|
Post by obsessive24 on Nov 20, 2011 6:03:50 GMT -5
That bit about rewatching the source and your clipping process is SO relevant to my interests. It's really good to hear that I'm "doing it right" with the clips, notes, and storyboard-esque process. Well, I think whatever works for you is "the right way". Everyone has a different process and I think that's really coming through in all the WIPs we've got posted here. I think the emotional flow of the Robb and Jon parts will be better once I fill in some of the black space with more of their storylines from the later episodes, because right now they are very minor presences who don't really get arcs. Hard to tell without seeing the clips actually on the timeline, though, so my first priority is to finish sourcing today.
|
|
|
Post by rokikurama on Nov 20, 2011 12:39:47 GMT -5
This is really fascinating, and I love the idea of editing as you go, pulling in one episode at a time. I'm very obsessive, and so in my first vid I waited until I had clipped every single episode of my series before starting to edit and almost lost hope that I would ever get to the edit stage! (I ended up with, ah, 65 GB of footage for a 1 minute vid) I'm trying your process of building the clip library and timeline at the same time for my Festivid, and it's much more fun!
I'm not sure if this is useful feedback or not, but the direwolves are really jumping out at me from your vid. I think it's because that beginning scene of all the Starks getting their direwolves comes right at the start and seems hugely important since it's intercut with the execution (and they're such beautiful creatures, zomg). I wonder if finding scenes of the Starks with their wolves might be helpful in tying some of the later footage into the overall emotional arc since I found myself actually looking for the direwolves as the vid went on.
|
|
|
Post by obsessive24 on Nov 20, 2011 12:42:33 GMT -5
Roki - I'm making this vid for my husband, and wolves are very much this thing, so I'm trying to pull them in as much as possible. And of course they're important for the characters/themes too. The difficulty is that there are far fewer viable wolf clips in the show, compared with their large presence in the books, so it sort of comes down to using what's available.
|
|
|
Post by rokikurama on Nov 20, 2011 12:47:17 GMT -5
Ah, true. Silly HBO--don't you know we all loved the direwolves? I think my girlfriend was actually more upset at Lady's death than Ned's!
|
|
|
Post by obsessive24 on Nov 20, 2011 13:37:36 GMT -5
Working Day 4Stream WD4 vid draft at Vimeo (password: WD4) Screencap of timeline at end of WD4 (Full timeline \o/) I exported to lagarith AVI as usual, but this time I want to do post-production colouring in avisynth/Virtual dub, so the next step is: Create an avisynth file. Eunice has a great run-through of the basics using AVsP here and the principles are the same, although for mine I just opened a notepad .txt file, typed in the following commands, then changed the filetype to .avs. avs script1. AVISource() - points to the source. 2. ConvertToYUY2() - something about colourspace. Not sure, to be honest, but I think I need to put it in, otherwise Tweak() won't work. 3. Tweak() - where I specify saturation, brightness, contrast etc. A good guide to Tweak() is here and lets you know the range of commands that are available. I just wanted to subtly intensify saturation/contrast, so the result isn't hugely dissmilar to source. Then I just dragged and dropped the .avs file straight into Virtualdub. I can run a XviD encode from there, but I don't like manually doing stuff in Vdub, so I ended up just re-exporting to Lagarith AVI, then loading that file into AutoGK as usual. Today I sourced the remaining 1x08 - 1x10, also went back to earlier episodes to find some clips to fill in gaps in the timeline. Time spent = ~5 hours. The most important decision re emotional flow was that I gave up trying to parallel Robb and Jon at the end. Robb's rise was the emotional centre this season, so I put him front and centre, and let Jon recede into the general family background. Brought in an Old Gods vs New Gods motif as I'd wanted, but just a light touch. There's not enough footage to really push that notion. As a first draft full timeline, I'm actually pretty pleased with how the narrative and emotional flow shaped up, and there's nothing that I think needs immediate change. The next step is to sit on this draft for a few days and let any "impurities rise to the surface", so to speak. I'll have a much clearer idea of what I still need to tweak after I watch it a few more times. I don't anticipate any significant structural changes. There are a couple of things that I'm still iffy about (is the Lyanna & Jon bit too much of a digression, there's no Rickon/Shaggydog at all because there's woeful little viable footage), but I think trying to "fix" these things is likely to make the vid worse, not better. Again, I'll think on it for a few days before making a final call.
|
|
|
Post by legoline on Nov 20, 2011 14:43:40 GMT -5
Okay. WOW.I was unconsciously holding my breath while watching this. Seriously. I mean. This isn't even finished and it's about a thousand times better than anything I could ever throw together The editing! Flawless! Frickin' flawless. The way it moves with the song...yes, I'm all flaily hands; I can't help myself. This is going to be so, so good. Just --> (I think you integraded the wolves really nicely, btw)
|
|
|
Post by obsessive24 on Nov 20, 2011 18:18:26 GMT -5
I'm so glad it worked for you!
|
|
|
Post by gnattery on Nov 21, 2011 6:36:57 GMT -5
I'm loving seeing your cuts get more and more accurate with each update. I have a bad habit of putting a few clips down, and then spending ages getting them exactly perfect, before I work on anything else. Need to remember to just throw things at the timeline and fill it up before I worry about hitting everything exactly right.
Also the new clips of Ned at the start look exactly like a section from my half-finished GoT vid. Snooze you lose. >_<
|
|
|
Post by winterjasmine on Nov 21, 2011 7:19:49 GMT -5
Eunice - that's a really good point, and something that I think I haven't fully articulated, even though I embrace it in practice, in vidding and in writing - especially writing. Whenever I have to write something - whether it was an essay for uni or a fic or a business paper - I draw up a rough outline, type up all the chapters/headings, and then I pretty much just close my eyes and type all that I know, and really just spew it out there with no form and format. (In the case of fic, pre-writing alcohol consumption also seemed to help...) I've found it's so much easier to go back and edit something that's already there than it is to put something onto the blank page. It doesn't matter how rough it is and how much you're gonna have to shuffle things around and trim the fat and make it look pretty, having something down feels a lot better than having nothing. I think that's also true for vidding - it works for my personal process, anyway. This is like NaNo... NaNo for vidding... what a great idea. Once November is done I'm totally going to try this with my next vid. Vidding, NaNo style. Heh Jaz I feel like I should have a tag, 'the contents of this post may or may not make sense, it is week four of NaNo after all.'
|
|
|
Post by obsessive24 on Nov 21, 2011 7:22:06 GMT -5
gnat - It's sort of a side effect of having an imperfect timeline preview. Sometimes the preview is jerky and stuttery even when I pre-render, and at other times it looks right in preview but then comes out a little off in the export. So I usually just try to fix it the next day after seeing the export, instead of trying to get it perfect on the timeline preview straight away. Which kind of ties in neatly to a question someone asked earlier somewhere about why I'd do test exports at all.
Jaz - exactly the same approach I took in the one year I did NaNo, in point of fact. ;D Subsequently took me three years to fully clean up that 50,000 words, LOL.
|
|
|
Post by winterjasmine on Nov 21, 2011 7:27:38 GMT -5
Lol! I'm beginning to think this years attempt is going the same way. I keep making additional notes about stuff to insert earlier. That file is growing at an alarming rate! Anyway, enough distraction, I've got another 4000 words to get down today Jaz
|
|
|
Post by obsessive24 on Nov 26, 2011 10:53:06 GMT -5
I posted the vid, but was just doing my export to DVD files (for con vidshows etc), and thought that would be useful to document as well. Exporting to m2v/wavSo now I've got a final vid that I want to convert to con format compatible m2v/wav. The specs are as follows: Existing input file: Lagarith AVI, 624x352, 23.976fps. Desired output file: m2v/wav pair, 720x480, 5500kbps, 4:3, 29.97fps (NTSC), 48kHz audio. If you have a working version of Vividcon's official conversion software LlamaEnc, it'll do everything automatically for you. Unfortunately, it doesn't always work with newer versions of Windows, so I have to resort to a more manual process. I use freeware HCenc. Firstly, I have to create an avisynth file to load into HCenc. This is the step where I need to get the aspect ratio right, because unlike LlamaEnc, HCenc doesn't automatically convert your vid to 720x480 at the right aspect ratio. So the script I use is: Script screencapI've already explained AVISource(), ConvertToYUY2() and Tweak() earlier. These reference the AVI file and deal with colour adjustment, not relevant to the DVD encoding process (except of course for AVISource().) LanczosResize(720,352) increases the width to the desired 720x. However, it's important to note that I don't change the height from the original 352x. Why? Because math. AddBorders(0,64,0,64) makes the total height to the desired 480x (being 352 + 64 + 64 = 480). This gives the letterbox bars on top and bottom, making the whole thing 4:3 whilst maintaining aspect ratio. And then I load the .avs into HCenc, like so: HCenc specsIt's another one of those things that look horrible and complicated, but you don't actually have to do much. Just leave things at default unless you know what you need to change, as below: Find your .avs as the input file and specify an output file name. Bitrate is important for con submission because too low and the video quality suffers; too high and the video will play jerkily. Vividcon prefers bitrate to be between 5500 - 6500kbps. So here I specified an average of 5500, and a maximum of 6000. Aspect ratio - specify 4:3, since that was how I approached it in my .avs script. You can do anamorphic 16:9 if you want to, but that would mean different numbers in avisynth with LanczosResize() and I'm not familiar with the details and it scares me. One last important thing: to get the framerate from 23.976 to 29.97fps, you need to tick the box under Miscellaneous called 3:2 pulldown. The theory behind this is long and involved and you can read all about it on Wikipedia, but as usual, let's just accept the math magic and move on. ;D Then just hit Encode and you get a shiny m2v file that is con-compliant! As for the .wav, I just directly exported that from Vegas. As long as the export loop is set at exactly the same specs as the AVI/m2v, everything should sync up perfectly. Don't forget to set the sampling frequency to 48,000Hz if it doesn't default to that already.
|
|
alba
New to the Pub
Posts: 12
|
Post by alba on Feb 19, 2012 19:52:01 GMT -5
This was really interesting. Thanks for sharing your process. I wish I used your program, so I could use the technical parts of this (I use Final Cut). I've been worried that I waste a lot of time, the way I do it. First I go through all my source material and make all the clips that I think might be relevant, which is apparently the direct opposite of how you do it. I save all these clips, so hopefully I can use them again in another vid. As you can imagine, this takes forever and I probably would have made more vids by now if this wasn't so labor intensive. Then I import them into my editing program and build up the timeline. But I do it in that picky way of making each part perfect before continuing on - again, the opposite. I always have to go back and make more clips, of course.
I'm going to try using an additional track for other optional clips. Like that idea!
|
|
|
Post by obsessive24 on Feb 20, 2012 7:39:04 GMT -5
Alba - I certainly know a lot of people who'd do it your way - build up a clip library first and only start editing once the clip library is complete or almost complete. You're certainly not the only person who vids this way. I guess the main takeaway is to know that there are other ways of doing it, and you don't have to do it any particular way - just whatever feels best for you and works for your program.
Certainly if I'm making a vid with a more defined narrative, or for lots of source, then I'd first rewatch the source, take notes and storyboard before I start editing, but like you said it's a much more time-consuming process. Depends on the vid.
|
|