FCP Worflow
 

OK, so I've been doing some research on workflow and here is the understanding that I have come to:

Transcode footage to ProRes through compressor at 1080p 30fps.

Is this correct? It makes sense to me, but I have been reading about color issues that need to be corrected by Neo Scene?

Is there a tremendous gain in quality by transcoding to ProRes HQ?

Why is everyone determined to convert their clips to 24p? Without slowing down the footage won't you get missing frames if you just convert to 24p since their is no pulldown information?

If anyone has a solid workflow solution I would be extremely grateful because I know for a fact that editing H264 footage in FCP is a PITA!

one word.... Cineform

Thanks Ray. I'll look into it and do some more research.

Just curious, what are the benefits of Cineform over ProRes HQ? Also, is it best to convert the footage to 24p before starting the edit process or should you just finish editing, make a final render as ProRes and then convert that to 24p?

Here's a tutorial on the worflow of transcoding to Proress. I like folllowing this guy's blog. He's got some good stuff. Check it out.

tutorial on workflow - Final Cut Prose Episode 2 - Editing Canon EOS 5D Mark II Footage on Vimeo
blog - pro ? active ? ly

And also, I'd love to hear an answer to Michael's question as well. I've been wondering about that for a while.

24p
 

One reason to use 24p is that the rest of the work is in 24p already. This is true with me. I'd love to use this camera for shots, but I am scared of the headaches associated with 30p->24p conversion.

Beyond this, the aperture issue is another big headache.

one more thing
 

The promise of the Nanoflash when used with this camera is also very enticing.

Minimal as far as I understand if all you do is edit on Macs but apparently useful if you want to exchange files to/from Windows PCs. ProRes has only been available since Final Cut Studio 2 was released a couple of years ago so there are seem to be plenty of enthusiasts for Cineform who have been using it on Macs for years who will continue to recommend it despite the fact that FCP now ships with its own high quality intermediate CODEC.

The tutorial on Vimeo is pretty good. However there is an easier way and that is to create a droplet out of compressor. So when you want to convert your media you simply drag and drop the files on to it.

For me I always edit in the frame rate that I am going to be airing.

Sadly this usually means an over-night conversion of footage to whatever frame rate you want. (In my case 25fps) as it can be anything up to 20:1 even on a fast machine like an 8 Core MacPro if you want a good quality conversion.

Staircase, Stairstep, Jaggies on Raw footage - Problem Solved
 

I thought I'd tack this on in this thread. I am now getting into serious analysis now that I own the camera.

My shock was seeing staircasing on footage, and I was just believing that "well people said it's not perfect" with the aliasing and moire, etc.

However, I made this discovery...

Here's an example of the horrid stairstepping I was getting:
http://www.holyzoo.com/content/dslr/...aggies_Yes.jpg

And here's an example of the raw file:
http://www.holyzoo.com/content/dslr/...Jaggies_No.jpg

I finally figured out my error. In FCP, the raw 5DMkII footage has Upper field dominance. And I was assuming it had None (progressive). I had first thought it had to do with transcoding it or not transcoding it, since my transcoded footage worked fine in a progressive timeline.

Anyway, I am planning on transcoding everything as a rule, but I know sometimes I may want to just shortcut and edit the raw footage. And now I know it's critical to get the field dominance right if I'm editing the raw footage in FCP.

-steev

We are mostly creating stuff for the web so stick with the 30fps. As it's mainly real estate we film we also don't generally have a lot of movement so can change the framerate without transcoding (we just get a 'cool' slo-mo look).

There is a new short course on Tutorials - Online Training - lynda.com specifically on using FCP with the Canon 5D Mk II Photography Video Workflow: Final Cut Pro + Canon 5D Mark II - Online Training Library - lynda.com It's a pretty basic introduction to FCP but covers the whole workflow. I have created a FCP Easy Setup for the Canon 5D Mk II as shown in that course. Everything is setup correctly to edit the footage natively e.g. 1920x1080, 30fps, upper field dominance, 44.1KHz audio etc A zip file of the Easy Setup is attached. Just drop the unzipped file into '/Library/Application Support/Final Cut Pro System Support/Custom Settings', restart FCP & then the setup will be available to you.

There are many other excellent FCP training videos available on Lynda.com & you get unlimited access for a month for $25. However if you just wanted to take "Photography Video Workflow: Final Cut Pro + Canon 5D Mark II" with instructor Frank Rohmer then you can sign up for a free 24 hour trial Free Trial Signup

What I would love to find is a quick and easy way to batch add time-of-day timecode to the movies so that they don't all start at 00:00:00:00. Makes for a very colourful timeline if you have dupe detection on. ;-)

Anyone know of anything that can do this? Even if the tool was used to add the TC as it copies the clips from the CF Cards to the HD that would help.

For my two-penneth, I transcode to XDCAM HD which leaves me with files about the same size as the originals and edit in 30p. Then I save out at 30p and 24p when I've finished. The files as sharp and clear and take colour messing (Color or Looks) without problems on my late 2007 MBP.

I did some experiments with some footage early on and thought that the XDCAM stuff looked almost exactly the same as the ProRes (Not HQ) transcoded files which were about 5 times the size of the originals, so I go for the XDCAM and smaller file sizes.

I know a chap who produces great looking stuff on Vimeo from his EX1/Letus and he edits in DCP in XDCAM HD, so if it's good enough for him, then I'm fine with that.

Anybody feel free to comment with problems in my workflow that I may not be aware of, all comments are helpful as you guys tend to know your stuff!

Avey

Nigel, what's the point changing your timeline to those settings? I thought FCP, automatically adjusted all footage to its exact settings, because it's an "open timeline". I could be wrong, but that's what I've always been taught. I'm definitely gonna test out that easy set-up, though for sure.

I thought I'd add to this. I'd say, for quick editing, you don't have to change any settings at all and yes FCP automatically prompts you to let it automatically change it to the proper sequence settings to match the native 5DMkII files.

So I don't have a need for that setting to be in my setups really - the idea is to be quick anyway.

But 2 last points.

A) If I was editing quickly with H.264 sequence settings, I'd be certain to switch the compressor when I'm done editing and ready to export. I surely don't want to export a master back out to H.264, unless that's exactly the format I intend on distributing. I prefer ProRes or DVCPRO HD to export a master - and it's also a LOT faster to export.

B) Again, as has been said many times, editing in H.264 is a chore in FCP. Scrubbing and playback are pretty frustrating even on my 8-core Mac Pro. So it's now a rule to transcode before serious edit-intensive project work.

-steev

I transcode all my H.264 files to proress422(hq) in compressor, then bring them into FCP. Proress looks slightly better, and it just works better overall, like you guys said earlier. In my sequence settings, should I have the field dominance set to "None" when working with proress? and should the audio settings be 48, 44.1? I thought FCP knows best which settings to use. Haha. It would be nice to know if I should change any of that up.

Just watched one of those Lynda.com training tutorials on this and they say top field dominance and 44.1 for the sound.

Avey

On the Lynda tutorial they are working with the native H.264 files straight off the camera. Douglas was talking about files transcoded to ProRes.

I am no great FCP expert (I leave that to the wife:-) but I thought that this was a quick way of ensuring that all the relevant settings were correct. Otherwise what is the point of any Easy Setup preset?

As I said I am no great FCP expert so I was just doing what the trainer on the video told me to. I am assuming that he is a FCP expert. He at least seemed to know what he was talking about.

We have been having little niggly problems with editing the native H.264 files off the camera (stuttering clips & FCP crashing on exporting any file amongst other problems) so switched to using ProRes. This has worked pretty well but even using Qmaster & sharing the work between two dual quad core Mac Pros it still has been taking a hell of a long time to convert.

I was a bit dubious about the benefits of Cineform NeoScene but having downloaded the trial today & done some testing then I have to admit that I am now a convert. NeoScene works just as well as ProRes in FCP (& even iMovie!) but the conversion from native .MOV file to NeoScene is really fast. In fact it's just a bit less than real time which means getting footage into FCP is taking only a little longer than when we were using the the Canon XH-A1 HDV cameras. So for just $129 NeoScene seems worth the money. I haven't yet tested NeoSceneHD but as that costs $499 then ReMaster & First Light will have to be something really special to justify that price tag.

Can anybody here explain to me why progressively shot footage needs to have a top field dominance in FCP? This doesn't seem to make sense. And then if you convert to pro-res and take that footage into FCP now it has no field dominance. What gives?

I would also like to know the logic behind this ^^^