The Tyranny of Gear

If you participate on a music forum, one of the first questions you'll be asked is "What does your studio setup look like?" It's a fair question, but I always feel that my answer seems pretentious. Partly, it's people trying to decide how far along you are in the learning curve. To get to the advanced gear, you've presumably used and discarded a whole lot of beginner gear. Which is often, but not always, true. Problem is: you can throw away a lot of gear without learning much about music. I'm a good case in point. I've learned a lot about gear, I know a very little bit about music, I have some positive experience as a performer, but after ten years of hands-on practice I don't claim to know that much about synthesis. Still, one needs to answer, so hear it goes.

At heart, I'm an acoustic musician. That said, my personal preferences don't constrain my view of music. I'm not interested in playing or listening to hard metal, but I have enough music theory to appreciate that it's valid and interesting music that doesn't happen to appeal to me. There's a difference between biased and ignorant. "Extreme" mostly writes music that I don't care for, but it's valid music. As a half-trained musician, I think they are brilliant and challenging composers, and I thoroughly appreciate the fact that More Than Words is simultaneously a beautiful composition, a song I enjoy, and an exercise in flipping the bird at conventional music critics. More power to them, and thankfully other people are buying their albums! Why tell you this? Because if you are going to read opinions, you should know where the author is coming from and what pre-dispositions they may hold.

So to get it out of the way, I'm going to answer the "gear" question, but I'm going to give a bit of history along the way. I'm also going to say what I would do differently now. You can indulge me or skip ahead, as you please.

My main instruments are acoustic guitar and vocals. I grew up abusing a Baldwin grand piano. I've graduated to abusing synthesizers, which may or may not be forward progress. I've spent some (disappointing) time on MIDI guitar capture (translate: I have the state of the art, and it stinks). I've dabbled in MIDI drums (Yamaha DTExtreme), but the truth is that I'm not a drummer. I talked myself into a DTExtreme kit because you just can't get the same dynamics out of tap pads. With enough passes, I can make something that sounds almost entirely, but not quite completely, unlike a real drum track. Which I can't do on tap pads. Probably not the fault of the equipment. But even getting a very modest sense of how the world looks to a drummer is helpful. Meanwhile, my six year old son loves banging on the drums.

My earliest experiences as a performer were acoustic guitar, playing an un-amplified Ovation in front of 300-400 campers in a dining hall as a 12 year old. 30+ years ago. Tough audience. Ovation guitars have sadly declined since then, but back then they were the right tool for that job, because of the round back and the amplification it produced. These days most of my playing is done on a Motif XF (soon a Kronos) and on Composite Acoustics guitars. My intermittent performances are done for personal entertainment, in blue-collar bars, and let me tell you: if you can get a pipe fitter to put down their beer so that [s]he can listen to you, you're connecting. And that's a lot less about being a great musician than it is about being a performer. Harry Chapin was merely above average as a musician or composer. His voice was extremely rough. But he had a true gift for connecting with the audience. Someday I hope to be as bad a guitarist and vocalist as Harry Chapin, but in a very modest way and on a small scale I connect.

At some point I said "Hey, I need a keyboard. I play guitar, but I think on piano". And I took home a Fantom X and hated it, and got talked into a Motif 8, and stuck with it for lack of better options. Upgraded to Motif ES8 for the same reason. And I quickly found the limits of the Motif ES synthesizer and migrated over to PC and Sonar. Partly, I was interested in orchestral composition and I fell in love with the EastWest libraries. Along the way I tried, and returned or sold as used a whole bunch of gear. No claims that any of these are the "right" answers, but at least I can tell you how I got here so that you can decide for yourself if these outcomes are relevant to your situation.

So being perfectly ignorant, I walked into Guitar center, and I got talked into Sonar (good, for me, still in use), Presonus Eureka pre-amps (good, still in use) and Firewire-based audio converters (better than USB, but not good for me). Here are the key questions about A/D/A interfaces:

• How much do you want to spend?
• How important is CD-quality audio sampling? This rules out most keyboard-based samplers, including the Kronos.
• How often are you willing to re-cable? You can only play so many channels at a time, but re-doing all of the per-channel setup is a nuisance. This is a time vs. money tradeoff. It doesn't have a "right" answer.
• Given that USB flatly doesn't work, does your platform support FireWire? This may no longer be a major issue; I shifted to PCI capture so long ago that I cannot say.
• How many channels of sampled audio do you need? To achieve CD quality (48Khz) end results, you should sample at 96Khz. In the context of Firewire and USB that's a high bar. So when you look at devices, compare 96Khz channel counts.

For myself, I ran off the end of 4 channels at 96Khz (i.e. FW400) pretty quickly, found a good deal on eBay, and moved up to PCI-based MOTU gear. At the time, Firewire 800 didn't exist, so I was already over four channels at 96Khz on guitar alone (see below). On Firewire 800 it would be possible. Somewhere along the way I outgrew the Motif sequencer and moved to Sonar and sample libraries. Personally, Sonar works well for me, but the competitors are probably just as good.

Guitars


I have two old-style Ovation Elites, purchased back circa 1993, replacing a "beginner" Ovation that I had as a teenager. Bought the Elites back when the Ovation Elite line was high-quality, unrelated to the current Ovation Elite line (grumble). Great pickups for the day, balance and roundback very much a matter of taste. Perfectly acceptable instruments and okay for what I was doing. I still enjoy them today, but they aren't my go-to guitars anymore.

In 2010, I talked myself into the Composite Acoustics line. That's 17 years later, which either says a lot about the old Ovation line or about my stupidity; take your pick. In any case, I bought a 6-string, and later a 12-string in the CA dreadnought body. CA is a company that built great guitars but bit off more variations than they could produce. They were constantly screwing around with pickups and in-body mics, and never really settled on a signature configuration. Hopefully that will settle down now that Peavey owns them. The acoustic sound is first-rate, but I found that I liked the pickup sound from the earlier models (which had an in-body condenser mic) better than the later models (they changed vendors for the pickup). The bad news is that those bodies aren't made anymore. The good news is that a carbon fiber body will last more or less forever.

The main thing about the CA guitars (and some others) is that they leave me with four options for acoustic recording: the built-in condenser mic, the piezo-driven pickup, and an external X-Y microphone pair. My practice is to record all four simultaneously, and decide later how to blend them.

The only lesson here is that recording acoustic sound is a tricky business, and the more you learn the less satisfied you'll be with your early attempts. Today, if I were doing my "ideal" micing on one of my acoustics, I'd capture five sources in parallel, each at 96Khz:

• The piezo pickup built in to the bridge (or in some cases the body)
• The condenser mic inside the body
• A 90 degree pair (NT5's) of mics at or near the sound hole.
• The C1000S at the fingerboard.

Which is probably excessive. The main point is to note that this requires five channels at 96Khz, which rules out Firewire 400 and low-bandwidth USB.

If I had to limit myself to four channels, I'ld drop the piezo first, because that is the one that least accurately represents the sound perceived by the listener. It also tends to saturate first, and I'm not really interested in any of the "magic" produced within the newer generations of pickups.

The summary, such as it is, is that mic-ing an acoustic guitar really well is a tricky business. But don't get caught in the Cadillac mentality. I'm covering my bases because I make mistakes. If you find yourself needing to prioritize, buy a matched NT5 pair. Use one on the sound hole and the other close on the fret board, and you can still use them as a pair in other applications.

In general, acoustic capture is much harder than electric. It requires better technique. On the other hand, investment in technique has a long-term payoff, so that's not necessarily a bad thing.

Microphones and Pre-Amps


My first mic came on recommendation from someone at Guitar Center: an AGK C1000-S. This is a good general-purpose instrument mic. At the time, I didn't know much about microphones and I lucked out. Ten years later I still use this mic as an instrument mic, but I wish I had bought a matched pair. Of course, that's hindsight, and that mic served me for a long time before I realized my error. And that's about as good as an early "mistake" can get. Today, I'd buy a matched pair of Rode NT5's for instrument mics at half the price I paid then for a single C1000-S. The NT5's are great mics, and at $500 for a matched pair it's hard to argue for the savings in buying just one once you are in the market for a serious instrument mic. With very good reason, the NT5's are considered a "classic" instrument mic design.

But the truth is that I was well-advised; the C1000-S is a great mic. Two lessons. First, understand that there is no such thing as a mic that is first-rate for vocals and also first-rate for instruments. The applications demand different condenser diaphragms. Accept that, identify the best you can afford, and suffer with the costs (but see below). Mic prices have come down a lot over the years, but if you're still bumming cigarettes that's not a lot of comfort. On the other hand, if you're mostly playing out rather than recording, keep in mind that the venue PA system is a piece of junk, and the quality of the input has very little impact on the quality of the output. For that application, cheap (within limits) may be fine! Second: ask your sales person what they have personally used, where, and how. If the answer is "Uh....", find a better seller. If the answer starts with "well, I work in x studio, and we see people go in and out, and a lot of them like this mic for the following applications" then listen carefully, but listen both to what is said and what gets danced around. Remember that you can always come back another day, and the cost of buying on bullshit is essentially infinite.

At the same time I got talked into the Presonus Eureka tube pre-amps for vocals. My personal opinion is that these days you don't need a tube-pre for anything other than vocals, but for vocals the quality of the pre-amp really matters. At the time I purchased, the Presonus tube preamp units were ~$750 each, and the next step up was $5000 each. Definitely a knee in the curve, and at the ~$750 price point Presonus was (and remains) a clear winner. If you are just getting started, and you're on a budget, ignore this entirely and go for the lowest-noise A/D device you can afford. Presonus and Motu firewire/USB devices are both good at that level. These days I favor the MOTU 828mk3 hybrid, but it's a close call. Priority should be pre-amp for vocals first, for acoustic instruments third, and for anything else never.

A few years later, I bought a GrooveTube GT-57 large-diaphragm mic for vocals. Today, once again, I'd buy a matched pair, but one was fine for me at the time. At the end of the day, an individual vocal really is a monophonic source. The GT-57 was (and remains) a great mic. It's a crying shame that it no longer seems to be made.

There are a number of lessons here, but the "best" mics change so fast that they are hard to apply:

1. Good mics are worth it. No amount of cleanup can correct errors introduced at the source. Because they differ so widely in dynamic range, the requirements for voice and instrument are fundamentally different.
2. Don't worry about stereo micing at first. If you do, deal with stereo instrument micing first. At the end of the day, the human voice is a monophonic source.
3. Good pre-amps for voice are worth the money. There is no instrument that matches the dynamic range of the human voice, and in consequence it is very important to do a good job of compression and limiting at the source. This is a notable exception to the general rule. For any other input, the general rule is that transforms applied before capture are things you cannot undo. This is a good rule for voice as well, but the dynamic range of vocals are to demanding to apply it religiously.
   
   If you can't afford a first-rate tube pre-amp for voice, don't bother with second-rate. Your second choice should be to go with the cleanest A/D's you can find and use software compression/limiting. That's fine if you are getting started, and for quite some time thereafter. Whatever you do later, you won't regret the investment in cleaner A/D's now.
4. Good pre-amps for acoustic instruments may be worth the money, but that's more about warmth than anything else, and these days it can mostly be done digitally. If you go the digital route, you should definitely do 24-bit capture at the source.
5. Tube pre-amps for anything else are irrelevant, provided you do your recording at 24 bit. Implementing the pre-amp in software will lose some bits on the low end, but if you start with 24-bit capture it won't matter.
6. You will inevitably lose fidelity in the processing chain. This will manifest as loss of significant bits. CD audio is 16 bits, so you can't afford to lose any of those. Therefore, you should record and process at 24-bits and down-grade your sample width on output.

Remember that, in the end, there are lots of stages between the mic and the user's speaker, and each stage will introduce noise, distortion, processing, and so forth. What matters most at the end of the day is the sound that the user hears.

Fidelity of the up-front capture is secondary. If you succeed well enough to re-engineer the album you'll want higher fidelity, but the first task is to succeed well enough to need it. Later you'll regret that trade-off. You'll go back to early tracks and wish you had a better capture. When that time comes, remember that if you hadn't spent your money strategically back then, you wouldn't be looking back now.

Keyboards

I'm in transition here. For a long time I used a Motif ES and was happy with it, though I was mostly driving PC-based samples. Somewhere along the way I discarded a Fantom Rack in favor of PC-based samples. Recently I upgraded to the Motif XF, mainly because I wanted to be on the leading edge for competitive evaluation reasons. On balance I feel that the marginal value between the ES and the XF is nil. The samples are better, but once you play them through venue audio the difference isn't audible, and it certainly doesn't matter for driving PC-based synthesis. Meanwhile, the boot time when from 18s (measured) to 45s (measured), and overall responsiveness degraded. There is definite incremental improvement, but the key word is "incremental".

Meanwhile, I've pre-ordered a Kronos, and I'm expecting that to change how I view things. For synth action, I have a Fantom-G6, which is largely a reflection of dumb luck. I was looking for a semi-weighted controller and stumbled into a mint-quality used G6. The action is good, and for a marginal $400 what the hell, so I bought it instead of a controller. Collectively this is completely insane, and I have this setup mainly for comparative reasons.

Since I ordered the Kronos-88, I've been suffering with extreme jealousy while others report on the K-61. To ease the pain, I've been sneaking in to Guitar Center in Seattle to play with the K-61 every chance I can. If I had the K-88 in hand, and I didn't have business reasons to hold on to the Motif XF, I'd put the XF up for sale.

If I were buying now, I'd buy a Kronos with my preferred primary action (in my case 88-key graded hammer action), and (if I cared) a controller to fill in the "alternate" action. Choosing synthesizers involves a lot of religion. The Kronos is (in my opinion) the most flexible synth out there as of 2011. Individual patches may be better on this synth or that synth, but if you really care that much you are likely to end up in PC-based synthesis in any case.

Other


I have a Yamaha DTExtreme drum kit, but I'm not a drummer. I also have a TC-Helicon VoiceOne. Best in class and I should rant about it some time. I may, at some point, talk about wiring the Yamaha Kit to the Kronos, or I may not.

And all that aside, I rely mainly on MOTU PCI-based gear for my A/D/A capture. Once again, that's largely outside the scope of the blog.



So there it is, my gear-slut confessional. It's flatly amazing how much crap you can accumulate over time. The only merit to this pile of circuits is that I can (perhaps) do some side-by-side comparison.

Be that as it may, you now know how my scenario evolved. If that's helpful great, and I'm happy to answer questions (if any) in the comments. Feel free to ask, and feel free to say why, in your view, my "what I'd buy now" ideas are out-dated. I may learn something, and right or wrong the readers will learn something; there's no down side.

Welcome!

Some of you have a Kronos in-hand already. Some of you, like me, are waiting for arrival. Either way, the Kronos is a uniquely deep synthesizer. It offers a lot out of the box, but to get the most value out of it you need to learn how it works from the inside.

I'm an amateur musician, a professional builder of embedded systems, and a sometimes-cogent writer. I have a fairly rich set of tools in my studio. I've been doing both keyboard-based and PC-based synthesis for years, and I've learned a bunch of things the hard (and expensive) way, but at the end of the day I write and perform music for entertainment, not for a living. I have a reasonably successful track record for breaking complex systems down into descriptions that normal human beings can understand.

What I'm going to try to do with this blog is to expose my own learning experience with the Kronos in the hope that it will be helpful to others. I'll be reacting to comments and questions, and trying to create a blog that is useful to others. Whether I succeed is up to you to determine, but I'll try.

I've got a Kronos-88 on order, so for the moment I'm limited to the time I can put in on the showroom floor on the Kronos-66. That said, I've seen enough to start addressing some of the questions that I'm seeing on forums here and there.