DV Info Net

DV Info Net (https://www.dvinfo.net/forum/)
-   Final Cut Suite (https://www.dvinfo.net/forum/final-cut-suite/)
-   -   emailing/ftp raw video for editing (https://www.dvinfo.net/forum/final-cut-suite/482361-emailing-ftp-raw-video-editing.html)

Scott Routt July 24th, 2010 10:34 PM

emailing/ftp raw video for editing
 
My customer is shooting himself in front of a green screen and emailing me the raw footage for chroma-keying and additional editing. His clips are about 10 minutes long. Right now he's using QuickTime Pro to make a 480p movie and that's what he sends to me because of course its smaller than the original file.

Can anyone think of a better way to compress it for emailing and editing?
Thanks

Perrone Ford July 25th, 2010 06:00 AM

Email won't support files that size. You'll need to find another way.

Jase Tanner July 25th, 2010 08:54 AM

yousendit.com $5/month for files up to 2 GB

R Geoff Baker July 25th, 2010 11:59 AM

I'm confused -- no where have you told us what you are compressing to, what you are compressing from, what codec you are trying, what data rate you are using ... you've told us what tool you're using to compress but nothing else.

Maybe with more information we can offer advice ..?

Cheers,
GB

Perrone Ford July 25th, 2010 07:59 PM

Quote:

Originally Posted by R Geoff Baker (Post 1551881)
I'm confused -- no where have you told us what you are compressing to, what you are compressing from, what codec you are trying, what data rate you are using ... you've told us what tool you're using to compress but nothing else.

Maybe with more information we can offer advice ..?

Cheers,
GB

Frankly, any codec he'd be able to pull a decent key from will be too big to email more than a few seconds.

Jordan Hooper July 26th, 2010 02:54 AM

Not Compatible
 
Email isn't really the best way to share files of video size. For the purposes of keying you want to avoid compression. I use Dropbox to send my clients YouTube sized files, but that's final projects output to H264. If you live in the same city go there with an external drive. Otherwise the answer is "no" there isn't a suitable way to compress the files you're talking about.

Dave Partington July 27th, 2010 04:54 PM

Pulling a good key really means working with the original footage with as many pixels as possible.

Forget e-mail.....

Scott Routt July 29th, 2010 05:49 AM

To clarify my original post about emailing: I am having my talent use Cutesendit for file transport. My account will let him send up to 2gb. However, since he doesn't have the fastest internet connection, we try to reduce the file size. 480p does that for me, and its not bad. I'm getting a pretty good key, but as you can imagine, its not straight from tape quality. My finished output loses some depth and looks a little flat.

95% of my finished work is for the web. I'm getting some very quality by putting my .mov file through Compressor twice. First by compressing to ProRes and doing the deinterlacing there. And then taking that ProRes file and Compressing it with H.264. I made custom Compressor settings calling the ProRes (wash) and the h.264 (dry). So I wash and dry my movie files in Compressor. Its a bit of work but I'm real happy with the result.

I can't help but hope and feel there there's a way to give it some quality compression on my talent's end before he sends it to me. Size example. A two minute movie file is 440 Mb. A QuickTime 480p version of the same footage is 30 Mb. That's a big difference when you're using a service like cutesendit to transport it. I'm just trying to see if there's a better way than 480p to get the footage to me.

Perrone Ford July 29th, 2010 06:20 AM

Please explain this workflow:

1. What camera is your client shooting with (and at what bitrate)
2. Am I to understand that the client compresses the camera original before sending to you? And if so, what codec are they sending to you? Telling us 480p is not giving us any useful information
3. When you receive the files you are doing a deinterlace and compress to ProRes? At what stage does the keying happen?
4. You are doing a conversion to H.264 as a final for web delivery? Is this correct?

And a couple of questions:

Why is your client shooting interlaced? That's the opposite of what you want.

Would it not be possible to ship a drive back and forth between you? That would solve half the problem.

What software are you using to pull this key?

What budget do you have to throw at this to improve what you are doing?

Luc Nutter July 29th, 2010 09:38 AM

Don't Downgrade Quality Ever
 
From past experience, there's never a good reason to downgrade footage that still has to be edited, let alone doing it with footage to be keyed....

Have your client use a file transfer service such as Send Large Files for Free - Large File Sending for Free. There is a 2 GB limit and it's free. If his internet isn't fast enough, tell him to put it on a flash drive, head to his neighbors house or a local internet cafe where the internet isn't as slow, and upload it there. This is still in my opinion the best way to go.

Cheers.

Scott Routt July 30th, 2010 09:17 AM

Quote:

Originally Posted by Perrone Ford (Post 1553240)
Please explain this workflow:

1. What camera is your client shooting with (and at what bitrate)
2. Am I to understand that the client compresses the camera original before sending to you? And if so, what codec are they sending to you? Telling us 480p is not giving us any useful information
3. When you receive the files you are doing a deinterlace and compress to ProRes? At what stage does the keying happen?
4. You are doing a conversion to H.264 as a final for web delivery? Is this correct?

And a couple of questions:

Why is your client shooting interlaced? That's the opposite of what you want.

Would it not be possible to ship a drive back and forth between you? That would solve half the problem.

What software are you using to pull this key?

What budget do you have to throw at this to improve what you are doing?

1) Sony VX2100. I don't remember the settings and apparently I need to educate myself on your question about the bit rate. I set up the camera on 4:3 standard. Put it behind a teleprompter. Manual focused it for his distance. White balanced and adjusted the exposure based on the lighting set up. And I white balanced it.

2) 480p. My talent isn't an editor. We gave him an iMac and told him to open up the raw footage in QuickTime Pro and save the movie as HD 480p (all automatic based on QuickTime Pro default settings)

3) When I receive the file, I put it in FCP timeline, render it and color correct it and export it out as a .mov. I then bring that mov back into FCP and key it and build the video show.

4) Final delivery is h.264 for web.

Client shooting interlaced? (I've actually never thought of this)

Driving to transfer tapes? That would be an hour of driving each way and I would need files or tapes daily. The idea is to cut down on expenses and time by having no one get in their car.

Keying: I alternate between the Final Cut Studio 2 built in key filter and BCC's color key filter. Sometimes one works better than the other depending on the raw quality of the video. Regardless, I use an Hchroma blur effect after the key to reduced teh jaggedness.

Budget? That's funny. My profit on doing a 3 minute video such as this is $200 which I split with my partner. Hopefully we'll get a lot of videos and will get the funds to pay for equipment upgrades. Our end goal is to get a Tricaster and have the talent drive up and read 15 scripts by teleprompter and edit and chroma on the fly. But until we get there, I'm trying to do the best with what I have.

Most of the suggestions I'm getting on this thread recommend cutting the file in half and FTP'ing them in full quality. This might be what I have to do. What I'm doing now is working and working fast, but I'm not seriously thrilled with my final output.

Perrone Ford July 30th, 2010 09:41 AM

Ok,

Give me a little while. This answer is going to be somewhat involved because there is a LOT going on here that should be better understood by you and your client for best results. I'm in the middle of something right now but will update shortly.

Ok... here we go:

1. The Sony VX2100 is a miniDV camcorder. Do not concern yourself about the bit rate. All miniDV cameras shoot at the same 25Mbps bit rate.

2. I can find no evidence that the VX2100 has a progressive mode. Since it is interlaced, that video should REMAIN interlaced until your receive it. Deinterlacing done poorly will reduce quality quickly. This step needs to be done with care. Telling him to save it as 480p is a mistake.

3. What kind of timeline are you creating for this project? And when you say you "render it" what are you rendering it to? ProRes? Additionally, why are you then exporting as a .MOV (and what kind) and then bringing it back into FCP? That workflow just doesn't make a lot of sense to me, but maybe I am missing something obvious.

4. Final delivery seems reasonable, but we can get into that later.


Yes, the client is shooting interlaced from that camera, unless I've missed a progressive setting on the VX2100.

Understand about the difficulty of trying to drive to deliver media. Maybe there is a shuttle service you could use. If not, the method you have now is going to have to suffice I guess.

In regards to keying, sounds like you have reasonable tools, but you need to get the media in better condition first.

$200 for 3 minute videos is a good rate assuming you can get the workflow down to something that makes sense. You're making your job much harder than it should be it sounds like.

It sounds like your "working fast" is seriously damaging the quality of your input files. Not sure about the output.

R Geoff Baker July 30th, 2010 01:42 PM

You are shooting DV -- that is all your camcorder is capable of. The frame size is fixed, as is the bit rate. No options to worry about there. You should keep your footage in its original form, and you should build a timeline that is DV (and therefore SD). You should not concern yourself with deinterlacing until the very last export step to your final deliverable.

DV files are about 1 gigabyte for every five minutes of material. It doesn't matter if it is a DV.mov file, a DV.avi file or a DV.dv file -- all are the same material, the same bit rate, the original quality ... they are just 'wrapped' in differing computer formats. Have your subject transfer via Firewire to their HDD as a QT DV.mov file if that is most convenient. Upload the resulting file to one of the many on-line file options, the time to upload & download will vary according to your service. My personal home service transfers at roughly 12Mb/s -- about 1/2 real time, so a five minute file takes about ten minutes to download. However, like many services my upload speed is quite a bit slower -- it takes nearer ten times real time to upload, so a five minute file will take roughly fifty minutes to upload.

Keep it simple. Stick to the original format. Do the math to calculate transfer times based on your service access.

HTH

GB

Nigel Barker July 30th, 2010 11:13 PM

The Sony VX2100 records to miniDV tapes. Why don't you have the customer post the tapes to you? If time is of the essence then use FedEx or similar.

Scott Routt August 2nd, 2010 08:06 AM

Perone: "3. What kind of timeline are you creating for this project? And when you say you "render it" what are you rendering it to? ProRes? Additionally, why are you then exporting as a .MOV (and what kind) and then bringing it back into FCP? That workflow just doesn't make a lot of sense to me, but maybe I am missing something obvious."

Answer: My timeline is the default 4:3 FCP sequence.

The rendering situation: Keep in mind my set-up. Broadcast monitor is connected to my camera deck. When I go to load a 480p file to my sequence, FCP asks if I want to change the format to fit the sequence. If I say yes and then render, the movie will play in the canvas but I only get a still picture on my broadcast monitor- the movie doesn't move. If I tell FCP not to change the footage to fit the sequence, it will play on my broadcast monitor after rendering.

So why do I export it out as QTmovie? I used to work with FCP2 (non-intel) on a non-intel mac mini. Even though it seemed like more work, it created less future renders if I did all my color correction and exported the color corrected movie as its own file. Otherwise, every time I added text or other elements, I'd have to re-render the color corrections again. So maybe I should quit doing that now that I'm using Studio 2 intel.

Geoff: "Have your subject transfer via Firewire to their HDD as a QT DV.mov file if that is most convenient."

Answer/Question: Are you saying my client should export as a Final Cut Pro Movie? Or are you saying to do something like...

export using QuickTime Conversion
Format= DV Stream
Use DV NTSC 48khz
Options: DV Format = DV
Scan mode = Progressive

Nigel: You asked why not use the post office?

Answer: Because I have a partner who thinks its a selling feature that we can turn around a request for a spokesperson type of movie the same day it is ordered. I'm the only partner who is obsessed with final quality.

And to all of you who are helping me on this thread- THANKS!!! I've been doing this almost 10 years and the only way I learn is through my own mistakes and solutions and the information I get here.


All times are GMT -6. The time now is 05:43 AM.

DV Info Net -- Real Names, Real People, Real Info!
1998-2024 The Digital Video Information Network