Olympus OM-D E-M1 review updated with thoughts on RAW quality

_ET41526b copy

Now that ACR has preliminary support for E-M1 raw files – amongst a whole load of other cameras – in ACR 8.2 (available here for Mac and Windows), I’ve gone through and reprocessed a few to assess the RAW quality of the E-M1’s sensor; I expect to have more thoughts on this in the longer term after I have a chance to put the camera through a greater variety of scenarios. Sadly, my loaner went back yesterday, so further updates after this one will have to wait until my own cameras arrive in October.

The full updated review is here. MT

Comments

  1. What’s the fastest SD card that you use with the EM1? I’m trying to compensate for the narrow DR by shooting all Raw and I can’t find anyone that has tested it!

  2. How much work do the higher ISO EM-1’s RAW’s take compared to the APS-C & FF files? I’ve read conflicting opinions.
    I’m leaning towards Sony or Fuji non-reflex bodies because it seems people notice you less when they don’t look like a pro-DSLR – how do you find reaction to your EM-1’s? I really like the size/weight, AF speed advantages of the EM-1 at this point though.

  3. Charles Ide says:

    Greetings-
    I just ordered the OM-D EM-1 with the M.Zuiko ED 12-40mm f2.8 PRO, and now want to buy memory cards. Can the camera take advantage of the difference in speed between the Sandisk Extreme (45 MB/S), and the Sandisk Extreme Pro (95 MB/S)? If so, what is the difference? The Olympus web site doesn’t seem address this question.
    Thanks for hosting this excellent site. Because of your review I chose this camera.
    Charles-

  4. (A general remark to many photo corresponders : the wanted word (spelling) is “complEment…”, not “compliment”! :o)

    To my awareness, it seems that m4/3 –Panasonic/Olympus– is doing much with LENSES in contrast to at least Nikon with its “DX” (APS-C). And to my mind, the GH3’s **increase** in size is welcome, as my hands want some size (sometimes); it yet leaves m4/3 size & weight advantage to come via lenses (and of course complementary bodies such as the E-P(L)5, GX7, & E-M5).

    IQ is a matter of appraisal, too. I’ve just seen some photographs looking fine from P&S cameras (i.p., some Pany TZ model) at 12×16 or so, and bigger from a 6mp D70). I’ve got one 2’x3′ print from my LX3 along with 20x30s (the 20×20 is same demanding size) from it & D40, which are nice, mostly SOOC JPEGs; yes, one can do better with what I’m using, even, before the **gear** becomes the limiting factor (in my case : skill & tools esp. in PP). I recall Ming answering that he’d print as large at 12×18″ with the 4mpx of the Nikon D2H.

  5. Thanks for the updated review, Ming.
    Do you have any plans on reviewing the Panasonic GX7 as well?
    Cheers.

  6. Thanks for the excellent blog Ming!
    May I have a question? I’ve got E-M5 too, because of you have been used the E-M1 may be you can tell me are there any difference between the SD card writing speed. I hate that when I take some picture continously then I can’t play back the pictures during the machine is working to save the pictures. And these the saving time usually is too slow (SD card type: Sandisk Extrem pro 16gb 95mb/sec, or 45 mb/sec).
    The other question is that it possible to set the ISO value automaticaly (depending the objective focal lenght. For example: 1/Focal lenght)

    Sorry for my English.

    Bálint

    • I was using the 95mb/s Extreme Pro cards; the camera still can’t play back in parallel with writing (as the Nikons can) but it dumps a full buffer very, very fast indeed. I tried it with a variety of cards – it would appear that the cards are the limitation, not the camera.

      You can set AUTO-ISO; the camera changes minimum shutter speed depending on the lens attached. But it’s not quite 1/focal length, usually a bit faster.

      • Glad I that is exactly the card that has dropped through my letterbox.

        • Thank you for Your quick answer.
          May be these question is a little bit strange, but it is important to me, because I had Nikon system before the Oly:
          So my last question is that there is a 2×2 dial “system”. At first glance that is not seem to be complicated, BUT: is it possible to programming the two dials, that one of them I can change the ISO, the other one I can change the aperture? (Without to use the lever switch (1,2 mode) or any plus button paralell)

          I really love this opportunity in my old system, and missing to me very much in the new one (E-M5).

  7. Ming, I’m really enjoying your reviews and critical info. They should have gifted you the E-M1. No? Strongly considering purchasing an E-M1 as a third body to compliment my two OMD E-M5’s. Will I need an adapter to use my Zuiko M43 lenses?

    • No, no adaptor required.

      If they’d given it to me, would people still think I was objective (regardless of the fact that I would be)?

      • Yes. That could have been done after the review. So, do you see yourself purchasing an E-M1? p.s. Thank you for the adapter note. One another note: Photographed a wedding exclusively with the OMD E-M5 {two bodies} recently. The only niggle / issue I had was the camera falling asleep and not waking up. It appears that you can only set the camera to wait 5 mins. before going to sleep, and there seems to be no override for this, which is problematic. Any advice for you or your readers?

        • As noted in the full review, I’ve ordered two E-M1s; one for video work and one for stills.

          Lag: I turn my cameras off between shots to conserve power anyway, so it’s a non-issue. I do believe you can turn off standby completely, though.

          • Oh, I missed that part of you ordering two. Good for you. I’ll most likely order one after a bit too. Thanks so much for your help. I’m visiting with an Olympus rep tomorrow to see the E-M1 in person and to resolve the narcolepsy issue. Cheers, Ming.

          • I never turn off my Canon bodies, they just go to sleep after a minute. I read in the E-M1 manual that it does the same thing. Is there really a significant benefit to turning it off?

            • The sensor is still powered on to some degree, since it’s required to form a VF/ LCD image with all mirrorless cameras. It seems to make a significant difference to battery life.

  8. My brother and I have both pre-ordered the E-M1 after reading multiple reviews and we both owned the E-M5. Really love where Olympus is going with their camera upgrades and bring out Pro grade lenses. Thank you so much for your detail review of the E-M1. Your thoughts and reviews contribute greatly in our purchasing decision.

    Wondering what the E-M6 would be since E-M1 technically is not the E-M5 replacement. I imagine it will have the same size as the E-M5 but will it have the same tech and more than E-M1? If so, it would be very difficult to pick which to buy at that point.

    • Wondering what the E-M6 would be since E-M1 technically is not the E-M5 replacement. I imagine it will have the same size as the E-M5 but will it have the same tech and more than E-M1? If so, it would be very difficult to pick which to buy at that point.

      Given that the E-M1 isn’t even available yet, perhaps the speculation is a little premature…

  9. Paul Jobson says:

    Thanks for the excellent blog Ming, keep up the great work. M43 Is looking more and more appealing but I wonder whether it is a dead end system. DSLRs still have a lot of room for development – Sony’s Alpha system has shown the way. The FF sensor has a lot of room for development with Canon rumoured to have an 80MP sensor under development but how far can Olympus push such a small sensor 24MP? Seems OK now but with 4K displays becoming the norm in the near future maybe not. I’m seriously tempted but just wonder whether it is a blind alley.

    • I think we need to consider a few practicalities to judge sufficiency:
      1. How big do you print?
      2. Is your shot discipline getting everything out of what you’ve got now?
      3. Do we really want to handle such big files?

      • Paul Jobson says:

        Good point on sufficiency. I’d agree that sufficiency is the best way to judge it. I tend to think in terms of absolute quality rather than sufficiency.
        To answer your points though 20X20 was the largest print I’ve ever printed although that was from a 12MP D90 ;-) Probably more likely would be an 8×10 though.
        I currently use a D800 and after nine months of deliberate concentration on my shot discipline I feel I’m getting the best out of it.
        Being an amateur processing time isn’t really an issue although storage is becoming one!

        In your opinion though – and accepting that the EM1 Is sufficient for most people’s needs – where do you see Olympus taking the system from here?

        • Hmm, tough question: I think the lens system is what needs attention first, and they’re doing that by filling out the pro zoom gaps; next up will be the special purpose glass like superteles and tilt shifts; somewhere in there we’ll probably need a second generation of PDAF-based sensor. Keep the size the same, though.

          • Paul Jobson says:

            So you this system being positioned as a replacement for traditional DSLRs with super teles etc. Intresting. I may yet take the leap but as a complimentary system to a FF DSLR at least for now.
            Thanks for taking the time Ming and keep up the excellent blog.

      • Yes, yes, yes! How big do you print? It’s a concept lost on many. I saw a post the other day from someone who said they wouldn’t buy the new E-M1 because the camera made it very difficult to shoot “selfies”. (I am not making this up) That guy probably shouldn’t waste his money on anything beyond a new iPhone.

        Regardless of what some of the FF obsessed crowd says, I think that 16 vs 22 mp is really not very different in most applications — as long as one uses a very good quality lens. I’ll trade the one stop high ISO advantage that my Canon 5D3 seems to have over the E-M1 for the Olympus IBIS and size/weight advantage.

        The other half of photo tech that is rarely discussed is the ever-improving state of printing technology. Some photographers don’t quite understand the difference between input and output resolution on large format printers. The big Epson and Canon aqueous printers can take an image file that has a resolution of only 100 dpi and internally upres the file to either 600 or 720 dpi output resolution, delivering beautiful results for gallery display.You can print the 4608 pixel-wide E-M1 files almost four feet wide and see virtually no pixillation from normal viewing distances. I make a living selling prints that size. Images that started as 12 to 22 megapixel JPG files. Most review web sites assume that you must feed a printer a 300 dpi image. That’s just not true with today’s large format printers.

        • If you really must shoot ‘selfies’, then it’s really not that difficult if you have some basic concept of how the camera works and what angle of view is.

          16 super-sharp MP are better than 22 slightly blurry ones; the stabilizer really puts the E-M5 and E-M1 into the next level simply because it’s so much easier to attain the full image quality potential of the camera; something which is very difficult with, say, the D800E.

          The confusion is between PPI and DPI: one pixel is not one dot. With a 12-color printer, one pixel is really 12 dots – so 300PPI is more like 3600dpi!

          • Timur Born says:

            To be fair, any camera without a fully articulated screen makes it hard to shoot towards yourself. This is *not* only a problem of selfies. Try to take a picture of a kid sitting at a table which is facing to a wall, or a kid looking the railing of a balcony. One can practice getting the angle right, but that’s something that works better with higher resolution sensors that allow more cropping.

            And when I got the E-M5 new I took five consecutive single shots of me and my son looking over the balcony railing, four where grossly overexposed and only the last was not (all withtin millimeters of the same framing). May have been a firmware bug with accidentally using Face Detection plus center-weighted metering, but still the shots where useless because I only noticed afterwards (no screen to check settings or anything while shooting).

            WIFI may help to workaround some awkward angle situations, but it also needs another hand/hold for the external screen.

    • 4K UHD is a resolution of 3840 pixels × 2160 lines (8.3 megapixels, aspect ratio 16:9) and is one of the two resolutions of ultra high definition television targeted towards consumer television. Full Aperture 4K which is not even a standard yet is 12mpx.

      If a camera can do 16mpx, you still cannot see every individual pixel on the screen.

      4K Televisions and normal consumer prices are still 5 to 7years away and broadcast TV over cable can barely do standard HD today without screeching to a halt. Blue-ray cannot handle 4K as it stands today.

      We can already print more dots than we could with film and the number of people who actually print is going down because there is only so many pictures you can hang on the wall and we are a mobile global society to there is only so many pixels you need for a cell phone or tablet.

      The whole pixel war is very overblown by consumers and photo geeks alike. There are very real world practical issues that make even 24mpx useless. Unless you are doing something super special, such as some billboard on the side of a building, you probably from a practical point view can barely take advantage of 16mpx. Many of the really nice LED displays in Times Square NYC, are not even HD.

      By continuing to ask companies for more megapixels we do a disservice to ourselves because there are so many other areas that should be focused upon these days and megapixels is not one of them. We have plenty of pixels…really we do

      • Agreed. Hand holding at 2 seconds via the 5 axis IBIS is a far more interesting frontier for photography for me than adding more megapixels at this time. When new lens designs start out-resolving modern sensors, it may be time to revisit megapixels, but Moore’s Law has all but eliminated any difference in IQ that the larger sensors have enjoyed over m4/3.

        Short depth of field is where the haters have been beating their drums loudly. As if the only way to get short DOF is a bigger sensor — it’s not. Yes a larger sensor gives a more shallow depth of field, but so does a larger aperture. High end M43 lenses almost always have a larger max aperture than similar (or larger) size lenses for APS-C or 35MM. The other, less talked about way to get shallow depth of field is to simply move away from the subject and shoot with a longer lens (duh). They never seem to mention either of those.

        The other side is that it’s often desirable to have more DOF at a given focal length.

        • Actually, I believe there’s a right amount of bokeh for every subject; too much removes context. I almost always have to stop down with full frame. Do I find the extended DOF of M4/3 limiting? No, hardly.

          • This. I fully understand the advantages of the full frame sensor but I realize that in practice I already stop down a lot on APS-C. Using m43 was a bit the same as using a P&S, you tend to work more with the lens wide open. There are some applications where a very shallow DoF is better but for me most of the time I want longer DoF.

            • That actually means M4/3 works to your advantage as you can go with a larger aperture for the same DOF; more light gathering ability. And the lenses seem to be much better as a whole than FF/APS-C as they were all purpose-designed for the system and telecentric…

  10. Steve Montalto says:

    Promising initial results. Very bummed that you had to give up the loaner already!

  11. Nice to see there are no immediate issues with ACR processing of these raw files. Usually that is the case, but there seems to be a lingering controversy regarding raw files from Fuji X series cameras with their unusual color filter.

    • I think Adobe just hasn’t bothered to spend enough time on the demosaicing algorithm, or Fuji isn’t willing to release all the secrets to the sauce. The E-M1 should be pretty similar to the E-M5, and the overall tonal palette is pretty consistent.

      • Rico over at Fuji Rumors ran a blind poll comparing RAW converters and Fuji’s JPEG engine. Basically they’re all now doing a decent job, most outperforming the JPEG engine but there’s no stand-out winner. Capture 1 did the best on average but never had a majority i.e, it’s a close call. Aperture also faired well across images.
        We all know how Bayer affects IQ, well Fuji’s heralded 6×6 CFA has a greater affect, as most anticipated. I’d imagine that X-Trans would benefit greatly from increased MP but for now Fuji and 3rd party RAW converters are working at full capacity, there’s no laziness nor conspiracy. X-Trans just produces flat images for me and you can’t expect the sharpness that people are accustomed to from RAW Bayer.

        • I don’t mind the flatness; the edge acuity is a bit better, but the artefacts are what kills it. ACR is definitely the worst; the problem is the rest do not integrate neatly into a PS-based workflow, nor do they offer the same tools.

Thoughts? Leave a comment here and I'll get back to you.

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 26,663 other followers

%d bloggers like this: