Quote:
Originally Posted by Dave882
Hi Allan,
I believe you are correct in that it is going to be a bit of a compromise. But the reality is that I just don't have the computing power to process that much data in one go. Just for interests sake, I loaded all the subs (1989 in total after manual removal of obvious duds) and let it give me the estimated stats - it needed 256GB for processing (way more than i have to play with) and from experience, would take an enormous amount of time. And thats binned 2x2.
When I stack sets, APP actually recognises when you load 'lights' that have already been integrated and deals with those accordingly. From my experience the only real problem I've encountered is if you've got some sets with differing amounts of integration. The problem there is that the stacking program will prejudice against good data just because its in a smaller set. But this is easy to solve by ensuring all sets are roughly the same size.
All in all, I'm happy with the compromise as the testing I've done on smaller projects seem to show very little difference - but yes I will accept that with the right computer running the whole set in one go would theoretically provide a better result.
Thanks for the very interesting link. Although judging by the responses the juries probably still out - with those with APP experience saying it definitely works, and some others with valid concerns...
|
Thanks Dave,
256GB of RAM - that's incredible.
Most desktops don't even have room to fit
that much on their Motherboards.
I have 32 GB - I thought that was a lot.
Yes - it seems that the stacking methods are debated strongly online.
cheers
Allan