Accountneedsrealnameupdate
Forum Replies Created
-
Yes it is the same, as long as you set it to square pixels and not anything else.
Glenn Stewart
1k Studios -
The source is dBeta (telecine from film/HD master) and to save drive space we digitized in photojpeg, it looks better than DV with a smaller file size and the decklink cards will pump it to SDI for monitoring without the need for a disk array. I’d rather recompress in software than tie up (and wear out) our decks redigitizing.
Thanks for the advice, I was just hoping someone had a trick to make the photojpeg work.
Glenn Stewart
1k Studios -
Same as above, but I’d highly recommend getting some delivery specs from your potential clients, will they accept DV/DVCam/DVCPro or will you need to rent a beta deck? (if so sp or dBeta?), what audio/black levels do they expect? You can spend the earth on equipment if you want but there’s nothing particularly limiting about final cut pro and a G5 as far as broadcast goes, I do recommend an uncompressed video card and a decent production monitor first of all, probably a scope if you can afford it but you can always do it in software if you’re conservative. If you’re in LA you’re already half way there, there are plenty of places to hire equipment or even bring in a firewire with your final sequence to lay off to tape, they’ll have scopes and if you’re good at picking brains you can learn a lot from them.
Glenn Stewart
1k Studios -
As above but in my experience you’ll probably get a decent result by just bringing the footage in, putting it in an ntsc comp and doing a field render without conforming to 24 first, after effects will add pulldown, it won’t be a nice 3:2 pulldown like the previous method but it will look ok and you don’t need to worry about audio. Make sure you interpret your source correctly if it has fields already (upper or lower field first). As the previous post stated it’s easier if you can deliver it with fields, if you need progressive some kind of plugin will be necessary (you can always get some result by playing with frame blending or a staight progressive render but it won’t look very nice, after effects will just duplicate frames and motion will look stuttery)
Glenn Stewart
1k studios -
If you’ve got the camera data, and it’s right you now need a way to position the objects you add in AE in the right place in 3d space. Ask the person doing the Boujou to include some of the trackers in the export, if there is something close to where you’re adding the new objects that has been tracked that would be best, otherwise a couple of points on the ground plane or other objects will help you position you’re new objects. I’ve done the same thing using PFMatch and it works well. When you import the camera the other reference points will come in as nulls in 3d space, unfortunately every time I’ve done it all names assigned in the tracker have been lost. Scrub your footage and make sure the nulls stay in the right place relative to the background footage before you waste any time replacing stuff (they should look like they’re stuck on the footage, you might want to paste in a small solid with the same position value as nulls disappear during ram previews).
Best of luck, it’s a tricky thing to do but if you can get it to work it looks really convincing.
One thing to watch out for, a lot of the 3d trackers have an option to export all of the tracking points, it seems like that would be useful for rebuilding the scene in AE but I found it to be pretty unmanageable as there are literally thousands of them and it’s a real pain finding the ones you need.
Also the nulls may come in with keyframes which you don’t need if the object they come from doesn’t move within the scene, for most points the camera should do all of the movement.
Hope I haven’t confused you even more,
Glenn Stewart
1k studios -
We use 960×540 for Standard Def ntsc, and 1024×576 for pal.
Glenn Stewart
1k studios -
That’s a tough one without a bit more information, what was the source? Is the interlacing a result of a video source (if so PAL or NTSC? DV or SD?), 3:2 pulldown or a field render?
I’d imagine if it’s just video you’d be better sticking to one field but you could always do each frame twice, effectively doubling your frame rate. Why didn’t you deinterlace in After Effects before exporting?
I’m just curious, don’t want to sound like I’m jumping on you for asking an impossible question.
Glenn Stewart
1k studios -
I would probably rename the footage files on the mac first and reconnect them before moving it over to the pc. You shouldn’t have to reconnect every layer though, close the project and rename the offending files, when you open it might find them anyway (recently OSX has been uncannily good at finding renamed files, useful in this case but kind of scary too). If they are unconnected double click on one of the sources in the project window (don’t cmd+H) point to the renamed psd and all the layers should reconnect at once.
Best of luck, I have this problem a lot and want to whip our Mac-Bred animators!!
Glenn Stewart
1k Studios
PS if you’re using the collect option make sure all the bins in your project window have legal names too as after effects will mimic that structure when it collects the files. -
With due fairness to Roland I think that tutorial is a bit out of date, the convert audio to keyframes assistant is a much easier way to do this than using motion maths scripts, by default they’re disabled in recent versions of AE aren’t they anyway? I’m not saying they’re of no use but I know I’d do the same thing a lot differently now, I’m sure Roland would say the same.
Glenn Stewart
1k Studios -
Also check out the text effects presets gallery in the help, there are loads of effects there, I find it a good place to look for ideas, although I usually end up rebuilding them from scratch as some of them are too hard to take apart.
Glenn Stewart
1K Studios