Best practices for HV20 footage acquisition?

[znelson32]

I've scoured the forum here but there's so many posts I can't find a simple answer to this question - sorry if it's been covered, if so please post the link.

What is the best practice for acquiring 24P (16:9) footage in Windows (vista) via firewire connection?

I've tried Windows Movie Maker - but the dvr-ms files are pretty useless in anything but WMM. I can capture the footage but I can't convert it to .avi because WMM can't load the file (it crashes). Captures, plays in media player, but can't be loaded by movie maker - great MS software!

I've tried Premiere Pro 2 but I don't know what settings to use to get the best picture quality - video was shot in 24P (cinemode). I used the Premiere "plugin" to support the 24F Canon format - the results were ok but there were big time interlace problems especially in fast-moving scenes. I have no idea how the footage became interlaced considered it was shot in cinemode. So - I would appreciate any help here with getting the settings in Premiere "just right".

Finally - I tried hdvsplit to acquire the footage but this results in badly artifacted mpg's so I don't think this is an option.

I think Premiere is the best but again, I'm not sure exactly how to set everything up. I certainly don't want an extremely huge file that's too bloated to play in real time - ultimately I'd like to store the footage as a tiff sequence.

Many thanks in advance - again sorry if this has been covered previously!

[Erik Bien]

Hi znelson32, and welcome to the forum!

The HV records its PF24 in a 60i stream with pull-downs inserted; you must inverse telecine the footage after capture to remove the pull-downs and get just the 24fps progressive frames (see Frequently Asked Question #11).

[znelson32]

Hi Eric - thanks for the response. I'm somewhat confused by the pulldown but I believe I can do this in AfterEffects CS4 pretty much automatically (one button push), correct?

My original question was more on how to get the footage into the PC via firewire - particularly using Premiere (I assume that's the best option?). I want the best-looking footage (don't we all?) and I'm sure my Premiere settings are probably wrong since I'm getting awful interlace artifacts in fast-moving scenes.

Regards,

Zach

[znelson32]

Sorry, I meant Erik. I wish you could edit your posts..

Anyway - I've found the best (pristine) way of capturing is to record live over firewire and capture using Vegas. Unfortunately, there's no way to modify the capture settings and increase the quality - but Vegas recognizes the HV20 and handles the 24P 16:9 footage perfectly unlike Premiere Pro 2 - maybe Premiere CS4 will handle it too which would be a win-win situation.

Has anyone found a way to prevent the camera from going into standby when it's in camera mode but you don't hit record? Maybe a firmware hack or use an empty tape cassette and press record?

[iThinkergoiMac]

Using an empty tape cassette and pressing record is a bad idea... don't try it. Just turn off powersaving in the menus... I forget exactly where it is, but it's in the system menu somewhere. That will stop it from going into standby mode.

Oh, and you can edit your posts... there's a button that says "edit" next to the "quote" button at the bottom of each of your posts...

[znelson32]

How does this workflow sound:

1. Capture in PP CS4 - HDV 1080P 24fps Sequence via Firewire from HV20 - no tape
2. Import into AE CS4 - interpret and guess 3:2 Pulldown
3. Render from AE as 16bit tiff sequence with stretch to 1920x1080
4. Use the tiff sequence for whatever.. rinse and repeat

This way there's only one compression stage (at step 1) and then another compression at finishing stage.

Now what's odd is I've locked the HV20 in 1/48 shutter 24F mode and when I do the 3:2 pulldown the fps rate changes, everything looks OK - but - when I inspect the frames in Photoshop (after step 4) I can see what looks like a blending between two frames.

For ex., say I do a quick pan across a Sprite can, it looks like there's two S's in the logo instead of a smooth blur between where the S was and where the S ended up. I thought this wouldn't/couldn't happen unless AE is guessing the pulldown cadence incorrectly - a bug?

[Lunchbox]

Your workflow creates more trouble than it should be.

1. You can't capture 60i video with a 24fps timeline. It won't automatic remove pulldown for you. the capture process also has no compression or re-compression involved. All compression is done during the time of recording to tape.

2. AE only "Guess" the pulldown. You shouldn't be surprise to see if it doesn't guess correctly.

3. It is definitely not necessary and a waste of storage and processing power to size the footage from 1440x1080 to 1920x1080. Similiarly, you are sizing up a 3-mega pixel photo to 5-mega pixel before you retouch the photo in photoshop.

4. not sure why you need to do that.