The Mandelbear's Musings

Just after Baycon cflute and I listened to the album on my livingroom stereo and Callie noticed that the bass was exceptionally boomy. The effect largely went away on my high-quality headphones or my studio monitors (which, admittedly, don't have a whole lot of bass).

Giving it another listen in the livingroom, I finally realized why: It's not that my speakers are particularly bass-heavy, it's all the junk piled in front of them. In particular, piled right in front of the tweeters. It's not that they have a lot of bass, it's that the treble is being attenuated. A lot.

I feel much better now that I understand it.

Don't know, but it's going pretty well so far. After some recording, some editing (I'll post about those separately), a stack of mail that included the CD I ordered last week, some lunch, and a nice five-mile walk, the Bear went shopping. The main targets were a a cheap MP3 player (Fry's has several on sale this weekend) and a pennywhistle in A. I started with Guitar Showcase.

GS often has woodwinds in their consignment shop, but no whistles today. What they did have was an M-Audio Delta 1010, for $115. Since this has a list price of $500 and a street price of something like $400, and since I'd been thinking about buying a Delta 44 for a street price of around $150, I snapped it up without thinking twice.

Next stop was Fry's. They did indeed have cheap MP3 players, at prices all the way down to $15, but the one I ended up with was an LG UP3 for $60. It has a sharp-as-a-tack little OLED display, 2GB of flash, plays OGG as well as the more usual formats, and lists Linux on the box under supported systems. When I got it home (getting a little out of order here) I was a little puzzled by the lack of a USB cable until I discovered the knob that slides out the connector that makes it into a USB stick. I was also puzzled by the fact that my box wasn't seeing it, until I traced the USB extension cable back and discovered that it wasn't connected. Duh. Only thing missing is FM, but I can live without that. Especially in these days of ubiquitous podcasts.

The final stop was Starving Musician. They had whistles in C and D; but not A. I bought a cheap little Yamaha descant recorder instead -- I had one, but can't find it. Probably lent out to somebody's kid. For $8.25 I can afford another one.

So all-in-all it's been a pretty good day. Happy Bear.

Added reverb to the vocals on "Someplace in the Net" (which improved it considerably); last night I also did some mixing on "Programmer's Alphabet" and "Paper Pings". Ran into some mysterious bugs in Audacity last night working on "Net", so I pulled down the latest updates from the CVS tree and rebuilt it. Several stupid mistakes (you really do need cvs update -d if you want to get all the new subdirectories) and a number of apt-get's to fetch missing libraries later, I had a shiny new Audacity 1.3.3 -- the bugs appear to be gone, and there are some welcome improvements in the UI.

In other news, my left leg isn't hurting, and the infected big toe on my right foot seems to be improving. These are both Good Things.

Happy Bear.

(9:15 reverb on "Paper Pings". 9:35 reverb on "Programmer's Alphabet".)

As it turns out, the isolation transformer has some problems -- it gets overloaded with a full-range line out signal. (The technical term is saturated: the transformer's iron core won't transmit any more signal than it takes for it to be fully magnetized. This turns out not to be a whole lot.) So, although it blocks hum like a champ, it's worthless for line out.

Fortunately, my output mixer has a monitor output with a volume control. That appears to cut the output enough to not saturate the isolation transformer.

In other news, some of the older tracks I've been having problems with might be fixable if I apply Audacity's built-in compression tool to each part separately. We'll see. It may work on "RFC1149", "Stuck Here", and "Uncle Ernie's". I've been playing with "Wannabe", which has other problems -- I can do anything I want to the scratch tracks because, well, they're going to get scratched. (Problems include clipping on a most plosives, and a couple of horrendous "clunk"s apparently caused by banging the microphone against the guitar. Oops!)

Everything takes longer and costs more.

Today it was mostly "takes longer". It took much longer than I expected to put the mic stands back where they belong, clear a space for the computer, shut down and disconnect the computer in the office, and bring it back up in the bedroom studio.

Then it took a lot longer than I would have liked to get the Cat6 cable run back in operation. It didn't work when I installed it -- I figured (correctly, as it turned out) that the problem was screwed-up plugs. I ended up cutting off both ends' plugs and connecting them to a patch panel on the office end, and a modular jack in the garage attic. Naturally this required several tests, each of which meant a round trip up and down the garage stairs. Par for the course.

This freed up the the old Cat5 run I'd been using, leaving me free to use it as I'd originally intended: to run audio for editing, so I don't have to waste two hours moving the box the next time. It took longer than I expected. Surprise. And I wasn't expecting the sound quality to be particularly good, but I wasn't expecting nearly as much 60Hz hum as I ended up getting, either. Not nice

(A few minutes later) An isolation transformer (Furman ISOpatch) helps a lot. And I'll be the rest of the problem is caused by the fact that there are two independent paths for what is really the same ground. (A few minutes later) Maybe not: shorting the two grounds at this end doesn't seem to help, and I know they're shorted at the other end. I may just be stuck with it for now.

If I had real differential signals this probably wouldn't be a problem. At least now it's down to the point where it's not really noticable when I'm actually editng. So I could edit now, except that I'll need to go to bed soon.

This article at Ars Technica gives a nice little overview of three open-source audio editors: Ardour, Audacity, and snd. It concludes with a few thoughts about why open-source audio apps are different from proprietary ones -- not necessarily better, but different. Along the way it introduces a few other pieces of the Linux audio toolkit.

I personally use and like Audacity, which is also available for Windows and MacOS and is undoubtedly the simplest of the three to get started with. You can download the lot as part of any Debian-based distribution; DeMuDi's live CD would be a good introduction, and I've found Debian testing (Etch) to be as friendly an install as one could wish for (see previous post).