I’ve just added the choice to build a pdf with either pdftex or vanilla latex and dvipdf. Although they’re pretty similar, sometimes you want one over the other (pdftex doesn’t work with pst-circ, for example, or pdftex can take png images). Note: I have not overhauled the image system to deal with this yet. This might mean I’ll have to convert on demand, or something, which might be for the best, since having more images than necessary might be detrimental to my hard drive space.

Also, I just learned about hyperref. I’ll have to keep that in mind for future experiments.

Fun fun!

I’ve been wondering how to apply agile-esque methodologies to things outside of programming, especially since I’m not that great of a programmer. To recap, agile (or what agile means to me) is the mindset that releasing content early and frequently yields better results than finishing and polishing something you release/perform once in a great while. It would seem like this sort of methodology would work with something like the performing arts, since performers are always in a state of flux: one continually gets better. (The discrete nature of traditionally infrequent performances mask that fact, at least to myself.). For instance, last year a visiting violin professor reorganized studio class so everyone would perform at least bimonthly, instead of only three times in a semester. Once you’ve accepted an unmovable release schedule, then one adapts and progress is more apparent through the frequent performances. So that’s my previous experience with an agile-methodology with music performance: where do we go from here?

Well, we do it. Here are some recordings that I made today, about half a month from juries.

To kick things off, here’s mostly chaos. I just overlaid everything on top of each other when I was recording, and here’s a few seconds of the end result.

Now, some Bach: eventually, I’ll overlay the Sarabande and the Double, which works since Bach wrote the two to be harmonically isomorphic (or whatever music theory people say when they mean equivalence).

bach-sarabande-27-11-09_1

A second try at the Bach Sarabande: bach-sarabande-27-11-09_2

I just started the Double, so no, I don’t suck that badly. Really: bach-double-27-11-09_1

And have some Mozart:

A quick run-through of the Mozart. Maybe I’ll add some piano overlay sometime. mozart-concerto-4-I-27-11-09_1

Cadenza: mozart-concerto-4-cadenza-27-11-09_1

And a less gimpy run of the Cadenza: mozart-concerto-4-cadenza-27-11-09_2

Have fun: remember, all content is under cc-by-nc-sa. The lead in/out times are there if you want to do your own trimming/noise removal.

To kick off the visual arts section of the blog, I’ll outline some of the inspiration behind anim3, the third serious animation I’ve started with an intent to release.

First, there is the widely read ‘Le Petit Prince’, or The Little Prince. Specifically, the world he inhabits served as a sort of model to the setting of anim3. However, the plot for anim3 is exceedingly simple (even, I think, compared to the plot of Le Petit Prince): I decided that a convoluted plot a la Elephant’s Dream wouldn’t be as easy to portray, and as this is my first time out on a complex setting/background, I wouldn’t want more complexity than necessary. Maybe anim5 will involve a more complex plot…

Now, for a bit from my past: I first got into CGI and programming* through a little program called Povray, which is a text-based raytracer, which takes some scene description language and renders out a scene, usually made with geometrical primitives. So I was well aware of this image from the Povray hall of fame,  titled Victoria’s World, which is based on another image, Christina’s World. The amount of detail is pretty amazing, but seasoned povray users can quickly pick out several things that mark it as made with povray, such as using an isosurface for the planet surface, the particular highlight patterns in the water,  and the sort of object instantiation in the flowers and birds that povray is good at.

Finally, the last discrete piece of inspiration that I’m going to add right now is a particular scene from Titan AE. I won’t elaborate, since it will be pretty clear what scene I’m talking about once I do finish it.

Well, that’s all for now. If I can focus myself, I’ll figure out how to do the character’s rigging this weekend.

* Technically, I got into programming before povray, probably with TI-83 Basic.

So, I finally got around to pushing out this update…

We have a second iteration of our coilgun, which was mostly put together and worked by my friend Clinton. For quick specs, we used thicker wire (21 gauge), more wraps over less area/layers (who-knows-how-many-layers over maybe 2 inches on a maybe .2 inch pipe versus one layer over 6 inches of an about 3/4 inch pipe), and more voltage ( 2 dual-parallel caps in series). With these combined, we upped our speed from around 2m/s to around 18m/s. We also switched our ‘ammo’ from rare-earth magnets to plain old screws with their heads sliced off. No pics on this iteration, but we do have video of Clint shooting a defenseless binder.

We’ve been working on a multi-stage iteration; more details soon.

So, just to follow up with my gripes about Ubuntu, it turns out that upgrading was the cause of my troubles. Doing a couple of clean installs (first to find corrupt libraries killing init within 3 seconds of boot, second to find I had installed x86 inadvertently, and the final one to get x86_64) got me to a working install, and I must say, it’s pretty good to be back out from under XP. Once I get the chance and funds, I’ll have to get me two copies of Windows 7, especially that 30$ student edition (baring any revelations that said copies are crippled in some way) so I have to reformat. I also learned how to ‘burn’ an iso to my usb disk (why yes, it is an actual tiny usb hd), which means I won’t have to burn another disc, hopefully forever. Which works out, since I had just run out of discs to burn (although it’s just smaller than your average DVD image: I’ll probably have to upgrade my drive to an actual flash dongle).

Also, I’m working on another software project. I’ll post details once it does something more than display pretty boxes in the wrong proportions.

Just so you know I’m not dead or anything.

So I’ve been having troubles with Ubuntu for the last few days: first, upgrading the desktop causes a cyclic restart after logging in: sound is broken again, as I remember putting good hours in making that work: and now I couldn’t find the command to upgrade without a gui. For someone concerned with gaining some server market share, that particular command is not very easy to find. Mind you, I’m not going to jump ship: I’ve also tried to concurrently install Mandrivia, and that’s not working out too well either.

EZLO so far has been using a fastcgi backend, with lightty up front. However, this hasn’t been the best of worlds, since the dispatch.fcgi processes would die at a certain time in the morning, and I couldn’t figure out how why or how to find out how (I got as far as trying to parse through the process accounting). I just got fed up and decided to ditch fastcgi and run a full-fledged http server as a backend.

There was a post I read a while back (I can’t find it, though) which claimed that cgi and fastcgi are hacks, and that using the web server as a proxy for application specific web servers is the simpler, more transparent method. HTTP has been around for ages, in contrast with fastcgi, the young protocol that no one loves.

With a dual mongrel backend (mongrel being a ruby/c web server), I’ve noted that the speed of the site seems to have dropped, and memory usage jumped up; however, mongrel knows I’m in development mode, so we’ll have to see how it responds when I push it into production mode. The memory usage isn’t something I’m happy about, but I’m not sure what else I could do, since they’re actually just full-blown web servers. I just tell myself it’ll be more flexible in the long run…

Also, as I’ve been doing alot of reorganization of the directory structure in anticipation of a release, I’ve thought about how to package up EZLO into something distributable seems less than easy, meaning it might actually take an appreciable amount of effort to just package it up into a release. Just a note.

I have an Ott lite, if you haven’t already guessed: I actually have two Ott lite bulbs, and one ott lite stand. Mostly people would go buy another Ott-lite stand, but me being the ornery cranky old person that I am, I decided to hack up my own stand to save on money, since I managed to bungle my paycheck and I’m not gonna have money for a while now.

This is a simple hack/reverse engineer attempt, in that lamps are not usually complicated. However, being careful, searches for the lamp bulb technical specs turned up nothing on the internet. Amazingly, no one seemed interested in using these lamps in ways not originally intended, which I intend to change fairly soon.

First, trying to take apart the casing proved to be difficult, and I eventually gave up, since breaking the only ‘official’ lamp stand I had in the name of reverse engineering seemed a bit… overkill. However, it did seem like there was at least one transformer in there, so something was going on. So, I just took the stand into the circuit lab, and found that the lamp stand’s output into the lamp bulb had exactly the same characteristics as the juice that comes out of the wall: 60Hz, 120V.

Now that is a mystery: what is that transformer doing then? I doubt it’s doing anything with the power, since voltage isn’t changing. The only strange thing was that the output of the stand was more akin to a triangle wave than a pure sine wave. Perhaps it’s a current limiter?

The only thing that remains to be done is create an actually stand, or run more tests on the stand to see if it does indeed function as a current limiter. Stay tuned.

As noted before, Ze Local Awesome-blowing-stuff-up club, of which I’m the Benevolent Dictator for Life, has created the initial iteration of a coilgun. Behold!

I can already hear the complaints about the lack of video of the thing actually firing, but I have an excuse: our benevolent email-writer for life, managed to kill a vital component while trying to make a would-be version 0.2 (this also explains why nothing is connected to anything else). So, no video until we finish up version 0.2, which should be soon.

For those of us unfamiliar with *what* a coilgun is, and who are familiar with physics, a coilgun is essentially a big electromagnet. For those of us who are not in the physical conspiracy, but still want to know what they’re looking at, a quick physical primer: first, the big cylinders are capacitors, which store electricity, only to quickly dump all that electricity through the coil. With all this electricity moving around in circles through the coil, a magnetic field is created, which then draws magnets or pieces of metal through the coil, sending them flying.

Well, okay, you should forgive me: this is the first iteration of the documentation associated with the coilgun, too. I’ll try to update the descriptions and inner workings better as we go, especially since we managed to do very little math on this coilgun. We’ll probably do more and more math/analysis as we go, too.

So, finally, some gory details: we have a capacitor bank of 4 capacitors in parallel, rated for 200V and 1200uF. The coil was made with 24 gauge magnet wire, for a coil density of who knows what, since we weren’t keeping track, and we’ve already taken most of it apart. The coil is wrapped around some 6 inches of 1in diameter PVC piping, and we used a high voltage, high current rated thyristor as a switch. With the capacitors charged up to 195V, we shot a magnet around 70cm straight up into the air, which is not too bad for doing absolutely no math. Maybe I’ll talk about our throwing-together process in later posts.

For the next iterations, we’re thinking about using a high-voltage transistor instead of a thyristor, for better control of the current: once a thyristor is turned on, it passes current, and continues passing current until there’s not enough voltage to keep it turned on. A transistor passes current only when there’s some voltage applied at the base. (Note: I know this is a bad explanation, especially when I use terminology on several different levels. As I hinted at before, I’m trying to apply some agile methodology to blogging, and maybe it’ll stick). Also, we’re looking into thicker magnet wire, and a way to power the circuit directly from the AC grid.

And I’m sorry about the crappy photos: the camera phone idea is good, but the cameras still don’t hold up well when you want crisp details.

So, the coilgun fires! I will (hopefully) post with more information soon.