[Accountneedsrealnameupdate]

I think you might want to turn off the proof colours setting and use an adjustment layer to match your standard def output to what your seeing with the proof colours on. I really recommend Mark Christiansen’s After Effects in production book, he covers this pretty well but our workflow doesn’t need it yet so I’m not completely up on it. If you can’t get your hands on a copy let me know and I’ll try and summarize it for you somehow. Great book though, do yourself a favour and order a copy.
Glennser

[Accountneedsrealnameupdate]

It’s a pretty cool effect, I don’t know if Aharon minds others breaking down his projects for him but I’d say it’s just a matter of arranging the circles and parenting to the null, you can even see his key frames in the tutorial, scaling the null moves the circles in and out while rotation spins them around. The circles scale too, I’d suggest tying them to an expression controller so one set of keyframes controls them all. The colours could be coming from the red solid with a couple of four colour gradients and preserve opacity checked so it only shows over the circles.
Good work Aharon, hope you don’t mind me answering this question.
Glennser

[Accountneedsrealnameupdate]

I’d definitely recommend rendering lossless (or close to it) and then compressing in a second step. All of the better compressions use interframe compression which after effects does not do a good job of at all, I’ve seen huge differences in filesize and quality rendering to Sorenson (for example) from After Effects compared with compressing later in procoder or squeeze with the same settings. Especially when using two passes in the compressing software, ideally the codec should know not just what was in the last frames but also what is to come to maximize efficiency, after effects just isn’t set up to do that right.
Glennser

[Accountneedsrealnameupdate]

I noticed this start about six months ago, it seems to be related to the dimensions of the quicktime, HD sizes seem to suffer more than SD sizes.
It does suck as it’s one of our favourite codecs too.
Glenn Stewart
1k studios

[Accountneedsrealnameupdate]

I’m not sure about your workflow but we interpret the 30fps CG to 29.97fps not the other way around. If we’re giving the CG guy video to match to we change it to 30fps, he matches his stuff to that and we convert his renders back to 29.97. It sucks that neither 3dsMax or Maya support the real framerates but as far as I know there is no such thing as video at 30fps, no tape format that I know of supports it at least.
Glennser

[Accountneedsrealnameupdate]

I love the Miranda converters but that one list for over $1k here, the more standalone converters I see the more impressed I am by the multibridge, for the price it does so much. Sorry I know that’s probably not what you wanted to hear but if you are still looking for a deembedder I think the Graham Patten AEDM ones are a bit cheaper.
Glennser

[Accountneedsrealnameupdate]

We get this one too, the in point is perfect but somewhere in the first minute it drops a frame and everything else is a frame late. We have 5 bays and all of them have done it at one stage or other, sometimes consistently (in which case we just start the bars a frame early, by the slate it’s correct on tape, forget about insert edits though), sometimes completely random, however long the layoff it’s always exactly one frame off.
Anyone know why this happens? We first noticed it on the first generation decklink cards.
Glennser

[Accountneedsrealnameupdate]

I get where you’re coming from, we usually do everything in 10bit as it’s mostly short form edits but last year we had a project that involved a huge amount of video, we decided to go jpeg with the option of redigitizing 10bit for an “online” if necessary. Anyway we delivered the jpeg as it looked fine, the client came back with complaints about “jpeg compression artifacts” (one of our idiot producers mentioned the workflow to the client… stupid, stupid stupid!!), anyway we did some tests and to cut a long story short in every single case the artifacts were in the source (dBeta), no-one could tell the difference between the jpeg and the 10bit workflow. There are differences if you really look hard, and there is a slight luma shift that may be a problem but I say if you can get away with it go for it, just don’t mention it to your client.
Glennser

PS In our case it was a straight edit with no graphics, compositing or effects, working in a compressed timeline with lot of dissolves and effects, or with footage going through a few generations of exports/recompression could start to stress the codec past usability, I’m not sure. Also 75% is the sweet spot for the photojpeg codec and decklink, I’ve heard some nonsense about it being YUV at 75% and RGB at other values which doesn’t make much sense but below 75% it starts to break up and going higher just increases file size with very little benefit.

[Accountneedsrealnameupdate]

I agree it really sucks, I’ve had some terrible times renaming files by hand to add the extension, I don’t think it’s a decklink problem, just final cut pro doing things really badly.
PS I know I could use automator or some file naming application but does anyone know how to get rid of the “Do you really want to change the extension?” dialogue? On Panther I was able to get rid of it somehow but since Tiger I haven’t been able to and it drives me mental.
Glennser

[Accountneedsrealnameupdate]

When we saw that deck at NAB with the firewire on the back we presumed it was going to work like the DVCProHD deck giving us a native 125 MBps HDCam stream!! It was a real disappointment that it only gives DV25. We just did an HDCam job and it really hurts that you have to grab uncompressed, be batched to DVCProHD to edit but the uncompressed files are huge, it’s not a nice workflow at all. HDSDTi is the native stream apparently (although I’m not 100% and can’t find any references online right now), anyone know if sony are going to give us access to that apart from dubs?
Just my 2c
Glennser
1k Studios