From: "Tim" <ignored_mailbox@xxxxxxxxxxxx>
Sent: Sunday, 2009, March 15 08:00
Tim:
I'd suggest getting a HDD+DVD recorder, record to hard drive, break the
recording into chapters after filming, then burn off a DVD. It's then
an easy job to replicate that DVD, either burning off another copy from
the hard drive on the same unit, or copying the DVD on computer.
Bill Davidsen:
You can record directly to the computer and avoid one layer of copying,
with
possible interface issues. Then break up the big file to as many parts as
needed, and edit out the bits you don't want to make public.
With a hard drive / DVD recorder combination, the same sort of thing
applies. Record to drive, edit out stuff, make disc. Only one
generation is used. The system cuts apart MPEG streams, it doesn't
re-render unless you change resolutions.
Though, with a computer, it is possible to lose more than one
generation. If you transcode from one digital format to another (any
change, resolution, bitrate, format, etc.), you lose one there. You
probably will, as the bitrate from a camera is generally higher than
you'll use on DVD, so there's often a re-rendering at some point. And
if you have to render a transition (fades, wipes, etc.), you lose one
there, too. The hint, there, is stick to straight cuts.
Another truly painful thing about editing on a computer is that the
computer screen is generally different than a video monitor. It has a
slightly different frame rate, different gamma, etc. That's a major
problem, here, where video is 50 Hz and computers are usually 60 Hz and
higher. You're never quite sure if jerks and flickers are just your
preview, or how the end result will actually be. Likewise, if you mess
around with video levels, the computer monitor is giving you a wrong
preview. You'd need to learn how to read waveform monitors to get that
right, or simply leave such things alone (take it as read that your
monitor is wrong, the video is fine, don't tweak it).
You even have problems when recording if you have a little Logitech
style video camera and a microphone to the computer input. The audio
will drift away from the video causing loss of lip sync with remarkable
rapidity. 48kbps is pretty much standard for the video industry. So for
broadcast use and 50 fields per second you need exactly 960 audio samples
for each of those 50 fields. For NTSC it gets more interesting. You have
60/1.001 fields per second. So for every five fields you must have exactly
801 samples for four fields and 800 samples for the fifth. I suppose this
can be managed by sophisticated editing tools that notice the error and
resample the audio.
Cheap can be done. But it requires a thorough understanding of the issues
involved. (And the "stuff" I work with is decidedly not what a Fedora user
would consider "cheap" by any stretch of the imagination.)
In the trade "no budget" doesn't mean zero money, it probably means you
have little money, and that you don't have a preallocated budget.
Putting a DVD recorder, even a HDD+DVD recorder, into a place is cheaper
than putting a computer there. The rest you need regardless (camera,
microphones, whatever else).
And you may come up with a superior product to a REALLY cheap Linux
machine setup. {^_-}
In particular the benefit of a separate CD or DVD burner escapes me,
since there are easy to use tools for creating anything from VCD,
thru SVCD to DVD, and audio CD formats are easily created.
The benefit of standalone video gear is generally that it "just works".
There's no fiddling around trying to figure out what to do with the
software, which of the gazillion options you need or should never use.
No time consuming importing and rendering.
Record to hard drive - real time. Snip out the waiting around and
goofs, maybe half an hour if you only have a few to deal with (find them
at high speed, few minutes tweaking at normal speed). Burn your edit
from HDD to disc, ten to fifteen minutes.
Use a firewire connection from a real camera with microphones attached.
Or use SDI video connections with embeded or AES/EBU digital audio that
has been properly synchronized with the video at the source. Otherwise
you'll look like the crumby YouTube videos with unsynchronized audio and
video.
My experience, as someone who does editing far faster than anyone else
than I've watched, is that the computer adds an awful lot of time in
editing (it's cumbersome without a decent edit controller), and so does
rendering (it's not uncommmon for rendering to take half the time of the
final production).
But, if you're /that good/ at producing a recording, in the first place,
that you don't need to edit. And you're an organisation that sells off
quite a few copies of your recordings. Having a bank of five DVD
recorders patched up to your camera means that you can record live, and
distribute several discs straight away. I've seen churches that do that
sort of thing. The operators need to know little more than how to start
and stop recording.
That's a good way to capture your edit source, too, if you record 420
or better. But that uses a lot of disk fast if you are trying to do HD.
Of course, HD is not "zero budget" so I presume you're using SD. For
NTSC that bogeys out to about 2.5 megabytes an hour for 422 video plus
some modest amount for the audio.
{^_^} Joanne
--
fedora-list mailing list
fedora-list@xxxxxxxxxx
To unsubscribe: https://www.redhat.com/mailman/listinfo/fedora-list
Guidelines: http://fedoraproject.org/wiki/Communicate/MailingListGuidelines
