Instrumental Anime Project

[rose4emily]

BTW, Song, the e-mail address I actually check on a regular basis is:

christian D07 paro /\T gmail |)0T com (sans spam-repellant notation).

[jasper-isis]

Hey, hope we're close to the end now.

I'll be lying low because AP week is getting frightfully close , but definitely notify me before you release so I can clear out all the other banners in my sig for this project. Puhehe.

___

EDIT

Just for the heck of it, I'm going to throw out a couple of ideas, all of them Instrumentality-related.

Instrumentality: Electronica
Instrumentality: Percussion
Instrumentality: Solos Only
...
A Capella: the Anti-Instrumentality

[rose4emily]

A Capella: the Anti-Instrumentality

Priceless.

---

AP Week - ah the memories...

They're actually not that bad, once you've finished them. You'll want to spend the next week sleeping, but most people I know have also found that they're scores are better than they thought when they walked out of the tests (that's certainly how it worked for me). For someone as bright as you, they should be no problem.

---

Still waiting on the last narrative. That's okay, though, since I'm also engulfed in a pretty tight crunch time in my academic work - so I really wouldn't be able to work on this all too much right now anyhow.

Last weekend I found a Java program for lip-synching that was a lot like mine - but with a GUI and ability to match phoenemes rather than just going on volume levels (more or less what I proposed earlier as the "proper" way to do what my scripts provided a kludgy version of). Haven't tested it yet, but I might end up using that when I re-generate the existing narrative video and produce the narrative video segments that I haven't done yet.

I've also worked out a pretty simple concept for the widescreen narratives that'll give that part more "dimension" with multiple angles (a close-up on speaker, long view of stage with projected screen behind it, shot from behind speaker looking on audience...). Won't add much at all to editing time, but should keep the viewer from tiring of the perpetual talking head (after having seen quite a bit of that already). When editing it, I can generate the close-up for the full narrative, and the "screen" for the full narrative, and so forth - and then cut them together as though they were redundant footage from multiple cameras on one live speach, which is actually a lot easier than trying to do it all as a sequence of segments.

Still haven't recompiled my kernel to support the really huge file that I'm going to need when doing the join. That on its own shouldn't take more than twenty minutes of my time, and an hour of my computer's time, but I want to wait to do it until I have large enough a block of free time to fix it if I manage to screw something up. RIT's not a place where you can go for very long without your computer in full working order (though, given the nice weather over the past two days, I really wish it was).

I have cleared out a sufficient amount of disk space for the really huge file. Good thing I finally got my DVD burner working a while back, though I also cleared out about 40 gigs of .vob files from DVDs I own and have sitting along the back edge of my desk. I tend to copy the VOBs to my computer to play them (don't like the drive noise, and MPlayer sometimes crashes when I'm deinterlacing and try to skip forward or back on the physical disk), and then forget to delete them when I'm done. A little foolish, really.

---

If it's a while before I next post here, that's because I'm already over my head trying to keep up with a number of other things. That, and I owe my sister a letter (unless her having sent one with return postage somehow wasn't a hint). Or at least a phone call. It's been too long since I've talked to her, so that's what I'll be doing the next time I have a free hour sometime before midnight.

[Songbird21]

Still waiting on the last narrative.

You're kidding...right? Try one of the email addies you don't check regularly. It might be there. I had problems with the email addy you gave me so I got a dif one from Chris. Please lemme know. If it's missing it means I hafta do it over again.

[rose4emily]

Song:

I just checked my old Comcast account, and found the narrative (buried under 164 spam messages - hence my not using that e-mail any more).

Thanks.

[Songbird21]

buried under 164 spam messages

LOL.

[rose4emily]

I was serious about the spam...

How anyone actually profits from the stuff mystifies me, yet it arrives reliably and in prodigious quantities.

---

Anyhow, I now have all of the narrative audio, all of the video segments, most of the narrative video, and the opening credits. I still need to put together a few more narrative video segments. I've also (sadly) discovered that I botched the original end credits scripts, so I need to fix those and re-render the end credits.

---

After a bit of research, I've discovered that my Linux installation already supports files up to a Terabyte or something obscene like that, and that past issues I've had with files over 4GiB were really just specific programs that weren't compiled with LFS (Large File Support) in mind. Therefore, I now know I can write everything to raw YUV and PCM without risking a kernal recompilation or anything bad happening to my file system.

I also installed AVISynth on my laptop, as a "Plan B" for the compilation process, which I can try to get to work while the desktop is churning out an uncompressed copy of all the video segments. It has occured to me that performing the encoding process (as well as the muxing and indexing) on a Windows machine might get the AVISynth approach to work. Always good to have more than one plan for getting something done.

---

Anyhow, I still have my hands full with more math and programming than my mind was meant to handle (though I'm actually really good with the actual "Software Engineering" component of my program, the same can't be said of some of my other courses), so I'm not even guessing at when I'll have a chance to get this done, but I'm making progress when I can - and can say that I'm now the only person this project is waiting on.

In the worst-case scenario, though, my Final Exams will be out of the way in less than a month. I'll most likely have several days off before I start work this summer, so I can confidently say that this will be done in under five weeks. I'd like for it to be less, but it's hard for me to say whether that'll be possible.

[jasper-isis]

Thanks for the update! And... godspeed!

Does "godspeed" even fit this situation? Hehe.

[rose4emily]

Actually, a little question for you, Helen...

When I'm using AVISynth (plain AVISynth, not VirtualDub) to process video, I gather that I'm supposed to feed its output into either WMP for viewing or an encoder for encoding - but I haven't had too much success with guessing at how I'm supposed to do this.

I've found plenty of information on how to write the scripts, between AD's guide pages, the AVISynth website, and the AMV Wiki (which, I notice, you're one of the first contributers to) - but must have missed the part that explains the details of how I'm supposed to run the scripts once they're written.

Anyhow, I figured you'd be a reliable source on this one, since you seem to have some hands-on experience with the tool.

---

Best of luck with your escape from AP testing hell.

[jasper-isis]

Ah, just load them into VirtualDub and encode from there.

You might have missed this page:
http://www.animemusicvideos.org/guides/avtech/xvid.html

Basic steps, though they're just repeating what's in the guide:

- Open the AVS script in VirtualDubMod just like you would any regular video
- Make sure "Fast Recompress" is checked on the video menu
- Set up your compression settings as described in the guides (the page that I linked to)
- Go to the Streams menu and click on "Stream List"
- If you're using a video stream w/o any audio, then add the audio stream now
- If the audio is still a WAV file, then choose its compression settings by right-clicking on the stream name and choosing "Full processing mode," then right clicking again and choosing "Compression"
- Start the encoding processing by going to the File menu and clicking "Save as..."

That's basically it, though I might have missed something because I was hurriedly typing most of it up from memory. xP

If you want to reap the full benefits of using AVISynth scripts in VirtualdubMod, check out this page:
http://www.animemusicvideos.org/guides/ ... mod.html#5