One thing that’s always a challenge as a solo music-maker is being able to just goof around and try new ideas quickly. With just one keyboard and a single pair of hands, I can’t play a bunch of parts at the same time like the members of a band could. I’ve wondered if software could come to my rescue, and indeed I have used seq24 quite a bit now, but it’s really designed more for live arrangement of pre-written patterns rather than true live improvisation and performance. I think I’ve found a solution now, but I took quite a round-about path to find it.
The computer has revolutionised the way we make music, but it also begs a question: how much work do you do “in the box”, using software sequencers, effects, and instruments, and how much do you do with hardware and traditional instruments? When I started making music again last year, having a powerful hardware synth was a huge enabler for me — I really do believe that it, as much as anything, is the reason I’m still making music with Linux now after so many abortive attempts over the years. Now that I have a few tracks under my belt, though, I’m as surprised as anyone to realise that I seem to be working “in the box” more and more.
I’ve just bitten the bullet and upgraded this blog to the current pre-release WordPress 3.0 version; I’ve been playing around with it on a test environment for the last couple of weeks and I’ve been quite impressed with some of the new stuff. In particular, I’ve replaced the theme I was using (Pixel) with my own customised theme. It’s based on the new default theme, Twenty Ten, which is a cinch to work with — you can very easily set up a child theme with just a CSS file, and then copy across extra bits as needed if you need to replace them with your own customised versions.
I’ve still fiddling around with things, so don’t panic if you see any sudden changes. If anything’s broken, please let me know with a comment!
Three podcast releases, all involving me in some way, that I feel compelled to draw attention to:
- A new Partners in Lag — in this ep, we talk about Red Dead Redemption, all things Halo, Steam on the Mac, and the potential future for Linux gaming
- Episode 39 of the Open Source Musician Podcast — another community call-in show, where a bunch of us chat about a whole heap of things
- Episode 41 is out, too — this one features all of the entries for the Tunestorm02 challenge. Even though they were all recorded with household item sounds, in line with the theme of the challenge, there’s huge variety in the results, so it’s well worth checking out.
Google had been notably quiet since acquiring video codec developer On2 Technologies, but today it revealed what it’s been up to: On2’s VP8 codec is now open-source, with a Theora-style irrevocable royalty-free grant for anyone to use the patents that Google now owns that cover it. Google’s calling the initiative WebM, which combines the VP8 video with Vorbis audio in a Matroska container.
The software’s a bit rough right now, but it really does work. Google has already started re-encoding Youtube’s videos, and after grabbing a WebM-enabled Firefox nightly build (the WebM site has to Firefox, Chrome, and Opera builds, and instructions for Youtube) I was able to check out the joys of HTML5 Youtube for the first time. As for encoding, there’s a reference encoder and a set of patches for FFmpeg; using the latter, I was able to build a WebM-enabled FFmpeg, encode a video, and play it back within Firefox.
Maybe Google’s just doing this as a lever against H.264 — MPEG LA might be less likely to go jacking up its prices if there’s a serious competitor in the market — but I get the impression it they really want to see WebM take off. There’s a big emphasis on hardware support, with simultaneous announcements from TI, Broadcom, Qualcomm, ARM, AMD, NVIDIA, and others, and Google will of course be bringing support to Android. With Adobe pledging to support it in Flash, giving support for IE and Safari, WebM is going to work just about everywhere apart from iPhone OS.
What remains to be seen is whether or not there are potential patent challenges here, from MPEG LA or other parties. Google obviously doesn’t think so, and it has a long list of partners that also seem willing to take that bet, so time will tell if MPEG LA makes good on its threats.
This one comes right from the Where The Hell Were You Last Month department — playing around with PHASEX, one of the more feature-packed and better-sounding softsynths for Linux. PHASEX’s interface is a little confusing at first because it has some interesting and unique features, but at its core it’s essentially an analog modeling synth.
One very cool feature that I only just discovered, though, is the support for audio input. Lots of hardware synths have audio inputs, which you can use to run external audio through the synth’s filters and effects, and that’s exactly what the inputs in PHASEX do, too. There’s even an envelope follower, so it can trigger its filter envelope based on the level of the incoming signal. It’s exactly the kind of thing that might have been brilliantly handy when I was making that sample-based Tunestorm piece!
This, then, is a cello sound from Qsynth running through PHASEX — I’ve used the filters with a bit of resonance added (shifting the cutoff on the fly with the mod wheel), and a square wave for amplitude modulation to gate the sound.
mp3 | vorbis | 60 seconds
NOTE: The link below was to an older version of the Salamander, so the link now goes to a download page that lists the latest version. Also, LinuxSampler isn’t as temperamental as it once was when it comes to loading SFZ files, so you don’t have to follow the instructions below to the letter any more — just add a sampler channel, set it to SFZ mode, and load the SFZ file, and you should be good to go.
After a bit of a wait, what’s perhaps the ultimate free piano sample library, the Salamander Grand Piano is available! One of the guys on the linux-audio-user spent I’d-hate-to-think-how-long recording every note on a Yamaha C5 grand at 16 different volume levels with a pair of stereo mics, and the result — all 1.9GB of it — sounds lovely.
Getting it running, however, is a bit fiddly right now. Due partly to its heft, it’s distributed in SFZ format, instead of the more common GigaSample “.GIG” format. Linuxsampler supports SFZ in CVS, but it’s buggy, and the instrument needs to be set up just right to load without crashing Linuxsampler. Once you’ve installed Linuxsampler from CVS — a bit of effort, but fairly straightforward, especially since it comes with Debian package scripts — follow these steps, in order, to get the Salamander up and running:
The Tunestorm02 reveal was today, so I can now post my submission for it: a breezy little tune that, for want of a better title, I called “frozen summer”. In accordance with the rules, it’s made entirely from sounds sampled around my house — there’s our doorbell, a blown beer bottle, and a stretched out balloon that I strummed to get a bass sound, along with percussive sounds from hitting various things (mainly my coffee machine).
Software-wise, I used Specimen for the melodic samples, Hydrogen for the drum samples, and seq24 for sequencing, before recording and mixing in Ardour. I’ve also uploaded my Hydrogen kit in case anyone’s curious.
mp3 / vorbis / flac: 2 minutes 48 seconds
I’ve been home alone for lunch a few days this week, and in combination with a chance encounter with some supermarket bargains, I’ve revived my random food adventures! This time, the hilarious shelf-stable ready-meal of choice is Maharajah’s Choice Vegetable Korma, with Basmati Cumin Rice. With its promises of Authentic Indian Cuisine, how could I lose?