must been asked before but anyway :)

[martijn76]

quick question
we shot with red

found out that when using the qt ref file's in fcp they upgrade if colour is changed in radalerd

my question is :

for broadcast i need to deliver digibeta pal

im thinking anamorphic is the way to go

im thinking set sequence to 1920x1080 sq pix
import my qt 2k ref file's set them to 100% small overlap left en right (2048-1920=128 pix) en little black bar on top and bottom (1080-1024=556 pix)

now export to 10 bit unc 422 pal




2-1 aspect ratio is not 16-9 (16-9 being 2-1.125) putting my ref file's in a 1920x1080 sequence @100% wil make them 16-9 anamorph when renderd to pal

my boss disagrees about this methode because we lose little frame on left and right (he's also dop:))

i think it's the way to go because there is as little manipulation as possible pixel wise?


questions

is it okay to render from the ref file's? if yes does it make any diffence what compressor is used in the seq renderd from?

in other words does rendering the R3D file's to 4k tiff first get's you better quality when there's only downsizing to pal involved?

thanks a bunch

martijn

[Rob Lohman]

1) your method seems okay

2) the QuickTime reference files are high-quality (with the grade codec installed), but not ultra high-quality. Whether they are good enough depends on your footage, quality wanted versus time to process. The QT codec does not do a proper downconvert from the footage, so you may see a bit of aliasing or stair stepping (more so on the 1K than the 2K for example). If this is a problem (check before assuming it is), then you need to do:

3) highest-quality is through RED ALERT exporting a 16-bit 4K tiff/DPX and then down sample / crop that to your output resolution