In the video industry there are a lot of buzz words. These days no two words are bigger the raw and fourkay, or more commonly RAW and 4K. Another word that garnishes a lot of attention is uncompressed. I can’t begin to tell you how many conversations I’ve had with customers, smart people that have been in the industry for a long time, where the customer inevitably says something like “will this camera record uncompressed 4K RAW?”. Or how ’bout the “I shot my project in RAW ProRes.” Lets go beyond the buzzwords and gain some understanding together.
The first and most important piece of information that must be understood is that these are three seperate terms that have absolutly nothing to do with each other. Commonly they go hand in hand and are often seen with each other. This leads to the perception that they are linked. In fact you can have each of these term completely on its own. You can have RAW video that is nether 4K nor uncompressed. You can have uncompressed video that insnt RAW. You can have 4K video that is compressed.
Alright, now that we’ve got that important little bit out of the way lets talk about what each of these things mean. Lets start with the easiest, 4K. 4K is just a resolution. How many pixels high by how many pixels wide is the image. With standard def we had 640×480 and 720×480. With HD we have 1280×720 and 1920×1080. There is 2K, which is 2048×1080. Finally there is 4K. With SD there was a recording resolution and a display resolution. Most DV cameras recorded 720×480 even though SD displays were 640×480. That mostly went a way with HD, we shot 1920×1080 for delivery in 1920×1080. Then 2k came along and we shoot 2048×1080 for delivery in 1920×1080. Now with 4K there is also an acquisition resolution of 4096×2160 and the display resolution of 3840×2160. The consumer term for the display resolution is UltraHD. UltraHD is 4K for monitors. Most 4K cameras give you the option to shoot in ether 4096×2160 or 3840×2160. Thats what 4K means. 4K is exactly 4x the resolution of 2K. That means the files are exactly twice as big.
Now lets talk about the other super hot buzzword, RAW. First of all, even though i’ve been typing it RAW its actually raw. I’m only typing it that way because most of the world has began to recognize it being displayed that way. We can thank RED camera for that originally but now Sony and Canon and everyone else writes it that way. It’s not an acronym. It doesn’t stand for anything. Still photograhpers have been working in RAW for the last decade. Its become so commonplace in photography and computers have become powerful enough that most people who shoot with a DSLR, pro or amature, are shooting RAW. If you’ve ever edited a RAW photo compared to a JPEG you know why. Its amazing how working with the raw camera data gives you virtually unlimited ability to tweak the image with out degradation. Lets talk about what RAW is and then a little bit about why you would ever want RAW video. Imaging sensors are analog. They run on light. Then subprocess in the sensor converts the analog light information into digital ones and zeroes. This is part of the reason all sensor are not created equal. After the sensor creates the ones and zeroes what do you have? Nothing. You still don’t have an image. Think about the relationship between your eye and your brain. Your eye doesn’t “see” and image, its a sophisticated light hole. Your brain “sees” the image. If you’re brain puts the light together wrong then things like dyslexia can occur. Similarly when the binary comes off the sensor its not a picture, its binary. When we shoot RAW, still or video, we take that raw binary and we capture it to a memory card. Then we use a piece of software on our computers to reassemble the bianary into an image. The benifits of this is that nothing is locked in. There is no white balance, there is no noise cancelation, there is no black point or white point, there is no color tone, there is no sharpness or skin detail level. All of these permaters are assmbled by the software on the computer. Thats why you want good RAW processign software. The result is you have footage that is completly adjustable. A lot of RAW files feature 16-bit 4:4:4:4 color, thats right, four 4s. The post production applications are limitless.
So if RAW is so fantastic why wouldn’t i want it? Well it not the cost of the camera. Thank to Blackmagic we can get RAW cameras for $995. So why don’t i shoot everything video in RAW? First, those files are huge. RAW will cost you 1TB per hour. Thats a lot of flash memory cards, SSD, and RAIDS that you need in the field and in the edit lab just to store the footage. Do you remember when i said that non of the image peramiters were set in RAW files? Who do you think is going to set those? You could let a computer do it for you but that kinda defeats the purpose of captureing RAW. That means a human being has to color grad every single shot. I don’t mean color correct, i mean color grade. Multiple passes dialing in every setting of every shot. Thats a lot of man hours and takes skill. Lastly the computer needed to process all this is immense. I’ve tried to edit a single RAW frame on my 2012 dual-core MacBook and the whole computer came to a halt. So if you’ve got the time and money to quaduple your post production RAW might be the right solution for you. There is a reason a lot of Hollywood movies are shot in RAW but network TV show are usually not.
Lastly we have uncompressed. Uncompressed is a relativly simple term. That means that the zeroes and ones that the image sensor creates have been moved down the pipe to the image processors. At that point all the image peramiters are assigned, white balance, dynamic range, etc. But in the last step of being processed the data, which has now been made into a viewable image via the internal proccessor, is spit out of the SDI or HDMI port instead of compressed into a codec. Codecs throw out data in order to create a compressed image. Formats like ProRes and DNxHD are mastering level codecs that are virtually lossless. XDCam422, MXF, or AVC-Intra are codecs that are more compressed but still retain “broadcast” ready standards. H.264, AVCHD, XDCamHD, and HDV are formats that compress the video to a level where they look good as long as you don’t do much post processing or much more compression. Uncompressed video means the image parameters are already set but there is no compression scheme applied. Now you can have “compressed” RAW. This is a little tricky but cameras like the new Blackmagic Production 4K camera use Adobe CinemaDNG wrapping on the RAW data. This throws out some of those zeroes and ones we’ve been discussing. But the data doesn’t go through any kind of image processing so its still doesn’t have set image parameters. You still have all the same flexibility as RAW, albeit with a few less zeroes and ones to play with.Well i hope that somebody finds this information useful. Post by Eric C. Petrie