
A three way tie (well, almost)
Unfortunately I didn't happen to have three different
six megapixel digital SLR's for this test (I would have liked to compare
the Kodak DCS 760 but couldn't get hold of one in time). Instead we rounded
up the Fujifilm S1 Pro and (purely from an academic point of view, still
no news on a Pro D-SLR from Canon) the Canon EOS-D30.
Nikon D1x vs. Fujifilm S1 Pro (both at 6 megapixels)
Lenses used: D1x (Nikkor
17-35mm F2.8 AF-S @ 35 mm), S1 Pro (Nikkor 17-35mm F2.8 AF-S @ 35 mm)
Nikon D1x
ISO 125, 1/1.3s, F10, sRGB |
Fujifilm S1 Pro
ISO 320, 1/3s, F11 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Colour balance, as you can see, is very similar. I'll
give the edge (just) to the D1x, to my eyes (and on my calibrated monitor)
its colours are closer to those in the scene. Clearly the D1x has more
resolution than the S1 Pro, that's no real surprise, the D1x has 5.33
million pixels on its sensor, the S1 Pro only 3.3. The theory says that
the S1 Pro has more vertical resolution than the D1x but the proof of
the pudding (above) shows that the D1x's far higher horizontal resolution
more than makes up for that. Having said that, for a camera which sells
for around half the price the S1 Pro puts up a pretty good fight.
Nikon D1x (at 2.6 megapixels) vs. Canon EOS-D30 (at
3.1 megapixels)
"No fair!", I can hear EOS-D30 owners (and
Canon) crying... And no, it's not fair test of resolution, the D1x will
be downsampling over 4000 horizontal pixels to generate the 2.6 megapixel
image, but just for the sake of it here's a comparison of the D1x at its
medium resolution setting against the D30 at its highest resolution:
Lenses used:
D1x (Nikkor 17-35mm F2.8 AF-S @ 35 mm), D30 (Canon 17-35mm L F2.8 @ 35
mm)
Nikon D1x @ Size Medium
ISO 125, 1/1.3s, F10, sRGB |
Canon EOS-D30
ISO 100, 1/1.6s, F10 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Clearly the D1x's additional pixels serve it well, medium
size images are sharp and detailed without a hint of noise or artifact.
It goes to show how much more detail there COULD BE at the sizes even
the best digital SLR's output as their native resolution. Interesting
to note the difference in hue between D1x and EOS-D30 for the orange and
red crayons (last crop). Obviously the D1x got the colour right here.
Higher sensitivities
Below is a comparison of noise levels at different ISO
equiv. sensitivities, we've started at ISO 200 (or ISO 320 in the case
of the S1 Pro - as this is its lowest sensitivity) and end at ISO 1600.
Lenses used:
D1x / S1 Pro (Nikkor 17-35mm F2.8 AF-S @ 35 mm), D30 (Canon 17-35mm L
F2.8 @ 35 mm)
Nikon D1x
ISO 200 |
Fujifilm S1 Pro
ISO 320 |
Canon EOS-D30
ISO 200 |
 |
 |
 |
 |
Nikon D1x
ISO 400 |
Fujifilm S1 Pro
ISO 400 |
Canon EOS-D30
ISO 400 |
 |
 |
 |
 |
Nikon D1x
ISO 800 |
Fujifilm S1 Pro
ISO 800 |
Canon EOS-D30
ISO 800 |
 |
 |
 |
 |
Nikon D1x
ISO 1600 |
Fujifilm S1 Pro
ISO 1600 |
Canon EOS-D30
ISO 1600 |
 |
 |
 |
Noise levels are very similar for the D1x and EOS-D30,
the S1 Pro starts with more noise (at ISO 320) but maintains approximately
the same noise as the others all the way through to ISO 1600. It's fair
to say that all of these cameras are usable up to ISO 400 without any
(noticeable) noise, at ISO 800 with caution and at ISO 1600 in extreme
situations.
Notable improvement: Yes,
there's noise at higher ISO's but look, it's random noise, not pattern
noise as we saw on the D1. This will be of great relief to most D1 owners
/ users as they will appreciate how difficult it is to remove pattern
noise, but how natural and almost 'film grain'-like ranomd noise is.
|