K10D Firmware 1.0 Full Size Samples Disappointing!

[RiceHigh]

As many of you folks may get aware that the firmware 1.0 of the K10D has been out and the full size samples are posted at the Japanese DC Impress review site:-

http://dc.watch.impress.co.jp/cda/re...1/13/5035.html

Use Babelfish to translate the page if needed:-

http://babelfish.altavista.com/

Whilst quite some excited Pentaxians feel that the quality of the images are excellent, I found the difference. I have some quick inspections on the test shots (for both those taken in the field and in the "lab"), and the following problems and disappointments are not difficult to be located:-

1. Exposure accuracy is not good, actually, it can be seen that +ve/-ve exposure compensations were applied for most of the test shots, in a random way. Also, even for the same amount of EV compensation were applied for similar scenes, the exposure results are slightly different;

2. Image corner blur and CA are obvious. The resolution shown is obvious not on a par with the Canon 400D. Actually, I think that the D80 is better too. To compare against my K100D with the same lens, i.e. the DA 16-45, I found that my K100D indeed perform even better when viewed my pictures at 150%, despite the K10D has more resolution and despite that the K100D zoomed images will have "blocks" seen, but still the differences in quality can be seen, that is, the K10D images are not of good resolving power and there is more CA and blur;

3. Images look dull somehow for the first series of pics and colors are not quite natural for the greens and the blues. The green colors of the plant of the K10D ISO test shots look different too (when compared with those K100D test shots also by the DC Impress);

4. AF accuracy seems a bit off for the ISO test shots - look at the bear doll images.

.. well, I shall not go on further, just because I have seen just a few number of test samples. But, why not these should be very good instead?

I'm not trying to find problems (well, I know some of you guys may probably accuse me of), I'm just disappointed!

RiceHigh
http://www.geocities.com/ricehigh

[ennacac]

Tom

[errno_gmm]

hurf durf

roy

[gadgetnut]

*sigh

[HogRider]

[rhermans]

:sad: RiceHigh, could you give me a link to those "K100D test shots also by the DC Impress" with the plant, I'm unable to locate those.

[ennacac]

It is the K10D and I expect this is the site he is talking about:

<http://dc.watch.impress.co.jp/cda/review/2006/11/13/5035.html>

Tom

[milrodpxpx]

wham bam thank you maam :?

[pbender]

RiceHigh wrote:

2. Image corner blur and CA are obvious. The resolution shown is obvious not on a par with the Canon 400D. Actually, I think that the D80 is better too. To compare against my K100D with the same lens, i.e. the DA 16-45, I found that my K100D indeed perform even better when viewed my pictures at 150%, despite the K10D has more resolution and despite that the K100D zoomed images will have "blocks" seen, but still the differences in quality can be seen, that is, the K10D images are not of good resolving power and there is more CA and blur;

While the other things you cite may indeed be problems with the camera's firmware, I have to question this particular comparison.

The issue I have is the problems you mention above may not exist in the RAW image, but may instead be introduced by the JPEG compression algorithm.

I don't know if the compression was done by post processing the image or by performing the required compression within the camera. If this was in camera compression, then it may not perform as well as an off camera compression routine will.

Without the raw images to compare with, I can't say one way or the other what is going on.

Paul