Even though things like PhotoShop & AE & Speedgrade can output up to 32-bit, Premiere is 10-bit. There's a link at the bottom of this post that goes into this issue in more depth. it allows you to import video files that were recorded at 16-bit) but it only outputs 10-bit. I mean, sure, maybe someday someone will build a 2500K video camera to be able to broadcast video on a 200 foot wide by 20 foot high screen where they want the resolution (pixel density) to be "equivalent to real-life" for anyone standing four feet away from the screen, but thankfully, 8K should be a reasonable limit for our basic needs on desktop monitors and regular televisions, I think.Ī few moments ago, I talked about 16-bit color depth. This is good because it will be an eventual limit on the nuclear arms race that is video resolution. Incidentally, on a positive note, 8K video is the approximate point where there are very limited returns in going with a higher pixel density (based on standard viewing distance) thanks to limitations in the average human eye. I'm sure that 8K video will have problems too. I'll bite my tongue and refrain from going on a rant about standardization for now. But 100x200 pixels is not a common screen size, so it's a poor example. That works out to 1,200,000 bits per frame. If each pixel is expressed in 16-bit resolution, then you need 20,000 x 16 bits to show each frame of the video. Let's also assume that we're working with a hypothetical screen size of 100 pixels by 200 pixels. The number of bits means a power of two since it's a binary system, so 8-bit is equivalent to 2^8th (or 256 different choices to represent ever type of colour known), 2^16th is 65,536 choices, and 2^24th is a number that I don't actually know off the top of my head, but it's a lot larger than 2^16th (which is a number that I memorized as a small child math geek, because it was the addressing range of the memory available in the Commodore 64). Also, there's a HUGE difference in range between those types. Be aware that 8 bit and 24 bit are also common color depths (more so in digital photography & graphics editing). Imagine this: let's say that each pixel on your screen can have a color depth represented in 16 bits. The Bit Rate refers to the amount of information being processed per unit of time.
I think I'm going to abandon this part of the discussion for now, and maybe make a separate blog post about it.
Except wait, some manufacturers (especially of hard drives) use the SI system (decimal) to skimp a bit, which means that a 2 terabyte hard drive is actually only 1.81 TB. And 2^10 is actually 1024, not 1000, which throws off the numbers for mega and giga and terra and so on (peta is next) because they're all just adding additional powers of two, not decimal multipliers. Now, alter the previous statement because computer stuff is expressed in binary at its heart, most of the time, and kilo is actually just shorthand for 2^10 or 2 to the tenth power, most of the time.
But ignore the fact that a billion can mean different things in different parts of the Commonwealth, etc. Kilo is generally a prefix that means a thousand, and mega is generally prefix that means a million, and giga is a prefix that means a billion, and tera is a prefix that means a trillion. Small "b" refers to bit, and capital "B" refers to byte. Generally, you should assume that there are eight bits in a byte (alright, computer scientists might argue about the exceptions).
I didn't want to confuse you, but I guess this is important. Let's talk about bits & bytes comparisons for a minute or two (you can skip this section if you want). If a piece of video or audio streams at 20 kbps (kilobits per second) and another streams at 40 kbps, and the two items are the same time length in duration, then the second file will take up twice as much space on your hard drive or other space as the first. It's the bandwidth, or amount of data that a piece of video or other types of media takes up, expressed in units of time. I also have a video link to a YouTube version of this information, which you can find at the bottom of this post.įirst, I'll assume everyone understands what the bit-rate is (but I'll explain it in more detail below).