This is a detailed, technical review of the Olympus E-300. It is based on a final, retail version of that digital camera, bought at my own expense, with Version 1.0, of the firmware upgraded to the latest v.1.3, and the text modified to reflect the changes.

• Die-cast aluminum frame
• Steel (?) lens mount
• Polycarbonate body shell
• Aluminum top cover

The top cover looks like either a bold fashion statement, or an afterthought, depending on your attitude. I'll have to get used to it.

The absence of the pentaprism hump above the lens reinforces the unusual look (note: even many digital-finder cameras have an imitation prism hump, to make them look more like SLRs).

While the look is high-tech, I don't like it too much. For the time being, the distinction of best looks among digital SLRs belongs to the Pentax *ist DS, I think.

In particular, the flat-top appearance can be considered nice or ugly, depending on your taste, attitude, and the day of the week. Regardless of that, the camera looks and feels more "serious" than others in its class.

By the way, the body is made in China. This made me a bit suspicious at first (stereotypes?), but only the label gives that away, not the construction or finish quality. A first-rate job here.

The camera is easy to hold securely (in spite of being left-side-heavy); the rubberized hand grip and thumb rest are of the right size and shape, yielding to the touch with the kind of feeling I like.

The E-300 certainly feels better than and not as cheap as the competing models, especially the Canon 300D Digital Rebel. At least from the outside you wouldn't guess this is an economy model.

If the camera width could have been reduced by 15 mm or so, the proportions would be better, and the size more enjoyable.

For sure, this is not a size revolution like that of the famous OM-1; we have to wait a few more years for that.

One would think that the Four Thirds standard, with the image size being half of that in 35-mm cameras, would allow for smaller cameras. Not so: the E-300 (not to mention the E-1) is still larger than any full-frame, 35-mm SLR, excluding some heavy-weight, pro-grade models.

Together with the bundled 14-45 mm zoom the camera weighs 865 g. This is almost 200 g less than the E-20, a noticeable, if not dramatic, difference.

Index

Lenses

Zuiko Digital 40-150 mm F/3.5-4.5
(Picture by Olympus)

Because this is an open standard, with well-documented interfaces and specs, we should see more third-party lenses for the E-300, more aggressively priced. At this moment Sigma already has three models.

Generally speaking, I would divide the available lenses for the 4/3 system into four price categories, correlated with specs and quality: cheap (up to $200), moderate (up to $500), expensive (up to $1000), and exotic (beyond).

So far only two zooms by Sigma fit into the first category, while the second one includes another Sigma zoom, two mid-range zooms from Olympus and, barely, the 14-54 mm F/2.8-3.5 introduced with the E-1. The expensive group has two Olympus zooms with attractive focal length ranges of 11-22 mm and 50-200 mm, and with wide apertures of F/2.8-3.5. Finally, the exotic division consists of two prime (non-zoom) lenses of professional grade and prices to match: a 150/2.0 ($2500) and a 300/2.8 ($7000). I doubt Olympus sells many of these, but they play an important role of presenting the E-series as meeting the needs of the most demanding users.

For reference, have a look at the full, current 4/3 lens list.

As I've already mentioned, most of the 4/3 lenses by Olympus are quite expensive, but the 40-150 mm F/3.5-4.5 (shown at left), aimed at E-300 users, sells at $280. The image quality of this lens is surprisingly good; adding it to the bundled 14-45 mm will cover the 35-mm equivalent focal length range from 28 mm to 300 mm.

OM lens adapter: On some markets and at some times Olympus offers, upon request, a free adapter allowing you to use Olympus OM-series lenses on E-system cameras. Do not ask me, check the Olympus user support. If not given away, it is sold at the outrageous price of $200. For more on this, see here. Other adapters are available for other types of lenses. Obviously, you can use non-4/3 lenses only in manual or aperture priority mode, with stepped-down metering.

(Picture by Olympus)

The picture shows how big the lens mount is, as compared with the body. You can also see the electric contacts to pass the information to and from the body. The viewing screen is positioned vertically, immediately to the right of the mirror, but hidden beyond the mount flange and not visible in the picture.

The bayonet action is smooth, precise, and positive, not a cheap job.

As a matter of fact, 4/3 lenses have their own computer chips with the embedded firmware, which can be updated from the camera.

The 14-45 mm zoom bundled with the E-300.
(Picture by Olympus)

Judging from my results, the lens is a good performer. In designing this lens Olympus made the right choice, putting optical performance ahead of the (relatively modest) specs.

The 14-45 mm focal length range (equivalent to 28-90 mm on a 35-mm camera) is more useful than one starting at 35 or even 38 mm, often encountered in the specs-driven electronic finder cameras. A 3x zoom starting from 28 mm EFL is more difficult and more expensive to make than a 5x one starting at 35 mm (which might be better received by some segments of the market).

The aperture closes down to F/22; at this frame size this is not trivial. While F/11 would be adequate for general shooting (and cheaper to make), F/22 comes useful in close-up applications, a nice thing.

On the downside, the lens is quite dark: F/3.5 at the wide end, and F/5.6 at the long end; this is a 1.36 EV difference. This affects not only exposure, but also the brightness of the image in the viewfinder.

The compromise has its reasons. Among the four variables: focal length range, maximum aperture, optical quality, and price, you can adjust almost arbitrarily any two; with some difficulty and extra limitations — three, but never four, period. And Olympus got it right in this case.

The make and finish of the lens (also made in China) are OK, with the zoom mechanism providing the right feel and precision. While it does not ooze quality like its more expensive counterpart, it is far from feeling cheap.

The front element does not rotate when you zoom or change focus (important when you use a polarizing filter), but zooming is not internal: the inner barrel slides in and out of the outer one in a cam-driven action. This is why I would advice against using any lens attachments heavier than 100-150 g (4-6 oz).

The zooming is mechanical, while focusing, done via a ring on the lens, uses electronic coupling. I was never enthusiastic about the latter feature, but it widely used by all makers, perhaps being less expensive.

This lens (and its 40-150 mm "economy" sibling) does not offer a distance scale, which is present on all mid- and up-range 4/3 lenses from Olympus. Once again, an economy measure, but I can easily live with this.

It's a pity you cannot (as of December, 2004) buy the E-300 body alone. I would be quite willing to pay $300 extra to get the 14-55 mm lens, not really for a greater focal length range, but for a brighter viewing image and faster shutter speeds.

Index

Image sensor

(Image by Kodak)

In full-frame transfer CCDs all connecting circuitry runs in a layer beneath the photosites (light-sensitive elements), not alongside. This means that the photosites themselves may occupy almost all surface of the chip, with a greater effective light-collecting size. The image quality may potentially be improved this way, with less noise and not only.

These chips require a real, mechanical shutter to define exposure times (electronic switching, or "gating" does not work with them), but this is not a problem with mid- to high-end cameras which all have mechanical shutters anyway.

Those interested should have a look at the detailed and interesting discussion of the KAF-8300CE by Jason Busch at digitaldingus.com.

I have read some complaints about how much is this smaller than the APS size used in other digital SLRs. Let's check the numbers. The Canon EOS 20D chip is 22.5×15 mm. Some of the longer dimension is wasted anyway: of common print formats only 4x6" (10x15 cm) has a 3:2 aspect ratio capable of using the full 3:2 frame; most others are closer to the 4:3 ratio, like the "small" exhibition size of 30×40 cm (12×16"). Thus, the 20D suddenly becomes a 7MP (not 8MP, as 1 MP has to be cropped out) camera, with the usable chip size of 20×15 mm; the linear difference is only 15.5%. Hardly a big deal; and taking into account the full-frame transfer chip in the Olympus, the actual photosite size may the same or larger in Olympus models than in APS-sized ones at the same resolution.

Why all this hair-splitting? First, I'm addressing my articles to Readers who like to know such things, second, it simplifies the calculations. An uncompressed image from a 7.62 MP camera, with 8 bits (one byte) per color, will contain 3*7.62 = 22.86 MB of information.

Now, is this a progress from the 5 MP of the E-1? Not necessarily, and for many reasons. Generally, the current technology (and, to some extent, the laws of physics) defines some pixel count above which, for a given sensor size, there is little, if any, additional gain in image quality. I would guess this limit, for the 4/3 sensor size, is somewhere between 5 and 8 MP. The 6 MP lower-end cameras already crossed this line (with the mass market being largely ignorant, and following the "more is better" marketing ploy). Yes, I would like to see a 20 MP camera, but only with the full 24x36 mm (or larger) frame size.

On the other hand, 8 MP is 60% more than 5 MP, and this translates into an almost 30% increase in linear pixel resolution (square root of 1.6 minus one). Whether your pictures need that increase, and whether the camera is capable of using it (with the lens being possibly a bottleneck) is often hotly debated. My personal feeling is that same-sized prints (of, say, 12x16 inches or 30x40 cm) from the 5 MP E-1 are as good as those from the E-300 in terms of perceived sharpness. Still, 30% more of linear resolution is nothing to sniff at, and better lenses can take advantage of this.

I consider this important: so far all SLRs by other makers follow the traditional film proportions of 3:2. When printed on any standard paper size except for the smallest 4x6" (10x15 cm), those cameras waste about 15% of their pixels, not to mention that the image composition may be also impacted or even ruined. How come nobody mentions that?

I'm using an SLR largely because of the accuracy in framing and composition, and I like framing my subjects tight. I don't want to keep guessing how the results will be affected by cutting 15% off the image width. This takes precedence above many other factors relentlessly discussed by the pixel-counting crowd.

Greater color depth, potentially at least, allows us more flexibility in squeezing the image tonal range into the 8 BPC depth of most formats used for final images (JPEG, TIFF), helping to extract extra detail from highlights or shadows.

The full 12 BPC information, obviously, will be retained only if your image is saved in the raw ORF format, see below. Otherwise it is up to the in-camera circuitry to use the extra bits wisely in the conversion to 8 BPC. (JPEG bits are assigned with use of a logarithmic scale, so there is no loss of dynamic range, but granularity problems remain.)

Being able to save the image in one of the "deeper" non-proprietary formats (16 BPC JPEG 2000 or TIFF) would allow the photographer to take full advantage of the captured information, without necessitating use of any ORF conversion software. Unfortunately, all camera makers are still ignoring this option.

• Auto setting: ISO 100-400
• Manual settings: ISO 100-1600

To make the two highest settings available, you have to activate an option in the Setup menu. This is a sign that the maker considers them less acceptable than others. Still, I prefer a noisy picture to no picture at all, and high grain was always a fact of life in available-light photography.

It is possible that, like in some other cameras, the settings beyond ISO 400 do not actually change the CCD gain, being implemented only in image conversion. In any case, don't expect miracles at the highest ISO values.

Generally, I find all settings from ISO 100 to 400 good enough for any purposes, and even ISO 800 OK for low-light uses. ISO 1600, however, is very noisy, and usually will require s good noise removal program applied in postprocessing. For more on this, refer to my image samples and noise removal articles.

In some cameras this filter is combined with the low-pass one, used to add a little "fuzziness" do the image, to reduce the possibility of Moiré patterns; in the E-300 it may be also combined with the dust protector, but it is unclear if it really is. Anyway, this, and the mechanical constraints, make the anti-IR filter removal unfeasible.

Infrared sensitivity: My IR session with the E-300 shows that the exposure compensation factor when shooting with the infrared Hoya R-72 filter is about only about 7 EV — a surprisingly high infrared sensitivity! More, the IR images are less noisy than those from other Olympus cameras I've tried (E-10, E-20, C-5050Z, C-5060WZ). This means that, in spite of the lack of live image preview (missing in all digital SLRs except of E-10/E-20), the E-300 can be quite useful in IR photography.

When activated, it uses ultrasonic (25 kHz) vibrations to shake most of pollution particle types off the dust barrier (or low-pass filter), catching them (hopefully) on a sticky surface in the mirror chamber.

This is important. Imagine every single speck of dust from your every film frame showing up on all your next frames as well, accumulating over time, and you will see what I mean. Or check the plethora of dust-removing sticky swabs, blowers, brushes, voodoo sticks and prayer beads, used with all other (interchangeable-lens) SLRs; this is what was making me hold on to my fixed-lens E-20, hoping that some day someone will come up with a kind of a solution. Well, Olympus did.

The value of 25 kHz is a correction of the 350 kHz one which I originally posted repeating a piece of information quoted in one of other reviews. The correction follows the statement of Mr. Masaharu Hamada of the BCT R&D Department in an interview published at the E-Fotografija Web site in October, 2004.

Based on my initial experiences and a few reports I've received from E-1 and E-300 users, the ultrasound dust removal really works, although I would still limit exposing the mirror chamber to elements to minimum, and in the cleanest possible conditions.

Update of September, 2005: My monthly checks of clear blue sky frames have yet to reveal the first dust speck over the imager, and I'm switching between lenses quite often. The system really seems to work! It remains unclear how often do you have to send the camera to Olympus to have the sticky tape replaced, though.

The dust shake-off is activated for a second or so every time the camera is turned on. For better efficiency of the process, the camera should be in the normal upright position at that time.

At least two reviews I've seen say that the Cleaning Mode activates the ultrasound dust shake-off process. An obvious mistake.

Image processing

• Auto
• Manual
• 8 presets
• 4 user-defined
• Reference

The presets include incandescent light (3000K and 3600K), fluorescent (4000K, 4500K, 6600K), sunny (5300K), cloudy (6000K), and open shade (7000K). I am happy to see, at last, each preset marked with both the corresponding symbol and the color temperature; no big deal, but it took Olympus a few years. Better late than never.

The reference WB setting was originally introduced by Olympus in the E-10, but it begun to work properly only in the C-5050Z. This is done by pointing the camera towards a neutral (white or gray) surface and telling the camera (from the menu, that is): "this is what I consider neutral, white or gray". In the '5050 and '5060 this worked very well, giving best results under tricky conditions. On the E-10/E-20 and E-1 the feature has its own, dedicated button at camera's front; but, unfortunately, not on the E-300, one of first signs that this is an economy model. (You can assign the reference WB to the customized OK button, but this, for a number of reasons, is not as good.)

Four user-defined settings can be adjusted manually between 2000K and 10000K in a number of discrete increments.

Unfortunately, the captured reference WB cannot be entered into one of these slots. You could do it in C-5050Z and C-5060WZ, and I found it very useful (for example, setting up two different, metered presets for tabletop photography). The E-1 has four reference WB memory slots, too. The absence of this option in the E-300 is really a shame, as the feature can be implemented entirely in software.

Setting the white balance by reference causes the camera to actually take a picture, and then analyze it in order to come up with the right setting. At first is seems a bit strange, but when you look at this closer, it isn't: how else can you do it in an SLR, where the CCD itself is used to process the information?)

Another omission is the external white balance sensor. Roughly speaking, this sensor (present in the C-5060WZ and in E-1) allows the camera to tell if a scene is red because the object itself reflects red wavelengths, or because the light used to illuminate the object is red. In the latter case the color balance should be corrected towards cyan, in the former — it shouldn't. This looks like some corner-cutting again, but I'm not sure if and how it will affect real-life results. (This is not just the cost of the sensor: to use it, the camera needs a second, independent, color temperature readout circuitry.)

Note of May, 2005: While in most cases the auto WB in the E-300 delivers very good results, I have noticed occasional problems when images shot outdoors with large amounts of green are overcompensated into magenta. Sometimes I see this effect in just one frame of otherwise almost identical shots taken within a few seconds from each other: quite a random behavior and quite difficult to understand (a consistent overcompensation would be easier to explain). This happens even after I've upgraded to the Version 1.2 of camera's firmware, and I've never seen this behavior in other Olympus cameras I have used extensively in the past (C-3000Z, C-5050Z, C-5060WZ, C-60, E-10, E-20), except maybe, to some extent, in the E-1.

As I said, this happens quite rarely, but it can be quite annoying. I hope another firmware upgrade will address the problem — it is hard to believe Olympus is not aware of it.

The adjustment increment is my guess: I suspect the step to be the same as in the E-1; larger changes can be done by just switching to the next preset. (The camera's Advanced Manual is not advanced enough to bother us with things we'll never be able to understand.)

Instead of marking the corrections in degrees Kelvin, Olympus uses these steps (of undocumented value), and does it in a quite confusing way. Applying a positive correction of say, +5 steps to a setting of 3000K gives, as a result, the white balance of 2900K. Yes, I know, for some historical reasons a positive step means "make it more blue", but this is counter-intuitive: add 5 steps of 20 to 3000 to get 2900!

Note that the camera does not actually take three pictures here; just one, and then it applies three color corrections to three saved files. Therefore this has no real practical advantage; it may be aimed at those reviewers who evaluate cameras by counting tick marks in a feature list.

In most cameras, the random noise (occurring at all exposure levels) is filtered out at the stage of converting the raw photosite readings into pixels. All cameras do it. This is always a tricky process, regardless of how smart the filtering algorithms are, and you can always find a subject on which it will result in loss of detail.

While Olympus claims to be using an "improved" dynamic denoising algorithm, I'm still skeptical. This is also why I'm not paying much attention to any camera noise measurements, as they can only show the noise after that filtering, without any regard to the lost image detail.

The E-1 has an additional random noise filtering feature, using another, more advanced, algorithm, turned on from the menu system and incurring a cost: extra processing time. In the E-300 this option is not available.

For a general introduction to noise issues, refer to my Noise in Digital Cameras article.

• sRGB
• Adobe RGB

According to what I've read, the Adobe RGB color space is capable of capturing a wider range of colors than sRGB. As the translation into a color space takes place when raw image is being converted into RGB (JPEG, TIFF), the choice is irrelevant when you save images in the raw mode (ORF files); you can postpone the decision until the conversion is done in postprocessing. Once that is done, however, any subsequent conversions will not bring any improvement: once information is lost, it is lost.

The only time when I experienced (with my previous cameras, that is) visible problems with sRGB color representation was when shooting some flowers, especially in shades of deep violet and purple.

• Saturation
• Sharpness
• Contrast
• Gradation

The E-300 is missing an option to boost an individual color component, neither has it the "skin tones" preset (both present on the E-1).

New in the E-300 (and E-1) is image tonal gradation adjustment: normal, high-key, or low-key. Olympus is not really specific about how this is really achieved, but I would imagine that this adjustment lifts or lowers the middle of the tonal response curve, therefore moving the midtones up or down without changing the limits of the tonal range.

This may be a nice addition to available image adjustments, but I haven't use it much, preferring to do it in postprocessing.

• Light fall-off
• Geometric distortion

The Four Thirds lenses, in addition to the usual, expected data, provide the camera with their characteristics regarding these two flaws, especially hard to avoid at short focal lengths (wide angles).

I can bet this information is passed either as an interpolation table (linear magnification or light loss as a function of radial distance), or as a set of polynomial coefficients; both ways would serve the purpose just fine (actually, for lens vignetting the data has to be aperture-specific).

This information can be then used to "reconstruct" the right image. For light fall-off it can be done simply by applying a multiplicative correction to pixel brightness; for geometric distortion — by moving the computed RGB values for a given photosite to a different pixel, along a radial line. Simple and ingenious.

The geometric distortion correction is not available in the camera, you have to use the optional Olympus Studio software ($150) to use this feature; the included Olympus Master does not include it. I have tried it and, indeed, it works well, adjusting the degree to the focal length as needed; it works both with raw ORF and JPEG image files.

Note of April 20, 2004: While the originally shipped firmware did not allow for light fall-off correction in-camera, Version 1.2 does (it can be activated from the menu system as "shading compensation").

• ORF (Olympus Raw Format)
• JPEG (three compressions)
• TIFF (uncompressed only)

This may be the safest way to store your pictures for critical applications.

In real life, however, I found that saving images as low-compression JPEGs is all I need most of the time; I have yet to grow as a photographer to really need the raw storage option.

I'm wondering when camera makers will start using the JPEG2000 (.jp2) format, which has better image quality at the same compression rate (or higher compression rate at the same quality). The format has been around for a while, and all better image processing or cataloging programs support it; I think it is time.

• 3264x2448 (8 MP, native)
• 3200x2400 (8 MP)
• 2560x1920 (5 MP)
• 1600x1200 (2 MP, UXGA)
• 1280x960 (1 MP)
• 1024x768 (XGA)
• 640x480 (VGA)

The 3200x2400 pixel size puzzles me a bit. Why not just the full size, just 2% larger in each dimension?

See also Storage and Data Transfer below.

Index

Storage

The camera will recognize the FAT32 file system, which makes it usable with cards of more than 2GB capacity. (As I don't have a card with more than 2 GB, I'm not sure if the camera also does FAT32 formatting.)

A native-size image (7.62 MP, see above) compressed to a JPEG will be written as a file of about 2.9, 5.7, or 8.5 MB (assuming the compression factors are as specified by the manufacturer). This varies from one image to another anyway, depending on the amount of detail. On average, I'm getting slightly higher compression ratios than these shown here.

It should be noted that Olympus traditionally uses lower JPEG compression ratios (higher quality) than most of the competition. For example, Canon 20D (7.81 MP) compresses JPEGs down to about 3.6 MB ("fine", 1:6.5) and 1.8 MB ("normal", 1:13). While this may speed up the camera operation, some people may find these values a bit too high.

The TIFF files remain uncompressed (although there is a lossless compressed TIFF format in existence), using about 23 MB per image. Actually, with full-information raw ORF on one side and low-compression JPEG on the other, I don't see a need for this format.

Olympus raw format is, contrary to what I thought before and to what most reviews say, uncompressed. As only one color component per pixel (really: photosite) needs to be saved, ORF files are still much smaller than TIFFs. The raw file size is 13.4 MB (14,127,616 bytes).

7.62 MP (7,998,272 pixels) with two bytes (one color only!) per pixel means 15.24 MB; if, however with bit packing (which is not compression!), there will be just 1.5 bytes/pixel, i.e. 11.43 MB. Add to this any overhead information, and 13 MB makes perfect sense. Besides, compressed files would vary in size, at least a little.

• SHQ (Super High Quality)
• HQ (High Quality)
• SQ (Standard Quality)

Switching between these three settings is fast; it requires the usual button-plus-wheel routine.

The compression/size combinations are assigned to these settings as pre-defined or user-defined:

• The SHQ combination is predefined: the native (3264×2448) image size and 1:2.7 compression, period.
• For HQ also the native size is always used, but you may pre-set the compression to either 1:4 or 1:8 in the Settings menu.
• The SQ setting may be pre-set, again, to any combination of pixel size and compression, except that the largest dimensions are 3200x2400 pixels. This is not really "Standard Quality" (lowest) but rather a user-defined one, another possible source of confusion.

Actually, the first two settings are redundant, as they may be reproduced in the SQ mode, described below. Olympus included them to "simplify" the choice, but I don't see much simplification here, unless you set HQ and SQ combinations once and forever to your liking, and then only choose between the three options.

Although this is better than in previous Olympus cameras (the system in the C-3030Z was really bad!), I hope one day Olympus will just introduce two separate, easily accessible, settings: image size and compression. Setting the format in one place and defining it in another is not really a good approach.

My suggestion: set the HQ to full-size 1:4, SQ to 3200x2400 1:8, and forget about the whole confusion. Just switch between these three settings. (As I already mentioned, why is the full size not available in the SQ mode, but just 2% less in each dimension?)

The use, if any, of this feature will depend on your workflow. For example, one may imagine a photographer recording raw ORF files for publication, but also writing high-compression VGA versions to be emailed from a PDA to the editor as soon as the card is removed from the camera. Or a scenario where a lazy photographer (like myself) saves ORF files plus HQ JPEGs, but then keeps ORFs only for frames which need more than just a minor touch-up.

The USB port is, like in all other Olympus cameras I've used, hidden under a rubber door at the left-hand side of the camera body. The door is made slightly differently than in other models, and attaching an USB cable to the camera becomes quite inconvenient. A minor, but quite irritating quirk.

Index

Viewing system

(Picture by Olympus)

This said, Olympus surprised us by using a unique, side-swinging mirror in the E-300. This unusual design (used, I think, only once before: in the Olympus Pen half-frame in the 1960's) allows the designers to move the finder away from above the lens, therefore reducing the overall height of the camera. The downside is one more mirror in the viewing optical path.

This also means that the eyepiece is not directly above the lens, but at the left end of the camera body. Right-eyed photographers (a majority) will find this useful: the camera is easier to hold and your nose does not press against the LCD monitor. Small pleasures of life.

As an economy measure (with an added benefit of weight reduction), the viewing system does not have a traditional pentaprism, using a system of mirrors instead (sometimes referred to as Porro prisms, which is not quite the same in this case).

One of the first things I did was to compare the viewfinder with that of my old E-20. Surprise: the finder in the older camera is larger and much brighter (the latter, no doubt, due to larger lens aperture, but remember that in the E-20 only a part of the incoming light is used for viewing).

Still, the finder in the E-300 is quite usable with the included zoom lens (better with brighter lenses), and as good as finders in Canon 300D and Nikon D70, or in the E-1 (which was a surprise).

There is no separate button dedicated to this function (the E-1 has one), but the OK button on the back can be customized to do that.

These data are as published by Olympus. My own averaged measurements, performed at F=32 mm, are very close to 93% in each dimension, with accuracy better than 0.5%, which gives 86% area coverage. Close enough.

To compare the viewfinder image size against a 35-mm film SLR it would be more honest to use the same EFL (equivalent focal length); in those terms the E-300 can claim 50% at EFL=50 mm. To quote the Ladies' Man of Saturday Night Live: this is really small.

The Canon 300D claims a magnification of 80%, and the Nikon D-70 — 75%, both at F=50 mm again. These numbers may look not as good, but divide them by proper focal length multipliers (1.6 and 1.5, respectively), and in both cases you arrive to 50% — your viewed image magnification, at the same lens angle, is exactly the same in all three cameras.

Using the same method (actual, not equivalent, focal length), the finder in my old E-20 (with image just slightly larger) would have to be described as having a magnification of about 220% at F=50 mm, and finders of video cameras with a 4-mm sensor could boast a magnification of 1000% or more, while still providing a tunnel vision. This clearly demonstrates how misleading specs can be if they are misused, and in this case they are.

• Shutter and aperture
• Autoexposure mode
• AE lock
• Exposure compensation
• Metering mode
• Autoexposure lock
• White balance override
• Flash needed and OK
• Autofocus confirm/fail
• AF area mark

I am in the second group: wearing glasses, I'm unable to see the information display without moving my eye to the left.

Update of September, 2005: After nine months of using the camera I can barely remember that the information display is there at all. Usually I don't see the display, and I do not bother to peek to the side to do it. I consider this to be a really bad design, perhaps my most severe misgiving about the E-300 ergonomics.

Eyeglass wearers will be able to see the whole viewed area without a problem, but not the information display next to it — see above.

With smaller size and weight of the E-300 mirror (area and weight four times less while camera's total weight remains the same) this feature is less important than in a 35-mm SLR; still, it can be useful in critical work, protecting you from camera shake caused by shutter release action and mirror movement.

For reasons unknown, Olympus refers to this feature as "anti-shake" instead of using the generally accepted term "mirror lockup". The term can be confused by some buyers with anti-shake mechanism of some lenses from other makers' systems.

• External eyepiece cover
• Interchangeable eyecups

The E-1 interchangeable eyecups do not fit the E-300. The accessory VA-1 angle finder can be used (with a special adapter), handy for tabletop photography.

The E-300 does not have an internal eyepiece shutter. Instead, you can use an external eyepiece cover (EP-4, supplied), mounted in place of the eyecup, which has to be removed (see the remark above). Worse, the cover cannot be attached to the camera strap (a solution used in many film SLRs); you have to carry it separately.

Providing a better (slide-on) eyecup would cost Olympus 15-20 cents; a strap-attachable cover, nothing (just a different plastic mold). Boo.

If you never shoot from a tripod, fine: you will not see a difference. I thought, however, that the E-300 is aimed at a somewhat more demanding user. For me (infrared photography) an eyepiece shutter, or at least a better cover, would be a lifesaver.

Index

Shutter

The shutter travels vertically (i.e. along the shorter dimension of the image).

At this frame size and at 1/4000 s, the shutter slit width is just 0.6 mm, the same as in a 35-mm camera with a vertical shutter at 1/8000 s. It would be quite challenging to narrow the slit even further, still providing the required accuracy. On the other hand, faster speeds could be implemented by increasing curtain travel speed, see below.

On the opposite end, the longest available shutter speed depends on the exposure automation mode, lens used, and the ISO setting.

• Shutter priority and manual modes: always 30 seconds;
• Aperture priority: 30s at ISO 100 with an F/3.5 lens; dropping down reciprocally with ISO setting, and also being reduced when a lens darker than F/3.5 is used;
• Program mode: 1s at ISO 100, with the same ISO dependence as above.

The bulb setting allows for exposures of up to 8 minutes.

Some dedicated Olympus flash units allow to use faster speeds with flash; see the Flash section below.

Actually, my tests show proper synchronization with a non-dedicated flash (or, using the wording from the Advanced Manual, a "non-specified" one) up to 1/300s or so: just a trace of image cut-off at 1/320, and none at all at 1/250 s).

Index

Drive modes

The four-frame maximum holds in ORF, TIFF, and full-size JPEG formats compressed at 1:2.7 and 1:4. At the compression of 1:8 the camera will keep up with your shooting until the card is full. Similarly, reducing resolution to 5 MP you can shoot 15 frames or so with the 1:4 compression.

Not too shabby. Most of the people complaining about the E-300 sequential mode, comparing it against Canon 350D and Nikon D50 forget that the 1:8 (worst-quality) compression ratio on this camera is less (better quality) than "Normal" settings on competing models. Besides, at this shooting speed I would rather worry about focusing than JPEG compression artifacts.

64 MB is no longer as impressive as it was just a few years ago: fast memory is getting cheaper every month. The E-1 has a 128 MB buffer.

• Infrared, optional
• Wired, optional, needs optional battery grip
• Via USB cable, with optional Olympus Studio

Addressing a common complaint about the C-series cameras, the IR remote release delay can be now set to 0 or 2 seconds. This it useful in close-up work, except that the remote sensor is still placed at the front of the camera (another one on the back would help a lot).

Disappointingly, the E-300 body does not have a wired remote socket, which is provided only on the HLD-3 battery grip, allowing you to use the RM-CB1 cable release. Having to buy an unrelated accessory just to use a wired remote is something I do not like.

The E-300 can be also triggered remotely from the $150 Olympus Studio application, which also offers image download and a wide range of remote camera adjustments.

Index

Exposure control

• ESP
• Center-weighted
• Spot

ESP stands for "electro-sensitive pattern" which means exactly nothing, but suggests some kind of matrix (pattern) metering. Olympus keeps quiet about details, especially on how many sensors does the system use (in the original press release they used a term "3-zone", but this can be interpreted in a number of ways).

In some literature Olympus provides a graph which does not answer this particular question at all, showing only how the central and peripheral readings are averaged.

With no information available from the maker, I have to make an educated guess, and the outcome may surprise you: I suspect that the whole method depends on just two light sensors. One sensor, I think, meters the light at a well-defined, tight spot in the center, and one provides a diffused area reading. The combination of these two measurements is then used to determine the overall exposure, with the weights assigned to both results depending on the ratio. (A similar system, also referred to as "ESP", worked very well in many Olympus film cameras.)

This could explain why Olympus never mentions the number of sensors: two don't look good on paper (regardless of how well they may perform), compared with seven, fourteen, or even forty used by some competitors on the specs-driven market. They also never use the term "matrix" or "multipattern", as this would be clearly a misrepresentation. If you come across any details on the subject, please let me know, and I will update this information.

One or two Web sources I've seen mention five light-metering sensors in the E-300, without providing any other explanation. I'm not sure how reliable this information is.

Neither does Olympus specify where the light sensors are located. Based on a number of hints I am almost sure they measure light scattered off the viewing screen, but this is just another guess.

Obviously the ESP system in the E-300 is something entirely different than ESP systems in their non-SLR digital cameras, where the imager itself is used for metering.

Is the Olympus system less accurate, or less "smart" than multi-pattern ones? Not necessarily so. So far my results with the E-300 don't give me much to complain about (which is something). Still, I think real matrix metering could, in principle at least, be more accurate, providing a better protection against washed-out highlights.

• ESP and CW: EV 2 to 20
• Spot: EV 3 to 17

Note: some Olympus sources quote EV 1 as the low-light limit.

Keep in mind that with the bundled F/3.5-5.6 lens metering range shifts upwards, i.e. to higher light levels. Near the wide end (with the maximum aperture close to F/4) this range will start at EV 4, and this will be 1 s at F/4.0.

All this can be stated much more simply: with any given lens the E-300 AE range starts at one second exposures at full aperture. Did you really need me to tell you that? (Well, maybe you did...)

I was able to check these numbers (both those by Olympus and my re-phrasing; ESP metering mode) with use of a dimmer in my living room; they are right on the nose, just a bit on the conservative side.

(Note: while the metering range depends on the maximum lens aperture, the quoted ISO value is just the reference point at which the EV range is defined; it defines what actual shutter/aperture combination will be used.)

• Program (shiftable)
• Specialized programs
• Aperture priority
• Shutter priority
• Metered manual
• Time exposure (bulb)

The shift feature allows you to adjust the shutter speed or aperture from those set by the program; the camera will then adjust the other variable in the opposite direction so that the overall exposure remains the same.

While this is a useful feature, I don't like the way in which it is implemented: by turning the control wheel without pressing any button. This makes it too easy to activate the shift by accident and without even noticing it. (The tiny 's' next to the program 'P' indicator can be easily missed, especially in the finder display.) More, there is no quick way to cancel the shift: you have to turn the wheel in the opposite direction until the tiny 's' disappears (trust me, you do not want to do that in a hurry), or power the camera off and on again.

Shooting a series of pictures with the program accidentally shifted by, say, 3 EV, can be a disaster: instead of using the 1/125 s shutter speed, common for overcast outdoors pictures, you will be unknowingly using 1/15 s.

This inconvenience is aggravated by the fact that you do not usually see the data display in the viewfinder.

(From a picture by Olympus)

The specialized programs, referred to by Olympus as "scene modes", provide beginners with convenient, pre-packaged sets of settings for some typical applications.

There are fourteen these programs provided: (1) Landscape, (2) Landscape+Portrait, (3) Night Scene, (4) Night Portrait, (5) Fireworks, (6) Sunset, (7) Portrait, (8) High Key, (9) Macro, (10) Documents, (11) Museum, (12) Sports, (13) Beach and Snow, and (14) Candle.

Five of these are directly accessible from the mode dial (1, 3, 7, 9, and 12); for all others you have to set the dial to "Scene" and use the menu system. The latter offers miniature thumbnails of example pictures and brief explanations, which are often, if not mostly, not so useful, sometimes even funny. (Example: "Fireworks: Suitable for shooting fireworks at night". Thank you, I would have never guessed.)

The "Macro" mode should be really referred to as a "close-up" one, and its capabilities depend on the lens used. For example, with the bundled 14-45 mm lens the closest focusing distance is 38 cm which provides, at best, a field width of about 15 cm; calling it "macro" is not right.

Some of the specialized programs may be marginally useful, modifying image parameters otherwise not accessible from the user interface (for example: enhancing greens and blues in the Landscape mode). To make them, however, of real use to a non-tyro, they would have to be well-documented: an ambitious user would like to know what the camera is going to set.

I believe that people who buy a digital SLR will not use the "scene modes"; the feature makes sense in entry-level cameras, but not here (the E-1 does not have this feature). Once again, it is the press and reviewers who may be to blame: "this camera has 26 scene modes, while that one just four". And the market follows.

While for most uses ±2EV is plenty, I'm using a +4 EV compensation in infrared applications with the E-300; the wider range came handy!

I'm a bit surprised not to see the five-frame option; not that I really need it, but because it is so cheap to add. A possible explanation is that five frames worth of data (five raw images) will not fit into a 64MB buffer, requiring just 3 or 4 MB more.

Still, all I've ever used with other cameras was three-frame bracketing, so no complaint here.

Index

Focusing

Phase-detection systems have one major advantage over the contrast-detection ones: they "know" not only how much is the image out of focus, but also in which direction; this results in some speed advantage.

Generally, AF systems on digital cameras (especially non-SLR ones) lag behind those on film SLRs. And here I was for a nice surprise: the autofocus in the E-300 seems fast, responsive and accurate.

Importantly, some cameras I've used before (including Olympus E-10 and E-20, which I both liked a lot) would sometimes confirm autofocus while not having focused at all. Not so with the E-300: if there is no detail to lock on in at least one of AF areas, the camera will not be fooled; see also AF Release Lock below.

Update of February, 2005: After two months with the camera I may say I'm very happy with its AF performance: it is fast (the total shutter release lag less than 0.5 s), positive, and accurate. At long last, I'm enjoying in a digital SLR an autofocus system comparable to a decent film camera.

You may allow the camera to choose which sensor to use, or make this decision yourself. In the latter case, the adjustment is done via a button sitting directly under your right thumb.

The central sensor is sensitive to both horizontal and vertical detail; the other ones — to horizontal only. This is a limitation.

The sensor chosen by the system is indicated with a flashing marker on the viewfinder screen, and an audible signal is also provided, which I found so useful on the E-10 and E-20, missing it a lot on the C-5050Z and C-5060WZ.

Olympus does not specify at what maximum lens aperture this range is defined; I'm assuming that this is F/2.0, the same as in the specification of autoexposure range. This would mean that with an F/4 lens mounted on the camera, AF will work from EV 2: 4 seconds at F/4.0, ISO 100 — quite impressive and two EV values below the minimum autoexposure level anyway.

I wasn't able to verify this claim to the full extent with the bundled lens, but for EV 4, which is the lowest AE light level, the AF performed just fine, with room to spare, so I have no reason to doubt the Olympus specification.

• Single AF
• Continuous AF
• AF with manual override
• Manual

Manual focusing is done with a ring around the lens, but this ring is not mechanically coupled with lens mechanism; it just activates the servomotor which actually does the job. From my experience with the E-10/E-20 I found this solution not as precise as a mechanical coupling, although better than the one using buttons or rockers.

The direction of focus response to the ring movement is programmable from camera options.

I wish Olympus decided to include a programmable setup of how the AF lock button and half-pressing of the release work together, like they did in the E-1, where each of these actions can lock one selected variable, or both. While that setup looks complicated and confusing in the E-1 manual, it is no rocket science, and you have to do it just once, setting the camera up to your preferences. OK, so the code takes some room in the firmware — so get rid of the moronic pictures illustrating the "scene modes".

Index

• Single AF: Focus priority
• Continuous AF: Release priority

(Both can be changed.)

Note of April 20, 2004: Originally these settings were hardwired; firmware version 1.2 allows them to be changed to your preferences. The defaults make good sense anyway.

This has been greeted as not-so-good news, as it leaves available-light shooters without a focus aid. In reality, however, the AF works OK down to light levels below the autoexposure range, so there is no real need for an available-light AF assist.

I took a series of six frames with the built-in flash in a dark room, using the flash AF assist. All pictures were focused just fine, although sometimes the camera decided to use the assist twice before deciding that the focus was OK.

The camera will also use the deep-red AF assist on Olympus dedicated flashes: FL-50 and FL-36. I've tried this feature with the latter, and my pictures in a pitch dark room were focused very well.

Flash

The guide number of 11 m means that at F/3.5 the range is about 3 meters, while at F/5.6 — two meters.

The built-in flash can also be used as a fill-in when an external one is being bounced off the ceiling. It does not, however, offer high-speed synchronization (you need an FL-36 or FL-50 for that, see below); enabling this flash in shutter priority mode resets the shutter to 1/160 s.

When the Manual Flash option is set, the intensity of the built-in flash can be adjusted manually between 1/64 and the full output. This may be useful in some types of shooting, and to trigger a slave unit.

Unfortunately, various flash settings are scattered among three different menus: flash mode in one, flash compensation in another, and manual flash activation in yet another one. I don't like this.

FL-36 (Picture by Olympus)

Olympus dedicated flashes co-operate seamlessly with the camera's exposure automation (in program, aperture- and shutter priority modes).

I discovered that my three-year old Olympus FL-40 unit does not work with the E-300 — or rather works as a non-dedicated unit only. A quite unpleasant (and expensive) surprise: it worked OK with all previous Olympus cameras up to the C-5060Z and E-1 (sic!).

Am I being paranoid, or is this a case of "obsolescence by design"? The tiny FL-20 works in TTL-auto with all older Olympus models I have, and the older FL-40 works fine with the E-1.

You have to remember, that when traveling with the E-300 and en external flash, you have to carry two battery chargers: one for the camera's BLM-1, and another for the NiMH AA's used by the flash unit.

Third-party flashes (including ones by Metz and Promaster) which worked with previous Olympus digital cameras may or may not work fully with the E-300. Don't ask me via e-mail — as soon as I know anything, I'll post it in the E-series section.

Index

• Auto (on demand)
• Auto + red-eye burst
• Slow + red-eye burst
• Slow (first curtain)
• Slow (second curtain)
• Fill (always on)
• Off

First- and second curtain synchronization determines whether the motion streaks (image created by the ambient light) precede or follow the frozen part of the image (created by flash burst).

The fill ratio is adjusted via the flash compensation setting, see below.

The flash compensation is applied in addition to (not instead of) the overall exposure compensation. It can therefore be used to adjust the ratio of flash to ambient light.

I resort to this adjustment quite often, therefore I was quite unhappy to discover that it can be done only by going into the menu system (remembering which of the menus it will be). In the C-5060WZ you could do that by pressing flash and compensation buttons together, and turning the control wheel, a much more agreeable solution. Too bad.

Some of the dedicated Olympus flash units (FL-50, the new FL-36) can provide synchronization at any shutter speed. They do it by emitting a fast series of short light bursts, which cover, for all practical purposes, uniformly the time interval during which the shutter slit is traveling across the film frame. A downside is that the effective guide number of your flash will be significantly lower than in the normal mode, when the whole frame is exposed to the flash at the same moment.

While for most uses it is not a critical feature, it may be useful for fill flash in bright daylight, for example, in sports photography.

Power

While a set of four NiMH AA's is cheaper and easier to replace in emergency, the BLM-1, like other, similar Li-Ion batteries, provides lots of juice and has a low self-discharge rate.

If you have a close look at similar batteries used by Canon and Nikon cameras, you may strongly suspect that this is basically the same battery, only with contact placement changed just enough to make all three types mutually incompatible. I suspect a case of greed prevailing over common sense.

The battery is inserted through a hinged door in the camera's bottom, accessible even when the camera is mounted on a tripod. The E-300 no longer has a battery-securing latch, present in all previous cameras using the BLM-1, a possible sign of cost-cutting, but I can live with that.

The manual says that BLM-1 will work at temperatures down to -10 degrees Celsius (it also says that the camera itself is limited to temperatures above freezing).

Notes on lithium-ion battery use:

Olympus states that the BLM-1 battery is good for about 500 full recharge cycles. Other sources are less specific, claiming generally 300..500 cycles for this battery type. Additionally, the batteries age regardless of whether they are used or not: a life span estimate is 2 to 3 years. (Do not stock up!) Optimum storage: in a cool place, about half-charged.

Li-Ion batteries have no memory effect, so they do not mind being only partially discharged before being recharged again. To the contrary: a series of such "shallow" cycles is better for battery longevity than one complete cycle. Some sources recommend one deep discharge every thirty partial cycles or so.

Olympus warns against using third-party batteries in their cameras. Obviously, one may suspect they are not impartial in this recommendation (last time I checked at B&H, the BLM-1 was $55, and third-party replacements $35), but Li-Ion batteries are quite difficult to make, and there may be a technically valid reason beyond it. So far I had only one Reader reporting a cheap (really cheap!) replacement fail after just seven months of use, with no harm to the camera (C-5060WZ).

For comparison, a a set of four 2500 mAh NiMH batteries has a total voltage of 4.8 V and therefore stores 12 Wh of energy, which is, nominally at least, slightly more. The usable capacity of such set may, however, be lower than that of the BLM-1, as the voltage delivered under load to battery terminals drops with discharge, and may reach a camera-defined usability threshold before the battery is really empty.

How does this translate into real-life use? After fully charging the new battery I started exploring the camera, learning how to use it, and taking pictures. This, obviously, involved a heavy use of the LCD monitor and computer connection, all without an external power source. Two days later, after 564 frames taken (true, only a few with the built-in flash), I've got a low battery warning. At another occasion, a single charge lasted through a three-week trip, with slightly more than 200 frames taken.

As unscientific as this conclusion may be, I consider this to be a solid performance.

A fast (one-hour) charger is being sold separately by Olympus, but I'm not that desperate.

• AC power adapter
• Battery grip/holder

The external power socket is quite unusual (proprietary?), so, for a while at least, the AC-1 adapter will be the only one working with the E-200.

The HLD-3 battery grip looks very similar to the B-HLD20 for the C-5060WZ. It accepts two BLM-1 batteries (but will also work OK with just one), and provides two extras: a vertical-position release button and a socket for the RM-CB1 cable release. (The socket still has that awful, easy-to-lose screw cap.) There is also a release lock, and a storage for camera's battery compartment cover.

September, 2005: The HLD-3 arrived to the U.S. a few months after the E-300 and is currently priced at $95 (B&H). I would recommend spending your money on something else, unless you have to use a wired remote.

Index

Image playback

• Liquid crystal, color TFT
• 1.8" (46 mm) diagonal
• 135,000 pixels

(From a photo by Olympus)

The monitor is referred to as "HyperCrystal LCD", which means nothing, and putting this as a blurb just below the monitor itself just offends my intelligence. Welcome to Kmart.

There are already larger (2" or more) monitors in some cameras on the market, but this takes the valuable real estate on the camera back and uses more power, so I'm willing to live with the 1.8" size. Oh, well, maybe 2" would be nice, like on the Pentax *ist DS.

The monitor pixel count (it could be translated in about 320x425 resolution, if not for the fact that these pixels are not really pixels, see below) is OK, similar to that of other digital SLRs, but nothing to write home about. Even with my myopic eyes I could easily use twice that number, i.e., 40% more of linear resolution. This is not so expensive in terms of power usage, as most of it goes for backlighting, therefore depending on physical size, not the resolution itself.

Note: The "pixels" in the specifications actually mean individual red-, green-, or blue-colored dots, therefore not being actual pixels. Olympus follows here the misleading convention used by all camera manufacturers. On your computer monitor (LCD or CRT) every pixel has three such dots.

I would risk an estimate that the "nominal" 135 thousand pixels of this display is really equivalent to that value divided by three: 45 thousand, and this translates into a 180x240 pixel equivalent. (This is not just Olympus; the same correction should be applied to monitor pixel counts claimed by all manufacturers.)

• File name and number
• Exposure and processing parameters
• Brightness histogram
• Overexposed areas

Of these I find the brightness histogram most useful in verifying the proper exposure, although highlighting the saturated highlights can be also useful: of all exposure pitfalls, the burned-out highlights bother me most.

You can scroll through the image while it is magnified, and pressing the [INFO] button shows the full view with the enlarged fragment marked, very much like in image samples I'm showing.

It may be a subjective issue, but I find the Control Panel on the E-300 less readable than that on the C-5050Z and C-5060WZ. The black and white on blue display is not as clear as the almost-gray one on those cameras, and the text also seems to be not as crisp.

Why mess with something which worked just fine?

Update of February, 2005: After two months with the camera, I've got used to this display, but I still don't like it, compared to the "old" one.

Index

Controls

Generally, the controls on the E-300 look like they evolved from those on the E-1 (and, indirectly, the E-10/E-20). This is a good thing, as the E-1 handles well. The bad news is that in a few cases the changes from the E-10/E-20 are not improvements.

The first impression in handling the camera is that the control layout, while quite good, but not as good as that in the E-10/E-20. I would say that even the less expensive, non-SLR models by Olympus, C-5050Z and C-5060WZ, have somewhat better ergonomics.

This does not mean that the E-300 is bad in this aspect; I like it more than the Canon or Nikon models in this price range (and perhaps beyond).

Another noticeable difference is the absence of the monochrome (low-power) LCD control panel on camera top (the one included on the E-1 was very nice). Its functionality has been taken over by the "Control Panel" display on the TFT monitor.

I thought I would be missing the top display, which used to provide a visual feedback to the basic (but not all) adjustments. Actually, this is not the case. Pressing the Info button brings up the Control Panel as described above, and all adjustments are reflected there while they take place.

As in most of the upper-tier Olympus cameras, the controls are adjusted using the press-and-turn approach: pressing a button selects the setting, and turning the control thumbwheel changes its value. Actually, this has been slightly modified in the E-300: you do not have to keep the button pressed, the camera offers you a brief "grace period" in which you have to start to turn the wheel if you let the button go. This is more important than it may seem, as in this model Olympus uses also the arrow buttons for the selection, and holding one down while turning the wheel would not be too convenient: both are located at the right of the camera's back. The grace period allows me to use my right thumb for both parts of the operation.

Note of September, 2005: Version 1.3 of the firmware (last July) allow you to set this "grace period" to 2 or 5 seconds, or to infinity.

While the wheel is being turned, the camera provides a visual feedback on the monitor screen: a small window with the value being adjusted. The monitor activates itself automatically if it was off before the operation, and turns off again after a second or so of wheel inactivity. This largely replaces the functionality of the top-deck LCD, but there is no indication in which direction to turn the wheel; for non-numeric entries I often find myself going through the full cycle, while a single step in the opposite direction would be enough.

This works differently if you do the turn-and-press operation while the Control Panel is used to show all camera settings. In this case, pressing the button highlights the setting, and turning the wheel changes the value without removing the whole Control Panel off the screen. I like this mode of operation, because I can have the whole camera status at a glance; this reduces the danger of using some non-standard setting, forgetting to reset it, and then applying it to other shots without being aware of it.

Note of April, 2005: After five months of using the camera I can say that the wheel itself is a little too easy to move accidentally (in the program mode this would lead to unintended program shift). Having it partially covered and/or moved to the left would be better.

All this said, I find the control buttons in the E-300 not as precise and positive as those in the E-10/E-20 and E-1. Nothing bad, bot not the same feel.

The menu system is called to the monitor screen by pressing of a dedicated button. The menu organization and readability are, I believe, a step back from the '5050/5060: the fonts are less readable, and the blue background does not help.

I also have some issues with the menu structure. There are five menus: two for picture-taking adjustments, one for display, and two for general setup. The assignment of particular functions to these five is not always obvious, and sometimes — counter-intuitive; too often I find myself scrolling through the settings of one shooting menu only to find that what I need is in the other one.

Additionally, for no apparent reason, there is no consistency in the interface: some menu entries, when activated, offer a choice from a submenu, while others call the same single-value window as the one used in thumbwheel operations, except that now you adjust the values using the up- and down-arrow keys. I was trying to find some regularity in using one or the other approach, but the designer's choice seems to be quite arbitrary here.

I find the menu system somewhat less polished than in the previous Olympus models I've used. This is possibly a sign that the camera was rushed to the market to meet the Christmas shopping deadline.

My biggest complaint is that flash exposure compensation is available only from the menu system (no dedicated button or other external control). This is one adjustment I use often when shooting outdoors with fill-in flash, and I would be happy to have it more easily accessible, even at the expense of some other adjustment. (No, the customized OK button cannot be assigned to this function, even if not needed for any other purpose.)

I'm also not quite happy with the location of the OK button. After using the cursor keys, pressing it usually requires a repositioning of the camera in your hands. This did not seem to be a problem when I was learning the E-300 in the first few days with it, but the inconvenience seems to grow as you become more used to the camera.

Of course, the controls are a highly subjective issue, so treat my opinions here as possibly tainted by personal likes and preferences. My general impression, however is that while the control system in the E-300 is quite good, it is not as good as I have expected, and not quite as good as in the other higher-end Olympus cameras I have experience with.

Index

Image quality

My E-300 image sample page contains some images which may give you an idea of the image quality from this camera.

Generally, I can say I'm mighty pleased with its performance, even with the bundled zoom lens.

To those who may object to such unscientific terms: for me photography is a way to get "mighty pleased" with the results. With a background in physics and some hands-on experience in instrument optics I can read MTF plots, and I know how to do spectral analysis of noise. Still, a partial or incompetent use of hard data is worse than no use at all. As the bottom line, either I like the final results or not, and my results are pictures, not Fourier frequencies. Whether you may agree or not, at least read this article by Ken Rockwell, both informational and entertaining.

First of all, the "economy" lens seems to be optically good enough in the whole focal length range. There is no objectionable softening of image in corners (just a tad at wide angle), no objectionable chromatic aberration. At the widest angle the zoom shows an average degree of spherical aberration (barrel), more at close distances; not impressive, but not bad. Wide open, it also has a slight vignetting at all focal lengths, which will not be objectionable in most situations, but in some it will.

By "good enough" I mean good for most of I'm doing, including prints up to 30x40 cm (12x16"), although at this size I would rather use the lens slightly stepped down, to improve resolution and remove vignetting. If you want a better lens, go for the 14-54, F/2.8-3.5 zoom, which sells for $450-500. Well, you get what you pay for, there is no way to outsmart the system!

The Kodak imager performs well: the dynamic range seems OK, and I had problems finding any traces of purple fringing around high-contrast transitions. (I have seen some chromatic aberration, but this is a different effect, due to the lens optics exclusively.)

The colors are usually as good as I have learned to expect from Olympus: eye candy (in the good sense of the word). Contrary to my expectations the E-300 provides a better and more consistent automatic white balance than E-1, verified by shooting with both side-by-side; this in spite of the E-1 having an additional, external WB sensor.

Now to the not-so-good news. Similar to some other recent Olympus (and not only) cameras, I believe the default in-camera sharpening is too strong: I can find white "bounces" at some high-contrast lines. This is a common problem with mass-market cameras, designed to impress a casual user performing a casual image inspection, and the E-300 is not as bad as many non-SLR cameras I've seen, but still. Setting the in-camera image sharpening to minimum (-2) brings this to an acceptable range. (I was quite surprised to see some reviews suggesting juts the opposite, but I stick to my opinion here.)

Another complaint: the dynamic denoising algorithms (not to be confused with the "noise filter on the E-1) are also set, I believe, too aggressively. This can be seen in some grass areas, where the process can remove image detail, leading to a fuzzy effect. Again, the market is to blame for this (along with the camera maker), more willing to accept these artifacts than any noise in large, smooth areas of the image. Unfortunately, the aggressiveness of the noise filter cannot be controlled in camera or in raw image processing in the Olympus Master. (It is, I believe, adjustable in the Olympus Studio, if you have a $150 to spare.) I think this problem can (and should be) addressed in a firmware update.

A hotly debated issue is that of the noise at ISO 800 and 1600. True: while ISO 800 is usable, if not spectacular, ISO 1600 — barely so, preferably with a noise-removal program. Still, we are talking ISO 1600 here, a really dim available-light shooting, not something I would need often.

Note of September, 2005: There is also the elusive problem with the Green/Magenta shift, showing rarely but unexpectedly, and difficult to reproduce, as already described in the Image Processing section. It happens rarely, but when it does, it is ugly. I don't see it lately: was it fixed in Version 1.3 of the firmware?

The bottom line is that, complaints aside, the overall image quality is very good, enough to satisfy a semi-serious user like myself, and even a serious one.

Index

Documentation

In addition to the Quick Start poster, the box contains a printed copy of the Advanced Manual (in the U.S. three copies: English, French, and Spanish, but none in the U.K.; I'm not sure about other countries).

This is (in the States, at least) a welcome departure from the Olympus' recent policy of providing the full manual only on a CD, but it doesn't take an Einstein to figure out that this is what the buyers want.

Advanced Manual is really a misnomer: Basic Manual would be a better name for this well laid-out and nicely printed booklet of about 200 pages. The booklet is slightly larger than the previous ones; a good balance between portability and readability.

Regarding the contents, I never had a high opinion of Olympus documentation, and this is no exception. I have to admit that the quality of the writing and coverage is improving from year to year, but still not enough. The language of the English version is better than before (with some room to improve, and remember that English is only my second language!), but the coverage still remains shallow and sometimes misleading.

The major problem is that Olympus does not seem to believe that anyone with IQ above 80 and any experience in photography will ever buy the E-300.

The manual contains very few things you wouldn't be able to figure out by just playing with the camera, while it lacks answers to many questions a photographer may ask about it. More, because the meaningful pieces of information are few and far between, they are easy to miss when our reading of the manual becomes just a casual glancing through the text which brings nothing of interest.

On the brighter side: some of the passages are quite entertaining. Let me just quote one example:

"The advanced shooting techniques used by professional photographers are drawn from years of experience. Now [...] you'll be able to take advantage of those same [sic!] advanced techniques simply by pressing a few buttons."

Poor professional photographers, having wasted all those years instead of just spending $1000 to buy this camera. This will never happen again.

Index

Software: Olympus Master

Staying in line with the long tradition of the Camedia Master (of which the Olympus Master is a slightly modified version, the program is still disappointing. Still, you may need to install it on your computer for any of the following reasons:

• Raw (ORF) image conversion and initial adjustment;
• Access to complete EXIF data;
• Upgrading of the camera/lens firmware.

In EXIF data readout, the Olympus Master does a better job than any general-use programs, I've tried, like ACDSee or Photo-Paint, which leave out or misinterpret some of the information.

As a general image management or editing application, however, Olympus Master does not stand up to many entry-level applications. For more, see a separate, brief review.

Index

The competition

Every major player in the digital camera field (except for Sony, who is not really a traditional camera maker) has at least one model competing directly (or almost so) against the E-300. Well, with the recent price drop, even the Olympus' own E-1 competes against its younger sibling.

The cameras Olympus is competing against feature- and price-wise are Canon 300D (a.k.a. Digital Rebel), Nikon D70, and Pentax *ist DS. From the samples I've seen, all three are capable of good results. All have bodies priced below $1000, with the Rebel being least expensive of the bunch (unfortunately, it shows).

(Yes, I checked out the Canon 350D a.k.a. Digital Rebel XT, improved upon the original, but I'm still not impressed.)

Two best cameras in the $1500 price range seem to be, at least in my subjective view, the Canon EOS 20D (finally Canon got almost everything right!), and Olympus' own E-1, not necessarily in this order. They are, however, more expensive than E-300.

The Maxxum 7D by Konica Minolta is too new to tell if Minolta broke its stretch of disappointing cameras with attractive specs (for the record: I'm a long-time enthusiast of their film SLRs), it is also selling at $1500 for the body alone. Still, I've tried the controls, and I like them a lot, more than on any other camera from any maker, and certainly more than in the 20D.

If you are reading this section in search of an advice which of the inexpensive digital SLRs I would recommend, you'll be disappointed. Most probably, every one of the cameras I've mentioned will deliver very good results, and your tastes and preferences may play a bigger role than any "objective" differences in cameras' specs and performance.

Index

Conclusion

This is a good camera, even better considering the price (April 2005: you can get the E-300 with the 14-45 mm lens at less than $750 from a respectable seller). After trying it out, I decided to keep it (and I have to get another one for my friend; he still wants one).

What I like about it is the general image quality, the solidness of construction, and wide range of adjustments available.

The controls are OK — not as good as in the E-10/E-20, but still. My biggest problem is with the placement of the viewfinder data display. I also wish the camera's width were 15 mm or so less.

On the image quality front, my complaints about excessive dynamic noise filtering and the occasional Green/Magenta shift should not obscure the fact that images from the E-30 are usually very, very good.

Most importantly, we have to see the camera compared to others in the same price range. I believe this one offers more than the competing models mentioned above, although my direct field experience with those is limited only to the Canon 300D/350D.

If you are a happy Canon, Nikon, or Pentax user, especially with an investment in lenses, I will not try to persuade you away from sticking to that brand. If, however, you are an aspiring amateur without such a commitment, then I can easily recommend the E-300 as an excellent choice you will not regret.

Index

Web References

As always, the popular digital photography news and reviews sites offer detailed looks at the E-300:

• Review by the Megapixel Magazine, detailed and informative.
• Review by Steve Jenkins at Steve's Digicams
• Review by Phil Askey at Digital Photography Review
• First Look by Jeff Keller at Digital Camera Resource
• First Look by Shawn Barnett and Dave Etchells at the Imaging Resource.

Other reviews worth your attention are listed below:

• Review at PhotographyBLOG, a right mix of detail and readability.
• Review at DCViews, relatively brief, but focused on results.

Most importantly, there are Web sites run by dedicated amateurs (like myself), with a wealth of information. At this moment I can point you to just one:

• Alfred Molon's newly re-organized myolympus.org (formerly: molon.de) has a fast-growing E-300 section. Alfred runs an operation in the best peer-to-peer tradition, and you will find there lots of information and links to other resources.

Index