I got the E-M5 II in for testing today, and of course the first thing I had to do was test the 40MP mode. Interestingly enough, Photo Ninja can read E-M5 II RAW files, including the 40MP file, but I don't think it demosaics the 40MP files correctly, as there is far more data in the JPEGs than there is in the PN decoded RAW...can't wait for LR support (or C1).
Anyway, Here are two quick shots, both with the 12-40/2.8 at f/5.6, with a standard 16MP file next to it. I uprezzed the 16MP file to the same pixel dimensions to make comparing the amount of real data in the 40MP file easier. With a good lens, there is a very real advantage. In my little experience, though, I have to say, the camera needs to be absolutely ROCK solid...any movement whatsoever yields a nasty image. The first image shows water, which at 100% on the 40MP file has odd artifacts, though if you resize that image to 16MP, you end up with no visible water issues (in this case), and an extremely sharp 16MP file (with more detail and sharpness than the standard 16MP file), so even in cases like this where there are some moving elements, you'd get a benefit, though you get much more of one if there aren't moving things. Anyway, on to the images.
Yowzers. Your test matches many others I've seen. When done "right" with the "right" scene and the "right" lens the results are pretty dang impressive. Thanks for sharing your usual careful comparison. I think come summer I'm going to be picking up an EM5II as a way to avoid going down the A7 rabbit hole. The EM5II 40MP mode limitations happen to not be an issue at all for 99% of my shooting, so it seems like a good fit.
kwalsh wrote:
Yowzers. Your test matches many others I've seen. When done "right" with the "right" scene and the "right" lens the results are pretty dang impressive. Thanks for sharing your usual careful comparison. I think come summer I'm going to be picking up an EM5II as a way to avoid going down the A7 rabbit hole. The EM5II 40MP mode limitations happen to not be an issue at all for 99% of my shooting, so it seems like a good fit.
Good thing about the 40MP mode - you can save the original 16MP raw and use it as a "background layer" for simple stuff like this water, since it doesn't have much detail to begin with. A bit of work, but should be doable, at least to avoid artifacts.
How good is the handling, ergonomics and buttons (pushing, layout) compared to the old version? I had the MK1 before but it was a deal breaker with the above issues. I am thinking about getting this new version if it is worth the price difference.
The buttons are a bit easier to press and the dials are in far better position. Still cramped, though, and the grip is tight. I'd plan on getting the grip with it again, like I did with the original.
As far as ergonomics go, after shooting with it for a day, I have the following observation on the new, more contoured grip of the E-M5 II. It's a more secure grip because of the little notch they've built into the front, but it's not nearly as comfortable. I'm talking about the bare camera, not with any add-on grips. That notch forces your fingers lower and tight to the lens, making it feel far more cramped than on the original. It also is a fairly hard grip, so there's no cushion on that portion. My fingers have started to go numb after carrying it for a while. (To be fair, I have a few nerve issues from my elbows down, so I've had this issue on several cameras).
At least for my hands, the horizontal portion of the BLD-8 (which thankfully Olympus is making available separately from the vertical grip for less money) would be a must if I were to purchase this camera. On the original E-M5, I found it to be a nice handling improvement, but not necessarily essential.
Hrm, good to know. I might just put the brakes on an EM5II purchase until I see what the next EM1 is like. I love the EM5, even with its tiny squishy buttons and minor grip, but if the EM5II is a regression, I'm sure I can wait a while longer for an upgrade
I played around with the em5II in the store, I also used to have the original em5. It is an improvement in every way, but it is still very much like an em 5. It isn't going to replace my em1, but I don't know how much more can Olympus do to improve this design. You can't have something this small, with this many exterior features and make it feel like you are holding the most comfortable camera. However, comparing it to those boxy Fujis, Ricoh, Nikon etc, the em5 is a pleasure to hold. Olympus did a fantastic job.
cputeq wrote:
Hrm, good to know. I might just put the brakes on an EM5II purchase until I see what the next EM1 is like. I love the EM5, even with its tiny squishy buttons and minor grip, but if the EM5II is a regression, I'm sure I can wait a while longer for an upgrade
Thank you for this Jman. I know the 40mp shots have there limitations and restrictions as subject matter, yet can see for the scenario that suits, this too be of great value, making the versatility of the OM-D series to work for many photographers.
Oh...nice surprise too: I installed the HighRes RAW Plugin provided by Olympus for Photoshop, which can process with 102MB RAW files from the high-res shoot, and these are 64MP, rather than 40MP (9216x6912). I don't know how much of an advantage this is. With the 12-40 on a perfectly still scene indoors, I see pretty good pixel level sharpness at 64MP. In outdoor scenes, absolutely ANY movement results in visible artifacts with the 64MP RAW. You need stock still perfection.
Here's a non-perfect example, with a 16MP RAW processed in Photo Ninja (the first RAW in the 8 shot sequence is saved too, under .ORI), the 40MP JPEG (center) and the 64MP RAW (right), normalized to 64MP. There might be very slightly more detail in the RAW, but it's really splitting hairs vs. the 40MP JPEG.
? If nothing's moving, there's a huge advantage to having more MP. I'm saying I don't know how much more detail is in the 64MP RAW than the 40MP JPEG. Both have massively more detail than the 16MP non-high-res shot.
What about cropping? An extra 24mp is sure to come handy no?
Also, lets say there is a light breeze and there is grass in the shot - at what size prints do you think one would notice "movement"? The reason I ask is because I know that you are a fuji fan and I have seen some 100% crops of the smearing one will see (foliage) with fuji files and Lightroom for example? Does that stuff actually show up in prints or do only pixel peepers worry about stuff like that? Thanks.
When I say I don't know how much it helps it's because at 40MP you're getting most of what the lens can produce in combination with diffraction effects. The 64 MP file is, in most situations, only marginally more detailed in actual detail vs the 40 MP down sampled file. I need to look and test more, but you can see the marginal increase in the crops I posted above (I edited my post above your original question if you missed it.)
The issue with movement in the High Res files is very different than any issues with Fuji (which have never really posed an issue for me). When there's movement in these images, it's rendered with diagonal hash like artifacts. If you shoot RAW with this mode, though, the camera saves a non-high red file too (the first shot in the sequence), so if you have movement that ruins a shot, you can always use the 16Mp original RAW And have no issues. When you do have a sharp lens, with a static subject, the high res mode does an amazing job.
It uses a similar principle but unlike E-M5II which relies on very precisely calibrated shifts, here you shoot handheld and the natural movements of your hand results in various slightly shifted version of the same image which are then combined to create the higher resolution (and less noisy) output image. The good thing is you don't have to worry about being rock solid. The bad thing is if you are rock solid, it may not work at all
Apparently they have stopped developing it any further so the profiles for newer cameras are missing. However for my A6000, I was able to use the profile for NEX-6 and it seemed to work well.
I may try that. I've tried the manual version with my a6000 to yield 90Mp files with Photoshop, doing what Photoaccute does, and my impressions are that the E-M5 IIs system is superior to that method (it certainly takes less time and effort).
Jman13 wrote:
The issue with movement in the High Res files is very different than any issues with Fuji (which have never really posed an issue for me). When there's movement in these images, it's rendered with diagonal hash like artifacts.
Thanks for the explanation - is this artifact (like the water scene) visible on your screen or only when doing 100%?
I am looking at the full picture on a 5K Imac and the picture is still too small for me to notice. Another thanks.
