Olympus E-1, Critically Annotated |
My other articles related to the |
The E-1 is a significant camera in the evolutionary process of digital photography technologies. This is the first interchangeable-lens SLR model ever (from any manufacturer), developed from ground up as a digital camera: unlike all competing models, it did not originate as a modified film SLR. | ||
This is not the first venture of Olympus into the digital SLR area. The E-1 bears some resemblance to two earlier digital SLRs by Olympus: E-10 and E-20.
Those, however, differed in one essential aspect: they did not offer lens interchangeability. (The latter, I believe, being a very desirable feature, is not without its drawbacks; everything has its price.) Many features of those cameras, including most of the ergonomics, migrated into the E-1, although interchangeable lenses and bigger sensor size necessitated a complete reworking of the camera internals. | ||
The E-1 is also the first camera following the Four Thirds standard. Developed jointly by Olympus and Kodak, this standard is based on a 4/3" sensor, and specifies the lens mount (including its distance from the image plane) and the communication between the lens and camera body. Any lens following this standard will work with any camera body adhering to it, regardless of model and maker. So far, however, no manufacturer other than Olympus makes cameras based on the Four Thirds standard, and it is not expected to happen in a predictable feature. The major players in the SLR field (Canon and Nikon) decided to stick to their respective, proprietary mounts, inherited from film cameras (although with some modifications in electronics). One of the reasons was to allow the buyers to reuse their existing SLR lenses (with some limitations); another — not to open their camera systems to independently made lenses, or at least to make it more difficult. Even Kodak and Fuji (both participating in the Four Thirds initiative) offer their SLRs with Nikon and Canon mounts only. Whether the E-1 and its followers will succeed, remains to be seen. If they fail on the market, it's not for technical reasons: the E-1 is a beautifully designed and executed camera. (If you are old enough, you may remember the Beta versus VHS controversy; the better player does not always win.) Regardless of that, the E-1 is, I believe, a milestone in digital photography, and will find itself a niche of dedicated followers. This article contains a detailed walk-through through E-1's specs and features, heavily annotated with my remarks on their meaning, significance, and background. Additionally, my English friend and an experienced photographer, John Foster, annotated my annotations (yes!) with his hands-on comments, based on almost a year of using the camera. John's annotations are always clearly marked as such. John also wrote a User Report on the E-1, posted as a separate article on this site. | ||
Construction:
Magnesium alloy cast |
Continuing the tradition of E-10 and E-20, at least as good. A tough act to match.
John: It has matched and dare I say, beaten the previous E series cameras. Although the E-1 was designed from a clean sheet with no 35mm lens baggage, a camera is a camera is a camera, and from a design point of view there are no massive leaps to be made. The result is the E-1 looks similar to 35mm-derived DSLR bodies. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Finish:
Crinkled black on metal, rubberized grip and thumb area |
Same here. The feel of the camera is better than that of the Canon D10 or D20, and much better than that of the Nikon D100 or D70.
John: The E-1 has been constructed to the highest standards, to satisfy the original target customer, the professional photographer. As it turns out and just like the famed OM system its main customers seem to be dedicated advanced amateurs; the only distinction I'm drawing here is that for advanced amateur use it is maybe a tad over-engineered. (Just a hindsight comment.) This does not mean E-1 is not being used by pros, it is; but not as many as Olympus would like. The standard of construction in many ways matches the considerably more expensive flagship offerings from Canon and Nikon, and is superior to its supposed target competition – mid range DSLR. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Weatherproofing:
Yes |
Gaskets and seals everywhere, pro grade.
John: Not weatherproof in the general sense, but meets industry standards for drip- and dust-proof. When I removed the viewing eyepiece for cleaning, I could see how careful Olympus is in making the camera D&D proof: even the bayonet-type eyepiece connection is sealed. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Size (WxHxD):
141×104×81 mm |
Compare this to the E-20: 129×104 mm (WH) — just 12 mm wider. Again, the "transition" cameras are generally larger.
As compared to film-derived digital SLRs, the E-1 has another advantage: the strongly asymmetric shape. If your nose is of average size or more, you will find Olympus easier to use; with the other models you'll be sticking your proboscis into the monitor screen all the time. John: Though I've not used the E-10 or E-20, I have handled them, and the finish on the E-1 seems of a higher quality, less "plasticky". Its heft, weight and ergonomics inspire confidence. For sure, this, as its predecessors is no toy. The asymmetrical shape gives space exactly where needed, Plenty of room between grip and lens; for example, for stubby fingers. The lens mount is placed well out to the side of the body making the lens sit nicely in your left hand. It is smaller than the main rivals but above all it falls to your hands; design kudos. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Weight:
|
This does not include the batteries. Almost identical to the E-20, which weighs 1050 g (37 oz) with the built-in lens.
John: The BML-1 Battery adds an additional 75 grams (3 oz.). The competing SLRs, based on bodies adapted from film cameras, weigh more (data w/o lens): Canon E-10 at 875 g, Nikon D100 at 780 g or D1X at 1200 g, Fuji S2 at 850g, this in spite of the fact that the E-1 seems to be sturdier, in addition to being weatherproof. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interchangeable lenses:
Yes |
Most people who say "I wish this camera had interchangeable lenses" do not see the other side of the bargain: CCD pollution. Dust tends to stick to the CCD surface (or to the front of the low-pass filter in front of it). There is a whole industry of contraptions designed to remove the specks (sticky swabs, brushes, whatever), and when you see the problem you may think that the fixed lens in the E-10/E-20 was not such a bad idea after all.
Olympus addressed the problem with a system where the IR-blocking filter is submitted to ultrasonic vibrations, supposed to shake off the dust which then will, or at least has a chance to, be caught by a sticky tape. This probably is a piezoquartz device, and the cycle is activated every time when the camera is turned on. The solutions seems to be quite effective. John: True; I don't know the technicalities of how it works, but it is very effective. The camera is supposed to be in the normal upright position when switching on — something I occasionally forget. Olympus says the device will remove most dust and dirt, and in addition the detailed, CD-based handbook devotes a section to do-it-yourself CCD cleaning, with appropriate warnings. I've had no need to undertake this yet. No evidence of dust from the standard blue sky test. Three Digital Zuiko lenses were available at the camera introduction; as of this update (November, 2004), seven are available from Olympus, three from Sigma, and a super-wide 7-14 mm from Olympus should hit the shelves in 2005 (start saving money now; this one will not be cheap!). A full list with basic details can be found here. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mount type:
The 4/3 standard |
This is an open standard announced by Olympus together with Kodak last year, an attempt to provide interchangeability between various makers' lenses and cameras. Obviously, the standard includes mechanical lens-camera connection, the flange-to-image distance, but also the communication protocol between the lens and camera (see below).
Fuji's participation in the standard is so far limited to verbal declarations; Sigma joined the bandwagon with three lenses so far, promising more. I am skeptical about any other major manufacturers (Canon, Nikon) building 4/3 bodies: they seem to be committed to APS-sized sensors, and they have all a long history of selling lenses with a proprietary mount to captive audiences. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Mount size:
46 mm (internal diameter) |
This is large, even in 35-mm camera terms. The first 35-mm SLR, the Exakta (1938-1972), had a bayonet with 39 mm internal diameter, and the most popular mount in history, M42 (Praktica/Pentax), had a 42 mm thread.
John: The mount is the same size as the Olympus OM one. In fact the OM bayonet flanges have been altered only marginally, probably to prevent direct mounting of OM lenses. Now, remember that the 4/3 format frame has the width of just 18 mm, compared to 36 mm for the film, and you will see why I'm surprised. Relatively, the mount throat is twice as wide in the 4/3 standard than in 35-mm film cameras. The relative film-to-flange distance is also considerably larger. This obliterates most of the claimed size advantage of the 4/3 lenses over ones inherited from film cameras, at least for focal lengths from wide angle to moderate tele. Except for the 300 mm (600 mm EFL), all announced Olympus lenses are roughly the same size as their 35-mm equivalents of the same EFL (effective focal length). John: Zuiko Digital lenses are bulkier but shorter than the OM Zuikos of comparable angle. There were good reasons for all this, mostly the need for smaller incident angles of light reaching the CCD. Smaller incident angles significantly improve the quality of the CCD response, wise people say. (The difference will really matter, I think, for wider lenses, but not as much as the focal length would indicate, because in 25-mm SLRs these have to follow the "inverted telephoto" design to accommodate the mirror anyway.) John: It seems to make sense as the CCD has more of a 3-D surface than film. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Control coupling:
Electronic |
Having no lens legacy ballast, Olympus can afford this approach. All data and control is transferred between the body and the lens electronically — more accurate and reliable than mechanical transfer.
In addition to obvious items, like aperture, focal length, and focus, there are some new ones, seen never before. According to Olympus, some lenses transfer the information about their optical distortion (an optical flaw where the lens magnification depends on the distance from its axis, causing the well-known barrel or pincushion effects) or about image fall-off (getting darker off-axis). This information is then embedded in the raw image data, then used either in-camera or in postprocessing to remove the effect. John: With E-1, this can be done in Olympus Studio, but not in Olympus Viewer, included with the camera. I am unfamiliar with the former, as I'm not prepared to pay a further £100 for the full version. My apologies. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sensor type:
Full Frame Transfer CCD: KAF-5101CE by Kodak |
While in older CCDs some of the sensor area between photosites is not light-sensitive (used for charge-reading circuitry), in this type there is almost no such loss: the photosites fill most of the chip.
These changes result in greater light sensitivity, lower noise levels, greater dynamic range, and less interaction between neighboring photosites (blooming). Of course, it remains to be proven if and how these potential advantages are translated into image quality improvement. As they say in the U.S., there are many ways to suck an egg. In all other aspects this seems to be a "classic" CCD array, like ones used in most digital cameras in the last few years, with a mosaic of RGB filters in front of photosites (other solutions include, at this moment, a CMOS array used in some Canon and Nikon SLRs, a four-color filter recently introduced by Sony, and the Foveon sensor, with each photosite responding to all three primary colors). Clearly, Olympus has chosen the most proven approach here, although in the best possible implementation. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Effective sensor size:
17.3×13.0 mm |
The often quoted size of 18×13.5 mm refers to the gross size, including parts of the sensor array not outside of the image area. Horizontally, this is a bit less than half of the 35-mm frame width. Compared with the E-20, the sensor is twice as large (linear), with four times the area.
Obviously, this leads to greater pixel pitch (spacing between pixel centers, horizontal and vertical). Combined with larger relative photosite size (see above), this opens a potential for better images. The individual photosite size in this chip is 6.9x6.9 microns. The sensor size affects the effective depth of field. According to the N times F rule, shooting with the E-1 at some aperture (F-number) is equivalent, in terms of DOF, to shooting with a 35-mm film SLR with the F-number doubled. (I am assuming that the angle of view is identical in both cases.) This means that while DOF is significantly greater than for a film camera, it is still possible to blur the background as desired in some applications. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Pixel count:
Nominal 5 megapixels, or 4.69 "binary" MP |
The stored image (in full resolution) consists of 1920*2560 = 4915200 pixels. In computer applications the "mega" prefix does not mean one million, like in the metric system, but 2^20 (2 to the power of 20), i.e., 1048576. Therefore in my book the E-1 resolution is 4.6875 megapixels.
Camera manufacturers prefer to use the metric definition of "mega": 5MP looks better on paper than 4.7 MP. John: It may seem that camera makers are beginning to admit that 4-6 MP is sufficient to match 35mm film. Or perhaps the public is, at last, seeing the megapixel race for what it truly is — commercial exploitation. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Native image size:
1920×2560 pixels |
Olympus chose to stick to the "digital standard" 3:4 ratio, as opposed to the 35-mm frame, 2:3 proportions. I think this was the right decision. The 3:4 ratio is generally more pleasing and better fits most of the standard enlargement sizes. The "purists" who stick to the 2:3 ratio tend to forget that it was a result of an accident: when Mr. Barnack built the first Leica back in 1913, he used two 3:4, movie-standard frames side-by-side. 3:4 is a return to the original ratio defined in view cameras.
John: The 3:4 aspect ratio is for me absolutely correct. 2:3 (the 35-mm letterbox size) inevitably needs to be cropped to achieve an aesthetically pleasing print that normally ends up near 3:4 in any case. Check it for yourself from your own cropped and printed images; 75% of mine come out at an unplanned 4:3. Cameras using the 3:4 ratio make a better use of the pixel count. For most standard print sizes (except of the smallest, 4x6"), the proportions are close to 3:4, not to 2:3. Cropping an APS-sized frame to these proportions leads to 11% of pixels being just thrown away; suddenly your 6MP camera turns out to be really a 5.3 MP one. Some people complain that the E-1 has "only" five megapixels, and not the 6MP of some of competing models with APS-sized sensors. Let's check the numbers, for a moment neglecting any factors other than pixel count: for example, Nikon D70 produces images 2000x3008 pixels (why the extra eight?). When printed in the 3:4 aspect, it gets an advantage of just 80 pixels vertically and 107 horizontally, 4% in each direction; 341 vertical rows of pixels (or almost 0.7 MP) are just thrown away. Yes, you can get bigger prints from the Nikon, by a whopping 4%. Go figure. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Color depth:
12 bits per RGB component | This means extending the dynamic range of previous Olympus cameras (10 bits per color) by a factor of 4×, offering a capacity to preserve detail in highlights and shadows to a much larger degree. How it is utilized, depends on the process of converting the raw image into the 8 bits/color format (JPEG, TIFF, whatever), whether done in-camera or in postprocessing from an ORF file. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Sensitivity:
|
The manual settings of ISO 1600 and 3200 are available only if this option is enabled from the system menu; otherwise you can go only up to ISO 800. While the manual does not elaborate on this, I suspect that speeds of ISO 1600 and 3200 are achieved not by increasing the CCD gain, but somehow at the image processing stage.
Interestingly there are no longer any settings below ISO 100; this is a more sensitive chip. The images I've seen at ISO 1600 are quite noisy. What would you expect? | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
White balance:
|
The presets range from 3000K (incandescent) to 7500K (open shade).
What I like most is the reference setting, where you point the camera at a white (or gray) card, which you want to have rendered as neutral. This feature was first introduced in the E-10 (where it was very flaky, not really useful), and then refined in the C-5050Z and C-5060WZ, where it works really well, giving me most pleasing results. (Actually this seems to be the only way to get the WB right with low-power incandescent bulbs, below 3000K.) In the E-1, you can save up to four such settings independently, to be later recalled. John: Auto WB is beautifully implemented. Override or adjustment is hardly ever needed. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
WB adjustment:
±7 steps in 20K increments | These adjustments can be applied individually to any of the presets (including auto and user-defined ones), and will be recalled when that preset is used. This way you can, indeed, customize the entire WB system in the E-1. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
WB bracketing:
Three frames, 50K, 100K, or 150K apart |
This is a nice-looking feature, but I don't think I'll have a use for it: this adjustment is easy to apply in postprocessing.
The camera does not actually make three exposures here, but just saves three RGB files with different WB applied. Obviously, the feature is disabled when the raw file format is used. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
WB sensors:
Internal and external | The additional, external sensor is not common in digital cameras. It is useful in deciding whether the image is, say, predominantly red because the subject is red, or because the incident light is shifted towards that color. Obviously, in the latter case the camera should apply color correction, while in the former it shouldn't. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Noise reduction:
Yes, both fixed and random |
Unless you are familiar with the difference (and even then), refer to my noise article first.
The fixed-pattern noise is reduced (and hot pixels effectively removed) by dark frame subtraction, a traditional and proven way to do it. Note that this, while removing the fixed noise, increases the random noise by about 40% (Olympus cannot just change laws of physics), so the reduction should be (and is) applied only at long exposures, where the fixed noise prevails. This usually means exposures of 2 seconds or longer. The random noise is addressed by algorithmic filters, weighted-averaging the signal from neighboring pixels. While the process is quite smart (the degree varies with direction and depends on the local detail), it may lead to surprises, filtering out not just noise but also image detail. I've seen cameras, praised for low noise levels, which were quite notorious in this aspect. I have also used a (very good) stand-alone noise-removal program with quite good results (noise removed without contour sharpness degradation), until once it successfully removed all shingles from a roof. This is why I think the random-pattern noise removal should be a defeatable feature. Olympus agrees, and in the E-1 you can turn it off. (When on, it also slows image processing.) If you are saving images in raw format, you can always turn noise filtering on during the image conversion. John: Noise reduction is not available for sequence shooting. The image processing time increases by fifteen times if this feature is active. I recommend to keep it off, to be applied later on a PC. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Color space:
|
This setting defines how the whole color gamut is represented by a mixture of three basic components.
Adobe RGB is can be useful if you prepare your pictures for print using the Adobe color management in PhotoShop. I haven't found much use for it yet. | ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Image adjustment:
|
The user interface is mostly external (buttons and control wheels); some functions are accessible from the monitor menu. At the moment of its introduction, the original E-10 offered what I believe was the best, most convenient set of controls available on any digital camera, regardless of price. The same set could be found on the E-20, and then, with minor changes only, on the C-5050Z and C-5060WZ optical-finder models. It seems that Olympus wisely retained most of that interface on the E-1, the only changes being small refinements. The basic functions are assigned to two control wheels at the right side of the body, very much like in the E-10/E-20. This is done in a reasonable and logical way:
The setting-access buttons do not require much getting used to. The full list includes:
Two more buttons work solo, i.e., without control wheels: exposure lock and control panel backlight, while focusing mode (single, continuous, or manual) is set with a separate lever. Last but not least, the camera back hosts the menu and display activation and navigation buttons, very much like almost any model from almost any manufacturer. All other settings, including user's preferences, are accessible from the LCD monitor menus. Those are organized in a no-nonsense manner, similar to that on the E-10/E-20, and quite unlike those on the newer Olympus C-series cameras. There is a number of quite detailed reviews of the E-1 posted on the Web. The ones I found most informative (although not without errors and omissions) are
|
My other articles related to the |
Olympus® is a registered trademark of Olympus Corporation.
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 2003/06/26; last updated 2009/01/10 | Copyright © 2003-2009 by J. Andrzej Wrotniak |