> What do you think of Panasonic's own samples?

> Ian

First, I would have to say that the leading image on their site (P1000228) is a peculiar choice. Not only does it exhibit the greatest evidence of processing artifacts, but it is also a strange use of the camera and an unflattering picture of the model. It's a strange use of the camera because one would not normally choose to take a portrait at wide angle. If I had taken this shot it would never have seen the light of day.

I think the three telephoto shots look very good at "actual pixel" resolution on my computer monitor, and they represent the way I would likely use the camera if I choose to buy it - telephoto shots of discrete subjects, primarily nature shots, making use of the OIS to gain a couple of stops at the optimum ISO equivalency rating. These samples please me enough that I am leaning quite favourable towards this camera.

To return to the portrait, however: This particular image is the only one of the four to be shot at wide angle. The lens was wide open, and the sutter speed relatively low (1/50th) compared to the other images on their site - though 1/50th should be no challenge at all to this camera, given the OIS lens and the wide angle of the zoom. I am very puzzled why these settings produce the significant amount of processing artifacts that they do in this image. The model's ear, lips, neck and eyelashes look quite "painted" to me. Given the conditions, why does this one look so artifical, compared with the others on Panasonic's site? It would have been interesting to compare a RAW image for this shot.

Also, if I had taken this photo, I would have underexposed by at least 2/3 stop so as not to lose so much highlight detail in the flowers. That would require some post-processing to lighten the rest of the image, but that's how I shoot almost all the time with a digital camera, since they're so unforgiving in the highlights. (Will we ever get a CCD with a "shoulder" like we have with film?)

I'm having trouble figuring out what it is that triggers the camera's processor to over-compensate. Ian, your shot of the souvenir seller (P1020020) is a good example - the hats on the ground are perfect, but the man's blue jeans look quite artificial, and even his face is somewhat processed-looking. Do you think it's because the plane of focus was on the hats? Is the camera's processor having trouble distinguishing between fuzziness due to plane of focus, vs. the noise that it's supposed to be eliminating? (BTW, you haven't lost much, if any, shadow detail in that photo - the data is there, and you can see it if you process the image to lighten the shadows.) The most "painted" part of your Gaudi museum shot are the out-of-focus rock wall in the lower left and trees in the background. The background of your wide angle sample looks like an impressionist water colour (what happened to the depth of field? I would have thought f7.1 would have been sufficient.) But this hypothesis doesn't seem to work for Panasonic's sample portrait, where the "painted" parts don't look to me like they're out of the plane of focus.

All in all, I'd rather have less processing in the camera and let my computer's filter take care of more of the noise.

Finally, I would have to agree that looking at the shots on the computer monitor using "actual pixel" resolution is probably not a real-world test for most photos, and that print resolution is probably a fairer test. But every once in a while I find a section of a photo that I paid insufficient attention to when framing the shot, that I want to zoom in on. So I'll continue to expect miracles from my cameras.