It's an optical paragon, according to Carl Weese. He thought it was better than Zeiss glass, used for landscape, putting it up there with the very best lenses available. Personally I prefer the softer bokeh from the slightly faster f2.4 one though.

Newborough Warren? Superb shots Simon, the light is wonderful.

Something like a 90 or 100mm macro is good for bugs and also quite useful for candid portraits owing to the length and wide aperture. The Da35 macro is more difficult for bugs because of the working distance, but I've seen great results with it and it seems great for flowers etc. And it's a terrific versatile focal length for other stuff.

If these features are expendable, why are you so hard line about returning the camera if everything doesn't work 100%? I would probably return the camera if I'd just bought it and the level was out significantly. If it was slightly out I might just learn to compensate, if everything else was working fine. You might not be so lucky with the replacement body!

I really don't think the k5 issues are caused by cramming too many extraneous features in. The k7 didn't suffer to the same extent but has a similar feature set.

A fast prime (Sigma 30mm / Pentax 35mm) will be great for individual 'portraits' or shots where only one person is in focus (e.g. a person talking while others listen) but I agree with Robin, for wide angle group shots indoors you need something a bit wider and reasonable DOF which will need flash (unless you have a K5 perhaps :wink. Bounce the flash off the ceiling.

Geometric distortion is a shame, but in some rooms you have no choice if you want to get everything in shot - the 10-20 would be OK but you might want to avoid the wide end! The kit lens might be just wide enough actually. Experiment in your living room and see what works for you.

Sony seem to have an aggressive stance on noise, banishing it at the expense of detail, whether it's in their DSLRs or compacts. I think there's something else going on other than the sensor itself.

The Q in those tests very nearly equals the dynamic range of the Sony NEX C3. The NEX-C3 has the same APSC sensor as the K5. On signal to noise ratio, the Q seems on a par with the Panasonic GF3, somewhere between the Canon G12 and the Sony.

Looks to me like Pentax engineers have done a great job with the imaging pipeline. Sony... not so good considering how much better the K5 is, using the same sensor.

However when the others discover whatever voodoo the Pentax engineers are performing on that little sensor, and apply it to their bigger sensors, they will outstrip the Q again.

It's a great testament to Pentax engineers that they have got such a fantastic result from that sensor. The dynamic range is particularly eye opening. Its much closer to the Sony NEX than the Panasonic GF3. Pretty amazing.

John, have you been sniffing the lens cleaner again?

My K7 noise minimisation tips:
- Shoot RAW and use LR3
- Go for strong chroma noise reduction (the coloured specks), and go lighter on luminance noise (which looks more like a film-like grain).
- Make sure you do not underexpose, as readjusting the exposure in PP will make noise really bad! If possible, slightly overexpose and let your highlights clip ever so slightly (you may be able to recover them in Lightroom).
- You can use layers in Photoshop to apply hefty noise reduction to parts of the image while leaving other parts unaffected... where there is detail the noise is less visible and the detail is reduced by noise reduction, but solid areas like skies show noise quite obviously while being less adversely affected by strong noise reduction.
- Do not pixel peep too much! By the time you have resized your shot for the web or printed it, noise which was clearly visible at 1:1 will not (in many cases) be visible.

Wow, lovely set. I do like the blur effect and the treatment, especially the first and last shots. Where is this amazing location?