Thursday, March 29, 2007

An Interesting "Unfair" Comparison: 5D Vs K100D

Here is a recent interesting head-to-head comparison for a Canon 5D against a Pentax K100D:-

(The original post is in Traditional Chinese, use Babelfish if needed.)

To make the test even much more unfair, the tester used a Canon 24-105 L lens on the 5D whereas a slow and long range Sigma digital superzoom 18-200 was mounted on the K100!

As it is stated in the beginning of the post, the difference in prices between the two comboes is four times! Someone did respond to the thread that it was considered to be "not making sense at all!".

But let's view the pairs of the test pictures objectively, then I think this test is actually very meaningful. Indeed, we can see how Pentax and Canon produces pictures which can look differently and in what ways they are looking different.

Before going on for what I have observed, you can first see the differences yourself carefully by inspection.

For what I can see, it is found that the resolution of the 5D pictures is much higher, there are more details in the pictures no matter viewed at full size or cropped. The colours are more eye-catching, especially for the reds. The bokeh are obviously nicer. The Canon pictures look brighter. Clipping of highlights are less obvious and the bright area transitions are smoother for the 5D, even for where highlights have already been clipped (that would be very obvious for the Pentax, just see the white sky for one of the test shot pairs).

But is that really that the Canon produces absolutely better images than the Pentax? I'm afraid not, against what most people would think. Actually, I don't like Canon's over-manipulation of the brightness tone curve where the highlight areas are with steep contrast curve, locally and regionally along the levels, and they are intentionally lifted to a higher level than it originally should be. As a result, one can see the highlights are quite bright and I would use the word "glowing" to describe. However, just take the example of those white or light coloured buildings in the test shots, I'm sure one would never be able to see such "glowing" buildings in *real-life*.

Furthermore, the Canon pics are with highly saturated colours and that there are actually more sharpening applied.

Both of the above ways of "processing" will actually make the Canon images not quite natural. Indeed, I can say the Pentax produces more natural pictures than the Canon. If a better lens is used, I'm sure that one must able to see better or even the best results.

Anyway, I know many people would like to love to see pictures which are with exciting colours, sharpness, brightness and contrast. But then whether these reproduced images are true to life and natural would be another problem :-)

Do note that the tester did set -1 for the contrast for his K100D, which I think it is the main reason for the lower contrast seen in the pictures taken with the K100D. Anyway, I think the tester should have a good reason for doing so, maybe he found that his Sigma lens was too contrasty, just say. Also, btw, the Sigma lens is not that bad as seen in the pictures, especially consider that it is a Sigma but yet the colour rendition is still okay.


Colin said...

All that test is showing is that Pentax in-camera conversions are rubbish and Canon ones are worse. Not really news. We regularly use a 5D with 24-105 and a K10D with 16-45, and the difference in image quality with RAW is like MF and 35mm.

RiceHigh said...

Not really. For the same camera used, I would say the in-camera Jpegs and RAW converted Jpegs "taste" similarly. That says one can see that they are of the same style, despite that the RAW converted pics can look slightly better (but I don't think they are hell and heaven as you've suggested).

Post a Comment

Related Posts

Creative Commons License
RiceHigh's Pentax Blog by RiceHigh is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.