A Simple Guide to Bitrates, Resolution, and other Techy Video Stuff You Don’t Wanna Learn About
Here’s the thing: video is complicated. All you want is a picture that shows detail, moves smoothly, and doesn’t fill up your entire (quite large) hard drive in one go. Is that so much to ask? Well, apparently it is. Open up any export manager and you’ll find about a hundred different possible options, including bitrate, resolution, frame size, aspect ratio, pixel aspect ratio, codec, container, format….the list goes on. Worse yet, not everyone agrees on which term means what half the time.
Furthermore, you’ll be told that such-and-such is the standard format for something. That may be true in some arenas, like a television station. Though, if you work in a television station, you’re probably not gonna get much new information from me. For the rest of us, there’s not a standard. We can say there is, but no matter how hard you try, you’re always gonna get a black bar somewhere. So you’re gonna have to take it for what it is: a bunch of info that you have to filter through and determine what you need for each project.
I’m gonna try and keep this as layman as possible, but there’s a lot of info to go over, so I wouldn’t dare promise brevity.
Resolution
This is one of the confusing terms whose definition depends on who you’re talking to, what you’re talking about, where you’re talking about it, what the weather’s like, and whether your left shoe is tied. In the realm of image editing, resolution can refer to the number of pixels per inch in an image (the higher the number, the more detailed the image). In video, resolution has a similar meaning, in that a higher resolution video display will give you a greater level of detail. However, while in image editing this is usually a single number (72dpi, 300dpi, etc.), video regularly refers to this as a dimension. 720×540, 1024×768, or 1920×1200 as examples.
Lines of Resolution
A video resolution will refer to vertical lines by horizontal lines (vertical x horizontal). Standard definition televisions in the U.S. use a standard called NTSC (remember that acronym). For practical purposes, TVs that conform to this standard (read: any old, non-HDTV) have 486 horizontal lines of resolution. Any video resolution you choose whose second number is at or above 486 will display on a standard television at a reasonable quality. Barring all the other things we’re gonna discuss.
HDTV
Naturally, a good thing couldn’t be left well enough alone, so when HDTV came along, you suddenly got a bunch of new formats to deal with. 480i/p (avoid at all costs, you’re getting ripped off), 720i/p and 1080i/p. The i/p refers to interlaced and progressive respectively. We’ll get to that in a bit. In the category of resolution, this introduces a couple new standards. First, there’s the lowest of the low-end HD standards: 480. If that number sounds familiar, it’s because I just mentioned a similar number in the previous paragraph. You know, the one talking about standard definition sets. While there are some vague improvements to this standard (hint: it has to do with progressive scans), I don’t care what you say. 480 is not High Definition video. The other two are 720 and 1080. These both refer to the number of horizontal lines in a video frame.
Note: It has come to my attention that some people like to refer to this number as “vertical lines” meaning that you would count the 720 or 1,080 lines vertically. However, the lines themselves are horizontal. The 720 refers to the number of lines that go longways across the screen. K. We good? Let’s move on.
Aspect Ratio
Aspect ratio refers to the ratio of width to height, usually represented as two numbers separated by a colon, such as 4:3 or 16:9. For most situations you deal with, these two will be the primary aspect ratios you use. Sometimes people like to refer to these aspect ratios in relation to a number 1. In that case 4:3 would become 1.33:1 and 16:9 would become 1.78:1. Annoying, I know. I prefer to deal with whole numbers, but apparently some folks don’t. P.S. These aspect ratios are frequently referred to as Fullscreen (4:3) and Widescreen (16:9)
Another thing I’ve recently discovered, slightly to my dismay, is that there are actually more than one kind of “widescreen” aspect ratios. Films in movie theaters are often shown in either 1.85:1 or a whopping 2.39:1. We’re talking ultra-wide screen. Sometimes your editor or other program will help you by including an aspect ratio next to a choice of resolution, but if not, there’s no shortage of tools online to help you get in the area you need to be.
Frame Rate
Frame rate refers to the number of frames that are shown per second. As you (should) know, video is basically nothing more than a series of still images played really fast in succession. The frame rate is usually pretty standard depending on where you live. In the U.S. the standard NTSC (standard definition) frame rate is 30 fps (frames per second), and PAL, which is the standard for regular definition video in Europe, uses 25 fps.
Note: Technically, NTSC runs at 29.97 fps. As with much of this information, this matters to a lot of engineers, but not too frequently with us average folks. In most situations, you won’t have to figure out which one to use. If you do, though, always go with the more specific choice.
Progressive v. Interlaced
The difference between progressive and interlaced tends to screw up the argument for which resolution is better, but here’s the short of it: for some reason known only to engineers, some display methods will show every other line of a frame at a time, but display them twice as fast. So with a framerate of 30 fps, you’ll see lines 1,3,5, etc. for 1/60th of a second, and then see lines 2,4,6,etc. for the next 1/60th of a second. These sets of lines are called “fields”. In one second you’ll see 30 full frames, but on a very technical level, they’re cut up into two fields which are displayed twice as fast. This method is called “interlaced” scanning. The reason for it has something to do with a technical limitation of video mediums in their early stages.
Of course, as time goes by, we get over limitations. This leads us into progressive scans. In a progressive scan, each frame is displayed in its entirety at a time. At a frame rate of 30 frames per second, you’ll see 30 entire frames, each one displayed for 1/30th of a second.
This concept is sometimes referred to as “temporal resolution” in forum arguments where a bunch of geeks argue about stuff that no one but other geeks care about.
*Incredibly Helpful Note: If you’ve ever been working with a video that gets jagged lines, or a “stairstep” effect, particularly when your subject is moving, this is typically due to an interlace problem. Computer screens handle video differently than televisions do. Playing an obviously interlaced video on a computer will look horrific, but output that same video to a standard DVD and play it on your old TV, it will look great.
If, however, your going to be showing it on a computer or an HDTV, you need to fix it. Most video editors have some form of de-interlace special effects filter. Simply drop that filter on your jagged video and your problem will clear right up. Problem solved.
Bitrate
Oh, bitrate. You’ve managed to elude me for so long, quietly screwing up my videos’ file size. But not anymore. Now I shall expose you and your file size altering mutant powers for all the world to see!
A video’s bitrate refers, surprisingly simply, to the number of bits that the video will process in a given period of time (usually seconds). In a bit of a twist, bitrate is pretty straight-forward. More bits per second means more information, means more resolution, means better looking picture. It also means bigger file size. Generally. But surely it can’t be so simple, can it? I mean, increase the bitrate, increase the quality and the file size sounds too good to be true.
That’s because it is. Well, kind of. While picking a frame rate or screen size/resolution or framerate, you generally have a strong hint to go by. Going to an HDTV? 1080 (or possibly 720). Showing this in America? 30 fps. Bitrate gets a bit more relative. You can have two different videos that conform to the same frame rate, resolution and other blah de bloo. But two different bitrates will yield two completely different file sizes, and two completely different qualities. You can also have subtle differences in bitrates, all other aspects of the video being equal, without screwing up playability.
As an example, where I work, I have to edit a bunch of videos of wildly varying lengths. Some are 2 minutes, some could be two hours. They all get output as video files to a CD, though. I generally export the videos at a pretty crappy resolution, at a constant bitrate of 1800 kbps (kilobits per second). Now, on a 700 megabyte CD, this will give me room for about 53 minutes of video. For a video that’s an hour and a half, splitting it into two discs makes sense. But what about a video that’s 53 and a half minutes? Splitting the video up because of 30 seconds of footage is entirely lame.
By dropping the bitrate down just a bit, to say 1600, or maybe even 1700, I can fit more video into the same amount of storage space. And to the human eye, the difference between 1800 kbps and 1600 kbps is negligible. But do be careful when changing bitrates, and never assume that dropping a bitrate won’t affect quality unless you really don’t care who sees it.
While there’s not exactly a “standard” for bitrates, there are some adequate guidelines that’ll put you about where you wanna be. From Wikipedia, here’s a basic chart listing bitrate figures for various video mediums:
As you can see, the 1800 kbps (or, roughly 1.8 Mbit/s) is slightly higher than VCD quality, but well below DVD quality. On the scale between the two, a drop from 1800 to 1700 isn’t that big, so given the circumstances, the drop is acceptable. Again, though, the above numbers should be used as a guide, not canon. For casual bitrate tweaking, this is acceptable, though if you’re looking for serious optimization for your specific project, you should probably set aside at least a couple hours to experiment with various settings.
Constant vs. Variable Bitrate
Another reason that bitrate isn’t as simple as it sounds is the difference between constant and variable bitrates. To put it simply, a constant bitrate is exactly what it sounds. Set at 1800 kbps, you will have exactly 1800 kilobits being processed every second.
A variable bitrate, on the other hand, is also what it sounds, but harder to understand. Through some crazy complex math that only (…who? Can you guess? ….yep) engineers care about, the video is encoded at a different bitrate at different times, depending on the needs of the current frame of the video.
Think of it like this. If you have a human being on the screen, or a complex 3D model, you want a ton of detail. Higher bitrate/resolution means higher quality and level of detail, right? But let’s say that during your video, there’s a part where the screen fades to black (as it almost certainly will at some point). For this particular moment in time, you don’t need a killer bitrate. I mean, what detail is there? There’s no lines, no shadows, no curves, no varying colors. Just solid black.
What a variable bitrate can do is determine what parts of the video need what level of complexity and adjust accordingly. So you’d have a high bitrate during the epic battle scene with 10,000 individual CG models, and a low bitrate when you’re on black, to conserve space.
Sounds great, right? Just toss it on variable bit rate and let the computer do the rest. Well….not exactly. Don’t get me wrong, in situations where you’re concerned with minimizing file size, variable bit rate is probably the way to go. But variable bit rate can also become highly subjective. And unfortunately, it’s not entirely automatic.
Perhaps the most interesting/annoying thing about a variable bitrate is that, often when you select “Variable” in whatever drop-down box you get, you don’t lose the option to choose a bitrate. “Wait. I thought the computer was supposed to be figuring this out for me?” Well, not so lucky. Because the computer still doesn’t know where you’re putting this to or what your optimal filesize is gonna be. While there’s always variations with any video, when using a variable bitrate, a single five minute video could range anywhere from ten megs to ten gigs of file space if you let it. Either by tossing out a bunch of data, or by including way too much.
So you have to set a guide. And therein lies the rub. With a constant bitrate you can almost always count on your filesize to be in the ballpark that you’re looking for. Variable, you’re not so lucky.
Short version: if you’re looking for a reliable output size for any given length of video (and especially if you don’t care about quality), go constant. If quality is important, and you’re concerned about minimizing disk space use, play with variable.
*Another Really Helpful Tip: Gizmodo has an excellent article with more detail on the subject, but simply put, a lower bitrate is the reason why, if you stream/download High Definition content from an online source, you may still end up with something that is a lower (often noticeably) quality than regular ol’ DVD video. It’s all about the bitrate.
Format/Filetype/Extension/Whatthecrap?
As if all of the above weren’t confusing enough, now we enter into the category I will here-to-fore refer to as Format. Video formats are a messed up category unto themselves, with too many choice. You don’t care about the practical differences between H.264 and VC-1. You just want the dang video to play! Well, that has the potential to get complicated. Don’t panic, though. Once you find your format, usually you don’t have to change until you change jobs.
Ok, yeah, not very comforting. Moving on.
To get started, we need to define a couple of terms here. To start with: codec. A codec is simply a type of standard used to code and decode video signals (get it?). Once again, the Giz has a great article on the subject of codecs, and you should really go read it now. But then come back.
As the article points out, though, codecs are only half the problem.
Containers/Wrappers
You’re probably more familiar with containers (sometimes called wrappers) than you are with actual codecs, as these are where movie files get their extensions. Things like .mov, .avi, .flv. These are packages that contain not just video, but audio, and other extra info, compiled into a single file. In order to play a video, your player needs to support both the codec and the container. Can’t make it easy, can they?
I’m not gonna repeat a lot of what the Giz article says, but I will mention, for those who can’t be bothered to read three gigantic articles today, that H.264 is generally gonna wanna be your codec for super high-quality videos. This codec is also known as MPEG-4 Part 10. You probably won’t hear it called that. Other than that, MPEG-2 part 2 (or just MPEG-2) is gonna put you in the DVD-quality range, and is also extremely widely supported.
Once you’ve got that, you start facing the major choices of format. You should be asking yourself by this point (and really, before this point), where you’re gonna show this video. If you’re going to burn it to DVD, it would probably be best to go with MPEG-2 and an MPEG container. If you’re looking to play this through Quicktime, or use it in one of the programs from the Final Cut Studio suite a .mov would be the best idea. Honestly, there’s really very little end to trying to figure out what format is best to use. I’ve personally found that .movs, and .mpgs are widely supported enough that I don’t have to worry about running into support issues. However, .oggs, though OGG is an open standard (which I like), I haven’t tried it out, so I’d be hesitant to use it. And .avis feel bloated to me, but that’s completely me, and not based on a lick of evidence. (Side note: .avis are also very widely supported.)
I’ve had situations where I carefully pick my resoltion, bitrate, framerate, codec and container to all be the most easy-to-get-along-with group of folks and still send it off to someone and get a call: “It won’t play. WHAT DO I DEEWWW?!” You’re not gonna get it right all of the time.
*Most important helpful hint of them all: Get VLC Media Player. VLC is a video playback program that supports every single video codec and container under the sun. Ok, probably not literally, but nevertheless, you will be hard pressed to find a time that you cannot simply open this application and run any video, no matter where in the bowels of the internet you got it. Carry it with you, install it on all of your comptuers. If you’re going to go show a video on someone else’s computer, bring an installer on a USB stick. I cannot stress enough how helpful VLC can be. As many times as I’ve had a project go awry (read: every project) I have a situation where some form of media will not play, I download VLC, and it works. It’s that simple.
In Closing….
Honestly, there’s a ton more info out there. I didn’t even touch on pixel aspect ratio, because quite frankly I’m still trying to figure that out for myself. But this should at least get you started in figuring out what the heck your video editor is asking you when you try to output your videos.
If I got anything wrong, or if something doesn’t make sense (as some of my analogies are apt to do), let me know.
Leave a Reply