variable framerate recording on a laptop?

[Frank Ladner]

I just had this alternative imaging idea and wanted to see what you guys thought.
(Immediately, I think it isn't doable since it isn't being done.)

Would it be possible to record onto a hard drive or other device (software controlled) at, say, 24 fps from a 29.97 source?

You would basically have your camera switched on, while the software (let's say it's capturing on a laptop) is sampling at a specified framerate (even overcranked, possibly) instead of 29.97.
Obviously, it would be a bulky setup.

Would this be impossible due to shutter synchronization?

,Frank

[Rob Lohman]

That would surely be possible if you write the software. BUT, it
will NOT be any different from a framerate change AFTER you
recorded the original signal. So, it doesn't change anything. It
only does it realtime instead of a two-pass process.

[Frank Ladner]

Rob: I think I see what you're saying. I was envisioning it as a constant stream of video that could have frames extracted at the specified fps, giving the even spatial qualities of that fps. If that makes sense.

So basically, what you're saying is that the capture software would only be seeing the 29.97 fps and could pull the 24fps from just those 29.97 frames each second? If so, then that wouldn't work. It would be just as you stated about doing it in post.

Thanks for the feeback!

[Rob Lohman]

I had the same thought as you a couple of years back. But think
about it carefully. Seeing it as a continues stream of video is not
a bad way of thinking about it. But even that stream has 29.97
INDIVIDUAL frames a second for NTSC or 25 for PAL. There is no
more or no less information in there.

So if I where to "capture" from that stream at 15 fps I would drop
the other 10 or 15 (for ease I'm using this number) fps. If I was
"capturing" at 60 fps (which is only possible in the digital world
through some form of interpolation) I will be creating 35 or 30
fps out of *nothing* (basically).

The point is there just is not more information to work with. The
easiest way to do this is to use an analog capture card that
supports capturing at higher frame rates. But the same rules
apply here as well. There simply is no more information. Yes, you
get a digital file running at 60 fps for example, but there is still
only information in there for 25 or 30 fps.

So your slowmotion what you are creating for example will not
be looking any better unless you come up with some new and
great algorithm for interpolation. But if you can do that, then it's
better to just have the algorithm working as a slowdown filter
in your editor or effects package after the capture has been done.

Slowmotion is the only problem anyway.

[Frank Ladner]

Rob, thanks for that explanation. I see exactly what you're talking about.

Guess I'll stick to using Twixtor for changing fps.

,Frank

[Stephen van Vuuren]

Frank:

I don't know if I would abandon this idea yet. First, check out this thread here.

Real-time capture that bypasses the DV compression is a real possibility.

One option is to find a old broadcast HQ head off ebay etc. and real-time capture higher frame rates etc. that way.

[Rob Lohman]

Stephen has a good point. You could make a higher framerate
device yourself (by perhaps modifying an existing camera or
adapting some other equipment) but this is a very technical
solution and requires lots of time and knowledge to pull off. The
idea you originally had would've been much easier to implement
if it where possible.

As always there are always ways!

[Frank Ladner]

Stephen & Rob: That is a good idea. It's a bit intimmidating to me, though. The time, I have - but not the knowledge.
I know a little electronics but not enough to build imaging devices of any sort. (The most I've ever done to a camera was open my Sony TRV-140 Digital 8 to install a 1/8" microphone jack since it did not have one.)

It would likely never happen, but it's fun to think about building an HD quality variable-framerate camera.

Just out of curiosity, where do Sony, Canon, Panasonic, etc. get their CCDs? ...Or do they manufacture them? And how expensive are these? It's probably not something you could just get a part # for and order one or two, but instead would have to buy in bulk for manufacturing.

Reading the different posts from all the people in these forums, I think there's enough collective knowledge here to come up with a full-fledged camera. But for now, the homemade Mini35 adapters are keeping us all busy enough. :-)

[Richard Mellor]

this guy did it for less than $4,000:
http://home.teleport.com/~gdi/vancecam.htm

I guess you can buy the ccd chip from rockwell for about $500

[Frank Ladner]

Richard: Thanks! I'm reading that right now. This is really interesting.

What about this quote, though:

"There are devices on the market that use an optical relay system to allow use of 35mm lenses on smaller format cameras while preserving the lens' depth of field, but a brief review of basic optics will reveal why this technique can't be considered for serious video-to-film work"

I wonder what specifically he's talking about in the Mini35 (and others) that wouldn't work for video-to-film situations.

[Rob Lohman]

I still find it odd there is no sample footage anywhere on that site.

[Frank Ladner]

I emailed Dan and here's what he said:


"Hi Frank,
Glad you like the VanceCam idea.
I finally figured out how to grab frames in Premiere, so I will be
posting actual frames from some scenes in my movie soon. I still can't
post any motion footage, because my IP only allows me 10MB (!) of space,
but I'm looking for alternative places to post some sequences.
Regarding the Mini35 and other ground-glass imaging systems, they may
look "good" on a TV or monitor, but when blown up for theatrical
exhibition, the image degradation will be really noticeable. Imagine
going up to an ASC Cinematographer on a movie set and suggesting
photographing the image off the ground glass of the Panavision
viewfinder and using that for the final film! My comments (and my whole
web page) is about shooting for eventual transfer to film, and in that
arena, shooting off a ground glass, no matter how good the glass is, is
nowhere near acceptable. In addition to the ground glass, you have more
lenses (the relay system) for the light to pass through, and every lens
degrades the image, so more glass is bad. Also, to get a good image
focusing that close, you really need specialized lenses (flat field) to
minimize distortion. Then there is the "hot spot" problem to deal with.
Again, on a monitor, it may be okay, but in a theater it would be really
annoying. One of the most popular DV to film transfer places
(dvfilm.com) strongly suggests NOT using these adapters, saying the
results are just not going to be satisfactory.
If you're doing things just for TV, it might be okay. But personally,
with all the time, trouble and money it takes to set up even a single
scene for a movie, to then "filter" that hard-earned image through a
ground glass!! I could never do that!
Someone ought to look into a system that uses an aerial image, rather
than a ground glass. Now THAT would be awesome!
Dan "

[Stephen van Vuuren]

<<<-- Originally posted by Frank Ladner : I emailed Dan and here's what he said:
Someone ought to look into a system that uses an aerial image, rather
than a ground glass. Now THAT would be awesome!
Dan " -->>>

aerial image???

[Frank Ladner]

Aerial image -

He basically wants to bypass the glass.

I have heard it mentioned here before, but I don't know how it is possible. I can't get past the idea of having to projecte the image on a diffused, translucent surface.

[Stephen van Vuuren]

Oh, I thought it was a typo. I got you - but I'm not sure how this would be done in front of a camera. Would you still not need some medium for it (fog etc.)?