Hi folks,
Here's a new one for me. Take any clip that has a few bits of information
below 0mV or above 700mV (luma, R, G, B, whatever). For example, here's an
external scope view, where everything green is above 0 and everything red
is below 0:
http://i.imgur.com/rp94vzB.jpg
Knowing that the particular clip in my edit would be fine to crush a bit, I
figure I'm fine with just dropping a Safe Color effect on to clamp at 0.
Using either the Avid Safe Color Limiter effect, or the Boris Color
Correction effect with Safe Colors checked, when I drop either one onto the
clip in my timeline, I get the desired result, which looks like this:
http://i.imgur.com/jKh4Cxg.jpg
I've got it set to clamp at 0mV low luma, 700mV high luma, and even 16/235
for my RGB. It does a PERFECT clamp across all my luma (as you can see) and
RGB channels, which is what I'm looking for.
Now here's where it gets weird. If I render/mixdown/export, it still does
the clamp, but misses a lot of stuff! In this case, it ends up looking just
like the clip without the effect:
http://i.imgur.com/rp94vzB.jpg
WTF? This happens with stuff going on above 700mV too.
Upon further testing, it appears that any of these effects work flawlessly
when you drop them in, but when rendered, they miss some stuff. If you
clamp without rendering, and park on an offending frame, zero alarms will
go off on the scopes. But as soon as you play/render/export/mixdown, the
effect goes from being precise to being approximate.
This happens with almost any clip (AMA, native DNxHD, etc.), on every
system I have (Mac, Symphony 6.0.3.2) and user setting I have (new
included). I feel like it's better with progressive material than
interlaced. I wouldn't even mind it so much, but the unrendered effect is
PERFECT. But what good is that, when it stops being perfect on playback,
render, export, or mixdown?
I am sufficiently boggled by this. Anyone else see this happening? To be
clear, I am talking about TINY infractions. But the pre-rendered effect
gets the outliers - that's the stumper.
[Non-text portions of this message have been removed]
| Reply via web post | Reply to sender | Reply to group | Start a New Topic | Messages in this topic (1) |
No comments:
Post a Comment