Posted on July 13, 2011 at 15:30 by Kevin Chung in Technology
This is the fourth in a series of five blogposts in which we discuss a new concept/method to preserve the visual quality of video. The series consists of the following topics: Interlacing versus Progressive, Format war, Telecine, NTSC vs PAL and Television manufacturers. The central motto: “simplicity is brilliance”. As said, this week part 4: NTSC vs PAL.
Previous posts in this series:
Part 4: NTSC vs PAL
There’re two distinctive large TV broadcasting standards: NTSC and PAL. Throughout its existence several revisions have been made to adapt for future technology changes such as HD (High Definition) broadcasting. However, when it comes to file exchange, having these two co-existing standards makes things complicated.
All kinds of NTSC-to-PAL and vice versa converter boxes have been introduced. How these boxes achieve that conversion is often not known: they are “black boxes”. Aside from using such converter boxes you can also use software to manually do the conversion. Unfortunately the conversion is never perfect, because of the specifics of both standards.
Resolution
Resolution and frame rate are the variables to consider. If you need to go from NTSC to PAL you need to resize from 720 x 480 to 720 x 576. That’s considered upscaling, which I’m not a big fan of, even it if is only a small amount like in this case. If you pass the Display Aspect Ratio with the new resolution eventually the frame dimensions will be correct.
But when you want to go back to a square resolution you need to take into account Aspect Ratio, Display Aspect Ratio and whether or not your source is letterboxed. This involves some calculations and often some don’t even care and resize it just straight away. That’s why some SD original footage broadcasted over an HD channel shows faces that are stretched out, as shown in the image at the right.
Frame rate
Now onto frame rate. It’s inevitable that conversions between NTSC and PAL will always harm motion in some way. This is because of their incompatible frame rates: 29,97 for NTSC vs 25 fps for PAL. Motion will become jerky. Why did NTSC want to slow down the frame rate by 0,1%, thus creating a fractional frame rate? It’s also very confusing to imagine how 29,97 frames per second are played. So the 30th frame in that one second is not shown entirely? In fact it’s different: there are 30 frames that are played at a speed of 29,97 frames per second.
Also for NTSC and PAL regarding their SD broadcast variants, anamorphic resolution is used, applying Display Aspect Ratio to stretch the video to 4:3 or 16:9 dimensions. Add letterboxing to it and its unnecessarily complicated.
Just use a full frame. The same counts for some HD broadcasters that use 1440 x 1080 as a base resolution and stretch it to 16:9. To me the existence of anamorphic is in some ways the same as upscaling if the playback equipment prefers it that way. An example: 720 x 480 with DAR 4:3 will be shown as 640 x 480 or 720 x 540.
All in all the examples above already show the standards are unnecessarily complicated. And converting between those standards for file exchange is just plain tedious if you want to do it correctly – which not everyone does.
We need just one broadcasting standard. To choose between NTSC or PAL can be unfair since there’s no scenario with only advantages. A better solution would be to define a completely new broadcasting standard which every broadcaster worldwide needs to use and just let go of NTSC and PAL. This would improve file exchange and avoiding many more conversions in the video processing chain.
Because it is just like Bruce Lee once said: “Simplicity is the key to brilliance”
