More perspective on video editing

I read Bill Machrone’s current PC Magazine column on PC non-linear video editing with a bit of bemusement. He talked about the difficulty he and his son have editing video on their PCs, and he concluded with the question: “How do normal people do this stuff?” and the misguided answer: “They buy a Mac.”
You don’t have to do that. In fact, you can do pretty well on a PC if you just play by the same rules the Mac forces you to play by.

Consider this for a minute: With the Mac, you have one motherboard manufacturer. Apple tends to revise its boards once a year, maybe twice. Apple makes, at most, four different boards: one for the G4 tower systems, one for the iMac, one for the iBook, and one for the PowerBook. Frequently different lines will share the same board–the first iMacs were just a PowerBook board in an all-in-one case.

And the Mac (officially) supports two operating systems: the OS 9 series and the OS X series. You keep your OS at the current or next-most-recent level (always wait about a month before you download any OS update from Apple), and you keep your apps at current level, and you minimize compatibility problems. Notice I said minimize. PageMaker 7 has problems exporting PDF documents that I can’t track down yet, and I see from Adobe’s forums that I’m not the only one. So the Mac’s not always the bed of roses Machrone’s making it out to be.

Now consider the PC market for a minute. You’ve got two major CPU architectures, plus also-ran VIA; 4-6 (depending on who you ask) major suppliers of chipsets; at least four big suppliers of video chipsets; and literally dozens of motherboard manufacturers. Oh, you want an operating system with that? For all the FUD of Linux fragmentation, Microsoft’s in no better shape: Even if you only consider currently available offerings, you’ve got Windows 98, ME, NT4, 2000, and two flavors of XP.

So we go and we buy a video capture card and expect to put it in any old PC and expect it to work. Well, it probably ought to work, but let’s consider something. Assuming two CPU architectures, four chipset manufacturers, four video architectures, and twelve motherboard manufacturers, the chances of your PC being functionally identical to any other PC purchased right around the same time are 1 in 384. The comparable Mac odds: 1 in 4. But realistically, if you’re doing video editing, 1 in 1, because to do serious video work you need a desktop unit for its expandability. No Blue Dalmation browsing for you!

So you can rest assured that if you have a Mac, your vendor tested the equipment with hardware functionally identical to yours. On a PC you just can’t make that assumption, even if you buy a big brand name like Dell.

But you want the best of both worlds, don’t you? You want to play it safe and you want the economy of using inexpensive commodity PC hardware? It’s easy enough to do it. First things first, pick the video editing board you want. Next, visit the manufacturer’s Web site. Pinnacle has a list of motherboards and systems they’ve tested with the DV500, for instance. You can buy one of the Dell models they’ve tested. If you’re a do-it-yourselfer like me, you buy one of the motherboards they’ve tested. If you want to be really safe, buy the same video card, NIC, and SCSI card they tested as well, and plug them into the same slots Pinnacle did. Don’t worry about the drives Pinnacle used; buy the best-available SCSI drive you can afford, or better yet, two of them.

Video capture cards are cranky. You want a configuration the manufacturer tested and figured out how to make work. Otherwise you get the pleasure. Or the un-pleasure, in some cases.

As far as operating systems go, Windows 2000 is the safe choice. XP is too new, so you may not have drivers for everything. 98 and ME will work, but they’re not especially stable. If I can bluescreen Windows 2000 during long editing sessions, I don’t want to think about what I could do to 9x.

And the editing software is a no-brainer. You use what comes with the card. The software that comes with the card should be a prime consideration in getting the card. Sure, maybe an $89 CompUSA special will do what you want. But it won’t come with Premiere 6, that’s for certain. If I were looking for an entry-level card, I’d probably get a Pinnacle DV200. It’s cheap, it’s backed by a company that’ll be around for a while, and it comes with a nice software bundle. If you want to work with a variety of video sources and output to plain old VHS as well as firewire-equipped camcorders, the DV500 is nice, and at $500, it won’t break the bank. In fact, when my church went to go buy some editing equipment, we grabbed a Dell workstation for a DV500, and we got a DV200 to use on another PC in the office. The DV200-equipped system will be fine for proof of concept and a fair bit of editing. The DV500 system will be the heavy lifter, and all the projects will go to that system for eventual output. I expect great things from that setup.

The most difficult part of my last video editing project (which is almost wrapped up now; it’s good enough for use but I’m a perfectionist and we still have almost a week before it’ll be used) was getting the DV500’s video inputs and outputs working. It turned out my problem was a little checkbox in the Pinnacle control panel. I’d ticked the Test Video box to make sure the composite output worked, back when I first set the board up. Then I didn’t uncheck it. When I finally unchecked it, both the video inputs and outputs started working from inside Premiere. I outputted the project directly to VHS so it could be passed around, and then for grins, I put in an old tape and captured video directly from it. It worked. Flawlessly.

One more cavaet: Spend some of the money you saved by not buying a Mac on memory. Lots of memory. I’m using 384 MB of RAM, which should be considered minimal. I caught myself going to Crucial’s Web site and pricing out three 512-meg DIMMs. Why three? My board only has three slots. Yes, I’d put two gigs of RAM in my video editing station if I could.

OK, two more cavaets: Most people just throw any old CD-ROM drive into a computer and use it to rip audio. You’ll usually get away with that, but if you want high-quality samples off CD to mix into your video production, get a Plextor drive. Their readers are only available in SCSI and they aren’t cheap–a 40X drive will run you close to $100, whereas no-name 52X drives sometimes go for $20-$30–but you’ll get the best possible samples from it. I have my Plextor set to rip at whatever it determines the maximum reliable speed may be. On a badly scratched CD sometimes that turns out to be 1X. But the WAV files it captures are always pristine, even if my audio CD players won’t play the disc anymore.

Editing my second video…

You know it’s a different kind of church when you see one making music videos. You’re probably not too surprised to hear that’s the kind of church I go to. And you’re probably not too surprised to hear I’m involved.
I spent a healthy chunk of time Monday editing video. A local radio personality recorded a version of “Mary Did You Know?” a few years back. I know, that doesn’t sound good, but his version is pretty powerful. I’ve heard several versions of it, and I think I like his best, and I’m not just saying that because I know people who know him. I’m also not just saying that because he gave us permission to use the recording. If that version wasn’t good, I’d have assembled a band to re-record it–one of the guys in my Bible study group plays guitar, and another one of them plays drums and has a recording studio in his basement.

So anyway, I’ve got a song I can legally use, and we secured permission to use a couple of different movies about Jesus so we’d have some footage to put to the video. And I gave myself a crash course in Premiere. Put the emphasis on “crash,” because I did bluescreen 2000 at one point. I muttered something about toy operating systems and got back to work. I hope Adobe eventually gets a clue about Linux–there’s plenty of proprietary, high-end video stuff out there for Linux, but nothing in the prosumer arena yet. And I do believe that if you build it, they will come.

After too many hours, I had something halfway workable. Since I was dealing with professional footage, I had a giant headstart. My partner in crime, Brad, had written up an outline that I more or less followed. There were one or two minor points where I didn’t agree with him about where the video fit, so I changed them, but I’d say I went with his outline 75% of the time, if not much more.

So I called Brad and asked him if he wanted to come over. I figured out how to get my DV500 to output to my ancient Commodore composite monitor, which was a good thing, The video was showing up much too dark on my computer screen, but when I exported to NTSC it was beautiful. I’d been playing with levels trying to get it right; I ended up just undoing all of the changes.

What I had can’t be considered finished product; the transitions are pretty lame where there are any at all, and I had a couple of gaps where I didn’t have any video that fit so I threw in a Rembrandt painting. Then I noticed that it didn’t matter what you did to the color on a Rembrandt painting; it still looked far better than any video I’ve ever seen, so I went looking for other Rembrandt paintings to put in. So the video was substantially done, but there’ll be minor changes.

It blew Brad away. I’ll admit, I learned from our first video, so the big mistakes that were in the first video aren’t in this one. And Premiere has great tools to help you avoid those mistakes–you can set the timeline to show every single frame in the video, and to show the waveform of the audio, which takes the guesswork out of transitions and lining things up.

At the end of it, Brad turned to me. “Dave, you are an artist. Do you know that?”

I’m not so sure about that one. Brad’s my ideas man. He tells me what he sees in his head, then I try to find a way to somehow put it up on the screen. And every once in a while I’ll get a better idea. Those are usually 3-4 seconds long. So then I revert back to his. And the result is something that looks decent. Plus a number of the things that happened were just accidents. I had some video of Jesus and the disciples walking through a field with some sheep in the background. I threw it in for lack of anything else to put there. Then about the 10th time I’d played through that sequence–you do a lot of playback during editing–I noticed that during the line “Did you know that your baby boy was Heaven’s perfect lamb?” Jesus happened to look down–towards a lamb walking past. I’d be pretty impressed if someone else put that subtle detail in there. But this was an accident. Or, more likely, it was God doing me a favor.

It’s been a lot of work, but a lot of fun.

And, incidentally, if you ever find yourself having to do any video production, Premiere 6 is an excellent product. I really dislike Adobe as a company, and I wish there were a better product out there than Premiere 6, but I sure haven’t found it. At $250, the Pinnacle DV200 bundled with Premiere 6 is a steal. If you’re into home movies and already have a camcorder with a firewire port (or are considering one), a DV200 and a little time will give you the snazziest home movies on the block.

Back behind the camera again

Yesterday was interesting. I’ve been learning how to operate a video camera. I’m not talking a $399 camcorder from Best Bait-n-Switch here. Those are toys. I’m talking a real, live, camera, like you’d find in a TV studio or a sporting event. A camera that gathers up all available light possible, encodes the picture digitally on ridiculously expensive mini-DV tapes, and gives an absolutely gorgeous picture in most possible conditions, provided there’s a competent operator behind it who can find things compelling enough to shoot and keep them in focus. No autofocus here. It displays what you shoot with it, how you shoot it, whether idiotic or inspired.
I’m also afraid to take the thing anywhere because it costs half as much as my car. (No, I didn’t pay for it. And no, it’s not mine.)

I’ve shot video with it exactly four times now. One of those projects never saw the light of day. Two of the others won’t see the light of day until we get some semi-professional-grade video editing equipment. Yesterday was the first time my camera skills (or lack thereof) were on display, because I wasn’t taping–what my camera saw was projected, live, onto TV monitors throughout the building and two huge projection screens.

I hate TV, so I’m a bit surprised that I enjoy doing this. I guess it’s not video that I dislike after all–it’s how video is misused that I hate. Sticking a microphone in some distraught soul’s face after a tragedy and asking how they feel, then putting it on TV where some game show host-types can comment on it and act sad. Cheap, mindless, crude, lowest-common-denominator art.

The first thing I noticed is that the viewfinder is fine for finding yourself crudely, but under these lighting conditions it always pretty much looks in focus. I haven’t exactly developed an eye yet to tell from the viewfinder how things will look. We’ve got a little 9″ composite monitor hooked up to it to help–what you see there is roughly what you’ll see on TV. It’s a higher quality display than some cheap TVs, so if it looks good there, it’ll look good on any TV. Those projection screens are a challenge. They’re higher resolution than TV, so a slightly out-of-focus picture that still looks perfect on the monitor can look like garbage on the screens. It’s disconcerting to get what looks like a great shot, then cut to it and see it look awful on the screens. All I can do is make a quick adjustment off the screen and go.

I approach camera work from a print journalist’s perspective. When you’re recording events, you develop a gut feeling for what people find interesting, whether your medium is the printed or spoken word or still or moving pictures. That’s learned but can’t really be taught. From my design classes I know that the visual center of everything is slightly up and left of center. When you center something perfectly, it doesn’t look quite right. Also from design classes I know the rule of thirds–divide a picture into thirds, and those intersections of the lines are the points that people generally find interesting, so when you compose your shot, you want to put the important things in those areas. Once you know what you’re doing, you completely forget the rule of thirds and go with your instinct. I’m not there yet.

A lot of the camera work is a no-brainer. You find the person who’s speaking, then keep the speaker in the frame and in focus. But at one point the speaker focussed on an object, so it was pretty obvious–to me at least–what to do. Get off the speaker and focus on that object for a minute. The problem was, that object was directly behind the speaker, and huge! Switching focus smoothly from the upper three feet of the speaker to a 15-foot object a few feet above and behind him… Well, I couldn’t do it too well. I shot violently upward with the camera, and violently back, and I still didn’t have the object filling the frame. I moved a bit more smoothly back, and then somehow got it in focus. I stayed there for a minute, then I trained back on the speaker. In retrospect, I should have cut back even further, brought him into the frame, then slowly zoomed back in on him. I’ll do that next time.

But that was an inspired move, I think. The speaker wanted the audience to focus on that object. If they weren’t focusing on it before, when they saw the movement on the projection screens, I got them to pay attention to it, if only for a moment. People instinctively pay attention to things that move, particularly when they’re focussed on something else. That’s one of the reasons why animated ads on the Web are so annoying. It’s hard for us to focus on anything else when ads are blinking at us.

Maybe I’m being pretentious. I was shooting a church service. Many in my audience see 50 of them a year. A fair number of them see more than 100. But I don’t like being competent at something. I don’t even like being good at something, usually. I want to be one of the best, so I’ll try to do the things that the best do. Given the choice between looking incompetent trying to do something the very best would do and looking competent by not taking any chances, I’ll take my chances each and every time.