Codec Lecture

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
here we go here we go ready here we go take notes here we go today we're talking about you get to okay today we're talking about the simpleness of codex the last class I start off by showing a video but I think I'm gonna start by talking codex first and then we'll kind of watch the video afterwards as a supplemental but the main things that we're going to talk about is that some of this might be remedial for some of you but but obviously video files have different sizes and if you've noticed on YouTube you can click on a video and it can have 243 64 80 720 1080 different iterations of the same video we're gonna talk about kind of what happens and how it gets there and then also about different different places to put your video so the main thing that we need to know is that videos have two parts one I like to call as Sorenson does cuz we'll watch this video from Sorenson but they call it a that's a box huh I'm an artist they call it a container so a container is what the video is in some examples of containers are an MOV avi there's a PPI if there's tons when we watch their video they'll have a whole you know a larger list but basically these are two of the common ones so a container there's a box what goes inside the container is the codec so if I do a codec and maybe it's a nice little square or something and what a couple of codecs that you guys know h.264 mp4 is a container photo JPEG is can't be a codec but but it's it's basically a JPEG is meant for a still image photo JPEG as it don't you get too crazy but as a codec it is a bunch of still images compressed together into a video file still has to go into some sort of a box okay so we have h.264 so I'll put that one up there I might get down on mpeg-4 and people are still kind of a rapper the container what else you see there's a couple called vp6 I six some DP six is one maybe two six four you'll see stuff like the NTSC DV will just standard codecs so now let's talk about what does codec mean does anyone know what it stands for this will probably be on the quiz a codec do you see that okay it means it stands for compression that's in our compression decompression I wish I could write nicer I really do so as as a codec there's compression and decompression we're telling something first how we're going to compress an image or head compress a whole file and then how it's supposed to be unpackaged we're telling about what box were putting it in with what codec and then that delivers in the file format to tell whoever's watching it whatever player is going to play it how to uncompress it and read it so that's why they call it a codec so here's what we're gonna kind of focus in on for a good part of this discussion remember compression decompression that's what codec stands for okay so there are four there are a lot of ways that you can change the size of a file we know that HD files are super huge we talked about in class what size they end up being so if I were to draw a couple of squares I should probably rectangles the world is different now so if I do something like this and then if I were to number them what would you say sizes of pixels on this super high def okay 1920 across the top 1080 across the side that's the number of pixels so if I take this picture and I were to cut this with on 1920 lines across 1080 lines across the top this way I'd get all these little pixel dots right so if that's 1080 what should this one be that's that that'll be this one good 1280 in 720 this is often called 720p 1080p oh I got those numbers wrong tonight thank you good wait way to wake up good job so long the big number goes on top 2 1 1280 720 p so having those in there again 1200 pixels across the side 700 wente down now we said 720 by 480 over here and then often in the early youtube world we didn't have much choice in video when it went online it had like almost always had to be 320 by 240 super small this is considered standard def so this is like substandard def but before youtube came around that was our choice like we wanted to get something online we had to make it so small because the files were still large right remember when you had dial-up and you wanted to watch something to do anything even even like in the early broadband days you still could watch video at a decent framerate you know this size but five years ago if you pulled up something that big on the internet you would probably hit start downloading and then walk away for three hours until you came back and said ok now all the information is there so let's talk about how these can happen with what I like to say basically we're looking at four things that allow us to take something that's super large and super dense with information for things that we're going to talk about that allow us to do smaller file sizes so I need somebody to pull out a calculator and do some math while I draw pictures who's got it good calculator on the back I want you to add at least this one 1080 by 1920 and give me a number and so 1920 x 1080 I don't know how many pixels are in there so x 1080 by 1920 2 million 70 3600 that's a lot okay a lot of pixels times that by 30 what am i doing if I take this number pixels time to by 30 what am i doing what is 30 important for number framed in a second right we watch 30 frames in a second I'm on TV for the most part and so if I times that by 30 62 million someone in there huge that's the number of pixels that have to get from the source you know like you back out in the internet somewhere to your computer screen in every every second right that's 30 frames at that frame rate so that's per second of video now let's do the same let's do the same math down here on this end to 40 this is pixels per second does it work that way not entirely but for our reference of saying how much information is there this is one way to kind of calculate 62 million you know if we thought of each pixels a bit which it's not we would say that's a lot of information that's got to get from point A to point B if we do this one 240 by 320 800 that times 30 or not times that by 30 that's for a second so this is 2 million this is 62 million huge difference right for our little piece I could send you know nearly 60 seconds of video and use about the same amount of space as I would use is one second of the super high def so let's talk about four things that we have control over that we we have control over everything but for the main ones that we can play with when we're talking about making this file size smaller and more manageable first is the frame that's that's this part right here how big is that frame and and the frame rate when you to grab something if you put it up as 30 P chances are most the time YouTube changes it to 25 frames per second dropping five frames a second saves them 10 million pixel bits right right let's put that on the docket for another you know for another day note to self talk about drop frame same thing happens with 30 frames there's a 29 97 and it's called a drop frame and the quick answer is in the broadcast world back in the day and the broadcast world frame did not match actual seconds so if you were to watch something on broadcast news that was 1 minute long on tape in the real world one minute was not one minute it was off by just the fraction this to the second so after a 30 minute newscast you wouldn't really notice because by that time it was you know adding up to be a couple of seconds off video wise so long story short is that they incorporated early on they started using something called the drop frame which was when you get from 59 seconds and 29 frames instead of going from 59 29 to 60 or to one minute they jump ahead to extra frames and so it goes from 59 29 to one minute and two frames you lose one minute one frame and one minute no frames that's the way that they catch up with themselves so that by the end of you know a block of time it is exact with the clock time so I mean there's a lot it's not like they screwed up it's not like oh we messed up that doesn't really come out it's the fact that frames and the way that they're put together mathematically didn't mesh with the accuracy of a second because remember we have things we do 30p on our cameras a lot of people in film use 24 p over in the cross the pond in England the British the world over there they use a different size of frame and they used 25 P instead of 30 so there's all these little kind of recipes which we'll get into so that's the time that's a time issue okay so write this down for frame that's one of our things that we have we have access to is frame that is frame size noted here these are different frame sizes we already noted that that's gonna that's gonna cut out a lot of space that's also frame rate if we if we're sending the the high-def version but if we cut out five frames we save ourselves 10 million you know bits or bytes like it's hard to keep saying that because I know that it's there's so much more in a bit than a pixel but for our small and meager Minds its its relevance okay frame size frame rate other thing that we can do any ideas I four of them what something else that we can adjust that we can control that would lessen the amount of information that needs to get pushed okay so this would this would actually fit resolution right resolution is how high so so you could call that resolution it's a frame size frame frame rate good audio is one I'll put that one right over here on the bottom audio is one because you've probably noticed that you can shoot some really clean super fine audio and a high frequency or a high kilobits per second that's how they measure audio kbps man I wish I could really write better I need a scribe so kbps kilobits per second similar idea in that we have this many pixels per second this is like how much information goes per second does anybody know what like good solid CD quality kbps is you see it on your iTunes all the time when you rip something 196 right in there yep so like 196 182 somewhere in there is is a good clean sound it means that that's how much every second of audio is getting that much information attributed to it if I cut that down to say 96 kbps I'm taking nearly half of the space and I'm saying you know let's say for five seconds it's just a tone like a B if I cut it into half 96 kbps I'm giving those same five seconds half the space in the file that they used to have something's got to get degraded something's got to disappear from that right so audio is one way you can drop it down if you keep your audio in this 96 to 196 range it's gonna be fine if you go much lower than this you'll notice a difference like a big difference if you go higher it's nothing you know so hot it oh that's super great quality but you'll be using up a lot more space in your file it goes all the way up to you know three hundred and something is like you know good clean sound nowadays they're pretty happy with that and then the high fidelity stuff you can go as high as you want at this point your ears especially like 186 is like okay that's good and cleaned I'm happy with that this I think you'd start to hear just the slightest kind of degradation to it a little hollowness but really you'd have to be a pretty good audiophile to be like that's that's not pristine but that's a good range so that's a good one what else we got two more that we can adjust no it doesn't technically I'll get into that in a minute okay uh so we got frame we've got oh let's go color down here color what can I do with color I'll give these numbers so you guys can mark them one is frame four is audio three is color now let's do this two is color side I'm leaving room for three up here here that's right okay color what can I do with color to save space to save room save file size black a white is one option because that gives me what two colors that means every one of these two million pixels with each frame I decide whether it's going to be black or it's going to be white now if I change that to grayscale I can go up to like sixteen different versions of black to white right but knowing that I only have to know those colors or employ those colors then when it comes time to render off one frame so we take a look hey let's make one picture one frame the computer can take a look at it and say Oh color wise I only need to use you know 16 colors to tell the whole picture here and I can use the same colors over again and then in the you know in the different iterations so if I were to rearrange this we've talked about our frame size so I'm gonna get rid of these for a second and instead I'm going to draw the same frame same size over again so there's frame 1 its frame 2 is frame 3 let's say in this frame my bicycle is just rolling backwards it's coming off the screen and that's this that's this this is my sad story it's my drama bicycles rolling backwards into the street it's at it and I also have on my on my scene a real nice tree that's up here in the corner kind of the as a a reference point it is a happy it's a happy train happy little tree I'm so not Bob Ross so looking at this picture if we talk about color there's a point in every frame that has to be process where the computer takes a look at it and says frame 1 frame to frame 3 what's changing does it need to print the entire picture again to that do I need to pull all 62 minute or I guess it would be all 2 million pixels in there again it doesn't need to if it already knows that in frame 2 in frame 3 the Sun and the tree don't move if they don't move and if their pixels don't really change it can take a mass majority of these pixels and basically write a pointer or something in the code that says you know when you got to put these pictures in if like the next 30 seconds go ahead and look back at this one right here because that's what it is it doesn't really change sure a couple of pixels might change here and there if the tree is like waving but the computer in the compression world says how much can I get away with yes it's so in here what would it be what would be basically recalculated for every frame the bike itself and that's the only thing that it's gonna really need to worry about for each one is it's gonna take a pic the pixels that are involved in the bike and it's gonna say oh these are the ones I need to be super accurate on because it's moving all these pixels yes from frame 1 to frame 2 it says what changed oh this area down here I know that what used to be you know background seen is now blue so I'm gonna paint that blue but it gets to be really accurate whereas the other 3/4 of the picture like you were saying not really changing so when it's doing the compression part of it I mean the camera grabbed it all right our camera has everything when time you go into editing and you start putting it together and then you export it and we're doing that compression part of it and also when it goes to youtube so there's a couple of compression points that's when it's saying I don't need to do anything with this because it's not changing that means in this frame instead of having 2 million new pixels like these 2 million I can probably get away with doing like 700,000 like a quarter of like a quarter of the amount does this make sense you guys followed me and I see some nodding head so that's good if you get lost catch me because I'm not good at math so I'm not good at teaching math but like yes are you losing quality from one brain to the other in a way yes that's the compression part what if you watching maybe we can pull one up we have time but if you watch on some of those really compressed pieces on YouTube you can actually see a background that kind of starts to degrade and get pixely so instead of seeing like a really pretty picture you start to see like the black background or some sort of piece that's not relevant and it start get junky so all of a sudden you can actually see like big squares instead of what should be like fifty pixels you can actually see a larger block because the computer said okay I've taken an average over the last couple of frames and this one for this frame is gonna be you know this color and so it's doing its own math and it's throwing out all that information I don't need to know you know the other million in that you know 1.3 million pixels I don't need to know them because I've already painted them once so that's compression does that make sense that's that's part of this oh you guys I totally jumped it okay so that's part of the picture which is number three picture so that whole discussion we just had was for picture because the picture is changing and when that picture is changing like we talked about that's what it can decompress and throw compress and throw out color colors that idea that in this exact pixel is it like you know the shade of metallic blue that is you know super special or can I just call it blue we are talking about the number of colors black and white if two colors grayscale will be like you get sixteen or thirty two different shades of gray you can do up to like back in the day you'd see a lot of these kind of posterized images where they would take a really nice picture but then they would just pick like maybe 300 colors to tell the whole picture that's that's where that's where color comes in is they we we could knock out a whole lot more information by saying well so we're for instance look at my shirt how many shades of blue do you think around my shirt right now with lights and with shades there's quite a few I've probably got fifty different blues or more that the camera is picking up but when I compress it and I put it up to YouTube or I can press it to put it somewhere else computer has a choice to say oh I can give you the same picture with like three shades of blue right there's this little quality feels a little crazy but you could paint this shirt with only three shades of blue so that's that's color discussion of you know following the moving parts that's picture we clear good because I really jumped the gun there I'm so sorry okay these four things will be on the quiz frame picture color audio four things four options for controlling the size if we take this huge file and say it's a one minute film a one minute film at 1080p so it's huge and if we kept the same frame the same resolution the same size same frame rate and we kept the same picture or we had it draw the picture every frame like we can tell the computer when it compresses to do everything this you know look at every pixel don't ignore anything and if we told it to give us a million plus colors I don't want 320 colors I want you to give me a you know every color you can and I want the best audio ever super-huge file results right probably if it's if it's high def you could probably get one minute and make it easily a gig gig and a half a lot of lot of information how do we take that and and make where do we balance that and make that a really nice-looking picture that nobody else notices oh yeah I'm drawing a scale guys like that they look like skill I win good good there we go what do we put on that scale what goes on one side what goes on the other let's do it house okay disregard that it's underneath these four things what are the two things were comparable yes all for this what balance is the other side what are we trying on one hand we're trying to get a smaller file size because it's got to go across the Internet's it's got to go from point A to my house and it's got to go fast because I'm impatient so on one side it's got to be file size but all these other four things and land up on the other side and determine the overall quality okay you're with me you're right there you're right there good okay yeah I know you're like what so essentially we're we're always going to be balancing size of the file with the quality of the video that's the key to codecs and compression and decompression that balance size and quality I need to have a small file size if I'm going to be putting it over the internet and up to YouTube etc if I'm gonna email it to my mom it's gonna be super small but I wanted to see the best that she can see it's easy enough for me to say alright I'm gonna adjust this by you know all four things I'm gonna totally adjust so my frame rate my frame size I'm gonna knock it down to the very smallest one we had over here 320 by 240 and my frame rate I'm gonna go from 30 frames a second to 15 frames a second right so I'm throwing out half the frames I'm throwing out more than half the size I jump over the picture and I tell it don't think about the picture at all during this 1 minute you get you you just put that one frame up there and make sure you have all the pixels in frame number one for the all the time you get to the end whatever is the left is left right it'll redraw what moves but anything else will just get left behind and then I tell it stick with 300 colors that's all I need for this and that's not bad audio way down to like 48 kilobits per second I could get a minute piece and like 1 megabyte size what did I lose quality gone there it would be unwatchable for me my mom probably would still think I did a great job so that's the balance I've got to find a way for every time we do a piece what's the smallest size for what I'm delivering and still keeps that quality for broadcast it's gonna be different for YouTube it's gonna be different for a DVD your hand over it's gonna be different so it's a size and quality now before I jump into the - like the whole wrap-up of the scene let me out here on the picture a key note that you will need to know keyword don't get confused the key word is key framing or key frame we talked a little bit about key frames when we did audio to remember that some of our audio adjustments we went in and we said here this is a key frame I want you to be at this level here and 10 seconds later I want you to be at this love we did a couple of keyframes manually for some of our audio similar but different this keyframe is what tells the compression and the decompression at this at this interval I want you to redraw every single pixel again so you'll notice this when you watch some videos where you'll see it and almost every second it comes back to crisp and then it degrades and comes back to crisp and degrades so watch that on some videos that's it usually they have like it it's preset it's like every 24 frames so your keyframe is that one frame that says reset go back to good if you like what's going on here every 24 frames you you draw you draw everything you draw the the Sun and the trees and a bicycle and a little boy in its house and it's mama you draw that every 24 frames or whatever is in the in the picture now I'm on the next frame and the next 24 frames after that you can let the colors kind of go where they want and the pictures the pixels kind of do what they want not paying attention to that's the key to the keyframe now all of this wraps up into what we call codecs right compression decompression algorithms because these four things and many others create kind of like the recipe all right think about it we we obviously you're not gonna figure this out every time we're not figuring it out every time we're not sitting in there going okay this time I think I want 600 colors not 700 right that's that some people do their called compression astir encoders we have recipes there they are they are the recipes for putting things together h.264 is this you can set different sizes of h.264 in the frame but when it comes to color and the picture and the audio and the things that are embedded in it it's a certain set of settings does that make sense somebody jumps up with with a VP 6 or some other type of thing that they want to you know setting that they want to do and you probably notice that there are a million codecs out there DivX heard of X you've heard of the mpegs one two and three mpeg-2 they're all talking about some of these things kind of coming together in a certain you know certain set of settings and then they package those in a container or wrapper and they put them out on the big World Wide Web and that's why you've got to have remember that what's the other half of compression decompression yeah well compression what's the other half of compression the decompression you've got to pull it apart that's why if you download something that's DivX and you load it up in your quicktime it'll say I don't have the codec to read this meaning I don't have the recipe I can't uncoding sent to me because it's not that it doesn't know what these are these are written in a certain file types are written in a certain language and so it says I need you to download that kind of decoder ring so that I can now show you when I'm supposed to be showing you make sense the reason why 2 6 4 is stuck around because it did the best job at the time of balancing size and quality it said I can give 1080p pictures at 25 frames per second on YouTube and they can look fantastic and they can be small and everybody oh that's so great so it's been around for a long time are there things coming out right now or they're still balancing and doing new compressions and sending them out yeah there's still new code no new codecs that people are trying new recipes that people are passing out but you've got a you've got to have one on each side max do a pretty good job of saying we know most all recipes and they're all kind of built in here sometimes you have to grab DivX or some other you know simple ones but pcs aren't I mean Mac was built on that video foundation it was built for that pcs are kind of cobbled together and put together and you have to download codec packs anyway try that or had to do that there's a couple of them out there where you can get a whole pack at one time you just say put all these in my repository so I'll have them when I need them and I kind of come in groups that's one of them okay so if we wanted to follow that trail real quick look for a few seconds the camera right now that is shooting me is suiting to a flash card 16 gig flash card I know that it is doing its own set of compression decompression and I know that this camera is set up to use h.264 but it also adjusts besides these four remember I said there many others it's just many others one of those things that it just is a long form pixel so it's doing different size pixels and it's spitting out files that are mts files so when I go and put that on my Mac and I say look at this card my Mac doesn't know what to do with it it looks at it goes that's an MTS I don't know to do that I have to put it in another computer program another program to encode it to decipher what's going on here that computer program deciphers it and says okay you've got a one gig file from that 40-minute discussion you just did it was one gig on the card but if you want to edit it and put it into final cut or somewhere else I'm gonna need the encode or decode it and make it something you can use and I had one of these that I did where it was probably a probably a gig or two on the card is about an hour long drawing session you guys probably seen that picture up there Morgan Spurlock and the guy climbing out of the rut well when I said X you know export that kind of make that something I can edit with it gave me a codec called pro res for two - this for two - actually stands for some of this color stuff but pro res is proprietary to Apple it came out a 60 gig file it was like two or three gigs on the card I remember small but when I said turn this hour of footage into something I can edit with it with 60 gigs because it started filling in all this information it went back in and rebuilt these frames with full color because it because it means you know knew what it needed and for editing it doesn't know where I'm gonna put the Edit point so it doesn't need to use these key frames it needs to have everything everywhere so that when I get into the cutting I don't lose anything so when I basically rehydrated it off the card blew it big so that's one place compression happens but then when I edited that piece together I put it back out to youtube I compressed it again to something that YouTube would recognize which is back to the 264 but it's with more common settings that YouTube's looking for and that's where the key frames and everything else came in that answer your questions that's those are two places then it happens a couple places okay wow I've got you for like six more minutes I'm trying to think of there's anything that I've missed in this big discussion I've got oh you know what we should do let's watch this movie Sorenson did a movie that explains this a lot better open again it was wrong but it kind of answers this question and somebody calls you up and says hey can you deliver that you know that thing you shot can you give it to me as an AVI somebody says that's me I just shake my hand I'm like what kind of an AVI because there's so many things inside that box that's that doesn't help me at all if they say yeah just making an MOV doesn't help me at all because it could have any number of these inside and I may not know what to do with it so that's that's painful but it happens all the time if that happens just say Oh which you know what would you what kind of mov would you like what what codecs can you use do you have final cut do you have QuickTime and that narrow it down but that can be kind of painful so for this last little bit I want to show you my sweet computer hasn't been giving me grief from when I like pop it up and then try to connect to Wi-Fi it takes like another extra two minutes to even find the signal anymore even though I know it's on there maybe that's what it is yes go not per clip this is on full export this is the recipe when you're so right now when we say share to YouTube it puts together the YouTube recipe that's where it adjusts things to say size color picture and audio yes yes so if when we shared go to YouTube it knows this preset if you there's this if you notice there's a button that says send to compressor whole of the program specifically for fine tuning each of these and many others saying I want this frame a I want this frame size I want you to drop this many pixels but not this many all of that comes into that frame right if I know that I'm going to give questions is actually just because I you know played with it it is actually the difference between like share to youtube and what we've done for some of you that you know kind of that hasn't been working we do export QuickTime conversion spit it out it is the difference between 1920 and 1080 is what if you do an export and upload and I think this share to YouTube is 1280 by 720 so when I I can tell which ones do that because I can get on YouTube and say that I can only when I hit that button at the bottom to like upscale it I can only get to 720 on some those are the ones shared shared to YouTube ones that are exported and uploaded separately have a 1080 option right you want 1080 yes and right there all I'm okay there are movies coming out but you want the higher-quality one but if I wanted to do a DVD like an actual disc that would play on a desktop I've got to do a whole different combination down here making an mpeg-2 something how does that make sense good let's see if I can get into this and will make more sense boom okay see if I can find myself here due to t2 if we go back to all my tweets I'm sure we could find it at some point here okay encoding 101 three and a half minutes perfect time for us to watch this is Sorenson media they're here they're here in Utah they know their stuff but they're gonna talk and show some stuff from their program Sorenson hold Java Sorenson is to take the video adjust these things and give you the best thing so they have a program that says here are all sorts of recipes that you can try which one fits for you what do you need to deliver so they're for encoding and they explain the containers and the wrappers pretty well first let's start with the basics each encoded video contains two major parts codecs and containers or formats think of the formats as the shipping boxes and the pc-12 and many more examples of containers or formats the majority of the video on the web is in the Adobe Flash format flash can contain three different codecs this one's the Spartan bp6 and h.264 there are pros and cons to each codec but each will play universally on 98% of the web browsers around the world along with formats and codecs one more thing that you'll need to understand before encoding your video to the web is delivery method there are two methods in which to deliver your online video progressive download and streaming with progressive download video begins playing on a flash player load the entire video needs to be downloaded before you can finish the video this is the most pervasive method typically used for videos 110 minutes in late streaming video is easier to navigate through because you can easily click play and see to any point in the video the file then begins downloading from that point this meant that is typically used or long coordinated now that you know the basics of encoding for the weather let's go over a few best practices first start with the best source always use video that hasn't been compressed and coming as a garbage in garbage out process so just don't bring the garbage in the best practice is to use raw footage of your camera or uncompressed abis or movs from your enemies second choose the best format for your audience within sources squeeze 6 there are many different formats of choose from inside of every four matically decide the frame size and data rate of the video squeeze has intelligent presets that take the guesswork out of encoding this preset tells me that my output video will have a frame size of 640 by 360 with a data rate of 1500 kilobits per second using the h.264 codec this will give me the flexibility and also the quality that I need third don't forget about the audio an old saying says the sound is half the picture this is especially true for the low data rates of web video make sure you give the audio enough data to sound good this level should be anywhere from 96 to 192 kilobits per second for the web for it delivery where do you want your video to reside when instead of coding in squeezed you can deliver your video directly to your AWS account by adding your credentials and applying the preset you can also deliver your video to source in 360 our online content management system to FTP sites local directories and even applications all simultaneously with these intelligent publish presets now you're ready to start encoding best practices you I'm coming pro in no time at all all right thank you so much so that's the same thing we have when we go final cut we say share you can hit that drop down where it says YouTube and you can choose apple TV you can choose hard drive you can choose FTP you can choose any sort of outputs same same ideas their stuff presets are there and you choose the frame size in the frame rate and put it up any crazy questions do that right on time because also like a static okay these are the four things I want you to know for the quiz so you probably ought him down one two three four frame color picture audio the balancing of these size and quality what what is balancing there and then possibly what does a codec stand for and codecs what does it stand for will be in there and then like containers codecs
Info
Channel: Drew Tyler
Views: 78,997
Rating: undefined out of 5
Keywords: Lecture Video Codec codecs, wsn, drew tyler
Id: bsRWszG84Ko
Channel Id: undefined
Length: 40min 6sec (2406 seconds)
Published: Mon Nov 21 2011
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.