-
wielders of the silver spears
2010-02-24 09:22:32
SH114:
- Source rip: complete
- Music: complete
- Precleaning: complete
- Storyboard/planning: partial
- Clipping: complete
- Edit: 3:44/5:20
- Postproc: none
- Export: none
Thanks to a tiresome 45-minute plunge through the upwards of 900 cuts in the source pool, I got further than anticipated last night, and the task ahead for tonight is simple: finish verse 7, deal with the 'wall', finish the last chorus, and don't fuck anything up. If this goes as projected, I smash through the part of the song after the lyrics finish on Thursday and then have Friday for tweaking and render. I won't be able to bring this one in under 30 hours when the final mixdown is considered, but finally, this video is going to be done, and based on how it stands currently, is going to be done in a form I can stand behind.
This video has been turning, lately, into a masterclass in composition, specifically on the virtues and drawbacks of the establishing shot. Even if I was going to follow the anime storyline closely -- which I'm not -- this is a storyline that's kind of disjointed, and would suffer more by being compressed from 11-13 episodes down into 5 minutes. I'm having to use a host of locations above and below ground, in a variety of environments, and moving between them usually means establishing shots if there's any space for them musically or conceptually. Then, after the necessity is decided, comes the difficult task of meshing in footage that I essentially don't want in order to set up stuff that I do.
This is more troublesome because as an editor I have a habit of long standing of using environment shots without any characters in them just as a normal part of the video; these days it might be argued (if pretentiously and unconvincingly) that stuff like this represents Massenmensch or the transient nature of humanity or something like that, but in the beginning these shots were used because they looked impressive and there's only so much impressive footage on your average sub tape that doesn't have subtitles scribbled all over it. It is an exercise to the reader to determine how much this is still the case.
--Kai out
-
it becomes my leather
2010-02-23 12:06:49
SH114:
- Source rip: complete
- Music: complete
- Precleaning: complete
- Storyboard/planning: partial
- Clipping: complete
- Edit: 2:43/5:20
- Postproc: none
- Export: none
I didn't get as much done last night as I wanted to, due to starting late and other factors, but I got "enough" done; four more days like this, which is what I have, and the video will be finished and rendered out, and I'm over the halfway mark with less than two minutes until I get through the lyrics and into the do-whatever-I-want break at the end that's really going to tie the video together.
A big part of the obstruction was the first hit of the 'wall'; time wasted first rendering out, then fixing suboptimal stuff revealed by the render, then reprocessing and bringing that in. I'm doing the interim chunks as Huffy rather than Lagarith as a way to get around the random color-barred frames that I occasionally get in Lags source, effectively trading space for time because if I wanted to, I could go frame-by-frame on the large chunks and excise the bad ones, but I don't have the kind of time in the production schedule required to make that happen.
The only other significant difficulty with the process is that every time I check a clip to see if it fits in, Magix has to build an index for it if it doesn't have one already. This is crap and takes time; tonight, I expect to spend a half an hour getting this done for the remainder of the 600 to 800 unique cuts in the pool, which will vastly accelerate production speed and reduce net frustration over the course of the edit process. And hopefully, I'll also get another 45-60 seconds iced, and be in less shape to tear hairs out tomorrow.
--Kai out
-
servants three flesh and blood
2010-02-22 11:42:24
SH114:
- Source rip: complete
- Music: complete
- Precleaning: complete
- Storyboard/planning: partial
- Clipping: complete
- Edit: 2:00/5:20
- Postproc: none
- Export: none
Either this video is going to be finished by the end of the week, or it is not going out for competition. Yesterday was quite productive, and a welcome change from feeling miserable and blowing sludge everywhere; hopefully work will not be all-consuming this week and I can do about a minute a night in order to finish in some remotely timely fashion and get a package mailed Saturday.
A lot of the time spent editing, though, has somewhat surprisingly gone into tweaks and 'effects' work; in the chorus section that went down to where I ended up stopping, I was able to bring in one nine-second cut and a seven-second cut a few cuts after. This is an unusual feeling, and a good one, to have long cuts with good production values and a lot of internal motion flow with and set up the music, rather than having to smash and stretch and deform stuff to make it hang together. And, as mentioned, this was such a long cut, and one that ended up being so deep, that even with the video less than 40% done, it's not totally unrealistic to think about this sort of plotting as the rule rather than the exception.
The most significant challenge going forward is probably going to be just keeping the source volume 'in memory'. Looking at the plotting sheet, individual cuts appear to come in clusters from the same vob, which is probably somewhat unavoidable as regards cut-to-cut consistency in trying to build flow, but whenever that happens I feel pressured to spend more time scanning the source pool: the next cut may be good, but one from three vobs before or after may be *better* in the context of the video as a whole. The ability to settle for 'good enough' pushes the video towards finishing in a timely fashion; the aspiration towards optimal composition and editing hopefully will make it worth watching when that day of final mixdown arrives.
--Kai out
-
rejoice by suicide
2010-02-10 09:30:31
SH114:
- Source rip: complete
- Music: complete
- Precleaning: complete
- Storyboard/planning: partial
- Clipping: complete
- Edit: 0:35/5:20
- Postproc: none
- Export: none
After some bordering-on-non-work time spent timing the song to extremely loose guidelines, a brisk pace due to slow-paced music sees the first 10% of the video done. Maybe I could have pushed on and gotten through the end of the first verse bloc, but it was getting late and I had to get up early to get some work in before Snowmageddon North arrives allegedly sometime this afternoon. Things are looking good, and because of the huge source volume, I have the impetus to go in and mess around with mark-ins to set up more internal synch. The volume being what it is, there is always another shot to bridge the gap in order get things to line up, and at 1.2 spc with this kind of volume, it's probably going to be a good shot, too.
The lone aggravation in this is that Magix, as a rendering environment as opposed to a compositional one, needs smashed in the face. My options for checking flow and fidelity are MPEG2, which plays back smoothly with good audio fidelity and video that has been run over by a bulldozer, XVID, which plays back smoothly with good line fidelity but shit for audio, uncompressed, which has good audio fidelity but can't play back full framerate, and HuffYUV, which doesn't work at all. This environment, not to put too fine a point on it, is simply wonderful at making video that is useless for actual watching purposes. Thank the gods for Clontz and Lee, right?
One way or another, things will get pieced out, and the video will get made, with some combination of laggy editor preview and shit rendered nightly builds.
--Kai out
-
22,000 getto
2010-02-10 07:01:04
All-time usage totals: Downloads = 22000; Bandwidth = 864 GiB
It's not much of a milestone, as it's a little more than 200 downloads average over a sample of 83 videos, most of which have been on local for five to six years, but you take what you can get. If this is an additional 200-1000 anime fans who heard Sentenced or Borknagar or Gamma Ray for the first time, who never would have encountered them otherwise, and got into it, my work is done.
--Kai out
JOURNAL: Kai Stromler (Kai Stromler)
Current server time: Dec 23, 2025 20:14:26