I had a chance to play with the new Premiere Pro CS5 last night and I’m blown away. You can stop holding your breath on that one. I’m finally weighing in. The reason I’m personally jazzed is because I can now take the native video files from my HDSLR camera and edit them in real time. Real time! This is unprecedented. I’m surprised Oprah hasn’t sat down with someone at Adobe yet to discuss. Has she? Okay let’s set the stage;
HDSLR cameras started hitting the scene and CS4 was already on the market or maybe just around the corner. Too late to cater to the new hotness – and catering they needed. That version of Premiere was actually really good. It was the first time I’d used it since college, as mentioned previously, and it had come a long way. I was using it to edit footage from my new 5dMk2 but the process was frustrating. For starters, the files are so manly I couldn’t even preview the footage in Quicktime. So seeing what I’d shot was tough. Secondly, even the folks with the really butch rigs couldn’t edit the native footage. So everyone was converting the files to an uncompressed format to edit. Or creating proxy files. There were tutorials all over for ways to deal with the problem. It was a problem. Decompressing the files doesn’t take nearly as long as compressing, but it’s just one more friggin roadblock in the way.
Right, so the whole need to decompress is due entirely to the high bitrate h.264 format the camera uses to save video. You’ve maybe seen that name before, maybe not. This is considered a delivery format, because it compresses the data, much the way MP3 files are a lot smaller than the original CD. Or a JPG is a lot smaller than a TIF or…. BMP. Gross. Or a ZIP makes everything else that isn’t already compressed smaller. You might encode your video using h.264 to take up less space on the web for quicker downloads or to fit more on a disc or memory card. They use it on Blu-ray discs.
It’s great because it’s taking up less space and looking great doing it, but it puts a significant load on the computer to decode or decompress. If you’re doing it on the fly, like watching a video or scrubbing in Premiere, it hogs a lot of system resources. If your h.264 file has a really high data rate it can get pretty miserable and if the load is too great you don’t get smooth playback, or any playback. I upgraded my computer after we returned from Alaska specifically to alleviate the preview issue because I came back with a lot of video to shuffle through. But I was still forced to selectively decode all my videos to an uncompressed format before I edited them. All editing suites suffer from this problem regarding these cameras. Every once in a while someone will pop their head our and say it’s fine but I suspect they are lying. For the PC we like to decode to uncompressed AVIs. Final Cut on the Mac likes ProRes. This increases your file size by 10x for the added benefit of hogging lots of disc space and like I said, it’s an aggravating extra step. All bad, but at least it worked and everyone was sharing in the misery. Except now Premiere Pro CS5 has risen above.
Premiere CS5 was just released and the boffins in the lab have somehow optimized it to edit these HDSLR files on the fly without conversion, assuming you have enough ponies under the hood. I’m not saying the Zino HD could do it. Thankfully my desktop machine appears to have the stones. Premiere’s even been optimized to offload some live visual effects and transitions to the video card to end the need for rendering those, assuming you have the correct hardware. I do not. 🙁 But I’m not going to cry. These other developments are far and away exciting enough for me. And I’m not a professional so I doubt I’ll be spending any time or money to fix that.
Actually, I’m fairly certain that most pros would prefer not to be using h.264 in the first place and folks are calling out for a RAW format from Canon all the time. Most professional grade digital video cameras record directly to an uncompressed format to preserve as much data as the camera can capture. One of the ways h.264 saves space is to cut out stuff it thinks you won’t notice. Just like mp3s. It cuts out the highs and the lows to start. In the case of video this is extreme lights and darks. In the case of audio it’s high frequency and low frequency sound typically outside the range of what the human hear can detect but sometimes more. It’ll throw out detail too. So the camera is dumping a lot of info to save space on the memory card by using h.264 but unless you’re a pro you probably won’t miss it.
Anyway, I’m over the moon about this development. More than I probably have a right to given the amount of time I’ve spent taking video lately. But it’s a good thing regardless and it makes me eager to dive in next time. Not so much into the part where the software costs a lot though. Hopefully I can scare up an EDU copy from someone. The actual list of improvements is really long and not as immediately exciting to me but I’m sure it combines all the… rings, together combined…. to stop evil everywhere.
