Starting my first video...hopefully (plus a question!)

[badmartialarts]

This four-day weekend I'm going to start working on my first video.

Song: ATWA - System of a Down
Anime: Brigadoon

While I watched Brigadoon, I kept seeing scences that went pretty well with this song, so I decided to make it my first try at AMV creation. I don't forsee any need for expansive digital effects, just a few fades, and straight cuts. Maybe some fancy stuff for the parts where the style of the song changes...not sure. Currently, every time I sit in front of my computer, I put ATWA on continuous repeat, so I'll have a good idea of where every hi-hat, cymbal, lyric change, etc. falls before it happens. (I'll also be insanely sick of ATWA!) I've also been watching many, many AMVs to see how others time things (some do it well, some...eh...). Problem is, I'm such a perfectionist, I'm afraid I'll never finish this vid. And there's like 3 more ideas floating in my head, too (a comedic Brigadoon idea, a G Gundam action vid...and a really cool Initial D idea...if I can get my friends DVDs). So, this four day, when I'm not busy, I'll be feverishly editing away.

So, to the reason I posted. The guides on this site are pretty informative on technical aspects of video creation. But the personal side of video creation is barely mentioned. How do ye great and all-knowing anime music video creators go about sitting down with an idea (Hey, this song would go perfect with...) and actually pick and choose from your vast, comprehensive collections of footage the scenes that will comprise your testament to your genius? Since I'm asking for personal opinions here, I realize they may or may not help me. However, they may give me (and other neophyte AMV creators) something to work with to build future projects upon. (Particularly my G Gundam idea...12 DVDs of footage for a 3 minute video...gonna take a while to watch it all again).

[mexicanjunior]

How do ye great and all-knowing anime music video creators go about sitting down with an idea (Hey, this song would go perfect with...) and actually pick and choose from your vast, comprehensive collections of footage the scenes that will comprise your testament to your genius?

I'm not a great editor but my vids go like this...

lyric = scene

beat = scene change

wash, rinse, repeat...

[badmartialarts]

That question wasn't really clear (the perfectionist in me). To rephrase: When you are making a video, where do you put the cuts, and how do you select the scenes? For example, in a passage in ATWA:

"I don't see....ANYMORE!
I don't hear....ANYMORE!
I don't speak...ANYMORE!
I don't feel..."

Currently I want to make cuts between each statement and the trailing ANYMOREs. But the footage I find might dictate otherwise. I don't wanna spoil Brigadoon, so I don't want to talk about scenes I'd use for each line. The problem is the ANYMOREs. I want to create a distinction in mood, yet footage may not be available to cut to that makes a distinction in mood without compromising the meaning of the line...it might be the perfect place for a digital effect of some sort (a washout of color, maybe....). Basically, what I'm asking for on this thread is for other, more experinced AMV creators to perhaps share situations where they've been unsure of footage or cut or effect choices, and how they resolved them, making a better (or worse!) video as a result.

Gosh, does that make any sense? Maybe I'm asking for too much.

[dokool]

Currently, every time I sit in front of my computer, I put ATWA on continuous repeat, so I'll have a good idea of where every hi-hat, cymbal, lyric change, etc. falls before it happens. (I'll also be insanely sick of ATWA!)

This is a good start. To make an AMV, you must know your song.

So, to the reason I posted. The guides on this site are pretty informative on technical aspects of video creation. But the personal side of video creation is barely mentioned. How do ye great and all-knowing anime music video creators go about sitting down with an idea (Hey, this song would go perfect with...) and actually pick and choose from your vast, comprehensive collections of footage the scenes that will comprise your testament to your genius?

First, I make a script, with columns for scenes, notes (like effects or transitions), lyrics, and song notes. (IM me and I can send you an example of what a 'blank' script looks like)

Then, I watch through the anime... sometimes fast forwarding to look for a scene, other times (For FLCL, for example), I'll watch it all the way through. As I watch the show, I'll go through my script and pencil in scenes and chapters.

Sometimes it's a bit more accellerated. For RNITTBS, I fast-forwarded through Lain at about 20x (it was on my DVD-Rom), jotting down chapter numbers as I noticed scenes with Lain on the computer. Then again, for RNITTBS I didn't have a script, so that's besides the point.

Anyway, at the end of this "pre-viewing", you should have a well-worn script that's full of scenes. If your handwriting's as bad as mine, you should *immediately* transcribe it into your computer so you don't look back later and wonder what the hell you were talking about.

So, once you have this down, get to work. Now that you know which scenes you're using, you only have to rip certain chapters, and that saves space.

Of course, remember that as your script was written in pencil and transcribed into pixels, that means it's not carved in stone. If in the middle of your editing you realize that "this scene would fit so much better there" or "i want to change this", go ahead. Planning is good, but trust your instincts.

-DOKool

[LantisEscudo]

Well, I'm not one of the "ye great and all-knowing AMV creators" or even well known here, given how quiet I usually am, but here's the method I use, which works quite well for me. Hopefully you'll find some use in my describing.

First, I find a song with lyrics that can match the anime I want to use (I tend to single-anime videos for this reason). I mean more than just one or two lines, but several groups of lines. Often, listening to the song will suggest a series to use. Even better is when I can find a unifying concept for the entire video.

Then, I get drop my audio file into my editor and scan through the song, marking the beginnings of lines and instrumental cues I want to change clips on. I export that list of times to a text file, with [Instrumental] or the corresponding lyrics beside the times.

Next, I sit down and watch the entirety of the series/movie I'll be using, writing down every scene I might want to use in the video on paper, with episode/time notes so that I can easily find them when it's time to actually lay video on the timeline. This is probably the second most time-consuming step, and uses a lot of paper (my notes for a scrapped Fushugi Yuugi video took eight pages of college-rule paper, both sides, two columns per side). Keep these notes afterwards, you might find them useful in future projects. At least one of my videos ended up using the written notes from a previous dead project, saving me quite a bit of time.

Then comes the biggest step: actually picking the scenes. Using my paper notes and the text file of timings, I try to pick the best scenes for the segments of the song. I also storyboard any digital effects I want to use here (including whether to fade or cut between the clips). This is what takes me the most time, and is where more than one planned video has died when I realized I really can't make the video work.

Only then do I actually start ripping the video and putting it on the timeline in MSP (I don't have Premiere).

Once I've got the whole thing laid out and I sit down to watch the beta render, I always find that there's something I don't like: a bad transition, bad scene choice, a mistiming, whatever. When this happens, don't be tied to the plan. Change it if you don't think it's right. It helps to have someone else watch your betas, since they might see something that seems off that you don't.

I hope that helps some. I'm off, I have the rest of a series to finish taking notes on.

[risk one]

Not one of the greats (or known) either, but I tend to work with less preparation than most, I think.

On my last video I first made sure that I knew the music well, and started out from the basic make-up of the song. Which would be intro-verse-chorus-verse-chorus or something on a pop-song, but I had a classical tune so it was loud&complex-slow&romantic-loud&complex. Then I fitted a general theme to it bringing forth the element of the anime that I wanted show. All this still taking place in my head, btw.

And then I fired up premiere. Already? I hear you think, yes already. I think if you're going to make sketches and storyboards, you might just as well implement it on the timeline and see the results rightaway. I think it's a bad idea to play a song so often that you can dream it, because music (for me anyway) loses meaning if can't surprise you anymore, I think thats a big part of what makes music special and an even bigger part of what makes amv's special. So if you haven't heard the song to often you can still feel what the important elements are for someone who hears it for the first time (ie you viewer).

For each individual bit I just closed my eyes, listened to the music, and tried to imagine what the video would look like if I could create my own scenes. From that mental image you can really get a lot of information about what beats or melodic stuff to pay attention to, and which kind of motion would be cool. (Actually this is not so much what I did, as what I should have done).

[Chaos Angel]

I don't have a system. I just put down whatever occurs to me might work in whatever part, then tweak it so it works for me. I worry more about having the idea solid in my head and I worry about scenes and transitions and such like largely on the fly.

Frankly, it sounds like you already do more preperatory work than me, so I don't even know why I am bothering posting here.

[nailz]

lyric = scene

Where oh where did you ever find a starfish on a beach? Asshole.

[Zarxrax]

I find that just trying stuff is the best way. I can plan all I want, but it's never gunna work out how I planned it to. Not to say that you shouldn't plan... this is still very important... just that your plan can evolve quite a lot once you actually start working on the video. The best way to choose how to time scenes and stuff is to just put something down... look at it... decide if you like it or if it sucks. If it sucks, then change it. If it continues to suck then continue to change it. If it STILL sucks after a while then reevaluate exactly what you are trying to do there, or try to consider why its not looking good.