| === irvy is now known as irv | ||
| ScottL | hi troy_s , i noticed you did some nice work for ubuntu-women | 03:57 |
|---|---|---|
| ScottL | hey, i'm also doing to dive into blender and do some video i think | 03:58 |
| troy_s | ScottL: Thanks for that. They asked me to give it a kick so I did. | 03:58 |
| ScottL | i'm working on a song and i'll try to make a video of myself standing around my room using node compositing | 03:58 |
| troy_s | ScottL: Lyz published the wrong still photo though. Slightly modified. | 03:58 |
| ScottL | troy_s, well, it looks real nice, elegant even..refined :) | 04:01 |
| ScottL | troy_s, i poked around you website because i saw the blender tag and i saw one of your pictures that showed all the nodes | 04:02 |
| ScottL | troy_s, and this tied into other confluencing influences and i'm given blender a wack | 04:02 |
| ScottL | troy_s, trouble is my video card, ati stopped supporting it so the new drivers don't work with it | 04:03 |
| ScottL | troy_s, and the old drivers which do work with it don't work with xorg1.6 :( | 04:03 |
| ScottL | i installed karmic on a spare hard drive and viola, all the drivers and xorg line up and work and my blender works nice and smooth and punchy | 04:03 |
| troy_s | Yikes. | 04:04 |
| troy_s | Well... Blender has some issues when you get into motion picture work. FFMPEG is the name of the issue. Lol. It works fantastic once you get a custom compile up and running, but getting there can be a bit of a PITA. | 04:05 |
| ScottL | troy_s, in the video i'm hoping to have myself playing in front of the computer for the rhythm guitar and other "me"s poking in and out for vocals, lead, back up vocals | 04:05 |
| troy_s | ScottL: Ubuntu's ffmpeg is pretty badly behind. | 04:05 |
| ScottL | troy_s, will using frames of still instead of video help? | 04:05 |
| troy_s | ScottL: Relatively intermediate level as a guess. You basically need to do a few things. If you are interested in knowing what they are, ask. | 04:05 |
| ScottL | troy_s, i am interesting but, alas, for the moment i must attend to my extremely tired, whiny son...man, he's really tired, and therfore really whiny | 04:06 |
| troy_s | ScottL: Well you are dealing with a few things. First thing you need to do is get to a final picture lock aka final cut. Which means cutting all of your footage such that the timing is all perfectly established. You can use opacity to get a feel as to how the 'popups' will look. | 04:06 |
| ScottL | troy_s, keep typing i'll catch back up in a bit | 04:06 |
| troy_s | ScottL: Ping me. I'll step you through it. | 04:07 |
| ScottL | troy_s, cool, it'll probably be twenty minutes (give or take) | 04:07 |
| troy_s | ScottL: No rush. | 04:08 |
| ScottL | troy_s, oh, have you heard from cory lately...i pm'ed me a while back and i've tried to catch him on irc, email him (thought about smoke signals) but he's not responding :/ | 04:08 |
| troy_s | ScottL: He occasionally pings me on IRC when all the planets align. | 04:08 |
| troy_s | ScottL: But I think he is avoiding the whole scene as he likely knows he falls back into the timesuck mode. | 04:08 |
| troy_s | ScottL: And the family takes a beating. | 04:09 |
| troy_s | ScottL: (Not a literal one, LOL) | 04:09 |
| ScottL | troy_s, okay, i'm back...tell me about blender please | 04:26 |
| ScottL | troy_s, it's just weird that he pinged me and then isn't responding...and a little frustrating actually :P | 04:26 |
| ScottL | troy_s, oh, and i don't really like the new font :/ and people are wanting to change some of the studio logos | 04:27 |
| troy_s | ScottL: 1) Assembly. Snap your bits together into a loose string. Don't fuss over timing and don't start at the same point all the time. Work from beginning to end and snap it all together. If you take a time out, don't start at the beginning again - start at the middle. The point is very much forest through trees. | 04:27 |
| troy_s | ScottL: 2) Rough cut. Refine your timing. Use overlays with opacity where you intend to have popup Scotts so you can see the exact timing. Again, keep forest through trees or your beginning will get finished at the expense of the project. Worry about getting the _whole_ thing done. | 04:28 |
| troy_s | 3) Final cut. Lock the timing. Make sure you get things timed as you like as once you head into post production, you _cannot_ adjust the cut. This is referred to as 'picture lock'. | 04:29 |
| ScottL | troy_s, you know that makes a lot of sense and it's probably exactly what i /wouldn't/ have done without advice, heh | 04:30 |
| troy_s | 4) Post work. This will require keying your little popup heads as rotoscoping is... well... let's say well beyond your patience and ability at this point. This would require you to have shot against white (luminance key) or a solid colour (chroma key.) This will require using nodes generally. You take your strip that you overlaid in the VSE and key them. What results is a travelling mask that leaves little Scott head isolated aga | 04:30 |
| ScottL | troy_s, i was going to find a green bedsheet (or buy one specially for this) for the popup scotts | 04:31 |
| troy_s | 4.1) You will need help. You only need to worry about 4 when shooting. If you try to use a luminance key against white - it needs to be significantly different exposure from your tones in your skin. As in bright white or pitch black. Chroma key is easier if you can find a consistent tone. It needs to be lit relatively evenly to prevent godawful amounts of post work. When you get to (3), we can cover this in depth. | 04:32 |
| troy_s | That's about it. | 04:32 |
| troy_s | But make no mistake, what you are trying to do is _intermediate_ ability, which means you can save boatloads of time doing: | 04:32 |
| troy_s | A) Tests. Do a short test against your keying background. Make it work. Get a feel for it. Figure out how much work it is with scenario A versus scenario B and get a handle on how you want to approach it. Failure is good as it saves you time. | 04:33 |
| ScottL | troy_s, hah, that's exactly what i was thinking, do a miniature just to get the methods down | 04:33 |
| troy_s | B) Keep it simple stupid. This is your first project and it is damn intermediate. That means keep it damn simple. Ridiculously simple. So simple that you think it is ridiculous to even bother. Remember, you are doing TWO things here - Doing a simple project _and_ learning a pretty complex program in _two_ areas simultaneously. | 04:33 |
| troy_s | ScottL: You are _well_ equipped to do it. But you can't expect magic. Learning any program is hard enough, let alone a process and several techniques. | 04:34 |
| troy_s | ScottL: When you itemize all of the Scott Is Learnings, you will see that the list is quite long. | 04:34 |
| ScottL | troy_s, oh, i expect to take months to do it :) and craploads of reading on tutorials, forums, and websites | 04:35 |
| ScottL | i would also like to use blender to make a ubuntu studio movie later on...like lmms did http://lmms.sourceforge.net/screenshots.php | 04:36 |
| ScottL | look at the bottom-ish of the screen | 04:36 |
| ScottL | troy_s, that would be a good thing to have on the front of a reworked ubuntu studio website | 04:37 |
| troy_s | ScottL: Months translates into never finished. Which is why I would advise yourself to build up to it so that you end up with a 'I did this and I can do this' attitude from the start. Limping out with an unfinished project will likely kill all future attempts. | 04:37 |
| ScottL | troy_s, well, given my available time i would expect two to three months to complete it (of course it's a rather ignorant guess) | 04:39 |
| ScottL | troy_s, by the way, this is the article from the magazine that's inspiring me again to work with blender http://www.linuxjournal.com/article/9831 | 04:41 |
| ScottL | talks pretty heavily about nodes and such | 04:41 |
| ScottL | troy_s, seems pretty simple if you take each part alone but it does build up some good stuff | 04:42 |
| troy_s | ScottL: It is all pretty "simple", but so is Chinese to someone that knows it. That's one of the toxic terms. ;) | 04:47 |
| troy_s | ScottL: Did I ever point you to the UI Garden article on Intuitive is a meaningless term? | 04:48 |
| troy_s | ScottL: http://www.uigarden.net/english/easy-intuitive-and-metaphor | 04:57 |
| ScottL | troy_s, i think you did, but i'll read it again, you always pick up more when you revisit things | 05:12 |
| troy_s | ScottL: It cuts to the bleeding core of our primary issue in Libre software. We make gross assumptions that words mean things. ;) | 05:12 |
| troy_s | ScottL: The issue is, at its core, a nasty bit of philosophical semantics. Unfortunately, in order to _cross_ the divide, we really need to figure out how to deal with it. | 05:13 |
| troy_s | ScottL: If North American law can more or less define terms in a useful and usable manner via examples and culture, I'd hope that we can do the same for Libre design. | 05:14 |
| ScottL | troy_s, bedtime for me, thanks troy | 05:15 |
| troy_s | ScottL: Nacth | 05:16 |
| troy_s | ScottL: Nacht even | 05:16 |
| === kubotu_ is now known as kubotu | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!