here are some 100% crops of the three cameras at their respective base ISO settings - ISO 200 for OM-D E-M5 (left), and ISO 160 for the G5 (middle) and GX1 (right):

    Subscribe to see EXIF info for this image (if available)

    Subscribe to see EXIF info for this image (if available)

    Same comparisons with all three cameras at ISO 6400:

    Subscribe to see EXIF info for this image (if available)

    Subscribe to see EXIF info for this image (if available)

    These aren't perfectly matched for brightness, tone curve, or color, so it's difficult to draw conclusions about small differences. Maybe a different challenge such as pushing underexposed shadows at base ISO would reveal an advantage for one sensor vs another, but my takeaway from scrutinizing these files was that for practical purposes, there is no difference between these three sensors.
    • Like Like x 7
  2. MAubrey

    MAubrey Photographer

    Jul 9, 2012
    Bellingham, WA
    Mike Aubrey
    They're very, very close indeed. I'd say, though, that I prefer the noise pattern/structure of the E-M5 over the others...though that might simply be a result of the brightness, tone curve, and colour differences...
  3. drewbot

    drewbot Mu-43 Top Veteran

    Oct 21, 2011
    Toronto, ON
    Thanks for posting this.

    If this were DPR, there'd be a maxed out thread about incorrect analysis already.
  4. Jonathan F/2

    Jonathan F/2 Mu-43 Hall of Famer

    Aug 10, 2011
    Los Angeles, USA
    They all look good to me!
    • Like Like x 1
  5. noohoggin1

    noohoggin1 Instagram: @tomnguyenstudio

    May 21, 2012
    with all of them being this close....I'd say the winner is the most inexpensive body...which is the GX-1!! :D 
    • Like Like x 1
  6. springm

    springm New to Mu-43

    Aug 9, 2012
    Amazingly similar results! Thanks for the comparison. For a substantially higher price the E-M5 offers the advantage of in-body-stabilisation - something I truly miss with my G3 and my primes.
  7. jnewell

    jnewell Mu-43 All-Pro

    Jun 23, 2011
    Boston, MA
    Thanks for those. I agree...the differences are so slight that normal-world real-life variations in perfect focus and optimal exposure would probably cancel out any actual test bench differences.
  8. Linh

    Linh Mu-43 All-Pro Subscribing Member

    Apr 14, 2009
    Maryland, US
    So basically, this just confirms to me that the latest top end m43 bodies are pretty much at parity IQ wise. At least, for me. I'm very curious what the GH3 will bring. If it does not improve much, I'll be disappointed. Mostly for the fact that it won't put enough pressure on oly/sony for something better.

    Now if oly and panny would just end the life of that 12MP sensor....
  9. Just Jim

    Just Jim Mu-43 Top Veteran

    Oct 20, 2011
    Wonder what the GH3 will bring out if it the G5 is the middle tier camera. If the middle tier G5 is equal (maybe slightly better on acutance on the G5 at base iso from this sample) compared to the top end oly. Oly may have to drop the price sooner than expected.

    There's no reason to get rid of the 12mp, it's good to be able to bring in a fresh audience with cheaper models to proliferate the system.
  10. bwidjaja

    bwidjaja Mu-43 Regular

    May 30, 2012
    Amin, thanks for posting this. Does this mean that G5 and OM-D is very close in high ISO performance? Say around 1600-3200. If so, then the G5 might be my ticket to upgrade my trusty GH1...
  11. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    Based on this, I would say that G5 and GX1 (and by extension G3), are similar to the OM-D E-M5 in high ISO sensor performance if you shoot RAW.

    Sent from my Nexus 7 using Mu-43 App
  12. danska

    danska Mu-43 Top Veteran

    May 21, 2012
    Portland, OR
    Based on my limited experience with the EM-5 it has cleaner files @ 3200-5000 ISO (all I've shot up to) over the GX1. I found most of the GX1 files unusable at 3200, where a large amount of my EM-5 files at the same ISO were quite usable with minor NR. I wouldn't say that they are a ton different but the EM-5 seems to have at least a one stop advantage to my eyes. Interested to see more tests like this.
  13. KVG

    KVG Banned User

    May 10, 2011
    yyc(Calgary, AB)
    Kelly Gibbons
    I would say you really splitting hairs between all 3 cameras, but the Oly does have ibs and when you mainly use primes that is a advantage. I think I might get the g5(or wait for gh3) just because I shoot 16:9 often and the multi aspect sensor is an advantage for me personally.
  14. zapatista

    zapatista Mu-43 Top Veteran Subscribing Member

    Mar 19, 2012
    Albuquerque, NM
    Mike Barber
    I can clearly say that the best camera is the RX100. Oops, wait, that's another forum. Not much difference in these samples OM-D to G5 to GX1. To me, there's no real IQ difference between the G3 or GX1 either.
  15. With_Eyes_Unclouded

    With_Eyes_Unclouded Mu-43 All-Pro

    Apr 17, 2012
    From these (and other) tests, it seems the OM-D is superior at the highest ISO settings. Big deal, if you ask me. I shoot the OM-D and it wouldn't make any sensible difference over the others at such conditions. The biggest difference would be for very high DR scenes, and the added latitude of almost one stop you'd have to work with in post. But, at that point, for real life situations, the lens selection matters most, IMHO. It's not a race for pulling the most detail out of shadows. Such exersices are silly, IMO.

    What matters more for me is the way the OM-D renders noise, which is very pleasurable in my eyes at least. Also the way it handles highlights is very nice, IMO. Both these situations have a "film" element in them, I think.
  16. Amin Sabet

    Amin Sabet Administrator

    Apr 10, 2009
    Boston, MA (USA)
    I'll present some controlled testing from my E-M5 and GX1 when time/job permit.
    • Like Like x 2
  17. tanngrisnir3

    tanngrisnir3 Mu-43 Top Veteran

    Oct 18, 2011
    Based on what my eyes are telling me, the GX1 at base ISO beats the others for color reproduction, and at high ISO the G5 is the best.

    Why pay more?
  18. dwrk

    dwrk Mu-43 Regular

    Aug 10, 2012
    thanks for the comparison...i was looking at the g5, em5 and gx1 jpegs...the G5 kept the most fine details...they really did a good job here...can you take a look at it in the raw files too...

    also do you all know if the g5 or panasonic cam in general, is a global cam...ie selectable fps for pal/ntsc video capture? thanks.
  19. napilopez

    napilopez Contributing Editor

    Feb 21, 2012
    NYC Area
    Napier Lopez
    I think the E-M5 is very clearly the best here. Half stop to a stop better, if you factor in noise pattern and detail. I see it as the oly having a half stop over the G5, and the G5 a half stop over GX1. Which makes sense since I consistently see results of 2/3 to one stop better performance in RAW from the oly over my G3. I'm fine with the G5 as such personally.
  20. phrenic

    phrenic Mu-43 All-Pro

    Sep 13, 2010
    I really can barely see a difference. Amin, it would have been more interesting to make it a blind test and see what people think is the best/guess the camera before you knew which sample was what. ;) 

    6400 may be the tiniest bit less noisy on the OMD but I feel like the detail is abit worse, so is that just NR applied? I'm not an expert on IQ but I really doubt many would see the difference.
  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.