At work I'm on a MacPro mid 2010 tower upgraded to 12 core OS 10.9.5 running MC 8.5.2 with a NitrisDX. I have a safe color limit on a sequence set to 422 composite -20, 110 default and RGB limit set 16 to 235. I rendered this fine on the work system.
When I take the sequence home I'm on a mid 2012 MacPro 12 Core now running a DNxIO MC 8.5.3 OS 10.9.5. I find that when I unrender the safe color limit the black level is elevated. If I drop on a new safe color and use the same settings all is correct with no black level elevation.
Finally after comparing the two effects I noticed the Luma L parameter is 7.5 on the safe color that elevates the black when it becomes unrendered and the new safe color has a Luma L value of 0. Now I've never even paid attention to those parameters before but I have seen a similar issues over the years and had never put my finger on it. The first time it happened it was when I applied the safe color limit from an 29.97 project in a Pal Project. I had figured it had something to do with a lack of setup in Pal and just rebuilt the effect. Since then I have seen this behavior from time to time.
Given my system is virtually identical to my work system the only difference is at work it's NitrisDX and at home now DNxIO. I also have a NitrisDX on my home system but have switched to the DNxIO for this 4 K PIxel hunt I'm in the middle of. Before I had a Black Magic SDSDI card and I had occasionally seen this same issue.
Is there something in the Avid IO hardware that would account for a different interpretation of the Luma L parameter in the safe color limit effect? Given I never knowingly set Luma L to 7.5 I figure it must be a by product of the video output settings composite output where you can choose to add 7.5 ire on composite setup. Could that setting effect the default value assigned to Luma L in the Safe color effect?
Given the video output doesn't have the analogue output adjustments in Avid's video output when running 3rd party IO I'm beginning to think that having Luma L at 7.5 on a Nitris DX system gets interpreted as no black level elevation whereas the same setting when using BM and perhaps other manufactures IO hardware results in elevated black. Does this make any sense? I can't see any other explanation as to how it rendered to correct black levels on the NitrisDX system but doesn't render to correct black levels on the DNxIO system. At least I finally found the parameter that is causing it bit how come it doesn't cause elevated blacks when rendered on the Nitris DX system the sequence was created on?
John Moore Barking Trout Productions Studio City, CA bigfish@...