RSS feedIt’s hard enough living outside the U.S. if you’re trying to do anything technology based, let alone trying to deal with various video standards in use around the world. This is a quick “how to” on how to capture video from a PAL DV camera using Final Cut Pro 3 (FCP). The detail comes first, so skip to the end if you just want the FCP settings.
NTSC is the colour system used in Japan and the U.S. We won’t mention it again, except to say that the vertical resolution is less than PAL, and there are more colours than PAL, albeit less accurate colour reproduction due to it being analogue.
PAL is the colour system used in most of Europe and Australia, and it specifies the way colour information is stored with the video signal. The base monochrome (black and white) signal however is a system called CCIR, which PAL simply adds colour information to. SECAM is another colour system used in parts of Europe, which also uses the monochromatic CCIR system as a base.
Because PAL is just colour information, when we talk about resolution, we’re not talking about PAL, we’re talking about CCIR or PAL CCIR as it is sometimes known. But it gets worse still. CCIR has a vertical resolution of 625 lines, and leaves the horizcontal resolution up to the display device. This is because television is vertically interlaced. Every time the screen is refreshed, it updates the alternate scan lines. When television was invented, the phosphor technology was too slow to refresh the entire screen in one go, making the image flicker as earlier pixels faded out before the complete screen was displayed, so it only updates every other line on each refresh. Horizontal information is just a sequence of voltages, and it is the vertical complexity that forms most of the CCIR standard, hence the focus on vertical resolution, not horizontal.
So in the case of standard TV, the aspect ratio is 4:3, which gives 833 pixels horizontal. All new numbers you don’t often read about with video, but then the physical resolution of a TV bares little resemblence to video output resolution.
The trick for computer based video, is to ignore television (oh, I love that line, it works on so many levels) and stick with the aspect ratios. My Sony TRV15E PAL (according to the manual), spits out PAL CCIR, which is 625 lines. But all I really care about is the aspect ratio, because most of the time I’m trying to generate QuickTime movies of 240 vertical resolution. So taking 240, applying 4:3, we get a final resolution of 320 x 240, and you need to apply this formula to any vertical resolution you want to bring the video down from the camera in.
Don’t be tempted to bring the video down as full CCIR 625 lines or the down resolution of 720 x 576 specified in FCP, and then only convert to 320 x 240 at the final export stage. If you do that, any text overlays you have will be scaled, and you’ll lose quality, not to mention subtly affecting the aspect ratio. Always apply text to the final resolution video. No exceptions.
So on to FCP. When you create a new project, FCP uses the current “Audio/Video Settings” in the application menu, to decide how to capture video from your camera. The bottom two settings should be permanently set at FireWare PAL (or as appropriate) for Device Control Preset, and Apple FireWire PAL (720 x 576) for External Video.
The Capture Preset is the key setting, and should be set to a resolution of 320 x 240, or whatever your required 4:3 calculated resolution is. Click on the Capture Presets tab to create a new one, select DV PAL 48 kHz in the list, and click on “Duplicate…”. Change the name and description to include your resolution (e.g. DV PAL 320×240). Now, ignore what FCP tells you about PAL standards, and change the Aspect Ratio drop down to “Custom” first, and only then put 320 and 240 into the Width and Height boxes. Click on OK to save your new Capture Preset. You should now be able to log and capture PAL video at the correct aspect ratio. Note that whatever you put in the Width and Height boxes, will be the size FCP captures at. So if you have say 50 x 200, then you’ll capture your video as a thin vertical bar on the screen, and the video will be scaled to fit. Also, we’ve captured using square pixels, not typical rectangular CCIR PAL pixels, which means our video is no longer technically CCIR, which is important for the next step. Bottom line is: Capture preset = video is scaled to specified size and aspect ratio, using square pixels.
You can now use your 320 x 240 capture in whatever projects you wish, but most likely you just want to spit out a QuickTime of the same resolution, which means creating a Sequence Preset. Click on the Sequence Presets tab, click on DV PAL 48 kHz Superwhite and click “Duplicate…”. The Frame Size is probably defaulted to 720 x 576 CCIR 601, which is actually a 5:4 aspect ratio, not the square pixel format that our captrues are in. So ignore what FCP says, and change the Aspect Ratio drop down to “Custom”, then enter 320 and 240 into Width and Height. Then change the Pixel Aspect Ratio from “PAL – CCIR 601” to “Square”. Finally change the name and description to include a reference to PAL and 320 x 240, and click on OK. In both dialogs, there’s a “Multimedia Large (4:3)” item in the Aspect Ratio drop down, which you can use instead of the “Custom” settings if you wish.
As a final step, in the “Audio/Video Settings” dialog, make sure your two 320 x 240 presets are selected for Sequence Preset and Capture Preset, then click on “Create Easy Setup…”. Enter a name and description, such as “Videoblogging PAL 320 x 240”, and click “Create…”. Next you’ll get a file save dialog, so just keep the folder FCP chose for you, and enter a file name for the setup, such as again “Videoblogging PAL 320 x 240”. Click on Save. You can now select all the relevant 320 x 240 settings at any time by going to the “Easy Setup” menu item in the application menu, and selecting the setup you want from the drop down.
Any questions? Just post them here, or drop me an email. A lot of this may well be relevent for NTSC as well. Anyone know?
Comments (6)