Will Head
By correctly framing your shot while shooting, you can ensure all the action and titles can be seen, no matter on what medium your film is viewed.
Before you start shooting, it’s worth considering on what medium the final edit will be viewed, as that can affect many aspects of the video-making process.
Traditionally, most video output was watched on a TV set, but with the advent of video-sharing websites such as YouTube and Vimeo, more and more content is going straight to the Internet, bypassing television altogether. This creates a problem, as the viewing area of TVs can differ, while it’s fixed for Internet content.
The reason for this dates back to the early days of television. Cathode-ray tubes (CRT), which were used in big bulky sets before plasma and LCD panels became the norm, were difficult to produce to any degree of accuracy initially. This meant that the image on display would differ from set to set. In order to allow for this, manufacturers introduced overscan, which was an area of the display that wasn’t meant to be shown. By allowing the overscan area to bleed off the edge of the screen, all sets appeared to show a full image, even though some wouldn’t show as much of the actual image as others.
To compensate for the differences between various sets, video makers used safe areas to ensure that any action or titles would be visible to all viewers. Two areas are defined – a larger action safe area and a smaller title safe area. Any element that’s essential for the viewer to see should be framed within the action safe area, and any titles or captions should be placed within the title safe area to be sure that they’ll be readable. Typically, the action safe margin is 3.5% of the frame size and the title safe margin 5%. Most editing applications and professional cameras will let you overlay safe area guides on your shot, so you can quickly check what falls within them.
In theory, overscan shouldn’t be a problem for LCD TVs, since the display is made up of an exact number of pixels, so there’s no overscan area. However, in order to accommodate older material that was made for sets with overscan, some models still don’t show the full image unless you adjust them manually. As a result, it’s best to shoot and edit with overscan in mind if there’s a chance the final footage may end up being viewed on a TV. If your video is only ever going to be viewed online, then you can place items as close to the edge as you like, but unless it’s essential, you’re still probably best sticking within the action safe area.
The final output also makes an impact on whether to shoot your footage in standard (SD) or high definition (HD). Obviously, if you’ve only got SD camera equipment then you’ve got no choice, but most cameras on the market today can shoot in both.
DVD is limited to standard definition, so unless you have a Blu-ray writer, your final output will be limited to SD, too. When it comes to online, it depends where you’re planning on putting the final video. If it’s going to be embedded and hosted locally on a website then standard definition is more than adequate – SD is still larger than most website layouts, so the final video will be scaled down further to fit in the site.
If the intended destination is a video-sharing website then many can show HD footage. Sites such as YouTube and Vimeo offer decent HD quality, but also make it easy to embed a smaller resolution version as well, which takes a lot of the hassle out of the decision making. All you need to do is upload an HD version, and the site will handle processing and delivering the different versions for you.
In terms of production, sticking simply to SD has its advantages – capture files tend to occupy less space and are less demanding to edit, so you don’t need as fast a Mac to process them. If the final destination is DVD or embedded on a website then shooting in SD could be the right choice.
In some cases, you can even shoot in HD and then switch to SD at capture time, which provides some flexibility and future proofing. If you shoot HD but edit in SD, you can go back later and re-capture in HD. If you only shoot in SD then you’re stuck at that resolution.
Many HDV cameras provide a DV output option, even if you’ve actually recorded your footage in high definition. The camera handles downscaling the content on the fly as it’s output, so as far as your Mac is concerned, it’s just attached to a regular DV camera. The only problem you might encounter with this approach is that any errors from tape drop out will be much worse.
Despite having similar names, HDV and DV are quite different compression systems. DV records every frame in full, while HDV is more similar to the compression used on DVDs, where some frames are stored in full and subsequent frames only record the differences until another full frame is captured.
If a problem occurs during recording and all the information isn’t saved to tape – which does happen occasionally – the effect will be greater if you’re recording in HDV. With DV, you’ll only lose the frames in question, but with HDV, you’ll also lose all footage until you come to the next full frame. If you’re using a dual recording setup that captures both to tape and solid-state media, such as the Sony HVR-MRC1K Compact Flash memory recording unit we looked at before, then you can minimise the risk of drop out, since you’re recording two copies of everything you shoot.
If your editing machine is fast enough for HD then it may make sense to process the entire project in high definition and then downscale it to standard definition when you export. Editing in HD and outputting to SD provides additional flexibility over editing just in standard definition. If you’re not happy with the composition of a shot then you can zoom in without losing detail, since you’ll be downscaling the final output anyway. If your footage is 1080 resolution, the frame size will be 1920 x 1080 pixels. This is roughly twice the Pal DV resolution of 1050 x 576 pixels, so you can zoom in slightly less than 200% before losing quality in the final output. If you tried to do the same with SD footage, there would be visible artefacts from zooming the footage in the output.
It’s even possible to pan and zoom shots over time, effectively recreating a manual pan or zoom in the edit. Given the time it takes, as a rule, it’s better to perform these functions while shooting, and if you need to use them to hide an edit or fix the composition, it’s an additional tool at your disposal.
As always, there’s a downside to this technique, which is you can’t create an HD output file from your edit – you’re stuck with SD. To keep your options open in the future, you may want to create a high-definition export of a project, even if it’s currently only destined for SD. If you then need it in HD in the future, then it’s there. Zoom into any of the footage during editing, though, and while the effect won’t be noticeable at standard definition, it will become instantly obvious at full resolution output.
In an ideal world, you’d shoot exactly the footage you need to create the edit you’re after. This rarely works out, though, and various factors mean there are things you need to fix later in the edit. Planning – or finding out from the client in advance – where the final video will be used can mean one less thing to have to worry about later on, and occasionally even provide additional tools to create the best edit possible.
