|
Olympus E-M1X
More than a better E-M1.II |
|
|
My other articles related to the |
|
Table of Contents: | |||
|
The late rumors had it right: the long-awaited January announcement by Olympus was not about the E-M1 Mk.III, but about an entirely new model: the E-M1X. Olympus makes it clear that the new model, aiming at the top of the market, "will be situated alongside the OM-D E-M1 Mk.II", so I'm assuming that the E-M1 line will be kept alive and we will see the Mark III later in 2019. Just for the record: the official, full name of this camera is Olympus OM-D E-M1X, but I'm not ready for a tongue twister like "oh em dash dee [pause] eee dash em one ex mark two" or similar. The E-M1X is not an update off the E-M1.II, but a new model, built from the ground up, with most components either new, or changed from its predecessor. Olympus provides rather scarce information on that, most of it available not in official sources (press releases, statements, manuals) but scattered in unofficial ones (interviews, leaks, hints, rumors, third-party research). This is my semi-literate attempt of reconstructing the list from all available (and, presumably, reliable) sources. Most of the listed items will be discussed further down in this article.
With all the specs and features added or improved in the E-M1X (as compared to Mk.II), there is one, overshadowing them all and affecting all users: the form factor. This is the first Olympus camera featuring an integrated vertical grip, and it shows at the very first glance.
|
Note: all images shown here are from promotional materials © by Olympus Corporation.
| For this publication, the tonality has been adjusted to show shadow detail.
|
| 
E-M1 Mk.II
| 
E-M1X
|
Both cameras are shown with the MZD 12-100 mm F/4.0 Pro lens.
| Pictures have been rescaled to show body dimensions properly.
| Now, some people find a vertical grip essential when shooting in portrait orientation. The integral, non-detachable kind is more resistant to abuse, thus becoming the choice of many professionals. This type has one more, often overlooked, advantage: the upper part of the body does not have to house the battery. This space is wasted when you use a detachable grip. While the grip duplicates some of camera controls difficult to access in the portrait mode, it does not, obviously, provide a second viewfinder. You end up holding the camera higher (compared to landscape mode), and almost entirely to the right of your head (with your right arm extended too much). Hardly a good ergonomics.
|
| 
The E-M1X in landscape mode
| 
The E-M1X in portrait mode
| If you have large hands, strong arms, and healthy spine, you may be tempted with the E-M1X — but see if the degree of convenience gained in shooting verticals is worth the price in bulk, weight, and awkward shape you'll be paying, even in landscape orientation. This becomes the proverbial elephant in the room, overshadowing any improved specs or new features. While personally I'm a gripophobe, some of my best friends are gripophiles; therefore I'm not going to argue one way or another. An omni-aspect μFT? Better-polished stone tools are not the right solution to technological challenges. We are still waiting for a real multi-aspect camera, not requiring rotation to switch between portrait and landscape. (We will call this omni-aspect.) For cameras using electronic viewfinders, this is no rocket science. What we need is the following:
The image-processing software can then collect images in the original aspect ratio, but also in the inverted one, and anything in-between (for μFT: from 3:4 to 4:3, which includes 1:1), taking advantage of the full lens image circle. Outside of that range, a simple crop of the appropriate "whole" frame (μFT: 4:3 or 3:4) can be used, exactly like it is being done in current cameras (with one notable exception, see below). The existing lenses will work with the "new" sensor just fine (except that any rectangular baffles have to be removed, but these are not common). Also, rectangular or petal-shaped hoods may cause vignetting for vertical frames. The beauty of this approach is that the only really unavoidable increase in body size it requires is due to the new sensor height (for μFT this would be an extra 4.3 mm). I also suspect that the EVF screen height can be pushed up without much of a problem, and, last but not least, I can live with vertical frames shown on a horizontal monitor. Details aside, I would expect this add 5mm or less, if anything, to the body height; nothing comparable to what the vertical grip brings in. "So, why did not Olympus do it right the first time?", you may ask. Four possible reasons come to mind.
I hope this will not last forever and we will see some multi-aspect cameras within a year or two. Actually, the scheme presented above is an economy version of the more elegant (and more expensive) true omni-aspect solution, in which the sensor is a square with side length equal to the lens image circle diameter, and for any aspect ratio, all frame corners belong to the circle perimeter. (As the photosites outside of the circle are not used at all, they can be non-functional dummies, if this makes the chip cheaper or speeds it up.) For μFT this means a sensor of 21.6×21.6 mm, almost exactly twice the original area. Of that, only 80% has to be covered with working photosites, see the remark above. Of course, using a somewhat smaller square is always possible, but this is, again, the original "economy" solution we started from; possibly with a different aspect ratio cut-off. I won't be going into details here; I hope to write more on this subject soon. Let me just mention that a version of this concept has been used in the Lumix DC-GH5S by Panasonic, if only at 4:3 and wider aspect ratios — only the frame width, not height, was extended; understandable in a camera aimed primarily at videographers. Update of September, 2019: The article is ready — Time for Omni-Aspect Sensors? Now that we have Jumbo out of the way, we can start looking at the remaining specs and features of the new camera. First and foremost, this is a Micro Four Thirds camera.
Micro Four Thirds The Micro Four Thirds (of just μFT) is the standard, developed by Olympus and Panasonic, defining the sensor size, lens mount, and lens-to-camera interface, so that cameras and lenses conforming to it are guaranteed to work together. It is intended for mirror-less interchangeable-lens cameras. That's why the lens-to-image distance in μFT is smaller than in the original Four Thirds (or FT) standard, used for (mostly) Olympus SLRs, now extinct. Four Thirds lenses can be mounted on μFT bodies with an appropriate adapter, and the autoexposure will work properly, but (EXCEPT FOR the E-M1 and E-M1X), the autofocus will not. The opposite is not true: μFT lenses cannot be mounted on FT bodies. The μFT imager size is the same as in Four Thirds: 13×17.3 mm, it is a bit smaller than the common APS-C format (14.8×22.2 mm); much smaller than the so-called full frame (24×36 mm), but much larger than the common 1/1.7" sensor used in better compacts (5.7×7.6 mm). For many people, including myself, μFT is (like APS-C or even 1") a sensible alternative to the "full frame" (24×36 mm) sensor size, offering more convenience and portability in exchange for some loss in image quality (mostly noise amplitude, a metric I do not consider very meaningful). Last but not least, at this moment μFT seems to offer the widest variety of quality lenses. I'm not going to preach here; you can only decide on your own, and only after having used the system for a day or two. The imager seems to remain unchanged since the E-M1 Mk.II. This is the hybrid-AF type, originally introduced in the E-M1 Mk.I (see here for details) and further upgraded in Mark II (all 11×11 phase-detection AF spots are of the cross type). Anyway, this seems to be a quite expensive chip, and Olympus uses it only for E-M1 models. The filter in front of the imager does not provide anti-aliasing capabilitY. That would impair the resolution, so Olympus decided to handle an occasional Moiré effect in software, rather than fuzzying up all images, just in case. The filter, obviously, still cuts off almost all (above 99.9%) of infrared light. The resolution of produced RGB images remains unchanged at 20 MP; one may find it a bit of a disappointment, especially when matched with some of the outstanding optics by Olympus or Leica/Panasonic. I'm not saying that 20 MP is not enough for a full-frame (i.e., uncropped) image at a standard viewing distance equal to the image diagonal. For that, 7 MP is already fine. Still, sometimes we may want to get more detail, maybe for higher cropping, maybe for pixel-peeping — or just for bragging rights.
The available ISO range is now 64 ‐ 25,600; an extra ⅔ EV at the bottom compared to Mk.II. The imaging engine is TruePic VIII, also running on multiple processors. Here Olympus shares with us just one piece of information; the E-M1X release notes refer to "Dual TruePic® VIII Dual Quad Core Processors". A very unfortunate wording, I would say, with 'dual' used twice, for different objects and in two slightly different meanings. The most common understanding of this disclosure is that the E-M1X is running two CPU chips, each having two processors. each processor with four computing cores. This gives us a total of sixteen cores, capable of simUltaneous processing. It sounds like a lot of computing power — and probably is, although that depends on a number of other parameters, unknown to us mere mortals. Worse, I'm totally confused about what TruePic refers to: (i) software, i.e. firmware, possibly its parts; if so, which ones?; (ii) processor(s), or (iii) some combination of both. This would be an academic question, if not for version naming: TruePic VII of the original E-M1 became TruePic VIII in Mark II. but it was not renamed again in the E-M1X. This suggests no major changes (in whatever TruePic is). This seems to be in a contradiction to claimed major improvements to autofocus, image stabilization, and pixel-shift resolution enhancement. To keep the processor cooler, thus reducing the noise and helping to prevent overheating failures, the designers found room for some heat ducts. This was made possible by the larger body; also note that the space used for battery in Mk.II became available for other uses in the E-M1X. Nice, but let's not get overly excited: this may become meaningful only in rare situations with high power usage (for example, high-speed serial shooting). See the Power section for more on that. Olympus was the first camera maker who came up with an effective solution to the problem of stray dust specks sitting on the image sensor (more exactly: on a filter in front of it) and showing as black spots in pictures. The cleaning is done by a brief burst of ultrasonic vibration applied to the sensor/filter ensemble. This shakes dust particles off, to be captured by a sticky strip below. This operation is performed every time the camera is powered on. [???] This AF system was introduced six years ago with the original E-M1. It combines phase detection with contrast detection AF, both done on the (appropriately modified) image sensor. Its basic architecture and features have been described in E-M1: The Full Monty, section Autofocus — still up-to-date. What is so special about this system is that it works well with legacy (Four Thirds SLR) lenses. Too bad it is available only in three cameras, all E-M1 models. For details, see Dual AF in the same article. Since the original E-M1, the manufacturer has been tweaking their AF algorithms. Mark I was fine with the S-AF (Single AF) mode; Mark II improved upon C-AF (Continuous AF) and added the C-AF+TK (tracking) option. In the E-M1X the latter part was, according to Olympus, re-worked from the ground up, with some Artificial Intelligence process used to "learn" the patterns at the design stage. How well this works, remains to be seen. At the moment this boils down to giving the user a choice of objects to track: cars/motorbikes, trains, and airplanes — in addition to a generic trackable. Do I sound skeptical? Perhaps. But having been around the block a coupla times, I've seen the term "AI" being misused at least as often as legitimately used. Olympus uses a piezoelectric motor for tiny sensor movements, to compensate for camera shake in five dimensions: three rotations and two translations. (For a few of the Pro-series lenses they use a hybrid approach: a dedicated, moving lens element also participates in the process, helping to achieve more stabilization.) Panasonic also introduced a similar, dual IS system in some of their recent μFT cameras; unfortunately (and ironically) both remain incompatible. Olympus says their IS system is the best on the market, especially after improvements in E-M1X: new gyroscope (orientation sensor) and new software algorithms. What they claim is "the world's highest1 image stabilization performance approximately 7.5 shutter speed steps of compensation, fulfilling photographers' expectations for capturing split-second images". Leaving technical aspects aside for a moment, the second half of this sentence makes no sense and carries no information whatsoever. Is the ability of taking "split-second images" a result of image stabilization? Really? One Olympus Visionary quotes the maker as saying "most other lenses can achieve about 7 EV stops [while] M.Zuiko 300mm F4 IS PRO [...] 6 EV". I wasn't able to find the original statement on the Web. 7.5 EV means handholdable exposures 180× longer with IS than without. An attached footnote says that this refers to the MZD 12-100/4.0 Pro lens (hybrid IS!) at the 100 mm long end (where the effect is most visible — note by JAW), and following the so-called CIPA standards. Those seem quite permissive to me, though: for the same lens Olympus quotes an IS benefit of 6.5 EV on the E-M1.II, while my own procedure, see description, gives 4.3 EV at 100 mm, 2.9 at 35 mm, and 2.4 at 12 mm. Go figure. While 4.3 EV is a far cry from 6.5 EV, it is still an outstanding result, and Olympus may still be the world champion, without any need for juggling or inflating the numbers. And if the E-M1X, indeed, improves the IS by even 0.5 EV (my rough guess, unsubstantiated with any data), this is nothing to sniff at. What I don't like, is that Olympus supports its claim with just one data point, and the best-looking one (this was the case with Mk.II). Showing a range of results for a given set of lenses and/or focal lengths would be more meaningful, not to say more honest. To add all this up: the E-M1X has (most probably) the best IS system in the industry, which may be not getting the recognition it deserves because of communication issues. Like most ILCs, this camera has two different mechanisms to control exposure time (over which image signal is collected). These are both referred to as shutters. This shutter (electronically-controlled; focal-plane) works like one in DSLRs, except that it stays open while the camera is powered on (viewing!). When the shutter is released, it will first briefly close (for data buffer flushing and other housekeeping), only then being ready to perform its prescribed cycle (as discussed here, for example). After which, the shutter re-enters its viewing/ready state again. This is a typical ILC behavior, no surprises. The available speeds range from 60 seconds to 1/8000 s, the latter value setting it apart from lesser models (Nikon D1 is, I believe, the record holder here, at 1/16,000 s); the fastest flash sync (single-burst) speed is 1/250 s; again, more desirable than the typical 1/125 s. The expected shutter life is 400 thousand releases, which is as good as they come, twice the pro-level 200 thousand claimed for the E-M1 Mk.II. (Entry-level models are rated at 100 thousand; Canon EOS-1D X Mark II at 400,000). Here light collecting is started and ended electronically (a process known as gating), with the sensor being illuminated full-time. This is referred to as the "silent" mode. Additionally, there is also an option to open the shutter electronically, and to close it mechanically, thus reducing the motion blur caused by the initial acceleration of shutter blades (which is, by my estimate, from 0 to 3 or 4 m/s over 1 or 2 mm). Olympus calls this mode "anti-shock", whatever.
A fully electronic shutter offers higher speeds; in this case, up to There is a catch here, though. When the image data is being read from sensor photosites, this is done not simultaneously, but sequentially, row by row, with time offset between rows defined not by the shutter speed but by the sensor readout speed. This means that every row is exposed a bit later than the previous one — like using a focal-plane shutter with a very narrow (just one pixel tall) slit. The geometric distortion of moving objects, long known for focal-plane shutters, is back! Except that this time it is more of a problem, because, surprisingly, it takes more time for the readout line to cover the image height, as compared to curtain travel time (which is just a tad less than the fasted single-burst flash exposure). A typical top-to-bottom scan time is about 1/15 of a second (see, for example, Canon EOS R); less often half of that. The E-M1X is four times as fast, at about 1/60 s (same as E-M1.II), which means four times less distortion. I believe only recently the Sony Α9 was able to best that at 1/160 s.
Roll effect Reducing or removing the roll effect would help not just with image distortion, also allowing for faster serial shooting and affecting some more or less directly related camera capabilities (more on those later). The problem seems to be just in the computational complexity of the process, pushed into a single processing pipeline. A global electronic shutter is a much faster (and more expensive) solution, using multiple pipelines (one per photosite? on per row? this is unclear, although the latter would seem reasonable) to copy the image data quickly to another chip; then, while the copy is being processed, the image sensor may be exposed again. Obviously, this is not a trivial task. While global-shutter sensors are used in professional digital video, they suffer (in audition to high cost) from a number of technical problems, starting from noise, limited dynamic range, and losses in color fidelity. And even if/when chip makers come up with such a sensor, it may be missing some other unique function the camera maker may be depending on, like Olympus with 20% of green sites replaced with phase-detection AF receptors. For a relatively small company this may be a problem. See also this interesting article by Greg Scoblete. Sequential shooting With fast-travel shutters (both kinds), no wonder that the E-M1X offers some of the best sequential rates on the market:
|
|
|
As good as these numbers are, they are exactly the same as those for the E-M1.II (which, in turn, were much better than in Mk.I).
In this mode, the camera starts taking pictures once the shutter release is half-pressed. They are stored in a buffer, up to 35 frames long (that's more than twice the 14 frames in E-M1.II). The buffer is circular: if space is needed for a new frame, the oldest one is deleted. When the shutter release is fully pressed, the buffered frames start being written out to a memory card, white the camera keeps taking more pictures. This stops when you let go of the button, or when a preset sequence length is reached. We end up with a sequence of frames, some of them (up to 35) preceding the release. Thus, a photographer may wait with capturing some moment till just after it actually happens (as discussed in my Mk.II write-up). Anyway, except for the buffer size, this feature seems identical in both cameras. Neutral-density filter ND filters are used to reduce exposure without resorting to higher shutter speeds or smaller apertures (larger F-numbers). This is usually done in order to assure low depth of field or motion blur (water in landscapes!) at light levels that are too high for wide-open lens or for long exposures. In digital photography this can be achieved without using an actual neutral density filter, with technique which Olympus refers to as Live ND (used in some phone-cameras as well). To save our pretty little heads from overheating, Olympus does not provide any information how this works. Still, a lot can be deduced from the user instructions. First of all, Live ND works only up to ISO 800, in Manual and Shutter Priority modes. You can set the filter density to any integer value from 1 to 5 EV, which means the ND factor from 2× to 32× in 2× steps. The exposure time as set must not exceed 1/60 multiplied by the ND factor (1/30s at ND=2 ... 1/2s at ND=32). Then the camera uses electronic shutter to capture not a single exposure at a set T (as estimated for the NS filter used), but a number (ND) of frames, exposed at T/ND each. Superimposing these frames together at 1/ND luminance results in a properly exposed image captured over T. Done. Example: the scene requires 1/125s with no Live ND. We set ND=16 (EV=4) which results in 1/8s as the right shutter speed for that filter density. The camera takes 16 frames at (1/8)/16, rounded (back to the original) 1/125s each, aperture unchanged. Merged, they result in motion blur captured over 1/8s (slightly more, as we disregarded here the time between partial frames). As Live ND makes use of the electronic shutter, it cannot be combined with other such uses, like HDR. As expected. The first camera with this, quite revolutionary, feature, was the Olympus E-M5 Mk.II, shortly followed by Pentax (a slightly simplified version), and then the E-M1.II. Indeed, there is intelligent life beyond Canon and Nikon. This mode seems to be implemented identically as in the E-M1 Mk.II, and that is described and discussed in a separate article, with more examples in two follow-ups. In short, for every high-res picture that camera exposes eight frames in a high-speed sequence, every time shifting the sensor just one pixel in one of the principal directions (or 0.7 pixel diagonally). The resulting information can be converted into 80 MP or 50 MP RGB images; see the size/format table. Like in the Mk.II, the Tripod High Res Mode sets the flash sync speed to 1/50 s, limits the ISO to 1600, does not allow apertures beyond F/8. Image stabilization remains, obviously, disabled, (Also, some other features cannot be used concurrently, let me skip the details.) My image samples from Mk.II show the 50 MP images, lens permitting, with more detail than the standard, 20 M ones. On the other hand, the 80 MP versions do not seem to be more sharp, as much as I tried. Still, I'm not writing the 80 MP mode off: I suspect they may submit better to sharpening. (Don't keep your fingers crossed!) E-M1X samples by Robin Wong seem to produce very similar results; I have all reasons to claim both cameras are identical in this aspect. Now, the famous/infamous Handheld High Res mode, new in the E-M1 series, and long-demanded by the "why don't they..." entitlement crowd. In this mode, Image Stabilization is forced to Auto, flash disabled, and ISO limited to 6400, with aperture closing down to F/8. Only 25 or 50 MP JPEGs are produced, with (optional) raw files yielding 50 MP images. Here Olympus does something I never expected, a trick somewhat similar to one used in cell phone cameras. Instead of a single exposure, the electronic shutter fires sixteen times in a rapid succession. Because of camera shake, each of those partial frames shows some shift (translation and/or rotation) with whatever is chosen as reference. For every RGB photosite in every partial frame, the software identifies the matching pixel of the final, composite image, and updates that pixel's RGB components with the photosite's readout. (Some interpolation from the neighbors, like in the "regular" Bayer demosaicing, cannot be avoided, I'm sure.) Image Stabilization in Hi-Res? Please note that if the IS works perfectly, compensating exactly 100% of camera shake, a given point of the scene will alwys be mapped into the same photosite, so that this HR scheme will not work. What we need is an offset, different for every frame, added (if needed?) to the IS-corrected sensor position. Or, excuse me for sounding stupid (but this just might work!), reducing accuracy and/or magnitude of IS. The 16 sub-frames contributing to the final image are spread over one second or more. The IS system is usually not capable of removing shake over such periods with one-pixel accuracy; therefore it is quite possible that Olympus does not have to do anything to randomize the correction shift. How convenient. Most probably then, the handheld version works like the tripod one, except that the neat two-squares sensor shift pattern is replaced with one generated with camera shake (partially compensated by image stabilization or randomized). Think of a square dance after lots of Kentucky's best-known product. For a given pixel, the kind (R, G, or B) of the next photosite in its sequence is a random variate (with some degree of autocorrelation possible). This means that the number of R, G, and B photosites contributing directly (i.e. without interpolation) to that pixel will also be random. (For the tripod HR it is fixed at 2, 4, and 2, respectively.) Therefore for an R or B component there is a chance P(0)=1.0% of getting no direct input in that channel (for the selected pixel, that is) and P(1)=5.3% of getting input from just one matching photosite out of 16 used — still less reliable than two photosites used in the tripod version. (For G photosites these values are better: P(0)=.0015% and P(1)=.025%, as there are twice as many of them in the mix.) [Note: These numbers are derived from the binomial distribution under assumption of no significant autocorrelation in the receptor RGB sequence.] I suspect this may be why the system uses 16, and not just 8, frames here. For a sequence of eight partial images the binomial distribution gives P(0)=10.0% and P(1)=27.7% (R and B). This, I would guess, would be not accepted by the engineers (regardless of the method used in their analysis or measurements). This is as far as I would go with virtually no information on the subject provided by Olympus (or anybody else). This is just an ideal case, with camera shake allowed (even required!), but no moving objects within the scene, periodic patterns, water reflections, and other effects which often need work-around's (usually rejecting or replacing a partial frame or a part of it). The devil is in details, and I have no doubt that Olympus system designers had a lot to handle here. Now comes the funny part. Frankly, I usually dislike solutions like this, with too many variables outside my control (or even knowledge), with too many "this should..." pivotal points. But this thing — it seems to work, in most cases at least. A close look at the High Res Mode images by Robin Wong leaves no doubt here: the detail in HR (50 MP) samples is better than in SR (20 MP) ones, although (for me, at least) bot quite as good as for tripod-based SR, both in terms of apparent resolution and artifacts. The water-, dust-, and freeze-proof body is built around a metal shell (magnesium alloy?); the build and finish seem very good — as we have learned to expect from Olympus. Actually, even better. This was quite a surprise the first time I handled the camera.
Here is a table, listing body dimensions and weight for the E-M1X and both E-M1 models (which provide similar functionality, except for the vertical grip, but including splash- and dust-proofing).
|
|
We can see that the E-M1X is visibly larger than the previous Olympus flagship, adding 56 mm to its height (which was expected) and 10 mm to its width (which was not).
| The increase in weight is also quite large: 423 grams, or 74%! (of which 74 g is due to the second battery). Also note that the E-M1X is twice as heavy as the original E-M1. No, I'm not going to comment on this.
| 
The E-M1X body with the MZD 12-100/4.0 Pro lens attached
| According to the official specs, the E-M1X is splash-, dust-, and freeze-proof. (So is Mk. II, but the new model seems to be better in this aspect.) In particular, it should survive an occasional rain or a [???] These, known also as direct controls, follow the common "push button, turn dial, watch monitor" scheme, which works just fine in many cameras (including all other OM-D models). I was able to spend a few hours of quality time, checking out the controls. No surprises: the layout seems somewhat, if not drastically, improved since Mark II, both in terms of the design logic, and of accessibility and ease of operation. The latter makes a good use of the greater body width; it is up to you to judge whether it was worth the price of an extra 10 mm. Again, a highly subjective matter. There is no top-deck display (LCD or other), so when changing any camera setting you must watch the monitor screen — except for the exposure mode, which is shown directly on the mode dial. With more room available (10 mm extra in body width, no LCD), the designers could afford to keep the top deck relatively uncluttered with controls. The layout is very similar to that in the E-M1 Mk.II.
| Left of the eyepiece hump, there is an On/Off power switch centered around a fake knob, almost identical to that on the Mk.II, except that it houses three (not two) buttons. One is used to access the settings for sequential shooting, self-timer, and flash; another for light-metering pattern and AF settings, and the third one — to set up or enable bracketing. To the right there is the mode dial with a locking button; in addition to the usual PASM exposure modes it offers four custom slots. time exposures (including the Live Composite option), and a movie mode.
Olympus once tried to remove the mode dial in order to get space for a top-deck display. This was the E-3 of 2007. They never tried again.
| 
The top deck
| After E-M1 Mk.II got rid of the infamous Scene and Picture Story "modes", the E-M1X puts to sleep Art and iAuto. At long last. Still, why do I have to spend three grand to get a camera without all that bloat? At the far right we have more controls. First, two adjustment dials (for a button held down each dial can be set to adjust a different setting, if you wish). Then three buttons: [ISO], [+/-], and [], by default assigned to, respectively, ISO setting (dimpled for a tactile feedback), AE compensation and video recording . For the last purpose I would rather recommend going through the mode dial. This will also allow you to repurpose the said (red-dot) button. For exposure compensation (which I apply and change a lot on my E-M1 Mk.I and Mk.II), I prefer to use one of the dials alone, with no buttons pressed. Both dials, the shutter release, as well as the ISO and AE Compensation buttons are easily accessible when the camera is in the landscape orientation, but not in the portrait mode. That's why all five are duplicated — see bottom-left and bottom-right of both pictures below, respectively. (Other such duplicates will follow.)
Now, have a look at the front of the E-M1X. The two front control buttons from Mk.II made it to this model, but there are extra two below! Yes, these are "vertical" duplicates of the original ones; when the camera is held in a vertical orientation, they can be operated with the same fingers of your gripping hand (a tribute to Larry Niven) as the originals were in landscape format.
| Olympus (justly) assumed that many users of this camera will be using the two features by default assigned to these buttons: the depth-of-field preview and setting the white balance by reference (using the light scattered off a white surface); therefore they must be easily accessible in both orientations. Thoughtful.
| 
Another full frontal, this one with the MZD 12-40/2.8 Pro lens,
| The camera's posterior hosts its remaining direct controls — more than its top and front combined, in addition to the monitor screen (which we will discuss later). These are also the controls most affected by the transition from the E-M1 Mk. II — understandably, as this is the area which gained the most space. Above the monitor, to the left of the EVF eyepiece, there is a button to switch between EVF and monitor for frame preview, while a long press on it can be used to set up the auto-switching mode.
| The knob on the left side of the EVF hump is used for diopter adjustment. At the far right we have the rear control dial, already listed among top-deck controls. To the left of it, there is the [AOL/AFL] button, to lock autoexposure and/or autofocus. (It is duplicated at bottom-right.) How it works together with the shutter release. can be set for each focusing mode (AF-S, AF-C, MF) in a setup menu — almost like in the E-M1 (with two minor options added). The last remaining control (or rather two) in the top row is a two-position lever switch, encircling another button marked as [Fn] ("function button"). Both controls are re-assignable and independent of each other. The default binding (or "role") of the [Fn] button is AF target selection (see the Autofocus section). I'm puzzled, as the same function is hardwired to the AF joystick (a.k.a. "multi selector"), just below.
| 
The camera's back side is not crowded;
| The Fn Lever, dating back to the original E-M1, is as an example of a solution in search of a problem. "Hey, guys, look at this cool gadget I just designed. Now we must find out what to use it for". Out of the Fn Lever presets, just one deserves consideration: switching between two user-defined combinations of AF mode, target shape/size, and target position. For example, one for AF-S with a single-point, centered target, and another for AF-C, with the whole 11×11 grid. While this is remindful of a custom mode, it differs in how the settings persist, but let me skip the details now. The other options are:
Below the top row and to the right of the monitor screen there is a vertical row of controls. It starts from a cute, miniature joystick (8-way + click), refereed to as multi-selector. This control is intended primarily for AF target positioning and selection (but can also be used in most UI screens to move around). By default, its functionality is duplicated by the arrow pad, and a vertical-use duplicate is provided at the bottom. The [Info] button is used to switch between status display modes and UI screens in a context-sensitive fashion; quite often you will see a prompt to that effect. The cursor key cluster consists of four arrow buttons around the [OK] button. Again, the response to these depends on the context. In appropriate situations (AF target placement, UI screens) they, indeed, move the cursor or change a selection, but during the live image preview you have three options:
The last Olympus camera with buttons below the monitor was the E-3 of 2007. The new design is improved: the four buttons in the row are easier to tell apart. There is also a lever switch here. The buttons are:
I'm not sure I like having the [WB] button in this group. Perhaps placing it above the arrow pad with [Info] below and [Play] in the bottom row would be better? The lever allows the user to disable some of the controls, to prevent them from being activated accidentally. It has three positions:
Some of the buttons may have assigned a function different than the one bound to it as a factory default. This number includes:
In total, 14 buttons can be customized (plus, partly, the arrow pad): nine originals and five vertical-use duplicates. A duplicate may have a different binding than the original, but I would not recommend that. This means you may customize up to seven buttons (not including ones provided on Pro-series lenses), choosing from 37 different roles. This comes, obviously, at a price: button labels are no longer true. The button marked as [WB] may end up being used for Image Stabilization. Also, most (if not all) of default assignments make sense and I wouldn't change them without a good reason. Viewfinder and monitor Both serve similar purposes: real-time scene preview, review of stored images, and/or pToviding a visual interface to camera functions or settings. As a matter of fact, in most situations both can be set to show the same information (concurrently or alternating). Viewfinder The electronic, eye-level viewfinder in the new camera seems to be using the same microscreen as the E-M1.II: an LCD unit of 2.36 million dots (which is equivalent to 0.8 million pixels). The optical system used to view that screen has been improved, though: the effective ("35-mm") magnification increased from 0.74× to 0.83×, or by 12%. While not critical, the difference is clearly visible. The optics used to view the screen shows no chromatic aberration, even in corners, which indicates serious job done here. Two aspects are sometimes criticized about this finder. First: why LCD, not OLED? Admittedly, an OLED finder provides deeper blacks and a wider color range, which adds up io a display not only more enjoyable to view, but also more accurate (wider viewing angle is not important in EVFs). Nothing comes free, though: OLED display show more degradation with time, even when not used at all. Second: some current top-of-the-line finders boast 6M dots (or close) of resolution. Again, this translates into 2.5× more data processed real-time, up to 120 times a second. Yes, the refresh rate is up to 120 FPS, with the time lag of just 5 ms (down from 6 ms in Mk.II). Olympus gives this as a reason to stay with the LCD technology. No stroboscoping or tearing effects which I could spot. Frankly, I'm more than happy with this display. OLED or not, I yet have to see a better one. Monitor The external monitor shows 1.04 M dots on a 3-inch (diagonal) screen; both are adequate for the job and increasing them may be not worth it in terms of power usage. It has two degrees of freedom: tilt and swivel, like the the in the E-M1 Mk.II. I prefer the tilt-only kind, like in Mk.I: it is more convenient for most applications; using the swivel in portrait orientation is quite awkward. I'm afraid, though, that most of the market (and certainly most of reviewers) hold the opposite opinion. Screen visibility in open daylight can be, at some angles, quite poor. This may not affect image previews (for which the display will usually switch to the EVF),
|
| 
Here shown with the MZD 40-150/2.8 Pro
| 
So, this is better than a tilt-only monitor? Really?
| Screen UI and Menu System All Olympus μFT cameras use multiple user interfaces (three or more!) to perform the same task or adjustment. For some reasons, the designers think this makes it easier for the user. Mercifully, you can hide at least some of them, and the first thing I also do is to disable the touch screen. Supposedly, the fastest and most convenient access is provided by pressing a dedicated button and turning one of the control wheels. This, indeed, works in most cases, but some problems may arise: with a large number of assignable function buttons you may just not remember which one does the job, especially when using more than one Olympus body (more often than not they would have to be configured differently). In such cases (and not only) I resort to the nicely-designed Control Panelin which the basic picture-taking and image-conversion parameters are visible at a glance, with cursor keys used to select ones to modify. This simple, consistent, and functional design got somehow broken a few years ago and still waits to get fixed; see my detailed remarks about Control Panel and some other design issues (or what I consider as such) on Mark II. Dual Memory Cards The E-M1X inherits from Mark II two SD Card slots (not just one like in Mk. I). They can be configured in a number of ways, regarding what files are written out and/or in which order. Best choose one configuration and stick to it. Both slots support now not just high-capacity (SDHC and SDXC) cards, but also the UHS-II standard (156-312 MB/s); in Mk. II only one slot would write at this rate. As nice as it seems, this is important only (or, perhaps, mostly) if you are videographing in 4k. For still photogaphy, the high-capacity are fast enough while costing less. Connectivity Every next generation of cameras offers more capability to communicate with other devices or systems. On one hand, this is good, as it opens new ways to use your equipment. On the other, it creates more opportunities for something to go wrong. E-M1X, like other present-day cameras, can communicate with other devices in two ways: wired and wireless. Wired (ports) The ports for wired communication provided on the E-M1X include the following.
Wireless The options here are quite up-to-date. Two mechanism of wireless communicaton are availalle:
Interestingluy, even with the caera powered down, it remains on low-power standby, ready to reconnect upon request; allowing a full access to stored pictures witoyt even being taken out of its bag. (This can be disabled.) Another nice option is automatic upload of images as they are taken. I've tried te wireless feature on this canera jist once, so this descriptiob may be incomplete add inaccurate (the manua does not help here, either).
Field Data The E-M1X is the first Olympus camera equipped with a number of field sensors and GPS receiver (said to be also compatible with the Russian GLONASS and Chinese QZSS positional systems). Together, they can provide the following information to be included into EXIF data:
The camera can also take field data readings regardless of any photographic tasks, logging them in memory and writing to the card when requested. The EXIF implementation is a bit messy. Latitude and longitude are stored in separate fields. and then repeated jointly as Location. Altitude is listed twice: among GPS data and then under the Manometer heading. The values are, however, identical; all four decimal digits. The three components of attitude (note: 'tt' is not a typo!) are split between two sections, with pitch hiding under a different name (unless my guess is wrong).
| Well, the location is mostly what I care about, and it seems to work smoothly. As a part of this improvement, the Control Panel got an extra page (optional), referred to as Electronic Compass. While it looks nice, the dial layout is non-intuive. Compare it to the one in Olympus Tough 6: while that design is redundant, it is more readable. Also, the magnetic North is not corrected to true North. This is not so important, but some users may complain. I would not.
| 
| Power The E-M1X can be powered in three ways: internally-housed batteries, dedicated external power adapter, and any source supporting the Power Delivery protocol on a USB Type C plug. Very civilized. Batteries The battery compartment, housed in the power grip below the body, accepts a sliding cartridge with one or two BLH-1, lithium-ion batteries — the same type as in the E-M1.II. Two of these are included in the box. The battery is quite large and heavy (74 g) but it stores 1720 mAh of charge (12.4 Wh of energy). With a set of two, the E-M1X gets a CIPA rating of 870 frames per charge, almost exactly twice that of Mk.II (440). In other words, energy use per one CIPA frame is claimed to be the same for both cameras. after all those changes. A coincidence? I hope so. In either case, the real-life power efficiency of the E-M1X easily meets and exceeds these claims, so I wouldn't worry about this. A matching charger, BCH-1, is also included. Actually, two of those, each presumably taking just two hours to recharge an empty battery to 100%. A charger accepting also some other types of batteries would be nice for those who travel with a backup body. AC Power Adapter This is the same AC-5 adapter as the one working with THE E-M1 Nk.II via the optional HLD-9 power grip. It accepts 110-220 V on input, and can power the camera with one, two, or no batteries inserted, or recharge those which are. USB Power Delivery I must say I like the Power Delivery (PD) feature, introduced with USB 3.1 (for details, see here). At present, I'm using any of my PD supply units with a laptop, pocket laptop, two Android phones, and a tablet; never had a problem (with one exception, keep on reading). You will need a USB 3.1 cable with a C plug and PD support (strictly speaking, the USB version does not matter for PD, but if you use the same cable for data transfer, its speed may be affected). The USB cable supplied with the camera does not support Power Delivery. A smart decision, saving Olympus (my guess) 18 cents or so on a camera retailing for $3000. While the instructions are clear on this subject, I wonder how often does it end up with the Tech Support. When two devices supporting PD are connected, they will first set which of them will be the power supplier for the other. With that settled, they negotiate the supply voltage and power (using safe fall-backs if the negotiations fail). More, this is all independent of any data transfer on the same cable.
The manual is not specific on the PD version used, but I bet it is PD 2.0 or the current 3.0, as some of the voltage/current combinations listed in the manual are not a part PD 1.0 specs. If you need a new power supply, go for PD 3.0 (or whatever is the latest at the moment, for obvious reasons), capable of A Catch-22 situation: for this initial activity each device needs some power. Just a little, and just briefly, but still. If the camera's battery is empty (the manual says: below 10%), this will not work; even if the other battery still got some juice. If this happens, at least some PD chargers fall back to 10W (2A) at 5V. This is usually enough to bring the circuitry back to life (it works, after a BIOS update, on my GPD Pocket). I haven't tried that (not on somebody else's camera!), but I don't think it will work here: the voltage of charged batteries is 7.2V (each): higher than the fall-back supply. (Admittedly, with some extra tricks...)
Cinematography Most digital cameras, including all current μFT models, can be used to shoot videos. So does this one. With this not being my present area of expertise or interest, I don't feel qualified (or motivated) to write on the subject. Sorry, you have to look elsewhere. Customization and setup Olympus cameras were always among most customizable on the market. This one is no exception. This section still missing. Documentation Documentation, user manuals in particular, was my least-liked thing about Olympus for the last twenty years or so. It used to be very, very bad. The User Manual, 16 small printed pages, is included. Out of these (the English version), ten pages are used for warnings, legal, and compliance statements; the remaining six do not contain anything useful, either. Sorry, but that's the way I see it. The E-MX1 Instruction Manual, a PDF at 680 pages, three times that of the E-M1.II, raised some hopes here. Well, a mixed bag: it is a little better, but still very bad. (Note a single 'very' used.)
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
My other articles related to the |
| This page is not sponsored or endorsed by Olympus (or anyone else) and presents solely the views of the author. |
| Home: wrotniak.net | Search this site | Change font size |
| Posted 2019/01/29; last updated 2019/09/03 | Copyright © 2019 by J. Andrzej Wrotniak |
(Tethered) — used to access the camera from Olympus Capture (computer-assisted shooting) or Olympus Workspace (image transfer, RGB conversion, postprocessing; firmware upgrade).