It feels like a lot of people are confused about how Pixel Aspect Ratio affects their image. So if you're wondering why your widescreen footage comes out looking squeezed when it's rendered, read here.
First of all, Pixel Aspect Ratio (PAR) is NOT the same thing as Aspect Ratio. Aspect Ratio (like 16:9 or 4:3) is the ratio between the number of pixels within the image, i.e. 720 x 480 or 1920 x 1080. PAR is the ratio between the width and length of each pixel that make up that image i.e 1.33 or 1.0.
When we talk about widescreen or fullscreen footage, we really have to define what PAR we're talking about. For example, HDV footage is 1440 x 1080 with a PAR of 1.33. Well, everybody knows that the footage is supposed to be widescreen (16:9 or 1.78), but if you take 1440 and divide it by 1080, you don't get 1.78, but 1.33 (4:3), and if you try to view the footage, it'll look squeezed. This is because the pixels are not square, so you can't just divide the aspect ratio. 1920 x 1080 footage with a square (1.0) pixel however will give you 1.78 or 16:9 footage. So you really have to check what PAR you're working with.
Here's another example, NTSC SD video is 720 x 480, but both the widescreen and full screen version works at this resolution. Just try shooting something with your SD camera in 4:3 and 16:9. The difference in this case is again, PAR. SD 16:9 has a PAR of 1.2; SD 4:3 has a PAR of .9.
So as you can see, most camera uses PAR, both in SD and HD. Where this comes into play is that certain applications like Quicktime Pro and places like Youtube don't recognize anamorphic pixels. They just assumed everything is square pixels and so you end up with squeezed video. So you have to convert your video with anamorphic (non-squared) pixel into squared pixel. Here's how
Let's say I have an SD video shot at 720 x 480 with a PAR of 1.2. Now I want to compress a 16:9 video with square pixel. Simply take the ratio that I'm trying to achieve (16:9 or 1.78) and divide the length of the video
720/1.78=405
So I would render out the video at a resolution of 720 x 405 with square pixel and it'll be 16:9 no matter which player you used. The deal here is after you set it to square pixel, you're trying to keep the ratio of the width and height of the picture to whatever the ratio is that you're trying to achieve (16:9 in this case). If I want to shrink the video a bit for example I would use 640 x 360 because 640/360 is also 1.78.
Hopefully this will clear up some confusions.
First of all, Pixel Aspect Ratio (PAR) is NOT the same thing as Aspect Ratio. Aspect Ratio (like 16:9 or 4:3) is the ratio between the number of pixels within the image, i.e. 720 x 480 or 1920 x 1080. PAR is the ratio between the width and length of each pixel that make up that image i.e 1.33 or 1.0.
When we talk about widescreen or fullscreen footage, we really have to define what PAR we're talking about. For example, HDV footage is 1440 x 1080 with a PAR of 1.33. Well, everybody knows that the footage is supposed to be widescreen (16:9 or 1.78), but if you take 1440 and divide it by 1080, you don't get 1.78, but 1.33 (4:3), and if you try to view the footage, it'll look squeezed. This is because the pixels are not square, so you can't just divide the aspect ratio. 1920 x 1080 footage with a square (1.0) pixel however will give you 1.78 or 16:9 footage. So you really have to check what PAR you're working with.
Here's another example, NTSC SD video is 720 x 480, but both the widescreen and full screen version works at this resolution. Just try shooting something with your SD camera in 4:3 and 16:9. The difference in this case is again, PAR. SD 16:9 has a PAR of 1.2; SD 4:3 has a PAR of .9.
So as you can see, most camera uses PAR, both in SD and HD. Where this comes into play is that certain applications like Quicktime Pro and places like Youtube don't recognize anamorphic pixels. They just assumed everything is square pixels and so you end up with squeezed video. So you have to convert your video with anamorphic (non-squared) pixel into squared pixel. Here's how
Let's say I have an SD video shot at 720 x 480 with a PAR of 1.2. Now I want to compress a 16:9 video with square pixel. Simply take the ratio that I'm trying to achieve (16:9 or 1.78) and divide the length of the video
720/1.78=405
So I would render out the video at a resolution of 720 x 405 with square pixel and it'll be 16:9 no matter which player you used. The deal here is after you set it to square pixel, you're trying to keep the ratio of the width and height of the picture to whatever the ratio is that you're trying to achieve (16:9 in this case). If I want to shrink the video a bit for example I would use 640 x 360 because 640/360 is also 1.78.
Hopefully this will clear up some confusions.
