The actual quote came from here and went like this:
Cameron is just stating his opinion (and he was guessing at that -- he wasn't sure).
And why would we even think of lending credibility to it if he's at the forefront of 3D filmmaking, right? Classic. By the way, he's actually had one of his 2D films converted (the one you included in your quote)...and converted as 'right' as it can be....so he kinda' knows about that as well. Put it this way...some 'opinions' are worth more than others.
What is doing it right? Why couldn't you break the film into 3 parts have 3 different groups work on the job simultaneously and have it done in 2 months (8 weeks) instead of 6 months (24 - 52 weeks)? That's not rushing it it just means you are making a choice between spending a little more money and making your release date and waiting and delaying your payback time. If you borrowed the money to make the film (and I say this in supposition) I don't think the bank is going to wait on your schedule slip to get their money. This is why making schedule is so critical since the penalties could be more costly than spending the extra money now.
I'm not sure why I'm going into specifics, because you won't listen anyway...but here goes...
Obvioulsy, facilities have different groups of people working on a project. but it doesn't necessarily makes certain things go faster...like rendering. And these places have their rendering systems maxed out...which means if something will take 12 hours to render, that's the fastest that it can be rendered with today's available technology...no matter how many people you hire. Any effects/CG place goes through the same thing. If you have more people prepping the work, then you get more jobs waiting in line to render....or you divide the rendering over multiple jobs which lowers the processing power and lengthens the time needed for each. So another possible workaround...like with last-minute effects work...is to subcontract other facilities/houses and have them go full-bore on their part of it.
But one of the problems with farming out something like a 3D conversion to different facility/conversion houses is that many of these places use their own proprietary software/systems and encoding formats. They have their own ways of handling convergence adjustments, per-frame information, etc. i.e...they've designed and built it themselves, not bought the same software and computers that every other house has. For example..Pixar couldn't just hand over 1/3 of their animation work to Dreamworks or what have you because Pixar designed/built their own software and animation systems...it won't even open up or translate to another place's system. So for the sake of consistency and control, it's mostly best to have it in-house at one place...as a lot of adjustments and tweaks are made with the whole thing in front of you. But if different 'groups' are using different systems and such, it bogs things down because it's harder to put it all together to check and proof...and adjustments that can be made on one system are made a different way on another.
Plus, in one place, even if you have multiple groups doing different parts...they all pretty-much share the same media through one gigantic server. But if you do part A at a place in, say, LeBrea, part B in Palo Alto, and C in San Francisco...they phsysically can't share media at once...and these aren't files that you can simply email to eachother. So even though you have more people and more systems doing different parts of the work, you can easily end up spending more time/money trying to get everything on 'the same page', and even the tiniest adjustments can mean going back and doing/rendering the entire sequence again...20 seconds of screentime can easily take up 3 days of checking, adjusting, re-adjusting, re-rendering, conforming, proofing, and transfer/delivery. It can actually take longer to to get two parts from two houses to jive with eachother than it would just one of the places to do it all. And that extra time isn't productive time, either...it's like turning a humongous quicktime into a WMV just so that another system can read it, and it'll take at least a day to transcode...and if something's off, they have to go back and do it all again and spend that extra time translating.
When everything's 'in-house' it cuts down on a lot of that...but if your in-house rendering and workforce is already maxed out, you need more time. And that's when quality goes down because you are rushing and settling for less proofing/fine-tuning/etc. in order to make a hard deadline. There's just a point where more money and more people can't buy you time, and you even start to lose more time trying to organize it all.
I hope that's a bit clearer, but I have a feeling it still won't sink in.
t:
