OM-D vs 5D Mk2 Test

[Binone]

I've been using Canon equipment for years with my main camera being the 5D Mark2. I love the IQ I get from that camera and the Canon "L" glass. However, what I don't like is carrying that equipment around when I hike or travel. So, after reading all of the reviews, I bought an OM-D and, I must say I'm very pleased with the results, and I love not carrying something that feels like a sack of bricks when I go on a hike.

The big question for me is: Do I keep or sell the Canon gear? I decided to do a test to determine if there's any meaningful difference between images shot with the 5D2 vs the OM-D. I realize that this isn't a fair test because I'm comparing a 16MP micro 4/3 sensor to a FF 20MP sensor. But, so what, to me what's important is the IQ. Will I keep the Canon for use where I don't have to carry the equipment for any great distance, or not?

The only important factor for this is what the images look like - can I see a noticeable difference on a print. The largest I can print is 22 x 24, so more than that isn't important, but I also wanted to see what the IQ is if I really zoom in. So, here are the test parameters:

For both cameras: Tripod mounted and IS turned off. RAW files processed in ACR with default settings. No sharpening or any other editing made other than cropping.


For the OM-D: Panasonic 12-35 f/2.8 set to f5.6, ISO 400. Lens set to 50mm. Exposure metered at 1/2,000.


5D2: Canon 24-105 f/4 L set to f/11. Lens set to 50mm. Exposure metered at 1/640. I used f/11 because that would result in approximately the same dof as the OM. The Canon was also set to ISO 400. I thought about setting the OM to ISO 200 since I typically shoot it at a wider aperture than the Canon, but decided to keep as much as possible equal.

Exposures: Much as been written about the OM-D's ISO not being accurate. With a 2-stop difference in f/stop, the OM-D's exposure should have been 1/2,560. The resulting OM-D image is about 1/8 to 1/4 stop darker than the Canon's. So, there may be a difference of about 1/4 - 1/3 stop between the ISO calibration of the 2 cameras. I have no way of knowing which is correct and, frankly, it doesn't matter.


Here is the full image from both Cameras - Canon on top:




Full image from the 5D, reduced for web use.






Full image from OM-D, reduced for web use.



From the above, you can see that the OM-D's image is a little darker, but they are remarkably similar.

Taking a closer look, here's a crop from the center section of the image. Again, Canon on top:




This is a small section out of the center of the 5D's frame






This is approximately the same crop area from the OM-D. Again, no differences that I think you'd notice in a print.



I'm not a "pixel-peeper", but to satisfy those who are, and my own curiosity, here's an even tighter crop. This was enlarged in PS6 by the same amount for both cameras. Again, Canon on top:




This is a detail from the 5D shot. Notice the slip numbers on the white bins, also the shadow detail and noise in the blue hull.




Here's the same detail from the OM-D shot. Again, look at the slip numbers on the white bins, also the shadow detail and noise in the blue hull.

These last shots, which are significant enlargements (probably representing 200% view) do show a difference in favor of the 5D2. There's less noise visible in the blue hull and the numbers on the white bins are a little bit sharper. However, I doubt that these differences would be visible in a print, which is really what matters.


This test turned out a lot closer than I expected it to be. If I had to pick a winner, it would be the 5D2, but a margin so small that I doubt that the differences would show in a print. Also, since there's a 2 stop difference in dof between the 2 cameras, with f5.6 being roughly the same on the OM-D as f11 on the 5D2, I actually end up using the OM-d at a lower ISO than the 5D and that would have made the comparison even closer. In fact, I doubt that I would have seen a difference at all.

When it first came out, I tried a 5D Mk3 and found that in good light its IQ was so close to the 5D2's that there was no point in spending the money on it. So, I believe that the above comparison would hold true for a 5D Mk3 as well.

What this doesn't tell me is what difference there is, if any, at high-ISO in low light. I have shot the OM-D at ISO 6400 and the results were much better than I expected, but I haven't done an exact comparison, like the above. That's a test for another day. But for now, the camera I grab when I walk out the door is the OM-D.

 

[nzmacro]

IMO, if you have to pixel peep this much just to see a difference, then goodbye Canon if m4/3, lenses and AF work for you. I actually use NEX, but to see this difference is interesting, well its more of a lack of difference I guess.

Excellent to see and well done. I would certainly be looking at the 5D MK2 sideways and going..... hmmmm ;-)

All the best.

Danny.

 

[daddyo]

Thanks for posting this -- it is interesting, but not surprising to me. I believe you are correct regarding the details that would be visible in print. I'm sure that once you start getting to the 30X40 print size differences would become noticeable, but I also know that an E-M5 print that size would look outstanding.

I am very impressed with the image quality from my E-M5 -- especially given the size and weight

God Bless,

Greg

www.imagismphotos.com

www.mccroskery.zenfolio.com

www.pbase.com/daddyo

 

[JimLong]

I made a similar comparison a few months back. I have a Canon 5D with the 24-105 L. I had just received my Panasonic 12-35 for my GH2 and was, like you, hoping that this lens would give me results comparable to the Canon. Since I was only interested how the prints compared, that was the only way I evaluated the output. I printed at 8 1/2 x 11 on an Epson 3880, so the difference in resolution of the two cameras was not really a factor; and I didn't do any comparisons at high iso. My results were very much like yours. It was very difficult to see any differences and any differences that I saw favored the Panasonic. My son was visiting at the time with his 5D Mk III so I did a comparison with the 24-105 on that body and once again, any slight difference seemed to favor the Panasonic.

I could have probably found situations that favored the 5D bodies (high iso, high dynamic range subjects), but I was primarily interested in comparing the lenses. In terms of things like color rendition, micro contrast, etc., the Panasonic lens, in my opinion, is as good as or better than the Canon.

I am now waiting for the GH3 review on DPR to help me decide between replacing the GH2 with the GH3 or the OM-D.

Roger Cicala, of LensRental, did a "real life" comparison of the OM-D/12-35 and the5D Mk III/24-70 and came to very much the same conclusion.

 

[Binone]

I'm not sure where the big difference between the Canon and OM will be. And, I'm coming to the conclusion that there might not be one. However, I do know that the Canon's tracking focus works pretty well and, while I haven't tried it, the OM's is supposed to be poor. That said, for the kind of photography I do, that won't make much of a difference. It's starting to look like there isn't much point to keeping the Canon.

 

[NumberOne]

Thank you very much for posting a "real" comparison, specially when the opponent to the mFT system is a well known FF contender (combo)...

I also doubt most people could identify the camera/lens set when seeing (separately) the respective (up to A3 size) prints ...

However, these shots were taken with "fair/easy" light conditions.

Quite curious to see how things work out when you need to push ISO into 1600-3200 territory to overcome bad/insufficient light...

Best regards,
Pedro

 

[Steve_]

If you're using the OMD's stabilization you are going to get at least 2 stops every time. About the advantage of a good full frame. So for static subjects, noise in bad light isn't a comparative weakness of the OMD, in my experience.

I shoot 1/10th-1/13th of a second handheld with the 12-35 across the full focal length range all the time with really crispy results. If you can even see the subject, ISO 800 is usually enough. Maybe you can with a 5Dx and 24-105L, but I can't.

 

[Binone]

You're right - the lighting was great so even ISO 400 wasn't really necessary. At high ISO, I'm expecting a more noticeable difference, but this test surprised me, so that one might also. So far, images from the OM-D up to ISO 6400 seem quite good, but I haven't done a direct comparison.

I had been intending to keep the Canon for instances where I'm not far from my car, and just use the OM when I'm going to walk far. Now, I'm not so sure. The Canon equipment commands a pretty good price on eBay and I can really fill out my OM kit with that money. So far, I've only got the body, Panny 12-35 f2.8, 7-14 f/4 and the Oly 60mm f/2.8 macro. To complete the kit, I need a couple of primes, a long lens and another body.

 

[NumberOne]

Steve_ wrote:

If you're using the OMD's stabilization you are going to get at least 2 stops every time. About the advantage of a good full frame. So for static subjects, noise in bad light isn't a comparative weakness of the OMD, in my experience.

Nobody is talking about weaknesses of either camera/system, and the same goes for stabilization, which you can also include in many (Canon and third party) available lenses, FF and other.

I shoot 1/10th-1/13th of a second handheld with the 12-35 across the full focal length range all the time with really crispy results. If you can even see the subject, ISO 800 is usually enough. Maybe you can with a 5Dx and 24-105L, but I can't.

The rule of thumb of speed=focal length is quite overstated - read mistaken - for the available resolution of the actual digital photo systems... Even for very steady/trained hands I would advise using at least the double if you don't want blurred pictures... Yes, it's Pixel Peeping, but most of the defenders of that "golden rule" certainly don't see their shots at 100%, or enlarge/print sufficiently big... If they did, they would see what I mean!


Thus, IMO, 1600-3200 ISO territory is quite useful - read mandatory (or at least, advisable) - on many occasions, which not always include insufficient light.

Hope, you get my point!



Best regards,
Pedro

 

[Binone]

I've made large prints from images shot at ISO 1600 and 1/8 sec exposure with Canon's lens-based IS. I'm not sure that there's a big advantage to in-body stabilization vs lens IS, except if you have a lens without stabilization. The secret is to shoot several frames and usually one of them will be OK. However, if something is moving - people, cars, leaves blowing in the wind - whatever - no stabilization is going to help. You have to use high ISO. What micro 4/3 allows, however, is a 2 stop advantage in dof and the ability to build a fast lens without it weighing 5 lbs. So, I can shoot at f/5.6 and have the same dof as f/11 on a FF camera. So, that 1/15 sec exposure on the FF camera is now 1/60 on the micro 4/3 for the same dof, making the need for high ISO less important.

 

[NumberOne]

Binone wrote:

You're right - the lighting was great so even ISO 400 wasn't really necessary. At high ISO, I'm expecting a more noticeable difference, but this test surprised me, so that one might also. So far, images from the OM-D up to ISO 6400 seem quite good, but I haven't done a direct comparison.

I had been intending to keep the Canon for instances where I'm not far from my car, and just use the OM when I'm going to walk far. Now, I'm not so sure. The Canon equipment commands a pretty good price on eBay and I can really fill out my OM kit with that money. So far, I've only got the body, Panny 12-35 f2.8, 7-14 f/4 and the Oly 60mm f/2.8 macro. To complete the kit, I need a couple of primes, a long lens and another body.

In the same boat, here - weight&size savings without much IQ compromise - but expecting more data about the GH3 to decide; I'm also considering the lenses you already have, but adding the 35-100f2.8, too...

(Also looking at the Fuji XE-1, but still waiting for a RAW Software solution. I'm not convinced with the IQ I've seen - so far - and that many rave about... Looking at DPReview's samples of the X-Pro1 (same sensor) leads me to a similar conclusion/concern!)

Nevertheless, quite curious about a "Low Light/High ISO" comparison, if you have time - and patience - to post.

Best regards,
Pedro

 

[Binone]

I've got some travel coming up and I'm not loading myself down with a sack of bricks (aka Canon 5D and "L" glass). So, I'll only be shooting the OM-D. My objection to the GH3 is that it's almost as large as an entry-level APS-C body and, IMHO, that defeats the purpose. I've read that its video is great, but I only shoot stills, so that's not important to me.

I"m probably going to get the 35-100, but there was a rumor that Oly was going to introduce something in the range of 40-150, f/2.8 and I was waiting to see if that's real. It seems like it isn't real, so the 35-100 is probably my next lens.

 

[NumberOne]

Binone wrote:

I've made large prints from images shot at ISO 1600 and 1/8 sec exposure with Canon's lens-based IS. I'm not sure that there's a big advantage to in-body stabilization vs lens IS, except if you have a lens without stabilization. The secret is to shoot several frames and usually one of them will be OK. However, if something is moving - people, cars, leaves blowing in the wind - whatever - no stabilization is going to help. You have to use high ISO. What micro 4/3 allows, however, is a 2 stop advantage in dof and the ability to build a fast lens without it weighing 5 lbs. So, I can shoot at f/5.6 and have the same dof as f/11 on a FF camera. So, that 1/15 sec exposure on the FF camera is now 1/60 on the micro 4/3 for the same dof, making the need for high ISO less important.

Spot on, and the advantage that contra-acts to some of the shortcomings of the mFT system (narrow DOF is the most talked about) in other (photographic) situations, and that many mFT detractors don't even (know to) put into equation...



Best regards,
Pedro

 

[Steve_]

Don't underestimate the OMD's IBIS; it really does help offset the smaller sensor's contribution to noise for static subjects. And it positively smokes the 24-105L's.

Handheld, under weak indoor fluorescent lights on a high ceiling

But I just don't look close, do I?

 

[marike6]

In a daylight image like your example, the differences will be less clear. But in low-light, there will be a much bigger advantage for the 5D2.

To put it in perspective, on DxOMark low-light score:

5D2 1815 ISO

EM5 826 ISO

So that's over a full stop better high ISO performance. With the 5D3 it's more than that, and with a camera like the D800 or D600, there is almost a 2 stop difference in high ISO performance. For many types of photography that is huge.

The other issue is lenses. m43 has some good lenses, and more each year, but Canon and the EF mount has literally hundreds of lenses to choose from. And L glass, what can I say that hasn't been said? There is something truly satisfying about using such high end lenses, and it goes beyond sharpness. Color/contrast, wonderful metal tank-like build quality, blazing fast AF, etc. It's the same using the higher end Nano Crystal Nikkors. Talk about pride of ownership.

Lastly, if you're into video at all, the 5D2 and 3 are both excellent cameras with some wonderful features, and great looking images. My point is, I would urge you to hang onto your 5D2. Even though I replaced mine with a D800 which betters it in so many areas, there's a lot a still miss about my 5D2. Amazing camera. I also have an X-Pro1 and GH2, that I love to death. But of my three cameras, the one I would never part with is my D800. It can be a bit of work to manage in the field, but nothing worth doing is ever easy. Cheers, Markus

 

[NumberOne]

Binone wrote:

I've got some travel coming up and I'm not loading myself down with a sack of bricks (aka Canon 5D and "L" glass). So, I'll only be shooting the OM-D. My objection to the GH3 is that it's almost as large as an entry-level APS-C body and, IMHO, that defeats the purpose. I've read that its video is great, but I only shoot stills, so that's not important to me.

I haven't - yet - used the GH3, but with some lenses inside the photo bag, the size of it will be mostly irrelevant, I guess; in a previous encounter with the OM-D EM-5, I also found the camera to be a little to small for the "big" mFT zooms, and the same went to the size of most buttons. No deal breaker, of course, and nothing you couldn't get used to!

I"m probably going to get the 35-100, but there was a rumor that Oly was going to introduce something in the range of 40-150, f/2.8 and I was waiting to see if that's real. It seems like it isn't real, so the 35-100 is probably my next lens.

...It seems it's a great lens and the f2.8 a real plus for such a small/handy combo...

Best regards,
Pedro

 

[NumberOne]

Steve_ wrote:

Don't underestimate the OMD's IBIS; it really does help offset the smaller sensor's contribution to noise for static subjects. And it positively smokes the 24-105L's.

Handheld, under weak indoor fluorescent lights on a high ceiling

But I just don't look close, do I?

Come on... Don't put my words out of context!

I find the IBIS a great system and an advantage, but as I said before, stabilization can be achieved by other methods and systems, doesn't apply to every image situation, and the comparison of the OP had (and has) nothing to do with this matter.

My comments were just a "reminder" about that fact - nothing personal and nothing more.

Hope you understand...





Best regards,
Pedro

 

[Steve_]

Come on... Don't put my words out of context!

I find the IBIS a great system and an advantage, but as I said before, stabilization can be achieved by other methods and systems, doesn't apply to every image situation, and the comparison of the OP had (and has) nothing to do with this matter.

My comments were just a "reminder" about that fact - nothing personal and nothing more.

In the most condescending way possible. You also went on to state that people shooting at even the golden rule only fancy their stuff sharp because they aren't looking close, so I'm not sure how much of a test you've given any of these stabilization schemes or implementations.

I'm just reminding you that your "fact" may just be opinion. I've tried many IBIS and lens-based stabilization systems, and I find the OMD's to actually be worth more, by at least a stop. I've backed this up by showing essentially blur-free result at 2.5 stops below the golden rule, and with some focal length involved. I stated in my original post that this is only a factor for static subjects, but it is just that - a factor for static subject that does indeed help mitigate the noise disadvantage of the smaller sensor.

Several posters in this thread have wondered aloud how the OMD would fair at high ISO in a controlled test like that conducted here. My point is that a 'fair' comparison isn't fair at all for the majority of my work, which typically involves static subjects and terrible lighting. The combination of the OMD's very effective stabilization and the ability to couple it with lenses as short (both focal and barrel length) as the 14mm Panasonic does indeed allow me to keep the ISO down as compared to full-frame alternative setups. And even when you whip out the Oly 75, the advantage doesn't fade nearly as much as other schemes and implementations.

 

[Binone]

I've got to add my 2 cents worth here. First of all, in my original post, IS was turned off in both cameras. I find the Oly IBIS to be great, but when I mount the camera on a tripod, it doesn't matter whether it has lens IS, or body IS - both get turned off.

I don't think that anyone was minimizing the value of IS. I've got a 13 x 19 print that was shot at 1/8 sec and is extremely sharp. It couldn't have been done without IS. If you're shooting static subjects - it's great. If the subject is moving, then you need good high ISO performance, and the benefit you get from 2 stops of dof with micro 4/3.

 

[Binone]

There's an old saying that goes: "the best camera for any situation is the one you've got with you". I agree with everything you wrote about the quality of the "L" glass - it's superb. But, I'm no longer physically able, or willing, to haul a heavy body, and bunch of even heavier lenses, while I spend a day walking around. I find it very uncomfortable, tiring and, because I've got back problems, painful. So, prior to the OM-D, I used a NEX-7 with the Zeis lens, or a Canon S100 and, more recently a Sony RX-100 whenever I left for a walk. The Sony was great, but there's only one good lens for it. So, bottom line is I never had my best camera with me at times when I wanted the best IQ. The OM-D has changed that. I think that the Lumix 7-14 and 12-35 are the equal in quality of Canon's L glass. But they're just a fraction of the weight. So, now when I'm out walking around, I've got great equipment with me and my back doesn't hurt. Will my images compare to those from a D800 - probably not, but in most prints, where I'm not cropping out a small section of the frame, it may be difficult to see the difference. But, I wouldn't have the D800 and the heavy glass with me, so what's the point. You can't shoot everything from within a mile of your car.