Speaker 1: If you've ever looked at codecs, then you'll know that there's a large number of them, such as H.264, ProRes, DNxHD, but what does it all mean? But first of all, I'm going to start with the difference between a codec and a container. A codec is the method that's used to encode a video file, whereas a container is the file type, such as .mp4 or .mov. Now, generally, the higher quality of a codec, the larger the file size, but this isn't always true, especially when comparing new and old codecs. Newer codecs are able to maintain a much larger amount of detail than older codecs are, whilst producing a similar file size. In terms of making a film video or YouTube video, whatever, there are three main stages, and at each stage you'd use a different codec, potentially. You have capture, where you record the footage into a camera, editing, where you edit the footage, and delivery, where you export the footage and send it to wherever you're sending it. But what type of codec should you use for each different situation? Well, for capture, it really depends. Entry-level, budget, and even prosumer cameras will often shoot in a more highly compressed codec, because people who are just getting started in the world of filmmaking and video recording do not want massive files that they have to store. In the past, H.264 has been one of the most common capture codecs in this level of cameras. Now, when we step up to the next level, such as film, broadcast, or just generally professional work, that's when we start to see more higher quality codecs being used. If you're going to be doing a lot of color correction, color grading, visual effects, then the extra quality and information that you get with a higher quality codec can be really useful. Something like ProRes or Avid DNxHD will give you a lot more flexibility in post. This is where you start to see external recorders. You plug them into the camera and then they record the video signal that's output from your camera, and this way you can get a higher quality codec than the one that would be recorded in your camera. You can have less compression, you can often get more bits of color information, and you can have better chroma subsampling too. At this point, I should also talk about RAW. Now, RAW is not a codec because it's not encoded. It is sometimes compressed, but that is not the same as it being encoded. RAW is when the camera just dumps the information from the sensor into a file. The advantage of this is that you can change things like the ISO, white balance, and picture recording profile in post. But that does come with some downsides, and that is that the file sizes are massive. On a Blackmagic Pocket Cinema camera, you're looking at 3.4 gigabytes a minute. RAW is a topic for a whole separate video, but this is just to make you aware that there are non-codec options for capture. Next, we have the editing or mezzanine codec. Here, if you are doing color correction, color grading, VFX, if you're using a highly compressed file, then the compression will start to show. If you adjust the brightness too much, your image won't look quite right because there isn't enough detail there. Most codecs do quite a good job of hiding the compression, but as soon as you start tweaking the image around, then things start to fall apart and you can see where the bits of detail have been cut out in order to reduce the file size. Even if you're not doing a lot of heavy post-production effects, there are other disadvantages to highly compressed codecs too. In order to explain why this happens, you should know that there are two different types of compression, but in particular, we have interframe and intraframe. Highly compressed codecs tend to be interframe. Instead of the codec capturing all of the detail it needs to recreate each frame, it will take all of the detail it needs to make one frame and then the frames surrounding it, it will just store the changes between them. So sometimes, if you tell your editing program to look at a particular frame, it will have to look at other frames too to work out what it should be showing you. This can be very CPU intensive and result in choppy looking playback, even on powerful computers. Highly compressed codecs like ProRes on the other hand are intraframe, so each frame is compressed independently of the others and your editing program only needs to look at one frame to work out what it should be displaying. Now, all of this doesn't mean that you have to capture in a lower compression codec like ProRes. What you can do is you can capture in something like H.264 and then when you're editing you can transcode, which is basically converting one codec to another, such as in this case ProRes. This will create some slightly larger files, but you can delete them once you've finished editing and you will get much faster playback performance. But transcoding does mean you have to wait around whilst your computer converts the files from one codec to another. Unless of course you have Final Cut Pro X, which can transcode files in the background. So you can be editing away on your project using the original camera media, but then after a while, once Final Cut Pro has finished converting the files, then you can switch over to using the transcoded files. One of the many reasons why I like Final Cut Pro X. Now onto delivery, which is actually I think the simplest part. This is normally where your video file will be most compressed. If you're uploading your video file to a streaming site like YouTube, Vimeo or even Netflix, it does need to be compressed quite a bit. Because if a user doesn't have a particularly fast internet connection, if the file size is too large, then they'll have to watch in a really low resolution or with very dodgy playback, which is not what anyone wants to do. And as compression is mostly visible when you start doing lots of image manipulation, which doesn't happen when something is being streamed, you can get away with removing some of the less obvious detail. DVDs are normally around 4.5GB, so if you're exporting to one of them, then you'll need to do some compressing too. Of course there are exceptions to this, such as Blu-rays, which store 25GB on single layer or 50GB on dual layer. You can use a bit of a higher quality codec if you're going to be exporting to something like that. Choosing which codec you need to use can be a bit of a daunting task, but a bit of research can help you choose the right codec for each different situation. In a bit of a brief summary for Captcha, if you're going to be doing lots of heavy manipulation in posts like colour grading or VFX, then you want to use a low compression codec. If you're not going to be doing any of that, then you can use a high compression codec. When editing, you want to use a low compression codec, and when delivery, you'll normally want to use a high compression codec. If you found this video helpful, then please consider leaving a like and subscribing so as not to miss any new videos. See you later.