Frame grab workflow from the Sony XDCAM EX-1

From my last post, Matt Dial and Peter Houppi asked to see some frame grabs from my new Sony XDCAM EX-1.
On Sunday I shot some video of fall color and my daughter, freckles and all, Brenna. The day was cloudy and the light was low contrast.

Here was my workflow:

Transferred the clips into Final Cut Pro 6’s browser via “Log and Transfer.” It took only about 20 seconds to convert two minutes of video.

Loaded a clip into the viewer. Set an in and out point, then dropped the clip onto my timeline.

Navigated to the frame I wanted, then made a “freeze frame” by going to>Modify>Make Freeze Frame.

The freeze frame automatically loads into the viewer. I placed it on the timeline and set an in and out point. I double clicked the still frame clip to load back into the viewer. Then went to: Sequence>Render all.  Make sure “Full” is checked
Render out the clip (Command R)

Now export the freeze frame

Go to File>Export>Using Quicktime Conversion
Under format use “Still”
Under “Options” use “Photoshop”

Click “Options” and use “Best Depth” (I’m guessing here. There is a “millions of color” option, but I went with the default Best Depth.)

Name the file and export to your desktop.

Open in Photoshop and work the file like any digital still photo. The photo open as 5.93 meg file.

Some things I’ve noticed:

In Quicktime Conversion, don’t set the export to jpg. It adds compression jaggies to the frame grab.

Adding a light unsharp mask to the frame grab really brings out the detail.

I tried using David Leeson’s Voodoo Tool with ok results, but found my method looked as good without the upsizing to a 67 meg file.

If anybody else has suggestions for getting the best out of a fame grab do share…

5 thoughts on “Frame grab workflow from the Sony XDCAM EX-1

  1. Awesome to see you working with the EX-1 Colin! Brooks tapped me to figure these out when we got some this summer and I’ve been in love with it since!

    For frame grabs I use the same system, except skip the Freeze Frame – there’s no need. Just jog to the frame you want, then use the Export Using QT Conversion > Photoshop (or TIFF) and you’re good.

    I’ve used this over the “Voodoo Tool” since I started working with HDV. Leeson’s system has issues, and this way you get the actual original image size instead of your monitor’s best resolution – which is great on laptops.

    Sorry to hear they’ve cut back your video work. You’re not the only one. I just talked to Manny Crisostomo at the Sac Bee, and he just got demoted(?) to Senior Photographer again. They’re down to 1 video shooter there. 😦 Tough times.

    Love reading the blog! It’s been a while since I saw you last. Hopefully we’ll cross paths sometime soon!

    Take care.

  2. Thanks for posting these, Colin. The images look great. We’ve been using the screen capture -> Voodoo workflow with decent success, mostly for the simplicity. I’ll try exporting as Photoshop or Tiff as you and Aaron have suggested and try to compare the results. I agree that exporting as a jpeg is no good.

  3. The one thing I forgot to mention is that I shot the video at 30p 1920 X 1080 resolution. Shooting in progressive mode means no more having to de-interlace the video.

  4. Thanks Aaron.

    I tried it your way today and it worked a stated–though I didn’t think the frame was as sharp. My next question would be: Should I do a “full” render of the clip before exporting using “Quicktime conversion?” Does that full render happen during export, or do you still need to do it before exporting?

  5. Colin — just got back around here.. Thanks for posting these. I noticed just recently that our local FOX affiliate is shooting with the Sony XD Cam — I believe the version above the EX-1. It’s a bit strange to see a local TV station using the smaller cameras now. They aren’t too thrilled about it, mostly ego and being used to having heavier/sturdier equipment.

Comments are closed.