Camera: Canon Legria FS200 (SD, PAL, Widescreen)
Software: Sony Vegas Movie Studio 9 Platinum Edition
Problem: Widescreen aspect ratios
OK, so here’s what I have learned. Let’s see if I’ve got this right. There are two aspect ratios to be concerned with - the Display Aspect Ratio (DAR) and the Pixel Aspect Ratio (PAR).
Let’s start with the Display Aspect Ratio. In simple terms, we can think of this as the output or playback. Old fashioned CRT televisions have a DAR of 4:3. The newer television format, often described as “widescreen”, has a DAR of 16:9. The first figure in each ratio refers to the width and the second to the height. With 4:3, the width is 1.3333 times the height (4 / 3). With 16:9, the width is 1.7777 times the height. We will ignore others, such as cinema screens, for now. Most (all?) High Definition (HD) material is in widescreen, but by no means is all widescreen material HD. I am attempting to create Standard Definition (SD) output in widescreen, because my camera and therefore my source material is SD, not HD.
Next we have the Pixel Aspect Ratio. This is a little more complicated. When we shoot video on a standard definition camera, it is in either PAL or NTSC formats. These are different standards used in different geographic regions. PAL camera footage is always 720 x 576 pixels and NTSC is always 720 x 480 pixels. Strangely, neither of these ratios is the same as either of the standard display aspects (4:3 or 16:9). PAL, at 720 x 576, has a width of 1.25 times height, close to the 4:3 DAR of 1.3333 and nowhere near the 16:9 ratio of 1.7777. NTSC, at 720 x 480, has a width of 1.5 times height, somewhere in the middle. The reason all this can still work is that individual pixels do not have to be square. If we look at the ratio of an individual pixel (not the PAR, which is the number of rows and columns of pixels in a frame), a “square” pixel would have the same width and height and would have a ratio of 1:1. But pixels can be rectangular – with a different width to their height.
My camera has a widescreen (16:9) display. It has options to record in either 16:9 (widescreen) or 4:3. I have it set to widescreen because my television is widescreen, I will output to DVD, which is widescreen, and upload to You Tube, which is widescreen. I use PAL because where I live (New Zealand) all hardware is PAL and I will send DVDs to family and friends in the UK, where the standard is also PAL.
Now, it took me a while to get my head around this and I hope I’ve got it right. I understand that regardless of whether I set the camera to 4:3 or 16:9, it will always record 720 pixels by 576 pixels. How is that? The answer is – variable pixel widths! I have no idea how that works. Apparently the camera adjusts the pixel width to suit the format.
I believe it is best practice to only ever mess with the width, never the height (something to do with potentially being played back on an interlaced screen versus progressive).
So here’s the thing. I have my raw footage in PAL widescreen 16:9 MPEG 720 x 576. I bring this footage into my Sony Vegas Movie Studio 9 project, which I set as a Widescreen project when I started it, and by default, the aspect is all wrong! The footage appears in 4:3 aspect (with black bars down the sides, because it’s a widescreen project) and all the people look tall and thin. I can fix this in one of two ways. I can right click the event in the timeline and either:
a) select Switches and un-tick “Maintain Aspect Ratio” (which seems an odd thing to do); or
b) select Properties / Media tab, and change the Pixel Aspect Ratio to 1.4568 (PAL DV Widescreen) using an option from the drop-down list.
Now, finally, here are my questions:
1) Why do I have to do this for every clip? Where I have extracted more than one section from a clip and placed them in the timeline, I only have to do this for one event and it affects all events from the same clip. But I have to do it for an event from EACH clip. This is inefficient and it’s easy to miss one, which is frustrating when you find it after you’ve rendered. Why, when the source material is recorded in widescreen and the project is widescreen, do I have to do this at all? Is there a smarter way? Why does Vegas insist on taking widescreen footage and “squishing” it horizontally into 4:3, distorting it in the process, when it could just leave it the heck alone and it would look fine?
2) When changing the pixel aspect ratio on the Properties/Media tab, I would expect to change it to 1.422216 (because 720 x 576 is a ratio of 1.25 and I want it to be 16:9, which is 1.7777 – so divide 1.7777 by 1.25 and I get a scaling factor of 1.422216). But there is no option for this in the drop down list on the Pixel Aspect Ratio box. Using 1.4568 works and I guess I should just be happy with it, but it would be nice to understand.
Now let’s talk about output and rendering. I can render the project in any of several formats (avi, mpg, etc.) and it looks fine. It plays on the PC fine. Upload it to You Tube and it squishes it to 4:3, distorts it and adds black bars at the sides. Through trial and error, I have found that if I render as HD Mpeg using a format of "MainConcept MPEG-2 and a template of HDV 720-25p" (ridiculous, because the source material is not HD), I get a file extension of M2T. My PC won’t play that (even though it can create it), but I understand it’s an Mpeg format, so if I rename the file and change the extension to .mpg, it will play on the PC. I then upload to You Tube and it works! It picks up an inappropriate HD logo, but it works. Is there a better way? Can I render as something more sensible (not HD) and have You Tube accept it for what, to me, it clearly is - i.e. SD widescreen?
I would really appreciate answers to these questions and pointers if I’ve got any of the above incorrect. I now know an awful lot more about video than I ever wanted to, but I’ve started down the journey, so might as well see it through.
Software: Sony Vegas Movie Studio 9 Platinum Edition
Problem: Widescreen aspect ratios
OK, so here’s what I have learned. Let’s see if I’ve got this right. There are two aspect ratios to be concerned with - the Display Aspect Ratio (DAR) and the Pixel Aspect Ratio (PAR).
Let’s start with the Display Aspect Ratio. In simple terms, we can think of this as the output or playback. Old fashioned CRT televisions have a DAR of 4:3. The newer television format, often described as “widescreen”, has a DAR of 16:9. The first figure in each ratio refers to the width and the second to the height. With 4:3, the width is 1.3333 times the height (4 / 3). With 16:9, the width is 1.7777 times the height. We will ignore others, such as cinema screens, for now. Most (all?) High Definition (HD) material is in widescreen, but by no means is all widescreen material HD. I am attempting to create Standard Definition (SD) output in widescreen, because my camera and therefore my source material is SD, not HD.
Next we have the Pixel Aspect Ratio. This is a little more complicated. When we shoot video on a standard definition camera, it is in either PAL or NTSC formats. These are different standards used in different geographic regions. PAL camera footage is always 720 x 576 pixels and NTSC is always 720 x 480 pixels. Strangely, neither of these ratios is the same as either of the standard display aspects (4:3 or 16:9). PAL, at 720 x 576, has a width of 1.25 times height, close to the 4:3 DAR of 1.3333 and nowhere near the 16:9 ratio of 1.7777. NTSC, at 720 x 480, has a width of 1.5 times height, somewhere in the middle. The reason all this can still work is that individual pixels do not have to be square. If we look at the ratio of an individual pixel (not the PAR, which is the number of rows and columns of pixels in a frame), a “square” pixel would have the same width and height and would have a ratio of 1:1. But pixels can be rectangular – with a different width to their height.
My camera has a widescreen (16:9) display. It has options to record in either 16:9 (widescreen) or 4:3. I have it set to widescreen because my television is widescreen, I will output to DVD, which is widescreen, and upload to You Tube, which is widescreen. I use PAL because where I live (New Zealand) all hardware is PAL and I will send DVDs to family and friends in the UK, where the standard is also PAL.
Now, it took me a while to get my head around this and I hope I’ve got it right. I understand that regardless of whether I set the camera to 4:3 or 16:9, it will always record 720 pixels by 576 pixels. How is that? The answer is – variable pixel widths! I have no idea how that works. Apparently the camera adjusts the pixel width to suit the format.
I believe it is best practice to only ever mess with the width, never the height (something to do with potentially being played back on an interlaced screen versus progressive).
So here’s the thing. I have my raw footage in PAL widescreen 16:9 MPEG 720 x 576. I bring this footage into my Sony Vegas Movie Studio 9 project, which I set as a Widescreen project when I started it, and by default, the aspect is all wrong! The footage appears in 4:3 aspect (with black bars down the sides, because it’s a widescreen project) and all the people look tall and thin. I can fix this in one of two ways. I can right click the event in the timeline and either:
a) select Switches and un-tick “Maintain Aspect Ratio” (which seems an odd thing to do); or
b) select Properties / Media tab, and change the Pixel Aspect Ratio to 1.4568 (PAL DV Widescreen) using an option from the drop-down list.
Now, finally, here are my questions:
1) Why do I have to do this for every clip? Where I have extracted more than one section from a clip and placed them in the timeline, I only have to do this for one event and it affects all events from the same clip. But I have to do it for an event from EACH clip. This is inefficient and it’s easy to miss one, which is frustrating when you find it after you’ve rendered. Why, when the source material is recorded in widescreen and the project is widescreen, do I have to do this at all? Is there a smarter way? Why does Vegas insist on taking widescreen footage and “squishing” it horizontally into 4:3, distorting it in the process, when it could just leave it the heck alone and it would look fine?
2) When changing the pixel aspect ratio on the Properties/Media tab, I would expect to change it to 1.422216 (because 720 x 576 is a ratio of 1.25 and I want it to be 16:9, which is 1.7777 – so divide 1.7777 by 1.25 and I get a scaling factor of 1.422216). But there is no option for this in the drop down list on the Pixel Aspect Ratio box. Using 1.4568 works and I guess I should just be happy with it, but it would be nice to understand.
Now let’s talk about output and rendering. I can render the project in any of several formats (avi, mpg, etc.) and it looks fine. It plays on the PC fine. Upload it to You Tube and it squishes it to 4:3, distorts it and adds black bars at the sides. Through trial and error, I have found that if I render as HD Mpeg using a format of "MainConcept MPEG-2 and a template of HDV 720-25p" (ridiculous, because the source material is not HD), I get a file extension of M2T. My PC won’t play that (even though it can create it), but I understand it’s an Mpeg format, so if I rename the file and change the extension to .mpg, it will play on the PC. I then upload to You Tube and it works! It picks up an inappropriate HD logo, but it works. Is there a better way? Can I render as something more sensible (not HD) and have You Tube accept it for what, to me, it clearly is - i.e. SD widescreen?
I would really appreciate answers to these questions and pointers if I’ve got any of the above incorrect. I now know an awful lot more about video than I ever wanted to, but I’ve started down the journey, so might as well see it through.