Olympus e-20 review

File Flush / Display Timing

Timings shown below are the time taken for the camera
to process and «flush» the image out to the media card, the
timer was started as soon as the shutter release was pressed (shutter
click was heard) and stopped when activity indicator LED beside the storage
compartment went out. This means the timings also include the cameras
processing time and as such are more representative of the actual time
to «complete the task».

The media used for these tests were:

• 128 MB SmartMedia (TwinMOS)
• 512 MB SanDisk Ultra Compact Flash Type I card
• 1 GB IBM Microdrive Compact Flash Type II card

Image Type	Time to store (seconds)		Time to display (seconds)
SM	SanDisk	Microdrive		SM	SanDisk	Microdrive
2560 x 1920	RAW	17.2	17.1	16.8		1.7	1.7	2.5
2560 x 1920	TIFF	38.0	34.2	34.9		1.3	1.3	2.2
2560 x 1920	SHQ 1/2.7	19.2	15.5	17.1		1.7	1.7	2.2
2560 x 1920	SHQ 1/4	11.2	10.9	11.6		1.6	1.6	2.4
2560 x 1920	HQ 1/8	9.0	9.0	9.4		1.4	1.4	2.4
1792 x 1344	HQ 1/8	13.3	13.9	14.4		2.2	2.2	2.4
1024 x 768	HQ 1/8	5.9	6.4	7.1		1.6	1.8	2.4

As I expressed above these write times are really too slow for a ‘professional’
digital camera.

Compared to…

Here we can compare, using results from our review archives,
the colour reproduction and resolution of the E-20 to three other five
megapixel digital cameras: Sony DSC-F707, Minolta DiMAGE 7 and Nikon Coolpix
5000 (results coming soon).

Colour Chart Comparison

Colour charts are shot in daylight, Auto White Balance,
EV compensation +0.3 (all cameras), measured light ~10 EV. This test is
also dependent on the accuracy of the camera’s auto white balance, the
colour charts are shot in daylight but some camera’s white balance is
better than others. All cameras are given 20 seconds to «settle»
before the shot is taken.

Olympus E-20	Sony DSC-F707
Minolta DiMAGE 7 (converted to sRGB)	Nikon Coolpix 5000

In the table below we’re only measuring colour. The RGB
values were measured from a VGA reduced image (to average colours, remove
noise and eliminate JPEG artifacts) using the Eyedropper tool in Photoshop
with a 5 by 5 Average Sample Size.

Patch	Olympus E-20	Sony DSC-F707	Minolta DiMAGE 7	Nikon Coolpix 5000
Black	31,29,32	18,18,20	29,30,32	19,20,19
Middle Gray	92,92,96	80,89,91	85,89,97	91,91,91
White	201,203,205	176,189,190	220,224,226	209,209,205
Magenta	184,31,102	182,16,82	212,40,116	219,37,93
Red	182,30,43	178,9,4	208,44,56	207,26,23
Yellow	201,193,55	175,180,37	215,221,45	212,189,47
Green	25,125,76	8,135,66	26,136,58	0,130,68
Cyan	26,137,193	12,148,198	22,144,215	1,153,200
Blue	50,34,103	36,21,98	45,31,104	33,44,98

As you can see the E-20’s colour reproduction (on our simple patches)
is remarkably similar to the Sony DSC-F707. There are two major differences
(which aren’t clear to see on this test):

• The fact that the E-20 images are far less ‘contrasty’ than the F707,
  that is the overall tonal balance is ‘flatter’ with a better preservation
  of shadow detail by deliberately maintaining most detail above the black
  point (this can also make colours appear less saturated).
• How the E-20 handles ‘brighter’ colours (more saturated), unlike the
  F707 the E-20 proved that it didn’t «blow out» reds or any
  other colour which may otherwise affect the F707.

Battery life

We ran the camera through our new battery life test.
This test is designed to be fair and comparative to each camera and battery
type:

• Take 4 shots without flash
• Wait 2 minutes (50% of the time powering the camera off)
• Take 1 shot with flash
• Wait 1 minute
• Repeat

All batteries had been fully discharged and recharged before the test
and all cameras were reset to their factory default settings. Here are
the results:

CameraProsumer / SLR-Like	Battery	Power	Battery life	Number of shots
Minolta DiMAGE 7	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	1 hr 14 mins	125
Minolta DiMAGE 5	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	1 hr 56 mins	195
Nikon Coolpix 5700	EN-EL1	4.8 Wh	2 hr 08 mins	215
Canon PowerShot S40	NB-2L	3.9 Wh	2 hr 13 mins	225
Minolta DiMAGE S304	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 18 mins	235
Canon PowerShot S45	NB-2L	3.9 Wh	2 hr 28 mins	240
Nikon Coolpix 4500	EN-EL1	4.8 Wh	2 hr 27 mins	250
Fujifilm FinePix 6900Z	NP-80	4.1 Wh	2 hr 29 mins	250
Nikon Coolpix 990	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 30 mins	255
Nikon Coolpix 995	EN-EL1	4.8 Wh	2 hr 30 mins	255
Nikon Coolpix 5000	EN-EL1	4.8 Wh	2 hr 32 mins	260
Minolta DiMAGE 7Hi	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 33 mins	260
Canon PowerShot Pro1	BP-511A	10.2 Wh	2 hr 36 mins	265
Canon PowerShot S50	NB-2L	3.9 Wh	2 hr 39 mins	270
Minolta DiMAGE S404	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 39 mins	270
Sony DSC-V1	NP-FC11	2.8 Wh	2 hr 39 mins	270
Olympus C-7000 Zoom	LI-12B	4.5 Wh	2 hr 44 mins	280
Nikon Coolpix 8700	EN-EL1	5.0 Wh	2 hr 45 mins	280
Minolta DiMAGE 7i	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 46 mins	270
Olympus E-20	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	2 hr 48 mins	285
Canon PowerShot G1	BP-511	8.1 Wh	3 hr 00 mins	300
Nikon Coolpix 5400	EN-EL1	4.8 Wh	3 hr 06 mins	315
HP Photosmart 850	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	3 hr 12 mins	325
Sony DSC-S75	NP-FM50	8.6 Wh	3 hr 15 mins	330
Nikon Coolpix 8800	EN-EL7	8.1 Wh	3 hr 27 mins	350
Fujifilm FinePix S602Z	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	3 hr 29 mins	350
Canon PowerShot G2	BP-511	8.1 Wh	3 hr 32 mins	355
Casio QV-4000	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	3 hr 38 mins	365
Olympus C-5050 Zoom	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	3 hr 48 mins	380
Casio EXILIM EX-P700	NP-40	4.5 Wh	3 hr 48 mins	385
Sony DSC-S85	NP-FM50	8.6 Wh	3 hr 50 mins	400
Sony DSC-F717	NP-FM50	8.6 Wh	4 hr 02 mins	405
Pentax Optio 750Z	D-LI7	6.6 Wh	4 hr 03 mins	415
Sony DSC-V3	NP-FR1	4.4 Wh	4 hr 05 mins	415
Nikon Coolpix 8400	EN-EL7	8.1 Wh	4 hr 09 mins	420
Sony DSC-F707	NP-FM50	8.6 Wh	4 hr 20 mins	440
Fujifilm S7000 Zoom	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	4 hr 23 mins	445
Konica Minolta DiMAGE A2	NP-400	11.1 Wh	4 hr 26 mins	450
Canon PowerShot G3	BP-511	8.1 Wh	4 hr 32 mins	455
Canon PowerShot G5	BP-511	8.1 Wh	4 hr 33 mins	460
Pentax Optio 550	D-LI7	6.6 Wh	4 hr 36 mins	465
Minolta DiMAGE A1	NP-400	11.1 Wh	5 hr 02 mins	515
Leica Digilux 2	BP-DC1-E	10.1 Wh	5 hr 05 mins	515
Sony DSC-F828	NP-FM50	8.6 Wh	5 hr 28 mins	540
Fujifilm FinePix S5000	4 x AA NiMH 1600 mAh (GP)	7.7 Wh	5 hr 31 mins	555
Canon PowerShot G6	BP-511A	10.3 Wh	5 hr 38 mins	570
Olympus C-8080 WZ	BLM-1	10.8 Wh	5 hr 52 mins	595
Other test notes:
LCD monitor is left on during the battery test   The camera is powered off half of the time and left on half of the time (switch every 10 iterations)   There are now higher capacity AA NiMH batteries available, however in the interests of fair     comparison we will continue to use our 1600 mAh test set

Flash Performance

According to the specifications the E-20 has a very powerful onboard
flash, with a rated range of 0.6 — 8.9 m (2.0 — 29.2 ft) at wide angle
and 0.5 — 7.4 m (1.6 — 24.3 ft) at telephoto it’s as powerful as many
off-camera flash units. Our tests seemed to back up some of this, although
I feel the figures quoted by Olympus are a little hopeful. I also found
that at distance or with the lens set to wide angle the flash produced
a slightly uneven coverage (as you can see below).

Skin tone test: Good exposure, a little ‘cool’ but I wouldn’t call it colour cast.	Surprisingly the E-20’s powerful flash didn’t cover our white wall test, there’s a noticeable ‘spike’ of brightness near the center of the frame.	Color patch test: Good exposure, a little contrasty but this could be adjusted with some careful settings adjustment. Good colour, no cast.

Studio Flash

Realizing that quite a few E-10 users have hooked up their camera to
studio flash systems I decided to fire off a few shots with the E-20.
Setup was extremely ordinary, an Elinchrom
Style 300S with a 70cm softbox suspended over a ‘curved sheet’ of
card. Sync cable connected to the E-20, manual exposure of 1/100 sec,
F8.0. A little fine tuning of the head power output and the results are
very pleasing (clearly better still with a more comprehensive setup, this
was just a quick test). White balance was almost spot on, a little warm
but nothing to worry about.

Note: I would normal use F11 or greater but at F11 resolution drops due
to diffraction.

Continuous shooting

The E-20’s image buffer sits just behind the CCD in the cameras ‘imaging
chain’. This means that no matter what image size you select the maximum
number of frames remains the same (see table below).

Image format	IS/PS	Frame rate	Maximum frames
JPEG, TIFF	IS	3.6 fps	4
RAW	IS	3.6 fps	3
JPEG *	PS	5.4 fps	7

* In PS mode you can only choose
JPEG image format

Continuous mode Flush Timing

The next thing to test is the E-20’s write speed. We do this by shooting
a burst of four images and timing the red LED next to the storage compartment
which is an indicator of card access. The total size of the files recorded
is then divided by the amount of time it took to write. This is repeated
three times for each image type to produce an average throughput.

The media used for these tests were:

• 128 MB SmartMedia (TwinMOS)
• 512 MB SanDisk Ultra Compact Flash Type I card
• 1 GB IBM Microdrive Compact Flash Type II card

Image Type	128 MB SM	SanDisk CF	Microdrive
Time (s)	Throughput	Time (s)	Throughput	Time (s)	Throughput
RAW (3)	45.5	640 KB/s	47.0	619 KB/s	45.4	641 KB/s
TIFF (4)	145.7	428 KB/s	132.5	471 KB/s	134.2	465 KB/s
SHQ JPEG (4) *	42.4	239 KB/s	41.1	247 KB/s	43.0	238 KB/s

* SHQ JPEG at 1/4 setting (2.5 MB each @ ISO 160)

The E-20’s write performance is squarely down at the consumer end, with
a JPEG write speed around 240 KB/sec it’s almost identical to Nikon’s
$800 Coolpix 995. Compare this to Canon’s EOS-D30 which managed to write
JPEG’s to the SanDisk card at 665 KB/sec and to the Microdrive at 800
KB/sec. Aimed at professional and semiprofessional photographers, with
even larger file sizes and priced above $1500 we would expect the E-20
to offer faster write times than this.

For comparison:

Camera	Best JPEG write speed
Olympus E-20	240 KB/s
Sony DSC-F707	530 KB/s
Nikon Coolpix 995	370 KB/s
Canon EOS-D30	1,100 KB/s

Resolution Comparison

Shots here are of the PIMA/ISO 12233 standard resolution
test chart (more of which are available in our comparison
database). This resolution chart allows us to measure the actual performance
of the lens and sensor system. It measures the ability of the camera to
resolve lines at gradually higher resolutions and enables us to provide
a definitive value for comparison purposes. Values on the chart are 1/100th
lines per picture height. So a value of 8 equates to 800 lines per picture
height.

Studio light, cameras set to auto, all settings factory
default. Exposure compensation +0.7 EV for all cameras.

Horizontal resolution	Vertical resolution	5 degree diagonal res.
Olympus E-20
Sony DSC-F707
Minolta DiMAGE 7
Nikon Coolpix 5000

Measurable findings (three measurements taken for each camera):

Camera	Measurement	Absolute Res.	Extinction Res.
Olympus E-20	Horiz LPH	* 1350	1650
Vert LPH	* 1300	1650
5° Diagonal LPH	+ 1000	—
Sony DSC-F707	Horiz LPH	1450	1800
Vert LPH	1300	1800
5° Diagonal LPH	+ 1000	—
Minolta DiMAGE 7	Horiz LPH	* 1300	1450
Vert LPH	* 1150	1450
5° Diagonal LPH	900	—
Nikon Coolpix 5000	Horiz LPH	* 1350	1500
Vert LPH	* 1200	1500
5° Diagonal LPH	+ 1000	—

* Moiré becomes visible, + Chart maximum

Definition of terms:

LPH	Lines per Picture Height (to allow for different aspect ratios the measurement is the same for horizontal and vertical)
5° Diagonal	Lines set at 5° diagonal
Absolute Resolution	Still defined detail (below Nyquist frequency*)
Extinction Resolution	Detail beyond camera’s definition (becomes a solid gray alias)
n/a	Not Available (above the capability of the test chart)
n/v	Not Visible (not visible on test results)

* Nyquist frequency defined as the highest spatial frequency
where the CCD can still faithfully record
image detail. Beyond the Nyquist frequency aliasing occurs.

Overall the E-20 does well for resolution, a match for
the F707 for vertical resolution and very close for horizontal resolution.
This goes to show that it is possible to get good resolution from the
commonly used five megapixel sensor matched with a good high quality lens
and sophisticated image processing algorithms. Compared to the Nikon Coolpix
5000 things are a little closer, it looks as though the E-20’s larger
slightly higher quality lens manages to squeeze a little more resolution
than the 5000.

Olympus E-20 Specifications

Street Price (US)		$ 1900
Street Price (EU)		??
Body Material		Metal
CCD effective pixels		4.92 megapixels
CCD size**		2/3″
CCD Colour Filter Array		G — R — G — B
Max resolution		2560 x 1920
Lower resolutions		1792 x 1344 1280 x 960 1024 x 768 640 x 480
Image ratio w:h		4:3
Image formats		JPEG (EXIF) RAW (2576 x 1924) — final output is 2560 x 1920 TIFF (8-bit)
Quality Levels		1/2.7 1/4 1/8
Sensitivity equiv.		Auto ISO 80 ISO 160 ISO 320
Lens Thread		62 mm
Zoom wide (W)		35 mm
Zoom tele (T)		140 mm (4 x)
Zoom type		Mechanically linked
Lens Aperture		F2.0 — F2.4
Lens Construction		14 elements in 11 groups
Digital zoom		None
Auto Focus		Contrast detection
AF Illumination lamp		Yes, Infrared
Manual Focus		20 cm — Infininty (focus-by-wire ring on front of lens barrel)
Normal focus range		60 cm — Infinity
Macro focus range		20 cm — 60 cm
Min shutter		Program / Aperture Priority / Shutter Priority: 2 sec Manual: Bulb, 60 sec
Max shutter		IS (2560 x 1920) 1/640 sec PS (1792 x 1344) 1/4000, 1/18000 sec
Noise reduction		Yes, can be manually enabled
Pixel mapping		Yes, menu option
Metering		Digital ESP Center-Weighted Average Spot
Exposure adjustment		-3EV to +3EV in 1/3EV steps
Auto bracketing		3 images 0.3 — 1.0 EV in 0.3 EV steps
Aperture priority		Wide: F2.0, F2.2, F2.4, F2.8, F3.2, F3.6, F4.0, F4.5, F5.0, F5.6, F6.3, F7.1, F8.0, F9.0, F10.0, F11.0 Tele: F2.4, F2.8, F3.2, F3.6, F4.0, F4.5, F5.0, F5.6, F6.3, F7.1, F8.0, F9.0, F10.0, F11.0
Shutter priority		IS mode: 2, 1.6, 1.3, 1, 1/1.3, 1/1.6, 1/2, 1/2.5, 1/3, 1/4, 1/5, 1/6, 1/8, 1/10, 1/13, 1/15, 1/20, 1/25, 1/30, 1/40, 1/50, 1/60, 1/80, 1/100, 1/125, 1/160, 1/200, 1/250, 1/320, 1/400, 1/500, 1/640 sec PS mode additionally: 1/800, 1/1000, 1/1250, 1/1600, 1/2000, 1/2800, 1/4000, 1/18000 sec
Full manual		As above plus these shutter speeds: Bulb, 60, 47, 38, 30, 25, 20, 15, 13, 10, 8, 6, 5, 4, 3, 2.5 sec
Exposure Modes		Program AE Aperture Priority Shutter Priority Manual
White Balance		Auto WB 3000 K (Tungsten) 3700 K (Tungsten — Warmer) 4000 K (Fluorescent) 4500 K (Fluorescent — Warmer) 5500 K (Daylight) 6500 K (Cloudy) 7500 K (Shade) Manual Preset
Continuous		IS: 2560 x 1920 — 2.5 fps max 4 images PS: 1792 x 1344 — 4.5 fps max 7 images
Built-in Flash		Yes, pop-up (manually released)
Flash Range		Wide: 0.6 — 8.9 m (2.0 — 29.2 ft) Tele: 0.5 — 7.4 m (1.6 — 24.3 ft)
Flash modes		Auto Red-Eye Reduction Flash On Flash Off
Flash compensation		+/-2 EV in 0.3 EV steps
External flash		Hot-shoe (Olympus FL-40 flash) PC Sync terminal
Tripod mount		Yes, metal
Self-timer		Yes, 2 or 12 sec delay
Remote control		Yes, supplied Infrared
Video out		Yes, E-20P — PAL, E-20N — NTSC
Storage media		SmartMedia (up to 128 MB) Compact Flash (Type I / II — Microdrive 1 GB supported)
Storage included		16 MB SmartMedia card
Viewfinder		Optical, TTL (prism), 95% view
LCD		1.8″ TFT, can be tilted by 20 degrees down and 90 degrees up
Playback zoom		Yes, 2x, 3x, 4x
Operating system		Proprietary
Connectivity		DC-In, USB, Video Out, PC Sync Terminal, Remote control jack
Timelapse		Yes
Battery		1 x CR-V3 Lithium (non-rechargeable) supplied
Battery charger		2 x CR-V3 Lithium, or 4 x AA batteries (NiMH recommended)
Weight (inc. battery)		1170 g (2.6 lb)
Dimensions (inc. grip)		128 x 103 x 161 mm (5.1 x 4.1 x 6.3 in)

Timings & File Sizes

Performance wise I found the E-20 to be a fairly frustrating
camera. Four things limit the camera’s performance:

1. Storage media write speed. In our tests (below) the E-20 had CF write
   speeds which were approximately the same as a 3 megapixel Nikon Coolpix
   995. This is not ‘professional digital camera’ performance, and nor
   is it helpful on a 5 megapixel camera where each JPEG can weigh in at
   2.4 MB.
2. If you shoot with the ‘live view’ LCD you will have to wait at least
   10 seconds between each shot, this is because the live view does not
   return to the LCD until the camera has finished writing the contents
   of its image buffer. It doesn’t stop you from shooting with the viewfinder
   (until the buffer is full), but for anyone using the LCD for framing
   it can become very frustrating.
3. In the same manner you can not certain camera settings (such as image
   size / quality) nor enter any camera menus while the camera is writing
   to a storage card.
4. Despite its looks, target audience and price the camera can only buffer
   four images at a time, no matter what the image size / quality selected.
   Couple this with the slow media write speeds and you often find yourself
   with a full buffer waiting for the camera to write images away.

Those are the negatives. On the positive side the E-20 focuses very quickly
and has very low LAG times, if you shoot with the viewfinder you’ll be
able to fire off a shot every 2.4 seconds (which is quick) and most other
camera functions are quick enough not to be too noticeable.

Timing Notes: All times calculated as an average
of three operations. Unless otherwise stated all timings were made on
a 2560 x 1920 SHQ 1/4 JPEG image (approx. 2.4 MB per image).

The media used for these tests were:

• 128 MB SmartMedia (TwinMOS)
• 512 MB SanDisk Ultra Compact Flash Type I card
• 1 GB IBM Microdrive Compact Flash Type II card

Action	Details	Time, seconds (128 MB SM)	Time, seconds (SanDisk CF)	Time, seconds (Microdrive)
Power: Off to Record		8.5	5.6	7.5
Power: Off to Play	Image loading	9.1	6.1	7.6
Power: Record to Off		2.6	2.5	2.5
Power: Play to Off		2.2	2.2	2.2
Mode: Record to Play	Image loading	1.6	1.6	1.7
Mode: Play to Record		1.0	1.0	1.0
Mode: Format		42.5	2.1	2.9
Play: Magnify	Each step	~0.5	~0.5	~0.5
Rec: Thumbnail view 2 x 2		1.6	1.6	2.2
Rec: Thumbnail view 3 x 3		1.9	1.9	2.7

Action	Details	Time, seconds
Auto Focus LAG	At wide angle	~0.6
Auto Focus LAG	At telephoto	~0.8
Shutter Release LAG	Using viewfinder	<0.1
Shutter Release LAG	Using LCD monitor	0.2 — 0.3
Total LAG	At wide angle	~0.8
Off to Shot Taken	Using 512 MB SanDisk Ultra CF card, at wide angle	5.8
Shot to Shot	Using viewfinder *2	2.4
Shot to Shot	Using LCD monitor (wait for live view to return) *2	11.5

*1	Generally using continuous auto focus will at least «prepare» the cameras focus system to be close to the ideal focus point.
*2	Image quality set to 2560 x 1920, SHQ (1/4 JPEG). For the LCD monitor test record view was disabled, the second shot was taken only when the live view returned to the LCD monitor although you can press the shutter release after 2.4 seconds (assuming there is space in the 4 image buffer).

Legend
	Quickest
	Significantly slower

Auto Focus LAG is (roughly) the amount of time it takes the camera to
autofocus (a half-press and hold of the shutter release button), this
timing is normally the most variable as its affected by the subject matter,
current focus position, still or moving subject etc. This timing is an
average.

Shutter Release LAG is the amount of time it takes to take the shot after
autofocus, this timing assumes you have already focused (half-pressed
the shutter release) and now press the shutter release button all the
way down to take the shot. This timing is an average.

Total LAG is the total time it takes (not just the two above added together)
if you haven’t pre-focused, that is no finger touching the shutter release
button, press it all the way down in one movement, this new timing is
how long it’d take if you were in one of those spur-of-the-moment situations.
This timing is an average.

Overall Image Quality / Specific Issues

There were no nasty surprises with the E-20, it is after
all a ‘matured’ E-10 with a new high resolution sensor. One notable improvement
is the ‘relative resolution’, that is the resolution the camera produces
versus the resolution we would expect for that pixel count. To me the
E-10 was a slightly disappointing four megapixel, later four megapixel
digital cameras showed that it wasn’t getting all it could from the pixel
count.

The E-20 is a different beast, as you’ll see in later pages
it has very good resolution and is definitely ‘in the zone’ of what we’ve
come to expect from a consumer grade five megapixel digital camera. In
normal Olympus style colour is subdued but accurate, this means you won’t
loose colour detail due to over-saturation but you may like to ‘boost’
saturation in certain images to give them a little more zing.

The E-20’s only major image quality disappointment are
noise levels, for a ‘professional’ digital camera it’s ISO 80 exhibits
more noise than we’d hope or come to expect, especially when comparing
its images to other prosumer level five megapixel digital cameras. I see
no reason why Olympus couldn’t achieve the smooth and relatively noise
free images we’ve seen of cheaper five megapixel digitals.

Thanks to Olympus’s «pixel mapping» I observed
no dead pixels on the E-20 I had for review, we have seen dead pixels
on other five and four megapixel digital cameras which don’t employ an
active pixel mapping feature.

Bayer pattern artifacts

One thing I noticed on fine detail diagonals were dotted ‘Bayer pattern
artifacts’ (as seen in the 200% crops below). These broken line patterns
are an ‘error’ induced by the decoding of the red, green, blue pixel Bayer
layout of the original CCD colour filter array. While not a huge problem
in themselves the existence of this artifact should be noted.

Shadow noise

As you will already have seen the E-20 does suffer from some visible
noise, even at ISO 80. Below are a couple of 100% crops from a ‘normal’
ISO 80 daylight image which should give you an impression of the noise
visible in shadow areas.

File Sizes

Here’s a comparison of the average byte size of each of different image
size / quality combinations.

Image Type	File size *1 (~ = approx.)	No. on supplied 16 MB SM Card	No. on 256 MB CF Card
2576 x 1924	RAW *2,3	9,697 KB	1	25
2560 x 1920	TIFF *3	15,616 KB	1	16
2560 x 1920	JPEG 1/2.7	~3,400 KB	4	64
2560 x 1920	JPEG 1/4	~2,400 KB	5	92
2560 x 1920	JPEG 1/8	~1,200 KB	10	166
1792 x 1344	JPEG 1/8	~570 KB	25	413
1280 x 960	JPEG 1/8	~300 KB	49	791
1024 x 768	JPEG 1/8	~160 KB	75	>999
640 x 480	JPEG 1/8	~80 KB	164	>999

*1	All file sizes are an average of three files. As is the case with JPEG it’s difficult to predict the size of an image because it will vary a fair amount depending on the content of the image (especially the amount of detail captured). For example, take a photograph of a fairly empty wall and you’ll get a small JPEG, take a photograph of a bush with a lot of detail and you’ll get a larger image. File sizes here are closer to the later, the larger size of file you should expect.
*2	All ‘active pixels’ are recorded in the RAW file, however the final image produced by the Olympus RAW processing software is 2560 x 1920.
*3	Additionally a 7 KB .THM (a 160 x 120 JPEG thumbnail) is written with the same filename as the .ORF (RAW) or .TIF file.