Dylan Reeve
Forum Replies Created
-
I should have put more detail in the post possibly, it originally included a link to my website where I’ve detailed my thoughts a little more – but that link is no longer there.
Basically if you’re an editor who is familiar with both Avid and FCP, and can choose either for a given job, what qualities of each system would help you chose what to use on that job.
The network editing strengths of Unity are a big one for many users I believe, FCP can technically do the same, but the discipline required to manage that process well is intimidating (that said I’d love to know more detail on the system you’re using Mark).
Similarly the integration with Color could be a big plus on jobs that are going to require complex colour correction.
What I’ve wanted to do for a long time is actually try and setup some head-to-head test between the two to get a somewhat more objective idea of strengths and weaknesses. A number of specific job scenarios, well-appointed suites, competent editors… Put each through it’s paces and try to figure out where it excels and where it struggles. It is difficult in a area full of marketing hype to really know what works really well in any given situation.
But I’ve wanted to do that for the past couple of years and haven’t managed to so far, so I’m not seeking personal experiences.
Here’s an question that I’d be interested in some answers to…
In what features or functions does FCP perform worse than Avid?From my experience, I find rendering slower, and have had less realtime performance in FCP than I’d expect in Avid.
Similarly I find many of the simple geometry (scaling/moving) functions in Avid to be clunky and lacking in finese.
I’d rate Avid’s media management as it’s number one feature of FCP.
What about you guys, more familiar with FCP?
-
Well, FFMpeg now includes native DNxHD support. It could be worth a look to see if that can be used somehow to streamline the conversion process if it has to happen.
-
[Jeremy Garchow] “You can software raid a mirrored set in OSX. No controller necessary.”
Indeed, a software mirror is less likely to leave you high and dry later on. I’m still not a fan of treating a RAID 1 set as two different copies, it’s really not how RAID is designed to be operated.
-
It is very possible there is a delay in the switcher. There are some setups where the switcher introduces a delay (depending on how the M/Es are setup, and what DVE options might be in place). So, if that were the case there’s a chance that it takes the video two-frames to make it from the camera to the record VTR. The ISO are run directly from the camera to the VTR so they don’t have that delay.
At least that’s my guess. I’ve certainly seen that with Digibeta – one of the shows I used to work on was always out by a frame between the line record and ISOs.
-
[Dave Neyman] “It depends on the RAID. I have a Sans Digital RAID 1 system that makes a copy on each drive. I then have a single drive reader which allows me to access the data while the other drive is archived. It is not operating in ‘damaged’ mode.”
If it’s a genuine RAID then I assume you need a special interface to mount it (couldn’t just stick it in the computer or a regular enclosure). And while the RAID controller might not be complaining when you’ve only got run disk mounted it is technically in a failure state if it’s running with only one device.
Also, if your controller/reader dies and you can’t get a replacement that the same then it’s possible the data is unrecoverable (or at least not easily recovered).
-
[Adam Smith] “Not to derail the discussion, but is it possible to import the MXF as normal and at a later date rebuild the project from tape clones?”
Yes, I believe it is, in both Avid and FCP. But there are going to be practical difficulties. It will require that the clips, as well as have a reel number that relates to the tape they were archived to (probably not going to be available until after the original NLE ingest) and it will be important that those archive tapes include accurate timecode cloned from the original sources, and that it not be duplicated on a single tape.
In practice I think this would be a huge pain. Footage would have to be brought into the NLE and then laid out to tape, except it’s going to be really difficult to get the archive tape to match the source timecode (while technically possible in Avid, and probably FCP too, it’s seldom as easy as it seems).
In reality, the easier way, I think, is to layout source footage in ‘reel lengths’ in an NLE and then layback to new source master tapes. Then redigitise those tapes to use an edit source tapes. A better approach again would be using a P2-player to make those dubs.
I know this is a problem a number of program makers face with solid-state acquisition on commissioned show for a few large international cable networks.
-
Be aware, copying a file to a RAID 1 device isn’t really making two copies of it. The drives can’t be used as standalone devices. While it is possible to remove one of the paired drives from the enclosure and use just one, the RAID set it then operating in a ‘damaged’ mode which is not a great idea at all.
Also, the other danger with RAID is that the data on the drives is inaccessible without the RAID controller. While unlikely, it is possible that the enclosure’s RAID controller dies, in which case you need to find an identical controller (sometimes even same firmware revision) to be able to read the data from the drive again. In many enterprise situations RAID controllers are always acquired in pairs, one for use and the other as a spare.
In my opinion, if you want two copies of the media, you should make two stand-alone copies. Something like ShotPut is ideal for that.
-
I wouldn’t think it is possible. The bitrate has more implications than just a metadata header really.
I wouldn’t think that any DNxHD-supporting application would have any real issues with what bitrate the footage was, it’s all the same format essentially.
-
I’m not 100% on this because I haven’t done it for a long long time in FCP, but I believe you’d edit as normal in a 16:9 sequence, and then drag the finished sequence into a 4:3 NTSC DV comp.
I guess I’d then use the crop settings to get the right image height.
It seems a little odd to me though as 720×223 doesn’t seem like it could be 16:9 by any measure I can figure out.
-
I’m in with the others – H.264 is a pretty good delivery codec, but it’s very lossy. I’d certainly never use it for that purpose.