Originally posted by Class A Cool, shall be very interesting to see these compared in the same conditions.
Your setup sounds very good, so I'm looking forward to seeing your results (K-5 on its own and/or compared to K-7/K20D).
The K5 was here when I got home this evening, so I ran a few tests. I do not have time tonight to post K20 and K5 comparisons, but I can post some shots that show the firmware 1.01 performance in low light versus the 1.03 in the same light.
Here is the setup:
K5
FA 50 f1.4
LensAlign target on a tripod with a leveling bubble attached to the mounting plate. Note that this target has a provision for making sure that the sensor is parallel to the target.
K5 on a very sturdy tripod with a leveling bubble built into my Manfroto grip.
Lighting - 2 300ws studio strobes with 250w halogen modeling lights that are dimmable.
Sekonic L-358 handheld meter set to read in EV. Readings taken at target face.
Gossen Color meter 3F. Readings taken at target face.
Camera set to remote trigger with 3 second delay.
Lens reset to infinty between each shot.
AF assist never came on but focusing was slow and it hunted a bit at EV2.
EV was 2, color temp was just above the meter minimum at 2040K.
Some notes:
I have tested the setup with my 5500K CFL lights and it does move to BF a bit but it holds DOF. Note that my earlier tests with these lights against daylight show that they BF more than natural noon sunlight. I suspect that they are more blue than normal sunlight. I will confirm in sunlight as soon as the sun comes back out and I have a bit of time.
I shot a couple of quick tests of my lovely wife in very dim indoor conditions where I could not get the K5 to hold DOF using earlier firmware. The results looked good.
It SPANKS my K20, which will not hold DOF below f4.0 with this exact setup (and barely at that). I will post those shots a bit later along with K7 tests as well for comparison.
I noticed some very odd behavior after setting the initial AF adjust:
The camera varied all over the place at various apertures, showing BF at one aperture and FF the next. I reset the camera to UNSET and re-shot the sequence, and things were very consistent, but BF. I then dialed in the AF adjust again (+2)and re-shot and the results were much better. It still moves around a bit from some BF to some FF, but it holds DOF at all of the apertures I tested (1.4 through 5.6). I would suggest that anyone who sees inconsistent AF or no change after the firmware update try unsetting and resetting the AF adjust. I have no other explanation for what I saw.
Of course, the odd bevaior may come back as I do not yet have extensive time with the camera.
Some real shooting is needed to confirm results in daylight and under my normal shooting conditions, but at this point, I think it is a keeper.
Ray