This Morning's Treat!

[merlin]

 

 

 

 

Edited June 12, 2017 by merlin

[Chucktin]

A little too bright on the highlights? Especially the yellows.

 

Sent from my SAMSUNG-SM-N900A using Tapatalk

[merlin]

They look fine on my calibrated 22" Eizo monitor.  Thanks anyway.

[kimcarsons]

@merlin, you need to try that calibration again, buddy. Just looking at the histograms should show you that most of these images are over exposed.

[merlin]

Thanks for responding.  The jpeg of the daylily (#1) is exposed correctly, but the other 4 jpegs are, as you stated, overexposed.  Since the RAFs are not, I wonder why?  I probably did not notice since my monitor brightness is set to medium.

 

I processed the RAFs of the last 4, and these are now posted.

[kimcarsons]

Thanks for responding.  The jpeg of the daylily (#1) is exposed correctly, but the other 4 jpegs are, as you stated, overexposed.  Since the RAFs are not, I wonder why?  I probably did not notice since my monitor brightness is set to medium.

 

I processed the RAFs of the last 4, and these are now posted.

 

RAFs have 14-bits of dynamic range, JPEGs 8-bits. Part of the art of processing raw files is in determining how to squeeze that 14-bits down to 8. It looks like you had some 'highlight recovery' going on there too, in which the raw development software tries to fill in blown out regions with the color of the surroundings (resulting in large, detail-less patches of color). Flowers are tough for subjects with digital for multiple reasons, but one is that it's easy to blow out e.g. the red channel while there's still plenty of headroom in the blue and green channels. Recovery here looks bad too, but not as bad as when all three channels are blown.

 

Take home lesson: a properly exposed flower shot should look under-exposed before processing (and probably after too, if you want to maintain any detail in the petals).

 

Bear in mind that display calibration entails calibrating the brightness and gamma also. The ideal brightness is a matter of opinion, but 110-120 nits is a common value.  Any decent colorimeter should be able to do this. BUT this also entails having a standard level of ambient illumination. So if your workstation is in a room with variable lighting (i.e. daylight), all bets are off (unless you have one of the fancy colorimeters that stays plugged in all the time to measure ambient illum. and compensate--even then there are compromises).

 

As always, the histogram and your software's over/under exposure warning indicators are more objective guides than your own perception.