Quote:
Originally Posted by ArabianSwan
The problem with this method, as I came across few times, is that not always the signs are final or good. For example, a translator will start with the dialogue, finish that and upload it. Next comes the signs script, and since he did them seperatly, if there is a sign that reads the same as the dialogue, they might not be exactly the same. Editor may not notice it as s/he does the same, edit each file seperatly. Only when it is up for QCing... they notice it.
For that few times it happened with me I decide to just go form pre-filtered to final xvid or x264 encoding  |
Well, signs are supposed to be final by then. That's why you're proofing them before ^_^; ... yes, sign errors are nasty, but they should happen rarely enough to justify step 8). And besides, I always keep the former only-prefiltered lossless too, so should a sign change later on, you can simply repeat the process.
If your signs are mostly ASS and you don't trust your colleagues too much, you can keep the signs out of it - however if you're working alot with AFX typesetting in a sign-rich environment, you usually do NOT want to do the overlays again for each pass. They DO slow down encoding significantly.