Admittedly, "Going Forward" is a pretty vague title, but it is nevertheless fitting. You see, I've finally finished what I expect might be the last academic degree I'll obtain during my life. For the record, I got my BS in Electrical Engineering from the University of Michigan (Ann Arbor) in 1989. I got my MA in Philosophy and Ethics from Talbot in 1998. Those were accomplishments in themselves, of course, but I will always remember 2003 as the year that I finished my Ph.D. in Philosophy at the University of Southern California. Yes, that's right, dear reader, it's official: Phil is a certified smart person. That and $5 buys me coffee at Starbucks. It also qualifies me to ask "Ya want fries with that?" at any major fast food restaurant.
Getting back on track, now that my degree is over I'm going forward with many projects that have languished unattended over the years, foremost of which is catching up on musical technology and putting together my very own little home recording studio. I don't think I've mentioned it before on these pages, but my life has been intertwined with music for roughly thirty years at this point. Along the way I've gained proficiency (and lost it too in some cases) in a dozen or so instruments, earned my living as a composer/performer, and have been privileged to work in a few different recording studios.
It's been a dream of mine for more than a decade, though, to have my own studio. I don't mean anything terribly fancy, mind you, I just want some way to work on my composing, performance, and recording skills in my own home. I want some way to record my own stuff. Academia sucked up so much of my time over the last seven years that I had largely fallen out of touch with music and the music industry. Getting back into it now is both daunting and thrilling, and I've been disappointed to see that the kinds of problems I've written about previously haunt modern musical hardware and software as well.
You could have knocked me over with a feather when my wife told me what she wanted to give me for a graduation present. She knows that I have wanted to buy music/recording gear for years, but I don't think she truly comprehended (or comprehends) how much it means to me that she's granted me a budget for shopping as a graduation gift. Thanks, honey!
The first step, after picking my jaw up from the floor, was to start researching the current state of the industry. Back in the early 1980s, when I first started to want my own recording gear, a lousy cassette-based four-track recorder was about $500. Over the years I used to look at ads for sixteen or even thirty-two track multi-track tape decks and salivate, knowing that I'd never be able to afford their price tags (more than $50,000 for the latter at one point). That 1980s/early-1990s mindset was what I brought to the shopping research.
Imagine how pleasantly surprised I was when I discovered that a few thousand dollars today can buy literally millions of dollars of functionality in 1980s/early-1990s terms. Seriously, today's digital-audio workstation (DAW) software can do more than any of the hardware of old and at the barest fraction of the price. To cut right to the chase, I decided to go with a pretty basic setup. My primary interest is in orchestral composition, so I figured I would get a hardware I/O unit, DAW software, a software sampler, some self-powered near-field monitor speakers, and spend the bulk of my budget on good sample libraries.
To fulfill the first three of those needs, I chose a Mark of the Unicorn (MotU) 828 mkII, Steinberg's Cubase SX, and Native Instruments' Kontakt. I haven't yet purchased near-field monitors because I'm still researching and comparing them. I also haven't finalized all of my sample library choices, though I have already purchased Sam Trombones, Sam Horns (with the additional solo horn), and Peter Siedlaczek's Advanced Orchestra. I'll probably pick up the Garritan Orchestral Strings as well, though I'm presently waiting on some unspecified yet important announcement from the company—they promise me it will be worth my wait, so I'm waiting. We'll see.
Future purchases aside, the stuff I've already acquired gives me an incredible amount of power. I can record MIDI tracks and route them to VST instruments supplied by Kontakt, record audio tracks, perform an incredible variety of editing and processing on both sorts of tracks, do surround-sound mixing, embellish and print scores, and pretty much everything else under the sun. It is truly overwhelming what musical hardware and software can do these days. I'm like a little kid in a candy store!
Unfortunately, the same sorts of problems I've had in the past with hardware and software seem also to plague musical hardware and software. I suppose that I should be thankful that I've had only two major problems to date. For the most part, my musical gear has simply worked, but I've had some nasty problems with the MotU 828 mkII. To be more specific, I've had problems with (1) excessive and prolonged CPU utilization, and (2) audio distortion. The two problems seem to be related, and I've had wildly varying luck in dealing with MotU technical support as I'll explain.
The first problem occurs whenever I use software to change the 828's settings. The software that comes with the unit allows the user to adjust the channel settings, the sampling rate and buffer size, and a number of other parameters. Unfortunately, whenever I use the "MotU FW Audio Console" (or any other software) to change the 828's settings, my CPU utilization pegs itself at 100% for quite some time. If the "Enable full Wave support for legacy (MME) software. (Less efficient)" box is checked in that utility, then the CPU will remain pegged for as long as one minute and twenty-six seconds (1:26). If it isn't checked, then it will remain pegged for a mere 0:12 - 0:20.
This wouldn't be such a problem if it weren't for the fact that changing the 828's settings seems to be the only way to correct the audio distortion I've seen. You can imagine how much fun it is to have to wait for 1:26 after each setting change to see if it fixed the distortion problem or not. Even waiting 0:12 - 0:20 is bad enough, though that is much more livable. At this point, I'm pretty sure that it's an issue with the MotU drivers or the unit itself, but I get ahead of myself.
The second problem occurs whenever it bloody well feels like it. Seriously, I don't know what brings it on; all I know is that nasty, digital distortion (i.e., the sort of distortion that occurs when samples are dropped or clipped during playback) haunts my audio from time to time. Once the audio is distorted, the only way I've found to get rid of the distortion is to fiddle about with the sample rate and/or buffer settings of the 828. It seems like twiddling those settings somehow reinitializes the 828 or its drivers, and that clears up the problem.
It also bears mention that the nature of the distortion seems to differ at the various sample rates. At 44.1 KHz., 48 KHz., or 88.2 KHz., the distortion usually sounds like dropped samples or clipping. When there is no audio playing there is no distortion; when audio is playing, the distortion varies concomitantly with the amplitude (i.e., volume) of the audio playing. Honestly, it sounds much like the distortion one gets when digital audio exceeds the 0 dB level and starts clipping. I don't think that's the issue, though, because lowering the volume does nothing to change the situation.
In contrast, the distortion at 96 KHz. is quite different. When the audio is distorted at 96 KHz., the distortion always manifests itself as a buzzing noise that repeats at a frequency of 2 - 3 Hz. Rather than sounding simply like clipping, the distortion sounds almost like the rhythmic buzzing of a bee. It too varies in its intensity with the volume of the audio playing, but rather than being continuous it makes a "Bzzzz... Bzzzz... Bzzzz..." noise on top of the audio signal a few times each second. This sometimes occurs at lower sampling rates, particularly 88.2 KHz., but this is the only sort of distortion I get at 96 KHz. Isn't that odd?
So I did the stupid thing; i.e., I called technical support. The first guy I spoke with, Miguel, was—how shall I put this delicately?—less than helpful. To be more forthright, he was an ass. He blamed, in roughly this order, my firewire interface, my operating system (OS), the chipset on my motherboard, my OS configuration, my sound card, my video card, my drivers, my system's available firewire bandwidth, and pretty much anything but the 828 or its drivers. Those, you see, are above suspicion, regardless of any facts or reason. But what absolutely astonished me, and convinced me that I simply could not deal with Miguel, was his attitude.
You see, Miguel eventually settled on firewire bandwidth as my problem. According to him, the reason I was getting audio distortion was because my firewire interface simply didn't have enough bandwidth to handle the audio stream to and from the MotU 828 mkII unit. Let's do some calculations, shall we? At a 96 KHz. sampling rate, 24-bit audio will generate 2,304,000 bits of data per second. That's 288,000 bytes per second. Even doubling that for two channels (i.e., stereo sound), doubling it again for full duplex traffic, and doubling it a third time to simulate a ridiculous 50% transmission efficiency, that's still a mere 2,304,000 bytes per second. Basic firewire is a 400 Mbps interface. So, I'm supposed to believe that an interface capable of carrying 50 MBps can't carry a mere 2.304 MBps? I'm supposed to believe that my firewire interface is so lame it operates at less than 5% of its advertised capacity? Ah yes, that seems reasonable, doesn't it?
Worse, as I told Miguel, I have two separate firewire interfaces available on this machine, and I was getting the same problem with them both. Maybe it's reasonable to assume that one of them really is that awful, but isn't it a stretch to assume that both are? Still, I took his recommendation, went to my local Fry's Electronics, and purchased two different firewire I/O cards, one a D-Link DFB-A5 and the other the Keyspan FireWire Card that he said MotU recommends. Neither of these interfaces showed any better or worse results.
When I said that his diagnosis didn't make sense, however, how do you suppose he reacted? Do you suppose he listened to the calculations and conceded that I had a point? Do you suppose he cared that I get the same results with four different interfaces? If you answered "No" to both, give yourself a cookie. Miguel's response to my observations was as simple and blunt as it was ridiculous: "Sir, I'm not going to argue with you." Yes, you heard that right. He politely reiterated that my problem was due to a lack of firewire bandwidth, issues I raised notwithstanding. Yup, sure enough; it must be my firewire bandwidth. After all, Miguel said so. Sheesh. What a llama.
Lest you think I'm being too harsh, consider something that Miguel refused to recognize: my system would give me clean audio after I would screw around with the 828 settings. In other words, my firewire interface clearly had enough bandwidth! If it were merely a matter of insufficient bandwidth, then it would never be able to give me clean audio. But it must be my firewire bandwidth nevertheless, right? After all, Miguel said so, and he's not going to argue with me about it. I'll say it again: what a llama.
After my dealings with Miguel, I was pretty close to taking the 828 back to the store and getting something else. After all, why should I pay $750 for a device that won't work reliably? The problem I faced was that I really liked the 828, largely because it is more capable than any other similar unit on the market right now. It does everything I need (and more), and it does it beautifully when it's working properly. So I decided to give MotU support another try, and I kept calling back until I got someone other than Miguel.
The next fellow with whom I dealt, Nathan, got off to a much better start. Yes, he did have me investigate various things with my system, but he came quite quickly to the reasonable suspicion that maybe something isn't quite right with the 828 or its drivers. What's more, he seemed genuinely interested in solving the problem, rather than simply getting me off the phone after assigning blame to a convenient but preposterous scapegoat (e.g., firewire bandwidth). Imagine that! I found a technical support representative who is actually concerned with investigating and fixing the problem. How refreshing!
For the record, Nathan hasn't yet been able to help me solve my problem. But I'm not nearly so ready to return the 828 to the store anymore for two reasons. First, I'm working with a technical support guy who isn't an ass, who seems genuinely conscientious and concerned. And second, I've managed to find a reliable work-around on my own. I've discovered that when the distortion occurs, if I simply open the 828 control panel and rapidly switch the sample rate back and forth between 48 KHz. and 96 KHz., the distortion will go away pretty quickly. True, I then have to wait 0:12 - 0:20 (because the system is unusable with the CPU utilization pegged at 100%), but it's a work-around that I can live with. Once I've banished the distortion demon with that particular exorcism, it stays away for quite a while, and that's good enough for me.
Of course, I'll keep working with MotU to resolve the problem. And if Nathan turns out to be otherwise than he seems, then maybe I will return the 828 after all. I've still got another week or so to decide, and my local dealer has already told me about a pretty capable replacement device from Presonus. It doesn't have all the features of the MotU 828 mkII, but it has most of them. Either way, I'm now largely able to make music as I wish, and that's what matters to me going forward.
09/05/2003