You can see them bigger, I didn't know that. Thanks.

Please don't think I've taken a lot of time over these images, because I haven't. They were taken very quickly with little care for composition. The most time was spent importing them to Lightroom (no adjustments applied), exporting them as JPG's, then batch processing through Noise Ninja. I did have to make a noise profile first which took a while.

Thanks for your appreciation.

Stefan,

#3 shows more noise in the shadows than the other two but it renders the best colours for me.
#2 colours are acceptable with some grainy noise in the shadows which I find more pleasing to look at than #3
#1 colours are not good and a similar noise effect in the shadows to #2

Which is which? I haven't a clue except that the noise pattern in #3 is similar to what I see from my K20D in certain conditions.

Is the a700 the Sony alpha 700?

You can see all of my test pics here.

The first five images are those I've posted here earlier.

The remaining pics were taken tonight from roughly 21:40 to 21:52 (1 hour ahead of UK time). The last 6 were taken in near darkness with no EV comp. All were taken at quite slow shutter speeds so please, no criticism of focus, camera shake, or straightness of lines, I could barely see the subjects, let alone focus on them. The last two images are lit by a street light which came on part way through this exercise.

Yes, some images exhibit noise, even after being cleaned through Noise Ninja, but I personally can live with it as I accept that any camera has its limitations and I try to keep within its usable performance envelope.

What I have learned from this exercise is that the K20D can produce perfectly acceptable results at high ISO settings and I will have no issues about selecting ISO 6400 in the future if the situation needs it.

I'm with you Gwyn. Times they are a changin', and not for the better I think.

It's a surprise to me that this thread hasn't been locked long before now.

George, some folk on here just don't understand you, but I'm happy to say that most of us do

OK, now I understand what you meant and I misinterpreted your first comment, apologies for that.

You are right of course, low light will test a camera's high ISO performance to the max. and I can't think of many other situations worse than underground.

Nestos: I think this image is best put through something like Noise Ninja, then converted to B&W. It's never going to make a great colour image I think.

Before you speak (or write) make sure that what you say is more interesting than your silence would have been.

Well it is almost totally dark and the DA*300mm is an f4 lens and it did eventually achieve focus lock when I chose the most contrasty part of the scene. I could have switched to the DA* 50-135mm ( f2.8 ) but I wanted to use the same hardware as previously. I don't want to invite criticism of it being an unfair test . I think probably only the military have cameras that work faultlessly in this light, could even be Pentax .

Thanks George, that's good enough for me. Let's hope a few others agree.

I'm just processing some low (very low) light shots. Same subjects, but it's now so dark that the street lights have come on and the AF system won't lock on without a bit (a lot) of encouragement.

I'll post them here in a bit, noisy or not.

I would agree, but if you were to simply look at them as images (OK, the subject is rubbish but it's all I had time for today), and without knowing they were at ISO 6400, would you say they were noisy? I wouldn't.

I know these are not worthy of Pentax Gallery for instance but I'm trying to build a case for the K20D/K7 noise debate and show that it really isn't the huge problem that people (potential defectors) perceive.

800 pixels on the longest side is the biggest I can show from Picasa.You want more proof? OK.