[Marc Sabatella]

Probably because that's what DPR said in their review of the K20D.

Don't you think there's truth to that, though? I admit I haven't done any real experimentation, but just based on the one-line description of what these modes do (sharpness & fine sharpness), it seems practically inevitable to me that "fine sharpness" would affect noise (adversely) more so thn "regular sharpening - although to be sure, *any* sharpening is potentially detrimental to the appearance of noise.

[audiobomber]

Don't you think there's truth to that, though?

I assume DPR is correct about fine sharpness and noise, but I haven't tested either. Should be an easy thing to test for.

[mingdie]

Thank you all for the helpful advice.

I have done some tests and firstly I must say that I am impressed with the JPEG straight from the camera. The auto WB works well and when I dialled in contrast -1, saturation 0, sharpening +2 and used the natural setting it looked good to me.
Of course the person that need the JPEG thought the vibrant setting was better so it will be vibrant for him and raw for me.

The best part was that we had everything from bright sunshine to pouring rain and everything between over a couple of hours which made testing easy.

[deadwolfbones]

Real answer is 'it depends on what you're shooting and if the day is overcast, etc, etc', but the short answer for the K20D is: 80% of the time, jpeg "Bright" setting with +2 fine sharpness is going to get you brilliant results.

Recently I've been hearing that 'fine' sharpness enhances noise more, but regular sharpness adjustment in camera does too, in fact I'd be hard-pressed to see the difference with regards to noise. But 'fine' just seems to render hair, eyelashes and things like that more naturally.

K20D is a very fine jpeg machine.

Having seen Jay's results, this is what I'd use.

[audiobomber]

Of course the person that need the JPEG thought the vibrant setting was better so it will be vibrant for him and raw for me.

Did you show him the Bright setting? I did some more tests as a result of this thread and Vibrant seems less saturated but, too cool compared to Bright.

I'm using Natural with +1 Contrast and +1 Sharpness. Gotta love Raw+Jpeg huh?

[wasim_altaf]

When it comes to sharpness, there appears to be some difference between the in-camera processor and the supplied PPL software. The camera processed JPEGs look sharper than the ones processed by PPL. In fact, pushing sharpness to +4 in PPL doesn't bring it closer to in-camera jpegs (I've tried that out). Even fine-sharpness lacks sharpness, comparatively speaking, possibly at the expense of noise.

Here is an example. This is from RAW+JPEG -- first one is JPEG straight from camera, second one is later processed in PPL applying the same parameters.

My guess is that the DPREVIEW tests were done externally with PPL. As is, I find the camera JPEGs plenty sharp, enough to leave it at -1, default in 'Natural,' if I'm saving directly to JPEG.