More on Signal-to-noise

Paul Curtis · March 10th, 2008, 11:09 AM

>users are shooting skin tones at 60-65%, reporting that 70% still leaves blown hilights in specular reflections on a face.

That's what im looking at this week too. I've noticed some highlight rolloff that i find unusual.

Im wondering whether colour temperature is affecting this. Sensors are usually less sensitive in blue and under tungsten blue is lacking. so red is clipping early by comparison (even if you white balance for tungsten it doesn't change the sensors native response). So i've been using 5500k sources for experimentation to see if there's a better overall response.

cheers
paul

March 10th, 2008, 11:09 AM	#16
Paul Curtis Major Player Join Date: Mar 2005 Location: canterbury Posts: 411	>users are shooting skin tones at 60-65%, reporting that 70% still leaves blown hilights in specular reflections on a face. That's what im looking at this week too. I've noticed some highlight rolloff that i find unusual. Im wondering whether colour temperature is affecting this. Sensors are usually less sensitive in blue and under tungsten blue is lacking. so red is clipping early by comparison (even if you white balance for tungsten it doesn't change the sensors native response). So i've been using 5500k sources for experimentation to see if there's a better overall response. cheers paul