color grading DV
by ali jafri
on
Jul 10, 2008 at 12:08:36 pm
Can someone tell me if its a good idea to color correct DV footage. I'll basically edit on premiere and then use combustion to colour correct using masks etc before putting the clip back on premiere's timeline. Since I'm working completely on DV (using FireWire to capture and print to tape on a DSR-11 deck) I want to know how much generation loss I will incur. Is it even a good idea to color grade on Combustion? What would the best and most efficient workflow be?
--------------------------------------------
Some people see things that are and ask, Why?
Some people dream of things that never were and ask, Why not?
Some people have to go to work and don't have time for all that.
Re: color grading DV by Jeff Brown on Jul 10, 2008 at 2:14:20 pm
Is your project short enough to make file sequences reasonable? Only a question you can answer, but I generally use PNG sequences, one clip per folder (keeps access/reading quicker). That is a lossless but efficiently compressed option. You don't gain anything at the outset, of course; you still have a DV source, but subsequent operations are not in a compressed color space. Premiere will import PNG sequences, so you can render back to a DV timeline for the "conform". Plus, you end up with all the advantages of sequences in Combustion: network rendering, faster C* timeline interaction, the ability to do partial re-renders.
An even broader colorspace would be to use a 10-bit capable format such as DPX, but I'm not sure how well PPro would read those.
Also, to possibly speed your workflow: if you have a 2nd computer to net-render, you can batch the finished file sequences via C* to render DV clips; then no waiting for Premiere to do it.
Then again, you may find the difference negligible, and decide that DV in/DV out is just fine for your purpose. Test a clip or two and see what you think.
Re: color grading DV by Ali Jafri on Jul 10, 2008 at 5:42:07 pm
Thanks jeff :) i'll definitely try out a clip both in dv format and in still-image sequential files, lets see what happens. Basically i'll need to mask out some commercial brand names from a supermarket shot, correct some white balance issues, and reduce saturation where need be.
Which brings me to my next question. I'm shooting on a sony fx1 in dv pal mode using its picture profile function to boost colour saturation. I'm thinking since dv has such limited colour space to begin with i won't get much margin to boost saturation in post, i've also heard that a cut is always better than a boost. So, in some shots the colour could use some reduction. My question is: does this technique hold any ground? Am i going about this totally wrong? Is there a difference in setting a dv camera to boost colour saturation and using combustion for the same purpose? Which method would yield the best results?
--------------------------------------------
Some people see things that are and ask, Why?
Some people dream of things that never were and ask, Why not?
Some people have to go to work and don't have time for all that.
