Underwaterbob

I had a bit of cash, and have been working with a portable setup lately that is mostly Koala Sampler, my phone, a Samson Go mic, and a tiny DAC that is literally a USB-C jack to two 3.5mm audio jacks, one in and one out, and decided I needed some more hands-on sound design power.

In the spirit of maintaining portability, I looked to smaller devices. The problem I find there is that a lot of portable devices look like they're about as much fun to design sound on as a VST or app, so why not just spend a fraction of the cost and do that instead? I did end up working out some of the trouble I had getting Csound for Android working, which is great since there's almost infinite sound design potential there if you don't mind coding (which is how I mostly made music for the first 15-years I was doing it.)

I made a track with Csound and Koala, and decided I still wanted something more immediate than code. So, I bought a Microfreak.

I'm sure the copious firmware updates in the 4 years since it came out have something to do with it being way more capable than I had initially assumed (and at the time I was comparing it with Hydrasynth, which is significantly more capable.) It's got something like 24 different oscillator types now which cover a lot of ground from VA, to wavetable, to granular, and even samples now. They're all a bit limited since there are only 3-4 parameters per oscillator that can be tweaked, but Arturia (and presumably Mutable before them) made sure that those parameters are the ones you really want to control.

I've made more patches on it in a couple of weeks than I have on the Hydrasynth in a couple of years. (Mind you, I've made many more finished tracks with the Hydrasynth so far, so we'll see in the end.) It's super easy to dial in usable and interesting sounds. And yeah, you can "freak" out and make it sound like a university student first discovering LFOs, but I find that side of it decidedly less interesting. If I want cutting edge modulation experimentation, I'll code something up in Csound. The Microfreak just effortlessly does great leads, basses, and pads.

Like everyone else who has one complains: it'd be nice to have some onboard effects since it sounds pretty dry without them. Luckily Koala has quite a few decent effects these days. Really, the amount of sound design, and track creation potential between a $350 synth, a $15 app, and an $8 interface is making me feel a bit dumb over the desk packed with hardware I have at home. (Never mind the $1200 phone I suppose...)

Surely, this can't be a coincidence.

With all the talk of samplers since TE decided to release the ridiculously hyped K.O. II, I decided to finally pick up Koala Sampler. I've heard many good things about it, and for good reason. It's amazing! It's so immediate and fun and actually stupid powerful if you shell out (~$15 for everything) for the mixer, effects, and time stretch extras.

I dusted off my ancient sample collection and plopped them on my phone (Galaxy S23 Ultra) and am putting finishing touches on 3 tracks in just a few days, and just hauling it out to play with my daughter who gets a kick out of it. I even found a new use for my Samson Go mic which works with Android and has a headphone jack. It's perfect since the S23 Ultra doesn't have a headphone jack (fuck you very much Apple, Google, and Samsung) and the Samson mic is obviously much better quality than the (actually not that bad) internal mic.

My phone battery hates me. Though I don't really notice Koala being any more demanding than anything else. I'm just using my phone so much more.

The base version is ~$5 and very much worth it to check out if phone sampling is for you. I really recommend at least the mixer upgrade. It really adds a lot of functionality for another $5. The time stretch stuff that comes with Samurai (the name of the other upgrade) is decent as well, though certainly not as necessary if you mostly use one-shots instead of loops.

Namely the new EP-133. TBF, it's the cheapest thing they've made besides the POs. It's really the first thing they've made that's genuinely got me excited. I don't even have a sampler. I mean, besides a PC which is arguably the most powerful sampler in existence.

Cuckoo's demo on YouTube is pretty good, but I wish Loopop would get his hands on one. I prefer his manner of stoic tutorial-review over Cuckoo's also-OK giddy enthusiasm.

The title's a bit dramatic, but I was having trouble coming up with a good pun or otherwise.

Hot on the heels of his Daemon and Freedom duology, Suarez cranked out this near-future, techno-thriller in 2012. Which I'm sure made a lot of sense given his success with the former. Unfortunately, it fails to live to up to the non-stop, dumb fun of his first couple of releases.

Where Daemon and Freedom found glee in speculation of near future tech changing the face of the planet, this one is dour to the core. Some shady operation is making drones that kill people autonomously. Some other shady operation sets out to stop them. It's hardly spoiling much to say they (at least partially) succeed in spectacular fashion through a series of larger-than-life set pieces involving copious gunfire and car/plane/drone/boat chases. There's no comedy to be found here, intentional or otherwise. D&F at least had the utter ridiculousness of its happenings to alleviate the constant severity. This one ain't got that.

The characters are as cliche as they come. Hyper-competent super secret agents, scientists, engineers, and shady business people. A couple of them even fall in love, though thankfully the sex is limited to a line of text: "They made love." I really wouldn't want Suarez to push his writing chops too far in that direction given his proclivity for over-the-top action and technological exposition. Both of which are here in quantity.

Overall, I wouldn't call it a bad book. Just an entirely predictable, fairly mediocre one. It comes in pretty short around 300 pages or so I'd imagine if I had a hard copy. The technological stuff is dry, plausible, and not poorly written if you're into that. The action is well done, if somewhat less plausible, and keeps things moving.

3/5 autonomous killer drones

If someone asked me to guess what Korg was going to do next, I would not have said a full-sized, 61-key update to the Wavestate. Well, that's what they did.

I don't know how I feel about it. I like the synthesis engine: it's unique and versatile. It really nails those ethereal 90s digital tones with more modern modulation capabilities and sound quality, but it seems like such a niche thing, I don't know if it warranted a full-sized version.

Also, what's with the UI? They took almost the exact same layout as the original, and plopped a gigantic keyboard on the bottom. Now, there are huge blank spaces on both sides of the knobs and tiny screen. Korg really ought to have made the whole thing less deep and spread the UI out across the length of it. I guess they save on R&D by this route anyway.

Personally, I'd say spend the money on a decent MIDI controller, and just buy the VST if you really want those sounds. The hardware here doesn't seem like anything special, and the UI, frankly, looks awful.

I wonder if the Opsix or Modwave are going to get the same treatment.

Canadian teaching English in South Korea here.

Where are you, and waddaya at?

Csound is my passion! I've been programming bleeps and bloops with it for nearly 25 years now. Short of one of the other synthesis languages (I've been meaning to check out Supercollider for years) no software, VST, or hardware synth can do a fraction of what's possible with Csound.

Lately, I've been playing with wavetable synthesis in Csound. The cool thing about using Csound for wavetables, is that there are very few limits of what you can do with those wavetables.

For instance, a piece I recently worked on I wrote an instrument that used a sin wave from a table with 16384 points between -1 and 1 for its single wavelength.

Inside the instrument, I made an if statement that ran once per cycle and randomly either squared the value of a random point or took the square root of the value of a random point (and made them negative again if they were initially negative.) Since all the values are between -1 and 1, this means they never go outside of that range, but they do either get closer to zero if squared, or closer to -1/1 if rooted.

In the end, it means the harmonic spectrum slowly changes in an odd and random manner. The change could be sped up or slowed down by using fewer or more points since the randomization is happening once per wavecycle. I tried some other values, but settled on 16384 because 8192 was a little too quick, and 32768 was a bit sluggish. (Csound likes its powers-of-2, which isn't a strict rule since there are oscillators that will use tables with lengths that aren't such, but I kept it simple.)

Unfortunately, for all its complexity, the end result doesn't really sound too dissimilar to a plain old filter sweep on a harmonically rich waveform. You never know until you try I guess. Ha!