Luminous Landscape Forum
Equipment & Techniques => Mirrorless Cameras => Topic started by: Dan Wells on January 22, 2019, 11:44:37 pm
-
I've just reposted this (my own post) slightly edited from the battery thread, at John Camp's suggestion, trying to put a better title on it.
The full specs just leaked on 43rumors, and nothing bad, but nothing groundbreaking. The version 43rumors has looks Google translated, so my apologies if I mis-interpreted anything from Google gibberish, especially about external power... I hope that isn't the actual English version of the press release :)
Here are the power specs...
870 shots on two batteries (CIPA) - almost exactly the same as the E-M1 mk II, which is 440 on 1
2580 shots on two batteries under unspecified low power test conditions (probably similar to how people get 900 shots out of a Z7)
Will run on either 1 or 2 batteries (no mention of whether some fast modes require the second battery, so not clear if it can use them in parallel for higher voltage).
Charges batteries in-camera from either an Olympus AC adapter or USB-C - charges with power off.
Also seems to power the camera from USB-C (apparently without charging).
Here are some other interesting specs:
18 fps, 20 mp (widely publicized)
The specs say it's a sensor "with a new coating", although earlier 43rumors posts have said it's an actual new sensor. If it's the tried and true sensor, PhotonstoPhotos has the E-M1 mkII with nearly two stops less maximum photographic dynamic range than the comparably priced Nikon Z7.
If it's a new sensor, maybe it picks up a stop of DR and becomes competitive with the best of APS-C, while still lagging the best FF cameras?
About the same width and depth as a D850, although significantly taller.
Within a SD card of exactly the weight of a D850.
World's best image stabilization - they claim an extra stop above the already rock-steady E-M1 mk II. The best performance (per Olympus) is on the 12-100 mm lens at 100 mm (they're claiming ~1 second, by my count)
50 MP high-res mode handheld (80 MP on tripod)
Built-in "neutral density filter" with five settings - this appears to be a computational effect, not any sort of reduction in light reaching the sensor.
Has twice as much video bandwidth in Cinema 4K (4096 by 2160) as in regular 4K. Cinema 4K is a GH5 like bitrate.
The EVF is good, but not one of the most modern "super EVFs" - 2.36 million dots (like a Fuji X-T2 or E-M1 mk II), while some modern Sonys, the Nikon Z series and the newest Fujis use 3.7 million.
IPX-1 weather resistance - the first time I've seen any camera company claim an IPX rating. Looking at the IPX-1 specs, the Olympus E-M1 original and mk II, Nikon D850, Z series and D5 (plus predecessors), Fuji X-T2, X-T3 and X-H1, Pentax (most of them) and Canon 1D/1Ds/1Dx series are all close to meeting it... It's basically heavy rain falling vertically for 10 minutes, and none of the above should have any problem.
As far as I can tell, an interesting, but extremely niche camera... Unless they've made huge improvements to dynamic range (and have a perfect stitching algorithm), the 50 MP multi-shot will be far inferior to any number of smaller, lighter 40+ MP cameras with FF or beyond sensors. It's actually larger and heavier than the GFX 50R or the X1D (although the lenses are much smaller than medium format lenses). The Olympus PRO lenses are large and heavy enough that it's going to be bigger and heavier than a Z7 (including lenses) until you reach significant telephoto focal lengths, and even there the Nikon PF lenses will keep it close.
One practical application I can see is mid-range sports photography (willing to accept significant size, weight and cost, but not a 1Dx or a D5). It has fast AF and an extremely high frame rate, and its overall responsiveness should be superb. The problem is the Sony A9 (with all of the above plus a superior sensor 4x the size) at around the same price when on sale. Right now, the A9 is back up to $4500, leaving a bit of room for Olympus, but will Sony drop it back down to $3500 for the next promotion? Once it's been $3500, there's little reason to think it'll stay $4500.
The application where Olympus has everything else beat is anything where image stabilization exceeds all other considerations. If you're shooting video under absurd conditions (while skiing, etc.), the "gimbal included" E-M1x is going to be very interesting.
It'll probably also grab some E-M1 mkII owners, although many will wait for a mkIII that is closer to what they have in size and cost. I'm not going to recommend it to my E-M1 mkII shooting friend unless he really wants the extra stop of stabilization (or there's a sensor breakthrough). If it had come out while his son was still playing college baseball, the speed would have been a perfect feature.
-
What's the reasoning on why the 50mp handheld hi-res will be inferior to the 40mp FF cameras? Does it involve more than the smaller sensor size of m4/3?
Given these specs, I guess the only reason I'd buy an E-M1x is that I have a lot of m4/3 glass; but I wouldn't buy it anyway because it's way too big, which is the opposite reason that I bought m4/3 to begin with. I *would* like to see an enhanced sensor for GX8/Pen F style cameras, that would make them directly competitive in quality and speed with the Z6 and APS-C cameras, but for my uses, the current crop of m4/3 works well.
It seems to me (and I'm not a tech guy) that for the time being, digital interchangeable lens cameras are settling on two general areas -- ~24mp for fast cameras, ~50mp for high resolution. That may (probably will) change over the years, but if I had a 50mp camera, I doubt that I'd bother to buy a 100mp camera, because I'd see no improvements in photos the size that I print. So I'd wait to buy until my original camera began to wear out. I give credit to Olympus for trying to match the resolution of the top-end FF cameras like the Z7, with its hi-res mode, but again, I don't think most m4/3 shooters bought their cameras for high resolution. I could be wrong.
I currently shoot a D800e and the GX8. Given the news over the past year (like Panasonic going FF) I suspect there are a couple of Z cameras in my future, because that would seem to handle both the size problems I have with the bigger Nikons, and the outdated sensor situation with the m4/3. But I'm in no hurry to switch right now: I'll wait at least until I see the next moves by Panasonic and Nikon.
-
Thanks for the summary (and new subject-appropriate thread!)
It is not going to be for me, but one thing that does interest me is even better IBIS, which also I suppose enables that hand-held high res mode that might also serve me well in some situations (for even more cropping latitude). As an extreme case, a few days ago I needed to use ISO 1600 on my E-M5 for a hand-held shot of a very stationary subject(*) so I have a use for three or four more stops of IS. I want my virtual monopod to become a virtual tripod!
I wonder how the mix of improved video, high res. options, vertical grip, enhanced battery life in a package still less bulky than a 35mm format kit (with appropriate lens choices) would serve for events like weddings.
(*) 300mm lens and the lunar eclipse; us, I could have dragged around a tripod, but I really strive to avoid them!
P.S. You missed its “Z-6 killer” feature; not the 60 FPS mode, but the dual card slots!
-
There are two reasons I'm highly skeptical of the image quality of the 50 MP stitch mode approaching that of true 40+ MP sensors - dynamic range (and related) performance and stitching/subject movement.
The specs on 43rumors support the sensor (probably) being the standard 20 MP Micro 43 sensor - the article listed "new sensor coating", but not "entirely new sensor". If that is the case, the sensor underperforms just about anything else on the market (in an interchangeable lens camera) in dynamic range. It's somewhat over 9 stops when the best current cameras approach 12 stops. There are three factors there - older (non-BSI) design, very high pixel density and high minimum ISO.
If it's actually a new (probably BSI) sensor, it will almost certainly be better than the old standby, but it will still have significant disadvantages compared to other sensors. The minimum ISO has been confirmed to be 200, while the Sony 42+ MP sensor goes as low as 100, and the Nikon variant goes down to 64. Assuming all else were equal, there's a full stop of DR in favor of the Sony, and 1 2/3 stops in favor of the Nikon from minimum ISO alone. It will also have significantly smaller pixels, which may cost more DR. If it's a very modern sensor design, it could pick up a bit from advances in technology since the 40+ MP sensors were designed a couple of years ago.
The second issue is that, absent a completely locked-down camera and subject (photographing art on a copystand is one of the few cases where the artifacts might be practically eliminated), there will always be motion artifacts in any stitched image. Really good stabilization and computational processing can reduce, but not eliminate those artifacts. Olympus' stabilization is known to be excellent, and I suspect their processing is also very good, but it still won't be as good as a single exposure. Pentax's handheld multishot mode has significant artifacts, according to dpreview (https://www.dpreview.com/articles/7242776989/pentax-k-1-mark-ii-dynamic-pixel-shift-mode). Olympus may have a better solution, but artifacts are inevitable.
The E-M1x is almost certainly a very fast camera with a tacked-on high-res mode, not a pixel monster. The question is what the market is for a very fast, very durable camera with fantastic stabilization, but modest maximum image quality at $3000.
The first competitor I would worry about if I were Olympus is their own E-M1 mkII. It's already very fast and responsive, and the stabilization is better than anything other than the E-M1x. Will very many people pay twice as much, and accept a camera twice as heavy, for 18 fps instead of 15, one extra stop of stabilization, a couple of new computational effects (the "ND filter" is purely computational) and some sort of a handheld multishot mode? Especially if it's the same fundamental image quality?
The second issue is the Sony A9, which offers two stops of extra DR and otherwise much better image quality in a similarly speedy camera for a similar price (when the A9 is on sale). Yes, you trade off some stabilization and the best-in class weather sealing for the image quality.
How many photographers are going to find that the E-M1x threads the needle between the E-M1x and the A9 perfectly? For that matter, used Canon and Nikon sports pro bodies are also serious competitors (or even a brand-new D5 or 1Dx II with a used supertele).
-
I am pretty sure Olympus did market research before they gave the go ahead on manufacture. Maybe they have modest expectations. I don't see it, but I don't shoot m43.
-
There are two reasons I'm highly skeptical of the image quality of the 50 MP stitch mode approaching that of true 40+ MP sensors - dynamic range (and related) performance and stitching/subject movement.
... the standard 20 MP Micro 43 sensor ... high minimum ISO.
Of course this and all multi-shot modes are for stationary subjects (as is the IS that it relies upon); but within that category, we have to wait for test results to see how it compares to other 40+ MP options.
Also, I am not sure about that "high minimum ISO". If you are referring to the minimum exposure index setting available on the E-M1 Mk II [200 standard, but with 64 expanded low setting] that is not a measure of base ISO speed—there is no good reason to expect the two to be equal (they do not measure the same attribute at all and there are good reasons why minimum EI is almost always higher than base ISO speed) or even to differ by the same amount in different cameras. Instead, they are further apart with Olympus MFT cameras than with most other brands. The gap is based on a decision about how much highlight headroom above metered mid-tones to aim for in AE modes.
-
Well, it's out.
Disappointingly, it's the Same Old Sensor, complete with image quality that lagged the field in 2016. In response to BJL's question on actual base ISO versus minimum marked ISO, Photons to Photos shows its maximum dynamic range just under 10 stops, and it reaches maximum at ISO 200 (so the true base ISO is either similar to the minimum marked ISO or there's no other reachable setting that's closer).
It looks like the multi-shot mode is very clever in how it uses the IS system, but it uses 8 shots, so it's going to take a while to record everything (good luck with any wind at all). Even if the stitching were perfect and nothing is moving, it will look like a 5Ds (which also has just under 10 stops of DR), not a Nikon or Sony pixel monster with 12 or more. For some scenes, that'll work just fine - but there's a reason the 5Ds is often seen as underwhelming.
Once you get past thinking it's a high resolution camera and treat it as a pure speed demon, it has some interesting tricks up its sleeve. It's the first time we've seen computational photography heavily used in something other than a phone.
It has an AI driven AF system that the early reviewers say is uncanny. Right now, it recognizes cars, motorcycles, planes and (oddly) trains - I would have thought "athletes" would have been a much higher priority than "trains"? Is there a huge genre of train photography I don't know about? Its AI database is expandable, and I'm sure athletes is the next item on the list! I wonder if it will automatically select depth of field to blur the ump/ref (in honor of how the ump sees the game)? Without an AI target, it's still extremely capable - faster than an E-M1 mk II, which is already fast. Give it a target that it knows about, and it takes the next step...
There are other AI effects, notably the ND filter...
This may well be a breakthrough that isn't quite ready yet - what I'm reminded of is the days of the Nikon F5 (back when film was a going concern). It had a revolutionary RGB metering system that used an unfathomable number of metering zones, and was sensitive to color. Nobody really knew what to do with it back then - it exposed slide film a little better than other cameras (print film had enough latitude that you generally couldn't tell). It was a little less likely to be fooled by backlit subjects than other cameras, too.
I shot an F5 for quite a while (after its heyday - I bought mine used), and it was the most accurate metering film camera I ever used. It wasn't that much more accurate than other cameras, though - standard multipattern metering was already darned good.
The end of the F5's story, though, came in the digital era - that meter that knew about color, but couldn't adjust it, ended up paired with digital sensors where you can adjust color. Every camera today meters like an F5 (although off the image sensor instead of a separate metering sensor). The color information is what adjusts the white balance.
Could Olympus have the first generation of some technologies that will really make sense in a few years? Might AI-driven AF become more than a novelty for improving tracking at an auto race? Could the E-M1x be remembered as the first computational camera that was really a camera instead of a phone?
-
Could Olympus have the first generation of some technologies that will really make sense in a few years? Might AI-driven AF become more than a novelty for improving tracking at an auto race? Could the E-M1x be remembered as the first computational camera that was really a camera instead of a phone?
Indeed, that may very well be the case.
The question in the end will be that of UI and control IMHO. When you have a face and a train in the frame, how do you tell the camera what to focus on?
Cheers,
Bernard
-
It has an AI driven AF system that the early reviewers say is uncanny. Right now, it recognizes cars, motorcycles, planes and (oddly) trains - I would have thought "athletes" would have been a much higher priority than "trains"? Is there a huge genre of train photography I don't know about?
I would have gone for "athletes" as well, but yes, there is a huge genre of train photography, and if you go to large train yards, you'll occasionally see groups of guys wandering around with cameras.
http://www.railphoto-art.org
-
Yes, including some brave souls who photographed moving trains (from what looks like directly in front) with view cameras... I would think that was a great way to get run over by a train (hopefully, they took advantage of a curve and weren't on the track).
I hope Olympus goes for "athletes" soon (as far as I can tell, that's the obvious use for the technology), then perhaps "horses", which is a subgroup of athletes that needs different AI instructions - about four or five of Sports Illustrated's top 100 athletes of the 20th Century were listed with their position played (amusingly) as "horse". Secretariat, Man o'War, Seabiscuit, etc. I wonder whose face you want in focus when photographing the Kentucky Derby - the horse or the jockey?
Animals in general (including birds) would be a useful one, but a very broad thing to train the AI on - I wonder if the current processor is up to the task, or if we are still a generation or two from that. I wonder if the initial choices of machines were made because they are easier for AI to pick up than athletes?
-
Forty years later and we are still trying to figure out autofocus.
-
Forty years later and we are still trying to figure out autofocus.
That’s actually very funny. Sometimes I still think stuff it, I will focus manually. But then I have avoided auto focus until just two years ago.
This camera looks very interesting to me. Would love to play with one. Can’t see myself selling all my gear and moving to Olympus but still.
The name is a total loser though. I would ever be able to tell anyone what camera I was using if I bought one. Who can remember all those letters numbers and hyphens. Make a good banking app password.
-
The "virtual ND" mode gives me an idea for the future (it might already be implemented, but I doubt it). That mode apparently works by combining several successive exposures of different durations, so sort of HDR. Combining that with the high resolution method of taking multiple shots with slight (1/2 pixel) offsets of the sensor, one could enhance both resolution and dynamic range—the former within lens limitations, the latter as much as one desires with enough exposures to combine. And with the 7-stops of IS (increasing total time to take the various exposures about hundred-fold), some of this could be done hand-held. That would mess with the kit weight and size comparisons, if alternative kits require a tripod.
BTW, I see some basic physical reasons why it is somewhat easier to do IS in a smaller format: roughly, a doubling of linear format size requires about an eight-fold increase in forces required, so far heaver magnets or servo-motors for moving the sensor. Maybe that is why Olympus seems to have put a higher priority on this aspect of performance than other systems with bulkier sensors.
(Still waiting for this in an OM-D E-M1 Mk III — smaller, less expensive, but with an even longer name.)
-
“Combining that with the high resolution method of taking multiple shots with slight (1/2 pixel) offsets of the sensor, one could enhance both resolution and dynamic range …”
Not to forget true R-G-B capture on pixel level instead of demosaicing, like Foveon or 3CCD.
Regards,
Jaap.
-
I read somewhere that virtual ND mode is essentially what one can do with Nikon DSLR's multiple exposures: shoot n x 1-sec shots to get the same results as n-sec shot. Olympus has also Live view of that mode where you can see the effect in the viewfinder and LCD.
Sad to see that there is no top LCD.
-
Very interesting... I could see Olympus trying it. Two exposures in each position, three stops off, gives dynamic range at least equivalent to a Z7 at ISO 64. They're actually using eight exposures for 50 MP handheld pixel shift, so doubling that would mean they need 16 (I'm not sure if they could get away with somewhat fewer with clever combination)?
Two questions: Is it easier to stabilize the Olympus for a full second (it's an 18 fps camera, so it'll take around a second to capture all the images it needs) or the Nikon for 1/20 or 1/30 second? Exposures where the Nikon is around 1/20 or 1/30 are the Olympus' best shot, because its individual exposures (ISO 200 vs 64) will be around 1/100 of a second. If there's any less light, the Olympus has to use shutter speeds slow enough that it's no longer 18 fps (or higher ISO, which reduces DR). If there's more light, the Nikon has a shorter exposure, taxing its stabilizer less.
I agree with BJL that, given relatively equivalent exposure times, the Olympus can have better stabilization, because its sensor is much lighter and doesn't have to move as far. This isn't relatively equivalent exposure, though. The Olympus has to deal with whatever motion takes place during the full second of captures. Some of that can be handled through good algorithms when stitching...
The second question is subject motion... How many things really stay still for a full second? Still life and product photography, certainly. Architecture, if there are no people or windblown trees in the image. Landscape, if everything in the foreground is unaffected by wind (rocks, dead trees, etc). No living being does, whether plant or animal, including people, unless there's absolutely no wind or you're photographing a sleeping animal.
If Olympus (or someone else) can get this to work, they have one advantage over a big, slow sensor taking a single shot. They also have a native 20 MP mode with OK dynamic range at 18 fps, while a Z7 is always relatively slow, whether you need the big pixel count and huge DR or not.
Of course, it would also be possible to come at the same problem from the other direction. It might be easier to have a big, slow sensor with enough resolution to bin pixels 4:1 to quadruple readout speed. I could see a Z7 successor having a 60 MP sensor (recent leak) that has a 15 MP high-speed mode. Those huge "pixels" would offer less resolution than the Olympus, but not much less, and they'd come with enormous DR and high ISO capability.
I'd imagine binning pixels on a high-resolution sensor is easier than mechanically moving a fast sensor to generate extra pixels, but I don't really know.
-
Seems to me every company is trying to get a piece of the upper market pie: pro expensive cameras, pro expensive and large lenses (both primes and zooms). Sony A9, Fuji XH1, Panasonic new FF, now Olympus...
This is where the profit seems to be at the moment, Sony saw it first, others are following suit. Getting ready 2020 Tokyo Olympics?
-
Since we have discarded the limit to 35mm format, I’d expand beyond mirrorless too: then I’d say that in the high end, Nikon and Canon have been there for far longer than Sony, and, even Olympus has pursued the high end of digital cameras with some bodies and lenses since back in the Four Thirds DSLR days. I agree that their home country hosting the 2020 Olympics might be one motivation, but also just the prestige effect of a “flagship” camera or “tent pole” products, even though the main source of revenue and profits is down in the far more affordable mainstream.
-
Seems to me every company is trying to get a piece of the upper market pie: pro expensive cameras, pro expensive and large lenses (both primes and zooms). Sony A9, Fuji XH1, Panasonic new FF, now Olympus...
Fuji X-H1 is now dirt chip... Fuji reduced the price to : Fujifilm X-H1 + vertical grip + two additional batteries for $1,299... peanuts vs other flagships : https://www.bhphotovideo.com/c/product/1388298-REG/fujifilm_16568755_x_h1_mirrorless_digital_camera.html
-
Those huge "pixels" would offer less resolution than the Olympus, but not much less, and they'd come with enormous DR and high ISO capability.
nothing "enormous" was in Sony A7S* with 12mp sensels (so nothing enourmous will happen with binning 4:1 from 60mp sensels)
-
Not to forget true R-G-B capture on pixel level instead of demosaicing, like Foveon or 3CCD.
except Foveon does not capute R-G-B... it has to use subtractive math to get to R-G-B data... do not believe marketing pictures showing RGB colored layers in Foveon they are not what is really captured
-
There are two reasons I'm highly skeptical of the image quality of the 50 MP stitch mode approaching that of true 40+ MP sensors - dynamic range (and related) performance and stitching/subject movement.
.....
The second issue is the Sony A9, which offers two stops of extra DR and otherwise much better image quality in a similarly speedy camera for a similar price (when the A9 is on sale). Yes, you trade off some stabilization and the best-in class weather sealing for the image quality.
For low ISO the OM 1DII and 5DII already match the 40MP+ sensors in DR with their high res mode, at least according to the data supplied by Bill Claff
http://www.photonstophotos.net/Charts/PDR_HighResShotMode.htm#Olympus%20OM-D%20E-M1%20Mark%20II,Olympus%20OM-D%20E-M1%20Mark%20II(HR),Olympus%20OM-D%20E-M5%20Mark%20II,Olympus%20OM-D%20E-M5%20Mark%20II(HR)
The Sony A9 gives up a stop+ of low ISO DR to all of the 40+ mp sensor cameras (and presumably then the high rez OM) and the 24mp A73, perhaps partly due to the high frame rate
http://www.photonstophotos.net/Charts/PDR.htm#Nikon%20D850,Nikon%20Z%207,Sony%20ILCE-7M3,Sony%20ILCE-7RM3,Sony%20ILCE-9
Stitching and subject movment are real of course and limit the use of the feature.
-
How is the stitching adding DR? I trust Photons to Photos that it is, but I didn't realize that it was using different shutter speeds for the multiple exposures (which would seem to be the easiest way of accomplishing what is effectively an HDR merge in addition to the pixel-shifting). I wonder if the handheld multi-shot mode on the EM1x will be capable of the same trick?
I also didn't realize that the A9 was a unique sensor. I thought it was part of the 24 MP Sony-sensor pack, although with unusually fast readout and processing hardware. These results look like it's a different sensor that gives up some low-ISO performance for speed.
The improved DR by stitching makes the question of "stitch or bin" all the more interesting. High-resolution sensors are rapidly reaching the point where they could produce a decent file size by binning adjacent pixels. I'd imagine it would require modified hardware, notably either a different Bayer filter pattern or a way of summing nearby pixels of the same color - they wouldn't actually be adjacent, because there would be a differently filtered pixel in between.
If the summing could be done directly on the sensor in fast hardware, it should offer close to 4x frame rate and ISO for a given noise level. Some old Phase backs did this, admittedly going from ~1 fps to 3+ fps, not 5 to 20...
A really good "speed/resolution" camera that used binning in speed mode would, of course, need the AF system of a sports-oriented camera (speed and tracking) to work well at those speeds.
It would make for interesting times if the EM1x's multishot modes work very well and the Nikon Z9 or Sony A9r turns out to be 60 mp at 6 fps, but also 15 mp at 22 fps... Especially if it's also 60mp 16 bit, even if that's at 1.5 fps.
-
How is the stitching adding DR? I trust Photons to Photos that it is, but I didn't realize that it was using different shutter speeds for the multiple exposures (which would seem to be the easiest way of accomplishing what is effectively an HDR merge in addition to the pixel-shifting).
I do not know the implementation details, but even taking multiple exposures at the same shutter speed will give extra information that can increase the effective (perceived) dynamic range, for example by averaging which lowers the noise level relative to the signal level. (This is somewhat related to the fact the increasing the number of pixels while keeping the per pixel DR the same effectively increase "per image DR": the shadow handing and such seen at equal display size will be improved though the smaller apparent size of each pixel in the displayed image, and the consequent dithering or spatial averaging by the eye.
-
DPReview just posted their studio test shots for the EM1x -the interesting part is that 80MP (tripod) pixel shift was included.
While the magnification of the image is clearly 80 MP, the actual detail is not. In an absolutely ridiculous comparison, I was able to look at it alongside a Phase One IQ180 - sorry, not in the same league (but it shouldn't be - the Phase costs $19,000 used - and if you think an EM1x is heavy...). The Phase also has a base ISO of 35. Confirming that DPReview had used the 80MP mode, rather than the 50MP mode, the image size was just about identical.
Against a more reasonable comparison, a Nikon Z7 that is somewhat lighter and slightly more expensive, the Olympus holds its own. The Z7 holds detail a little better in every part of the image, but only a little (and it has a 1 2/3 stop ISO advantage).
The Olympus may have had a lens advantage (DPReview doesn't say what lens they used for each camera). It's almost certain that the Olympus was one of the very sharp Olympus PRO lenses - the 45mm f1.2 PRO? Was the Nikkor one of the 85mm G lenses on the FTZ? The f1.4 is quite sharp, as is a good copy of the f1.8, but I'd be surprised if either of these 8 year old lenses is as sharp as the upcoming 85mm Z lens will be. The best comparison would have been the 25mm PRO against the 50mm Z lens, but DPReview shoots their target with a portrait lens...
Stepping down a little to try and find something very close to the same detail level, a D810 works very well. In some parts of the image the Olympus seems to have a little more detail, in other places the Nikon does.
The pixel shift mode clearly has more resolution than an X-T2. I would have been very concerned if it didn't!
-
Against a more reasonable comparison, a Nikon Z7 that is somewhat lighter and slightly more expensive, the Olympus holds its own. The Z7 holds detail a little better in every part of the image, but only a little (and it has a 1 2/3 stop ISO advantage.)
Actually, the Z is almost a pound lighter (14.5 oz.)
The Z is also more than 3 centimeters shorter and almost a centimeter thinner. (The Oly is insignificantly wider -- 2mm.) The Z7 costs about $400 more, but a better camera-to-camera comparison IMHO would be with the Z6 with its 24.5 megapixels vs. 20.4 for the Oly. The Z6 costs about $1000 less.
Some of the weight and size difference would be negated by the bigger FF lenses.
-
Weight and size comparisons only make much sense vs other cameras with vertical grips, including adding accessory grips to potential rivals like the D500. (One aspect being that the weights then include the extra batteries.) Those of us who do not need or want that bulky vertical grip option are better off looking at the E-M1; either the current Mk 2 or the hoped for Mk 3.
Aside: the increase in size over the E-M1 is far greater than the increase in weight; is a lot of of it about more and bigger buttons with more room between them, for the sake of fast-paced operation?
-
I find these size and weight discussions/criticisms bizarre. The E-M1X is bigger and heavier than the other Olympus bodies in much the same way and for probably similar reasons to why the Nikon D5 is a bigger and heavier than other Nikon "full-frame" bodies. No one had a bird when the D5 was announced.
I have a E-M1 and use a grip now and then, and the combo is not far off the size of the E-M1X, so what's the big deal. The E-M1X is a niche product and if you're not in that niche you probably shouldn't buy one. I can't see the D5 being used for street photography much either.
If you shoot sports/action in m4/3s, then it's a good product. The combination of the E-M1X with the Oly 40-150/f2.8 is still A LOT smaller and lighter than any Nikon/Canon field-of-view equivalent, so for those who want to use this m4/3s system they can now do so with a better body.
-
If you shoot sports/action in m4/3s, then it's a good product.
It will be if the CAF is up to the task. The E-M1.2 CAF was a major disappointment (I own one, but did not buy it as an action camera) and from what I have seen so far, the E-M1X CAF does not appear to be able to cut the mustard. If that is the case then it is not a good product for action. It looks to me as another case of Olympus over-hyping the AF.
-
I find these size and weight discussions/criticisms bizarre. The E-M1X is bigger and heavier than the other Olympus bodies in much the same way and for probably similar reasons to why the Nikon D5 is a bigger and heavier than other Nikon "full-frame" bodies. No one had a bird when the D5 was announced.
I have a E-M1 and use a grip now and then, and the combo is not far off the size of the E-M1X, so what's the big deal. The E-M1X is a niche product and if you're not in that niche you probably shouldn't buy one. I can't see the D5 being used for street photography much either.
If you shoot sports/action in m4/3s, then it's a good product. The combination of the E-M1X with the Oly 40-150/f2.8 is still A LOT smaller and lighter than any Nikon/Canon field-of-view equivalent, so for those who want to use this m4/3s system they can now do so with a better body.
I guess the big deal (for many people) is that *all* cameras are sort of niche products, because it's impossible to find one that does everything the way everybody wants it to. I would say that the big niches are (1) resolution, (2) speed in the sense of shooting speed, buffer and continuous auto-focus, (3) system equipment and (4) camera size. M4/3 cameras have good resolution, decent speed, respectable system size, without being a leader in any of those things, but is outstanding in providing excellent cameras in small size, especially for street shooting and for travel/backpacking uses. I have no doubt in my mind that if Galen Rowell were alive, he'd be shooting m4/3. The new Oly goes against that m4/3 strength. It's quite large, yet really doesn't seem to be fully competitive with other large cameras -- the Canikons -- which are mostly FF with much higher resolution and have huge systems supporting them (the new Oly seems to be closing the gap in speed, though.) What worries people like me is that Olympus is attempting something that seems likely to fail -- going head to head with Canon, Nikon and Sony -- and we *really* don't want it to fail, because we own substantial m4/3 systems and we like those systems, and feel that Oly would be much better off emphasizing upgrades to existing (small) cameras. My own m4/3 cameras, the GX8s, which I think are the best combination even now of speed and compactness, are getting long in the tooth, having been announced in 2015. M4/3 needs better sensors, better flash systems, a few fill-in lenses (though the lens offerings are really pretty good) and so on. A big fat camera, not so much.
-
It will be if the CAF is up to the task. The E-M1.2 CAF was a major disappointment (I own one, but did not buy it as an action camera) and from what I have seen so far, the E-M1X CAF does not appear to be able to cut the mustard. If that is the case then it is not a good product for action. It looks to me as another case of Olympus over-hyping the AF.
You may be correct that it may not be as good as the top action bodies of Canon/Nikon but does it need to be? The pro sports shooters with arsenals of Canon/Nikon lenses are not going to switch to Olympus in large numbers regardless. It is going to appeal to those with m4/3s lenses who shoot action. Obviously Olympus thinks that there are enough of them to warrant building a body for them. They have a known (to them) sales history of E-M1s and E-M1-2 so they must have some belief that it will sell well enough. If it's lacking in a specific feature that you or others need, how is that any different than any other camera?
-
The key to your statement about the E-M1 is that you "use a grip now and then". The E-M1x doesn't give you that choice. Integrated grip bodies are big and heavy all the time, and that's a notable feature that's worth talking about, although it's not always a negative feature.
The Olympus PRO zooms are pretty good-sized lenses. An E-M1x plus an Olympus Pro 12-40 plus the Pro 7-14 is 253 grams heavier than a Z7 plus the Z 24-70 plus the upcoming 14-30 (and those are very comparable lenses - opinions will differ on which is "faster", since the Nikkors have a stop more DOF isolation, but the Oly lenses let in a stop more light)... Adding modest telephoto capability doesn't really change that - although there really aren't equivalent lenses. A Nikkor 70-300 G EG VR plus the FTZ adapter is almost exactly the same weight as the Olympus 40-150. The Olympus is almost certainly the better lens, although the Nikkor isn't bad. Alternatively, if you're willing to go without the zoom, the Nikkor 300mm PF on the FTZ is also just about the same weight, and is probably of comparable quality to the Olympus lens. I'd personally love to see Nikon make a Z-mount 70-300mm f4 PF.
Sony's A9 is very similar in size to the Z7, although some of the better lenses are heavier, with speed versus image quality tradeoffs that are closer to the philosophy of the E-M1x. While they're a reasonable cost/weight comparison, I can't imagine any image where a Z7 and an E-M1x are both the right tool for the job. You could concievably shoot landscape with an E-M1x or sports with a Z7, but why?
Of course, you can pick other lenses that make the weight difference huge - including anything longer than Nikon's 500mm PF and Canon's 400mm DO - or heavier options in modest focal length ranges.
Oly has a niche in relatively compact super high-speed cameras, and even the E-M1x is a great deal smaller and lighter than a D5 or 1Dx (but not a Sony A9, depending on lenses). An E-M1 mk III that brings some of these technologies to a smaller, less expensive camera could be a real hit.
-
You may be correct that it may not be as good as the top action bodies of Canon/Nikon but does it need to be? The pro sports shooters with arsenals of Canon/Nikon lenses are not going to switch to Olympus in large numbers regardless. It is going to appeal to those with m4/3s lenses who shoot action. Obviously Olympus thinks that there are enough of them to warrant building a body for them. They have a known (to them) sales history of E-M1s and E-M1-2 so they must have some belief that it will sell well enough. If it's lacking in a specific feature that you or others need, how is that any different than any other camera?
The appeal is in bringing "pixels per duck" to a smaller, lighter package with superb IBIS. This is a winning formula that pushed a lot of people to buy the 7D2 + 100-400 and then the D500 + 200-500 or 500PF. If the E-M1X matched the D500 focus a lot of folks would notice. But without that focus, the word gets around, and few will buy. Seriously, I don't really know of a significant shooting segment that would buy this package without the CAF.
I actually know a number of folks, including myself, with those large arsenals of CAnon and Nikon super-teles, that would seriously consider the E-M1X if the CAF was up-to-par. I personally know 8 people who have or are ditching Canon for either Sony or Nikon. I don't hear many openly criticizing the Canon focus system, but if your shooting birds in flight it isn't your first choice. If your shooting people it isn't your first choice.
The bottom line is that getting tack sharp focus is the most important factor for action and wildlife photographers. No other feature really matters if you can't nail the focus. But if you can, then the weight, size, image stabilization, grip, UI and pixels per duck start to play a bigger role.