Here is a small set of image samples from the Olympus E-1 digital camera equipped with the bundled 14-54/2.8-3.5 zoom lens. The set is not comprehensive — I've used the camera for just a few days, and the weather was quite cloudy; this is also why all pictures were taken at ISO 200.

As in my all sample pages, on the left are full frames, reduced and resharpened, then saved as "economy" JPEGs, with red rectangles showing locations of 1:1 (full pixel size) samples. These are shown on the right, without any postprocessing except for cropping and re-saving at low compression, that is. All images were recorded in the native, 5 MP size.

The focal lengths shown are actual (not 35-mm equivalent) values. To get the EFL, multiply the quoted values by 2.

Remember that all image flaws are exaggerated in these samples as viewed on your monitor. For 8x10", frame-height prints, the sample height will correspond to 1.14" (29 mm) on paper, while for 11x14" ones — to 1.57" (40 mm).

April, late afternoon

I shot these samples during a trip to Poland. All were taken within a few minutes at most from each other, and from the same vantage point. The firmware version used was 1.4, the latest one as of this writing. Auto WB was used in all these shots; the saturation was at the CM3 setting (blue emphasis), preferred by some users. A tripod was used in all cases.

Interesting: zooming in by just a bit shifted the white balance considerably towards neutral. One more argument in favor of using manual WB whenever you can. The lens still sharp, no vignetting.

Ouch! What is this? Seems like the WB system went bananas! It seems to be overreacting to minor changes in the scene contents. Remember that this is firmware Version 1.4, where the auto-WB is supposed to be already tweaked and improved! The lens, however, still seems to deliver, important for the "standard" 50 mm EFL.

This series of samples shows clearly two features of the E-1. Firstly, the standard 14-54 zoom is an excellent piece of optics. It clearly provides enough detail for sensors with larger pixel count (even if you will not see a difference in printed pictures).

Second, the auto white balance circuitry is not quite consistent; it can be easily fooled into overcompensating. Some people may find it strange, as Olympus cameras were known for outstanding color rendition, including auto-WB.

There is one difference, though: the E-1 is the first model by Olympus without access to pre-exposure information from the CCD (the E-1 and E-20, while being SLR, had this access because of the semi-transparent mirror prism design).

One may speculate whether the WB adjustment is based on some dedicated, behind-the-lens sensors (in addition to the external one), or on the analysis of the recorded information after it has been collected from the CCD, but before JPEG conversion. The former would explain the problem (being also consistent with strange WB behavior with film-SLR lenses wide open). We can only keep guessing.

Most of the serious users will not be affected by this. They either use manual WB setting (I've switched to this approach), or save raw images, in which case WB setting is postponed to the postprocessing stage.

Samples with the ZD 40-150 mm F/3.5-5.6

My page with samples from that lens also contains a set of images shot with the E-1 at the same session as those shown above.

Higher ISO settings

My friend and a fellow E-System user from Poland, Jerzy Wojewoda, kindly provided a number of samples showing the E-1's behavior at higher ISO settings. Here are some.

Sharpness, contrast, and saturation were at 0 in all cases. All images in this section were produced under firmware Version 1.3, except for the submarine (1.4).

What can I say? The autoexposure works just fine at these light levels, providing consistent results. (The -0.3 EV compensation was a precaution: because of reduced dynamic range at higher ISO, it may be safer to protect the highlights.) The image is smooth at ISO 200, and quite good at ISO 400; let us remember that any noise is much more visible on your screen than in print, especially magnified to full pixel scale.

At ISO 800 the noise becomes visible, but it is not really bad. I wouldn't resort to a noise reduction program, like Neat Image, not yet. The image at ISO 1600, however, although still pleasing in tonality, becomes very noisy. It will be usable for some purposes, but applying the Neat Image (or a similar program) in postprocessing becomes necessary for print sizes above 6×8 inches (15×20 cm).

Unfortunately, Jerzy did not come up with an ISO 3200 version of this image, so here is another one.

It is difficult to compare this sample against the previous one, as it is missing the mid-grays, where the noise is most visible (see the column left of center). A scrutiny of almost-black areas in both samples, however, shows that the noise is much higher at ISO 3200.

If you are desperate, however, the highest setting may save your day, noise or not. Here is a submarine Jerzy spotted during a sailing trip on the Baltic Sea near midnight (still some light in the sky at those latitudes in summer). He had to use a 150 mm focal length below any safe handholdable shutter speed, so he was, indeed, desperate that night.

Putting the noise issues aside, notice that none of the night pictures exhibit any chromatic aberration or color fringing around the brightest areas, which often is the case. The flare is also well under control.

Web resources

There are quite a few places on the Web with selections of E-1 image samples. Some of those may be much more representative than mine: shot at more standard, or even controlled, conditions. Some may be worse (shot off large-sized prints of photographs taken with other cameras, close to useless). Here in one example:

• Steve's Digicams. I may be repeating myself, but again: I consider this, on average at least, the best source, especially for sample comparisons between various cameras.