Garry Simmons

Blue Kiwi

Thu, 01 Jan 2004 00:00:00 GMT

Blue Microphones was kind enough to send me their Kiwi mic for a test drive at the same time they sent me their new Ball mic for review. Truth be told, I love microphones. I don’t think you can have too many of them, especially if they have a distinctive sound. This would be my first opportunity to use a Blue condenser mic and I was anxious to try the Kiwi.

The Kiwi is Blue’s top of the line solid-state mic. The Kiwi is a large-diaphragm multi-pattern condenser microphone featuring discrete Class A electronics with a transformerless output. The classic “lollipop on a bottle” type design and rich green paint certainly make it a striking and handsome mic. The Kiwi also distinguishes itself by offering NINE, count ‘em, nine pickup patterns. Besides the expected cardiod, omni and figure 8 patterns, you also get three “sub-cardiod” variations (between cardiod and omni), as well as three super-cardiod patterns (between cardiod and figure-of-8). Patterns are easily selected by hand using a knob on the back of the mic. The Kiwi capsule is a multi-pattern variation of their B6 Bottle microphone capsule.

The Kiwi comes packaged in a beautiful wooden case and also includes an elastic, spider-type shock mount at no extra cost. I particularly liked the manual that came with the mic. It was just the right length (6 pages) with a good blend of technical info and application suggestions. The Kiwi is not an inexpensive mic at $2299 list ($1999 street), but quality components and craftsmanship don’t come cheap.

The frequency response chart in the manual shows basically flat response from 20Hz to 1KHz. There’s a small bump centered at 2KHz and a somewhat larger bump centered around 12KHz (est.). The manual goes on to mention that the Kiwi has a reduced proximity effect so that vocalists can get right up on the mic (an inch or two away) without the boominess that proximity effect can add.

I primarily used the Kiwi to record female vocals and acoustic guitar with a fair number of percussion and clean electric guitar tracks thrown in. I also used it to record male vocals when doing a remote recording for a local blues band. For me, the signature sound of the Kiwi is one of clarity. Everyone remarked at how clear things sounded with the Kiwi. It’s certainly bright, but not in a harsh, edgy kind of way.

I had been using a CAD VX2 (a tube mic) for female vocals on my current album project. The VX2 sounded better than my U89 or C412s on the girls. The VX2 does a very nice job (IMO) of splitting the difference between the warmth of the Neumann and the brightness of the AKGs. The girls preferred the clarity of Kiwi to the slightly thicker sounding VX2, so we used the Kiwi on the remaining vocal tracks on the album. Add compression and a bit of low-mids for some extra body and the tracks were done. The (male) lead singer of the blues band loved the way the Kiwi captured his voice. A touch of compression (no EQ) and we were good to go.

The Kiwi sounded great on acoustic guitar in both cardiod and omni modes with little to no EQ required in the mix. I loved the tambourine and shaker tracks I recorded with the Kiwi. Lots of detail, without any grittiness. Clean electric guitar sounded great too.

All in all, I really liked the Kiwi. It has a clear sound that works well on a wide variety of sources, although it might not the best choice for a thin source that you’re looking to fatten up. It looks great, sounds great, is very versatile, and is reasonably priced as professional mics go.

Give it a listen! Blue has a winner on their hands with the Kiwi.

Blue Ball

Thu, 01 Jan 2004 00:00:00 GMT

The Ball from Blue Microphones is Blue’s first entry into the dynamic market after a successful string of highly regarded condenser mics. I was approached about reviewing the Ball at summer NAMM and took Blue up on the offer. I’m always up for test-driving new gear. The Ball arrived a few weeks later, along with a “Blueberry” mic cable (optional).

The Ball is quite a unique microphone. For starters, it’s a big blue ball, about the size of a softball. The Ball certainly continues Blue’s tradition of making colorful, visually striking mics with interesting names, not model numbers. The Ball comes packaged in a cardboard box with Styrofoam packing. You also get a manual/application guide that perhaps tries a little too hard to be funny. At least it’s a light read. Retail price is $279. Street price is about $200.

There’s more to the Ball than just a pretty face. The Ball is the world’s first phantom-powered dynamic microphone. The circuitry in the Ball provides a constant pure-resistive 50-ohm output load to your mic preamp whereas the output impedance of a regular dynamic mic varies with frequency. Blue claims that “frequency dependent variable resistance has dramatic effects on the transducer’s acoustic balance, phase coherence, noise specification and overall output” and that the Ball yields “exceptionally smooth and open sound previously unheard of in a dynamic microphone”. I’m not an electrical engineer and don’t claim to know a whole lot about the gory details of microphone design, so I rely on the test instruments on either side of my head to tell me what works.

I should quickly describe my philosophy about mics and how I use and evaluate them. All mics have a sonic personality or color. The trick, in my opinion, is to select a mic that has a sonic personality that flatters a given sound source. This means little or no EQ is required to “fix it in the mix”. Use a bright mic to liven up a somewhat dull source. Use a “warm” mic to fatten up a thin source. Size, pickup pattern and SPL handling also come into play. As a painter will have a selection of brushes, it is advantageous to have a well-stocked mic cabinet so that you have more colors to paint with.

Which brings us to the Ball.

According to Blue, the Ball was designed to have its bass boost (due to proximity effect) at a different point (about 125Hz) than most dynamic mics. This would indicate the Ball to be a good choice on sources that you would normally find yourself boosting EQ in the 100-200Hz region for a bit of body. Adjust distance to the source to tweak the amount of proximity effect and adjust position (on/off axis) to tweak the high end and you should be good to go. Of course, personal preferences and previous experience factor in a great deal as well.

My plan was to record the same performance on a variety of sources using the Ball and other, well-known dynamic mics. I hooked both mics up to my Earthworks LAB102 mic preamp (a neutral sounding mic pre), fed the Earthworks output to my Swissonic A/D converters and recorded straight to hard disk (24-bit, 44.1K). By recording the same performance to separate tracks, I would be able to compare the tonal differences purely on sonics and not be swayed by differences between takes.

I wasn’t trying to necessarily decide which mic was “better”, but rather gain perspective on how the Ball compares sound-wise to well-known, classic dynamics. I’ll let you decide on the “better” part as that is a function of the specific source and the sound you want to end up with. What works for me on a particular instrument won’t necessarily work for you, but at least I can give you some perspective on how the Ball sounds compared to some well-know mics.

I used the Ball on clean and crunchy guitar amps, bass amps, kick drum, and even tried it on female vocals and acoustic guitar just to see what its top end was like. The other mics used in the comparisons were the ubiquitous Shure SM57, the classic Sennheiser 421 and the Sennheiser 441. The short answer is that the Ball sounded like a good quality dynamic on everything I tried it on. Relatively speaking, the 57 had more snap in the upper mids and the 421 had deeper bass.

The Ball sounded quite similar to my 441 on my guitar amp, which is where it spent most of its time. I preferred the deeper thump of the 421 on kick drum, but suspect the Ball would sound very nice on toms or a smaller kick. The Ball worked fine on bass amp for a live blues recording I did, but I preferred the deeper sound of the 421 on a rock track (with a different amp). Of course the Ball didn’t have the detail and “air” that a condenser mic would have on vocals and acoustic guitar - no dynamic does - but it did a respectable job for a dynamic mic.

A few notes on odds and ends. The Ball is a large mic, so it’s not going to fit in tight places. The mic stand screws directly into the mic, which allows for some range of adjustment in one axis (similar to a mic clip). A small red LED on the front of the mic indicates the phantom power is present. Output level seems typical (within a couple dB) of the other mics I used.

The bottom line for me is that I found the Ball to be a versatile, well-made, affordable dynamic mic. The Ball has a different sonic personality than a 57 or a 421. I tried it on most things that I would normally use a dynamic mic on. While it never blew me away on any given source, the Ball always did a fine job of capturing the source and seems to have found a home in front of my guitar amp. I preferred the 421 on bass amp and kick, but I prefer a deep thump that may or may not be what you’re after.

I think it’s an exceptionally cool looking mic and got positive comments from all manner of people (musicians and general public) every time I brought it out. If you’re attracted to the looks and have phantom power on your mixer, take one for a test drive to see if the Ball’s sonic signature is what you’re after. I’m a sucker for mics with personality and now my mic cabinet has another color to paint with… Blue.

RME Digi 96/8 PST

Tue, 01 Jan 2002 00:00:00 GMT

I sold off my pair of Yamaha DSP Factory soundcards earlier this year and was looking for a simple, reliable soundcard with rock solid Win2K drivers - preferrably with WDM drivers for use with SONAR. I wanted decent sounding stereo analog in and out, Lightpipe in and out, S/PDIF in and out, and MIDI in and out.

I couldn't find anything that exactly fit my needs, so I decided to give the RME Digi 96/8 PST a try. RME soundcards had been getting lots of good reviews from users, so I contacted Tom Sailor (North American distributor for RME products, http://www.xvisionaudio.com ) about getting a Digi 96/8 PST for a test drive. The PST provides everything I was looking for except MIDI and WDM drivers. The RME MME drivers were supposed to be pretty fast, and my Yamaha SW1000XG has MIDI in/out, so on to the test drive.

What It Is

All manner of feature info, specs and manuals can be found on the RME web site (http://www.rme-audio.com), but I'll summarize the high points of the PST to save you the trip.

· PCI soundcard
· Price: $460 list, about $300 street
· Stereo analog ins and outs on RTS jacks (unbalanced)
· S/PDIF in and out on RCAs
· Lightpipe in and out (optical)
· CD-ROM audio input (digital)
· MME, ASIO, MAC and Linux drivers

I/O, I/O, Where do my cables go?

The digital connections on the PST are simple. Just plug in your S/PDIF or Lightpipe cables and get on with your life. The analog section is what separates the men from the boys though. The specs on the PST analog ins and outs are quite respectable for on-board converters (Dynamic range: 109 dB, A-weighted on the ins, 112 dB A-weighted on the outs).

The analog ins and outs use RTS jacks (one jack for the stereo in, one jack for the stereo out), so you'll need to use a Y-adapter or make custom cables of some sort. I had several long insert cables that I hoped to use, but the cables have large barrel Switchcraft connectors and there isn't enough room to put two of them side by side on the PST. So, I ended up with one nice insert cord and one Y-cord.

The analog input is switchable between –10 dBv (default) and +4 dBu nominal input levels using a jumper. The analog output has an attenuator with four settings (in the control applet) of 0, -6, -12, and –18 dB of attenuation as well as master volume fader. When fed a full-scale (0 dBFS) signal, the analog output will produce a +10 dBu output (with no attenuation).

What It Does

Here's a screenshot of the PST's control applet. All the options are clearly described in the manual. There's no mixer panel (aside from a master volume) as it really doesn't apply to a simple audio interface like the PST.

The Input section is worth talking about. Note that you can choose between Optical (Lightpipe), Coaxial (S/PDIF), Internal (for CD-ROMs) and Analog. The critical thing to notice here is that since these are radio buttons, you can only choose one of them at a time. Yes, that means you can't have the Lightpipe inputs and the analog inputs (or any other combination) working at the same time. I have a Swissonic AD24 to feed the Lightpipe inputs, and 8 channels of input is usually enough, but it would have been nice to use the other analog and digital inputs at the same time as well.

The rest of the options are explained nicely in the user guide that comes with the card. I don't feel a need to rewrite the user guide, but if you want to read it, it can be downloaded from the RME web site. I pretty much just left the rest of the options at their default values since my mixer can handle the full-scale output of the analog outs.

Installation

Installation was a no-brainer. My studio PC runs Windows 2000 (SP2). It's a Celeron 566 overclocked to 850 on an Abit BH-1 (BX chipset) motherboard. I downloaded the latest drivers from the RME web site and unpacked them into a folder. Then I shut the PC down and installed the card. When it powered back up, New Hardware was found. I selected the "Have Disk" option and pointed Windows to the driver folder I had just created.

I should note that all testing was done using 24-bits and a 44.1 kHz sampling rate. I can't (personally) justify the extra processing load of using 88.2 or 96 kHz sampling rates in a multi-track environment, but the PST will support those rates on the analog I/O if you're so inclined.

In Use

My primary multi-track application is SONAR. My primary stereo editing software is WaveLab. I also use Sonic Foundry's Vegas Audio and ACID Pro, but SONAR and WaveLab are my workhorses. I had no issues at all using the PST with WaveLab, Vegas, ACID or SAW Studio from IQS (RIP). It just works. SONAR, on the other hand, is worth some discussion.

SONAR prefers to use WDM drivers if they are available since they usually have very low latency. My SW1000XG has WDM drivers, so when I ran the SONAR Wave Profiler, it ignored the PST, and only profiled the SW1K. I then went into the Audio settings dialog and told SONAR to use MME drivers, even if WDM ones are available. My next pass through the Wave Profiler picked up the PST and I continued on.

The next issue with SONAR was the 24-bit settings. There are several ways to pass 24-bit data to/from a soundcard and the SONAR Wave Profiler got it wrong with the PST. Wave Profiler kept saying it wanted to use the "32-bit PCM Right-Justified" when in fact, the PST wants to use the "32 bit PCM, Left Justified" setting.

Finally, if you select the analog or digital inputs in the PST control applet, you get a warning message about the unselected Lightpipe inputs when you launch SONAR or muck about with the Audio options. It's no big deal to click on the "Use Anyway" button, but it's something of an annoyance.

Part of the reason I was interested in the PST were the reports of WDM-like latency from MME drivers in SONAR. It turns out the Hammerfall series has the really low latencies, but the PST doesn't do half bad. If you slide the SONAR latency slide to the Fast end and cut the number of buffers down to 2 (from the default of 4), you get an effective latency of 46.4ms. Not exactly sub-10ms response, but sort of playable if you're not too picky.

Then I happened upon some info from Hans Van Evan (http://www.cakewalknet.com ) regarding the PST and SONAR. I followed his recipe for tweaking the Audio settings and managed to cut my latency in half. The trick is to manually set the buffer sizes for all 44.1K and 48K settings to 256 (on the Driver Profile tab). Exit SONAR and restart and you should now have a latency slider that allows settings between 5.8 and 384ms (with 2 buffers).

Hans reports the lowest usable number on his system is 23.2ms. This matches my experience as well. Perhaps a future driver update will be able to go even lower, but 23.2 isn't too bad. Be aware that you may need to increase your latency setting as your system load increases.

Regarding sound quality, the analog ins and outs sound fine to me. I use the analog outputs for monitoring and have no complaints. I have the analog inputs hooked up to my mixer's group sends, although I usually use outboard mic preamps and converters to feed the Lightpipe input on the PST. The analog input has sounded fine when I've used it. I don't expect mega-buck converter performance from a $300 soundcard, but I have no reservations about using the analog input for any of my work. Better converters are certainly available if you're so inclined.

Summary

Aside from providing the I/O I need, all I want from a soundcard is for it to "just work". It should be transparent in use. No crashes, no driver foolishness, and good clean audio from the analog ins/outs. Aside from some setup tweaks with SONAR, the PST "just works". That's about the highest praise I can give. It just works.

Sure, I'd like to see <10ms latency WDM drivers for the PST, and perhaps a little more space between the I/O jacks, but those are about the only things I can bitch about. Highly recommended.

Yamaha MSP10 Monitor System

Sun, 01 Oct 2000 00:00:00 GMT

Yamaha has long been associated with studio monitors. Love ‘em or hate ‘em, the ubiquitous Yamaha NS10s grace the meter bridge of many a studio, great and small. With the introduction of the all-new MSP10 powered monitor and the SW10 powered subwoofer, Yamaha staking a claim in the ever-popular (and increasingly crowded) powered near-field market. I was lucky enough to get my hands on a pair of MSP10s as well as the new SW10 subwoofer.

MSP10

The MSP10s are two-way powered monitors. The 8" woofer is driven by a 120-watt amplifier. The 1" titanium tweeter is driven by a 65-watt amplifier. The bass reflex cabinet measures roughly 10.5x16.5x13 and weighs in at a hefty 44 pounds. The MSP10 lists for $749. The MSP10M features a sexy maple finish and lists for $799.

The rear of the MSP10 contains the power switch, the XLR input jack, a sensitivity knob, an 80Hz low cut filter and a pair of switches to tailor the high and low frequencies. Oddly, the labels on the frequency switches don't match the dB reductions they provide. The Low switch provides a –1.5 dB change at 50Hz in the "-1" position and a –3.0dB reduction at 50Hz in the "-2" position. The High switch provides a 1.5dB boost or cut at 10KHz in the +1 and –1 positions respectively. The AC power cord is hard-wired to the speaker.

The frequency response specs for the MSP10s indicate that they start to roll off at 60 Hz and are about 10dB down at 40Hz. The frequency response looks reasonably flat out to 20K. Although not as flat beyond 20kHz, the frequency response goes out to 40kHz. I don't have any fancy test gear aside from the two things on either side of my head, but my ears tell me the same thing the published specs do.

SW10

The SW10 is a powered subwoofer featuring a 10" woofer powered by a 180W amplifier. The cabinet is a bass reflex design with a single port measuring roughly 13"x18"x19". The SW10 weighs in at 57 pounds and lists for $849. The SW10 comes in black only (no sign of a Maple version).

The rear of the SW10 has a control panel featuring three XLR inputs, three XLR outputs (pass through), a phase reverse switch, a crossover frequency knob, and a volume knob. The power switch and hard-wired power cord are also located on the back.

Setup

Setting up the MSP10s was pretty much a no-brainer. Since the MSP10s only have an XLR input, I had to change the cabling between my mixer and the monitors (from RTS to XLR). The lack of an unbalanced, ¼" input isn't going to bother most people, but it's worth noting for the home studio guys that run a wild mix of balanced and unbalanced gear.

After listening to the MSP10s by themselves for a while, I decided to hook up the SW10 subwoofer. I ran the stereo output of my mixer to two of the SW10 inputs, then ran the corresponding SW10 outputs to the MSP10 inputs. Since the SW10 outputs are wired to the inputs, the MSP10s are still getting a full-range signal (i.e. the SW10 crossover isn't removing any lows from the signal feeding the MSP10s).

Tweaking the SW10 controls took a bit of listening time. My goal was to get the SW10 to seamlessly fill in the bottom where the MSP10s naturally rolled off. That way, I could listen with, or without, the subwoofer by simply turning it on or off. I ended up with the crossover set midway between the 80Hz and 40Hz marks. I'm guessing that was around 65Hz. I played with the subwoofer volume and eventually settled on the midpoint (there's a detent there), although it spent quite a bit of time just under that midpoint as well.

In Use

The first thing I listened to was a solo album I recently finished for an old bandmate. I knew these mixes inside and out (warts and all). My first impression of the MSP10s (by themselves) was that they were a little brighter than my Mackie 824s and didn't go quite as deep. The vocals sat in the mix differently and the rockers didn't kick quite as hard on the bottom end.

All monitors have a "personality". You always have to adjust your ears a bit to how any given pair of monitors sounds in your control room. After many hours of use (and A/B'ing with the 824s), I'm of the opinion that the Yamaha's have a crisp, detailed sound and the Mackies, by comparison, are somewhat warmer/softer. I still think the 824s get down to 40Hz with more authority, but am not sure they are as flat on the way down. Of course, small room acoustics and monitor placement play a big role in how things sound below 100 Hz, so your mileage may vary.

The MSP10s are very easy to listen to. I had no problems with ear fatigue after hours of steady use. I tend to monitor at moderate volumes, but the MSP10s are very capable of moving some air when cranked up.

I'm not going claim a preference between the Mackies and the Yamahas. They both sound very good, but have somewhat different personalities. Some source material brings out the differences more other material. I would be perfectly happy using the MSP10s in lieu of the 824s, but am not inclined to spend money to switch. If you are in the market to move up to powered monitors in the $1500/pair range, you should definitely make a trip to your Yamaha Pro Audio dealer and check the MSP10s out.

Things got more interesting once I added the SW10 subwoofer to the rig. I like bass. Not flabby, loose bass, but tight, deep bass. I want to hear and feel the low E on a bass (or low B on my 5-string bass). I want to feel the thump of the kick. Once I got the SW10 dialed in with the MSP10s, I was a VERY happy camper. The MSP10s sounded like a much larger system. The sub wasn't loud enough to make you notice it separately. It just seamlessly extended the bottom end of the MSP10s. Exactly what I was looking for... It was a sad day when I had to pack the system up to send back to Yamaha.

There really isn't much to complain about. I would have preferred removable power cords and the flexibility of having ¼" inputs. I like the front-mounted power switch and the downward facing connectors on the Mackies and wish Yamaha had taken a similar approach. These are really minor issues, but even the little stuff matters sometimes.

Bottom Line: The MSP10s are very worthy a contender for your powered-monitor dollar. Crisp, detailed sound. Easy to listen to. Affordable. What's not to like? Add the SW10 sub to the MSP10s and you've got a great sounding, full-range monitoring system.

Swissonic AD24 and DA24 Converters

Sat, 01 Jul 2000 00:00:00 GMT

Although the Swissonic name is relatively new (they were formerly known as MusicNet), the AD24 and DA24 are no strangers to the US pro audio market. These highly regarded converters have been available from Sonorus as the AudI/O AD/24 and DA/24. Sonorus was simply re-badging the Swissonic units (i.e. same box, different paint).

Now, Swissonic is bringing their products directly to you. These converters, as well as the entire Swissonic product line, are now being distributed by Swissonic America. Sonorus is also offering the entire product line, although they will have the Swissonic name on them (see

http://www.sonorus.com/press24.html

for more info). The rapidly expanding Swissonic product line also includes a USB audio interface (the USB Studio D) as well as other converter options (see ProRec review of the AD96 and DA96). More info is available from Swissonic America at infousa@swissonic.com

The AD24 ($749 list) and DA24 ($599 list) are 8-channel A/D and D/A converters featuring balanced analog I/O and the industry standard ADAT Lightpipe interface for the digital connection. The Lightpipe interface allows you to use the converters with a wide variety of digital mixers and soundcards. I wanted a high quality set of A/D converters to feed my soundcards (Yamaha DSP Factory and Creamware Pulsar), and had read nothing but raves from end-users, so I bought one of each as soon as the Swissonic units hit the US.

The AD24 and DA24 are half rack units. I personally like the half-rack approach since it lets you buy the combination of converters you need. I had a spare rack shelf, so I didn't purchase the optional rack shelf. I probably should have purchased the shelf too, as the holes in my "universal" shelf didn't line up perfectly. Let's take a look at the hardware, starting with the AD24.

The front panel of the AD24 features eight pairs of signal level LEDs, three buttons (with status LEDs) and the power switch. The signal level LEDs indicate signal present (green) and a clipping warning (red). There are buttons to select the sample rate (44.1K or 48K), the clock source (Internal or External) and a Calibrate button.

The AD24 manual recommends that you calibrate the AD24 (for best performance) each time the unit is powered on, or when the clock settings are changed. Calibration is easy. Just push the Calibrate button and wait a few seconds for the LED to turn off.

The rear panel of the AD24 contains eight analog inputs (XLR, +4dBu, balanced), the Lightpipe output, word clock input (on BNC) and the power connection. The AD24 is powered by a "lump in the line" style transformer.

The front panel of the DA24 looks identical to the AD24, but the buttons perform different tasks. The three buttons on the DA24 let you select Deemphasis (on or off), the sample rate (44.1K or 48K) and the resolution (16 bit or 24 bit).

The rear panel of the DA24 also looks similar to the AD24. It features 8 analog outputs (XLR, +4dBu, balanced), the Lightpipe input, word clock output (extracted from the Lightpipe input) and the power connector. The DA24 is powered by a "lump in the line" style transformer.

Installation

I'm going to go off on an installation-related tangent since most of us live in the messy world of project studios that contain a wild mix of balanced and unbalanced connections at varying signal levels on all manner of connectors.

Here's my Cliff Notes version of interfacing gear in the studio:

Balanced wiring uses three wires. Unbalanced uses two. Balanced does a better job of rejecting noise, but unbalanced is probably fine for line level (or hotter) signals if you use good quality cable, keep the cables short and avoid running the cables next to noise sources like power strips and wall warts (power transformers).

Balanced wiring usually uses XLR or TRS (stereo ¼" phone plugs) connectors. Unbalanced wiring usually uses RCA or TS (mono ¼" phone plugs) although some consumer soundcards will use an 1/8" mini-jack for line level signals.

Analog signal levels come in three major flavors: mic, line and speaker. Mic level signals are usually pretty low (voltage wise), although some condenser mics have a very hot output (approaching line level). That's why you need a mic preamp to amplify the signal. Line level is what most rack gear uses for ins and outs. Speaker level is the output of a power amp for feeding speakers. Low-level signals are affected more by noise than high-level signals, which is why mic cabling is almost always balanced.

There are two flavors of line level signals, commonly known as –10 and +4. The "-10" name is shorthand for "-10dBV" which means 10dB below a reference voltage of 1 volt. "+4" is shorthand for "+4dBu" which means 4dB above a reference voltage of 0.775 volts. In the real world, a nominal "-10" signal is 316 millivolts. A nominal "+4" signal is 1.23 volts.

Since the two levels use different reference voltages, the actual difference between them is 11.8 dB, not 14 dB. See Lionel Dumond's articles in ProRec that explain all you'd ever want to know about signal levels if you want more info…

The last thing we need to think about is headroom. This is a trickier issue in the world of A/D and D/A converters since there is no "standard" that everyone agrees on. Most soundcard makers publish specs on how their converters work. The things to look for are what signal level does an A/D converter need to produce a full-scale digital signal (0 dBFS). Likewise, look for info on what signal level is output by the D/A converter when presented with a full-scale digital output.

We need to be sure our analog input is hot enough to drive the A/D converter to 0 dBFS. If it isn't, you'll be wasting bits which adds quantization noise to your recordings. We also need to be sure that our mixer or powered monitors can handle the analog output of the D/A converter without distortion or clipping.

So what does all this have to do with the Swissonic converters in question? The AD24 and DA24 are definitely targeted at "pro" users. The analog ins and outs are balanced +4 dBu on XLRs. It takes a +20dBu signal to drive the AD24 full-scale. Likewise, the DA24 will produce a +20 dBu output on a full-scale digital input. If you do the math, you'll see that nominal analog signal levels (+4 dBu) equal a –16 dBFS digital signal.

It would have been more flexible if Swissonic had used TRS jacks that could deal with balanced and unbalanced cabling along with switchable -10/+4 signal levels. So I asked why they only offered balanced, +4 I/O on XLRs... I was told that in order to get the very best performance out the units, they had to pick an I/O standard and optimize the design appropriately. I'm no EE, but I've been told similar things by engineers at other soundcard companies. Makes sense to me…

So, if your studio uses +4 dBu, balanced wiring on XLRs, then hooking up the AD24 and DA24 is painless. Just plug them in and go.

My good mic preamps (Earthworks LAB102 and Daking 52270s) have balanced, +4 dBu outputs on XLRs. Hooking them up to the AD24 was simple. A handful of short (3') mic cables did the trick. The Sub outputs (and direct outs) on my Mackie 1604-VLZ mixer are balanced TRS (with a nominal level of 0dBu) and are capable of putting out +22dBu. So I hacked the female ends off four 10' mic cables and replaced them with TRS connectors. No big deal so far...

As I mentioned earlier, the AD24 will output a full-scale digital signal (0 dBFS) if you feed it a +20dBu signal (16dB above the nominal, +4dBu level). Some semi-pro gear can't put out a signal that hot. This means that if you don't find a way to bump the input signal up, you won't be using all the bits the converters can put out. Luckily, each channel of the AD24 can add about 10dB of gain to the signal if you remove an internal jumper. So I opened up the AD24 (2 screws on each side and 16 on the back) and checked out the jumper situation. It's no big deal to remove the jumpers (save them!). You'll spend way more time messing with the 20 screws holding the top on than removing the jumpers… Since my gear was capable of feeding the AD24 a hot signal, I left all the jumpers in, but it's nice to know that Swissonic planned ahead for people using –10 dBV gear.

There were similar issues with the DA24. Yes, I could have just hooked the first two DA24 outputs directly to my powered monitors, but I prefer having a volume control between my soundcard and my powered monitors. So I made an XLR-to-TRS cable to feed the DA24 output to the aux return inputs on my Mackie mixer and used the volume control on the mixer to control the level in the studio.

Unlike the AD24, the DA24 does not have any internal jumpers to adjust the gain. It will always output a +20dBu signal with a full-scale input. That hot a signal will possibly overdrive your standard -10dBV semi-pro input. The inputs on my Mackie 1604-VLZ list a maximum input level of +22dBu, so I had a couple dB of room to spare there. Be warned. If this is a problem for you, the Swissonic manual shows how to make cables with built-in attenuators (as well as bunch of other interface cables).

So, when it was all said and done, I spent a few hours with a soldering iron hacking up mic cables and putting different ends on them. Yes, it was a pain in the ass, but it wasn't the end of the world either, and you only have to do it once…

Listening

I know, I know… Cut to the chase. What do they sound like? Pretty damn nice if you ask me.

I used the AD24 for a few months while finishing overdubs (vocals, percussion, acoustic and electric guitars) on a solo album project for an old bandmate. My Earthworks LAB102 mic preamp provides multiple outputs, so it was relatively easy to record two versions of a given track; one through the AD24 and one through the converters on the DSP Factory. When I A/B'd the tracks, the AD24 tracks sounded more open and real. I don't claim to have golden ears, but I heard a distinct improvement, so I considered the AD24 to be money well spent.

I spent a little less time with the DA24 because I was basically too lazy to make the required cabling to hook it up to my mixer. Big mistake. When I finally got the DA24 hooked up, playback of tracks I'd been working on for months were now clearer and more detailed. The thing that struck me was being able to really hear the reverb tails.

What I really wanted to do was A/B/C the Swissonic units with the Frontier Designs Tango24 (somewhat less expensive) and something more pricey, like the Apogee AD-8000. Unfortunately, I was unable to track down either piece of gear for a head to head test.

Summary

I'm a practical guy. I tend to spend my money on mics, and my time finding the right mic (and position) for the source I'm recording. The sonic differences between converters are more subtle than the sonic differences between mics. However, once you've found the right mic, the right position and the right preamp, you want to capture the best possible digital representation of the signal that you can. Everything matters.

I don't spend money on gear to be fashionable. I don't buy into hype or voodoo. If a piece of gear doesn't sound better than what I've already got, it goes back. The Swissonic AD24 and DA24 are keepers. They sound great and are affordable. No, they're not the cheapest converters on the block, but they're nowhere near the most expensive either. I think they offer a great bang for the buck. Check them out.

Contact Info:
Swissonic America
407 Stony Point Road
Santa Rosa, CA 95401
(800)-613-2187
(707) 577-7691 voice
(707) 577-7692 fax

www.swissonic.com

infousa@swissonic.com

Programming for the Motor Mix

Mon, 01 Nov 1999 00:00:00 GMT

This supplemental describes how to program for the Motor Mix. I figured I'd give the hackers out there a head start on writing code to support the Motor Mix. I really like the unit and figure it can't hurt to get more people writing code for it. Maybe someone will write a user configurable mapping utility to convert Motor Mix actions into whatever MIDI data a music application might find useful (hint, hint).

Reference

First of all, you've got to be able to read and write MIDI data. I used Paul Messick's "Maximum MIDI Toolkit" book (and CD) to build a C++ class to interface to the Motor Mix. I'd never written a bit of MIDI code in my life prior this, and it was pretty easy. Some of the best money I've ever spent on a book (ISBN 1-884777-44-9).

Daisy Chains

You can connect multiple Motor Mixes together. The first one in the chain transmits on MIDI channel 1, the second one on channel 2, etc. You use the channel info in the incoming MIDI data to figure out which Motor Mix the user is twiddling. I don't know how many people will buy more than one, but you can chain up to 16 of them together.

How it Works

The Motor Mix sends Control Change (CC) messages. All your program has to do to respond to the Motor Mix is to deal with the barrage of CC messages you get. The manual comes with a MIDI implementation chart that explains what happens when every control is twiddled. Here are the Cliff Notes for Motor Mix developers. Note that all values are in hex.

When you press a button or move a fader, the Motor Mix sends out a "control group" message (BN-0F-xx), where N is the MIDI channel (zero-based), the "0F" indicates that it's a "control group" message, and "xx" is the "group". Channel strips 1 thru 8 are groups 0 thru 7. The Left, Right, View and Effect banks are control groups 8 thru 11. You get a control group message for every user action (button down, button up, etc.).

Buttons

If the user twiddles a button, you'll get a "press" message followed by a "release" message (BN-2F-yy). The "2F" indicates a button event and the "yy" indicates which button in the control group. The buttons are numbered from zero thru N within a group. The button id (the "yy" part) is offset by 0x40 for the "press" event. For example, the Mute button down message is BN-2F-02 and the Mute button release message is BN-2F-42. For my application, I basically ignore button releases and trigger all actions on button presses. How you implement the Shift key is up to you (i.e. make the user hold it down or use it like a toggle). I prefer using it like a toggle since it only requires one hand to operate.

Faders

The faders have a range from 0-511. Obviously, the Motor Mix can't send position info using a single byte. Fader position is transmitted using two consecutive messages. You get the MSB in a message like BN-(00 thru 07)-zz, where the (00 thru 07) is the control group (i.e. which fader) and the "zz" is the seven most significant bits of the position (0MMMMMMM). If you only need volume ranges from 0 - 127, then this is all the info you need.

Two more position bits are sent with the LSB message that follows the MSB message. The message looks like BN-(20 thru 27)-zz. Note the 0x20 offset on the control group part of the message. That's your key it's the LSB message, not the MSB message. The zz only uses two of the bits (0LL00000). You'll have to do some bit-wise operations to tack the MSB to the LSB to get the fader position.

Knobs

The knobs on the channel strip can spin continuously in either direction. They don't "stop" like typical knobs on a console. As such, you really have no idea what "position" any given knob is in. It's simply a way to turn stuff up or down from its current setting. When you turn a knob, you get a "knob" message such as BN-(40 thru 47)-xx. The range 40 thru 47 tells you it's a knob twist. Subtract 0x40 to figure out which channel strip the knob is in. The "xx" (0DRRRRRR) consists of a direction bit (the "D") and six rotation change bits (the "RRRRRR"). When the direction bit is set, it's being turned clockwise. I found I needed to write a smoothing algorithm for the knob messages to make LUI Plus "feel" better.

Encoder Knob

The Encoder knob is a continuous knob, like the ones on the channel strips, but you can feel it "click" thru its positions. The Encoder knob works just like the other knobs except the control id is 0x48 (not 0x40 thru 0x47). The Encoder knob can also be a switch (pushed down). This sends a BN-49-xx where "xx" is 0 for a release and 1 for a press.

That's it. Now you're all experts on dealing with input from the Motor Mix. Of course you'll want to control the LEDs in the buttons and write stuff to the display so that Motor Mix reflects the state of your program. And you have to be able to move and position the faders when the user tweaks them via software.

LCD Screen

Writing text to the LCD is matter of cobbling up a SysEx string. There's a header (F0-00-01-0F-11-00) followed by the MIDI channel of the Motor Mix the message is destined for, followed by a 0x10 (indicates you're writing text), followed by the position of the text (0x00 thru 0x4F), the text and finally the requisite 0xF7 to indicate end of SysEx. To make things more efficient, you want to write as little to the screen as possible and do the write in one command, not several (so you don't keep resending the SysEx header stuff).

Controlling the LEDs

Controlling the LEDs in the buttons is easy. You send a control group message (BN-0C-xx), where xx is the control group. Then send an LED command (BN-2C-yy) where yy is the LED you want to set. LEDs are numbered starting with zero within each control group. With no offset, the LED is turned off. Add 0x40 to the id to turn it On. Add 0x50 to make it blink.

Positioning Faders

Setting a fader position is pretty simple too. It's just the reverse of getting the info. You have to send out a pair of messages with the MSB and LSB in them. They are formatted just like the incoming messages.

Graphical Mode

The Motor Mix LCD also has a "graphical" mode that lets do some very basic "graphics" (bar graphs mostly) on the LCD display. It's basically the same idea as writing text to the display. You cobble up a SysEx string and let it rip. You could use these for things like level meters, gain reduction, pan position, and the Q (width) of an EQ. See the Motor Mix docs for more info on the different modes. I prefer seeing numbers to a coarse graphic, so I haven't found the graphical modes too interesting. Your Mileage May Vary.

OK, Now What?

So what does all this mean to you? The user does something to the Motor Mix. You have to figure out which control got twiddled. Based on logic in your app, you may or may not do something with that control. If you need to change the display or light an LED on the Motor Mix, you have to send out MIDI data to make that happen.

YOU - the developer - control every aspect of the Motor Mix. It does nothing on its own. That gives you all the flexibility you could ever want to use or abuse the Motor Mix any way you see fit. Remember, the labels on the buttons are just suggestions. You can map those buttons to anything your program can do. Of course it helps the user if the button label has *something* to do with the way you've used it.

Now go forth and write some code!

CM Automation Motor Mix

Mon, 01 Nov 1999 00:00:00 GMT

Let's face it. Working with many music/audio programs using a mouse can be tedious. I love my "studio in a PC", but sometimes I just want to reach out and grab a fader or knob to get the job done.

It's more than an old habit that won't die. Hardware faders and knobs make good user interfaces. It's the same reason that we don't drive cars with joysticks or fly airplanes with steering wheels. A modern mixing console contains hundreds of knobs and dozens of faders. The knobs and faders are as small as they can practically be, and crammed onto a board that measures in terms of square feet. It is ridiculous to think that we can place all those controls onto a screen that measures in terms of square inches - and control them all with a single finger.

Enter control surfaces. Manufacturers have been scrambling to provide programmable boxes of faders and knobs that will easily integrate into the computer-based recording studio. By providing the user with a set of intuitive, tactile controls, these companies hope to offer us the best of both worlds: the programmability and granular control of a DAW as well as the ease-of-use of a hardware-based system.

By now you've probably seen the full-page ads for the CM Automation Motor Mix controller ($995 list, more info at http://www.cmautomation.com). Perhaps you're curious as to what it will buy you. Motorized faders. Big display. Lots of buttons and knobs. It's a pretty sexy device. This article will hopefully answer all your questions about what the Motor Mix is - and what it isn't. I've even included a supplemental on programming for the Motor Mix in case you're inclined to roll your own Motor Mix support.

CM Automation Motor Mix

I've been using a Motor Mix since May '99. I got a beta unit to beat on in conjunction with MxTrax for the DSP Factory (more info at http://www.minnetonkaaudio.com). I saw the potential for the Motor Mix to control most of the DSP Factory's parameters, so I added extensive support for the Motor Mix to my LUI program and dubbed it "LUI Plus" (more info at http://www.simmonsinteractive.com). I'm a Motor Mix developer as well as user, so this is the inside scoop on the unit.

Hardware

The Motor Mix is fairly compact, measuring 10.5" wide x 12.5" deep x 5" tall. The Motor Mix has eight channel strips that are surrounded by four banks of buttons. The LCD display (2 lines x 40 characters) and the Encoder knob complete the front of the unit. That's a grand total of 8 motorized faders, 67 buttons, and 9 knobs. That's a lot of controls.

The back of the Motor Mix includes the AC connection (no wall wart!), a power switch, MIDI In and Out ports, a 15-pin serial port, and a contrast control for the display.

Most of the buttons are clear plastic, with a green or red LED inside them. The LED can be on, off or blinking. Most of the buttons have two labels near them, much like a scientific calculator. The idea is that you can use the "Shift" button in conjunction any other button to have access to more functions. Note that the button labels mean absolutely nothing. It's up to the application that is talking to the Motor Mix to decide what any given button actually does.

For the purpose of this discussion, the banks of buttons surrounding the channel strips will be called the Left bank (left of the faders), the Right bank (right of the faders), the View bank (left of the channel strip buttons), and the Effect bank (right of the channel strip buttons). Please remember that the button labels on the Motor Mix are only suggestions. The Motor Mix Developers Guide gives examples of how these buttons could be used (given their names), but these are only suggestions. What the programmer actually does with any given button is up to them.

The Left bank has eight buttons labeled (from bottom to top) Shift, Undo, Default, All, Window, DSP, Suspend and Auto Enbl. The Shift key is supposed to be used in conjunction with other buttons. The other seven buttons provide a variety of handy "verbs" (especially Undo and Default for editing).

The Right bank also has eight buttons that are labeled (from bottom to top) Escape, Enter, Last, Next, Rewind, F. Forward, Stop, Play. Obviously these buttons are expected to be used for selecting data and transport control.

The View bank includes left and right arrow buttons, a Bank button, a Group button and three "burn" buttons. The arrow keys are typically used to change the "window" of channels that the Motor Mix is controlling. If the Bank button is lit, then you move in banks of eight channels. The Group button is used to move through "groups" of instruments. The "burn" buttons include a Record Ready button, a Write button and an "Other" button. The state of the burn buttons is supposed to control what the "burn" button on each channel strip is doing.

The Effect bank (my name, the Developer Guide calls them Multi Controls) includes four buttons labeled FX Bypass, Send Mute, Pre/Post and Select. I find the "shifted" labels more useful (for LUI Plus) which include Eff-1, Eff-2, Eff-3 and Eff-4. The state of these buttons indicates what should happen when the Effect button (the button below the knob) on each channel strip is used.

The Encoder knob and display are located just above the Effect bank buttons. The Encoder knob is a continuous knob that clicks like a ratchet. The idea is that you use the Encoder knob to select a typical channel strip parameter, such as pan or an aux send level, and then use the knobs on the channel strips to change the parameter selected by the Encoder knob. The two character display above the Encoder knob gives you a way to tell what parameter is currently active (such as PA for pan, or A1 for aux send #1). The knob can also be used a button (push the knob down).

Each channel strip includes a 100mm motorized fader, five buttons and a knob. The buttons closest to the faders are labeled Mute and Solo. The other three buttons on the channel strip are not labeled with any special function name, but are called (by me anyway) the Burn button, the Effect button and the Select button. The background coloring around the buttons gives a clue as to how CM Automation envisioned the buttons being used. The next button up the strip (above the Solo button) is called the "Burn" button. It has a gray background to indicate it could be related to the state of the buttons to the left of channel strips (also with a gray background). CMA suggests using this button to control the "record ready" or "write automation" status of a channel. The Effect button (located below the knob) has a tan background and could be associated with the bank of Effect buttons to the right of the channel strips. This button would obviously be tied to editing effect parameters. The Select button (located) above the knob could be used as a switch that works in conjunction with the content of the LCD display since they are right next to each other.

What It Does

The Motor Mix is not your standard MIDI fader box. It can not be programmed or configured by the end user in any way. You turn it on and twiddle controls. The Motor Mix sends out MIDI data (CC messages) in response to user actions. It is up to the receiving application to determine what the user is doing, how to respond to it and what to do to the Motor Mix (change the display, turn on LEDs, whatever) in response.

The application can send messages to the Motor Mix to position the faders, control the button LEDs and to write text to the LCD display. Fader positioning and LED control is accomplished using CC messages. The LCD is controlled via SysEx messages.

The Motor Mix does not save any "state" information. What I mean by that is that the Motor Mix has no idea what should happen when the user pushes a button. It's simply a way to trigger events (via MIDI data) that your program can deal with. For example, if I press the Solo button on a channel strip, the Motor Mix simply transmits a bunch of data saying a button was pressed and then released. It is up to the application to associate that event with solo'ing a channel. It is up to the application to tell the Motor Mix to turn the light On in the solo'd channel and (perhaps) to turn the Mute buttons on for all non-solo'd channels.

Software Support

As I've stated earlier, the Motor Mix doesn't do anything particularly useful on its own. It requires your application be Motor Mix-aware. With the winter NAMM show approaching, I'm sure that many software companies will be releasing updates to their products that include Motor Mix support. The Motor Mix ads sure have a lot of logos on the bottom of the page, so support is on its way.

As of this writing (Nov. 99) the following software makers have committed to providing Motor Mix support. Be aware that the level of support can vary wildly, so be sure to check with the vendor first. The "Support" page on the CM Automation web site lists the following companies along with a URL to hit their web site for more info.

Cut to the Chase

In a nutshell, if your application has extensive support for the Motor Mix, it's a wonderful way to control a music/audio program using real faders and buttons and knobs. It's so much nicer to use a well-built hardware controller than twiddle controls with a mouse. I can't say enough about how much I like using the Motor Mix to control the DSP Factory. It really makes the DSP Factory feel like a mixer now. The cool factor with clients doesn't hurt either. They dig seeing those faders jump when you change banks.

If, the other hand, your application doesn't have built in support for the Motor Mix, or can't be programmed to do something useful in response to the messages the Motor Mix sends, you've bought yourself an expensive doorstop. Given the wide support that exists (or is coming) for the Motor Mix, that isn't likely to be the case for many people.

So how does the Motor Mix compare to the current crop of digital mixers? The obvious answer is that the Motor Mix is NOT a mixer. Audio never runs thru the device. There are no mic preamps, no equalizers, no compressors and no effect units. The Motor Mix is simply a way to control your software using a mixer-like device. Think of the Motor Mix like a specialized mouse for audio engineers. It's really aimed at people that are mixing inside the computer, not ones that are using the PC like a digital tape deck and feeding an external mixer. If your software fully supports a digital mixer (such as the Yamaha 01V) as an external controller, then maybe the added functionality (preamps, etc.) is worth the extra cost and bulk. Only you can decide what gear fits the way you want to work and your budget.

As always, research a product before spending your money. The Motor Mix is a fine value, but it isn't cheap. Make sure you understand the level of support your application provides for the Motor Mix since it can vary widely. I use my Motor Mix every single session and love it to death. If you've found yourself wanting to grab a real fader or knob instead of messing around with a mouse, you owe it to yourself to find out if your favorite audio app(s) support the Motor Mix. I recommend it highly.

Soundscape Mixtreme

Sat, 01 May 1999 00:00:00 GMT

When I first started paying attention to digital audio, the popular soundcards were the Turtle Beach Multisound and the DAL CardD+. Stereo analog ins and outs were the order of the day. Then the digital I/O cards arrived, and you could get various combinations of analog and digital (S/PDIF or AES/EBU) ins and outs on a card. Life wasn't too bad on the PC-DAW frontier.

Now, not that many years later, we have a bewildering number of soundcard choices. Multi-IO cards featuring eight or more ins and outs of various flavors of analog and/or several digital formats (Lightpipe and TDIF), plus all kinds of new sync options and converter options are available. Most of these multi-IO cards are powered by a DSP of some sort. The soundcard usually comes with a mixer applet gives the user varying levels of control over the signal levels and routing. Many users have wished for the ability to use any leftover DSP power to do audio processing for them, so that they could lessen the load on their host CPU. Which brings us to the subject of this review...

The Soundscape Mixtreme is the latest entry in the land of multi-IO soundcards. Soundscape (http://www.soundscape-digital.com) has been making digital audio workstations (the highly regarded SSHDR1) since 1993. The Mixtreme is their first venture into the Windows soundcard market.

I received the following package from Soundscape: one Mixtreme card ($549 list) with optional S/PDIF I/O ($149 list), two SS8IO-3 analog-to-TDIF I/O units ($599 list, each) and a variety of optional software plug-ins from Soundscape, TC Works and Wave Mechanics. Soundscape also sells the SS8IO-2, a Lightpipe to TDIF converter ($349 list) and the SS8IO-1, a pro quality analog to Lightpipe and TDIF converter featuring balanced analog I/O ($1695 list). There is lots of product info on the Soundscape web site.

Discounted package deals are also available. For instance, the Mixtreme plus SS8IO-3 lists for $999, a $150 savings. The Mixtreme plus an SS8IO-2 lists for only $749, $150 off the individual prices. Discounts on the Soundscape Audio Toolbox and Wave Mechanics reverb plug-ins are also available.

Hardware Overview

The Mixtreme is a PCI card that is powered by a single Motorola 56301 DSP chip. It features 16 channels of digital I/O using a pair of TDIF ports. A pair of RCA connectors provide wordclock (or superclock) in and out. A MIDI In/Out/Thru to 9 pin SubD breakout cable will provide for MTC in a future release of the software. There is no analog I/O on the card. Optional S/PDIF or video sync daughter-boards can be snapped onto the Mixtreme.

The Mixtreme card itself has a single TDIF port and a pair of RCA connectors. The second TDIF port and the MIDI port are mounted on a bracket that connects to the Mixtreme via ribbon cables (note the blue connectors on the card). The bracket can fit in the card opening for a spare slot, or you can use the "punch outs" on your PC case to mount them. Since I only ended up using one of the SS8IO-3s, I didn't bother installing the bracket.

Installation

My studio PC is a Celeron 300A (running at 450MHz) on an Abit BH-6 motherboard with 128MB of SDRAM. I use a pair of Maxtor UIDE drives for audio data. The review was done with a fresh install of Win98 and little else but audio applications installed. The majority of the review was done using Cakewalk Pro Audio 8.04 and WaveLab 2.02.

The manual that comes with the Mixtreme has detailed installation instructions. I found the installation to be pretty easy. I downloaded the latest drivers (1.03 at the time) and had them waiting in a separate folder on my hard drive before installing the card. Insert the card in a PCI slot, power up your PC, and Windows will detect it. Use the "Have Disk..." button in the Plug-and-Play wizard to point to the drivers you want to use.

The Mixtreme supports both MME drivers and ASIO drivers, but the MME drivers were not on the CD that came with the card. I needed to download the 1.03 software to get them. You can run the Mixtreme under Win9x or NT. I tested the Mixtreme using the MME drivers under Win98. As of this writing, version 1.04 of the drivers has been available for a while. No doubt 1.05 will be out soon enough.

I also downloaded and installed a variety of Mixtreme plug-ins. I installed the Soundscape Audio Toolbox, the TC Reverb, the TC Dynamizer, and the Wave Mechanics Reverb. Each of these plug-ins can be demo'd by simply downloading and installing them, but it takes a password (specific to your Mixtreme card) to unlock them for full use. Soundscape was nice enough to provide me with passwords for all the plug-ins as well. More on the plug-ins later in this article.

IO, IO, It's Off To Work We Go...

At its core, the Mixtreme is a multi-channel (16 ins and outs) digital I/O card. There are lots of digital I/O cards out there, so what makes the Mixtreme different? The most obvious physical difference between the Mixtreme and other multi-channel digital I/O cards is the use of the Tascam TDIF format versus the somewhat more common ADAT-compatible Lightpipe format. This makes the Mixtreme a natural fit for anyone with one or two Tascam DA-x8 MDMs or a Tascam digital mixer. If you find yourself wanting a Mixtreme, but already own gear that talks Lightpipe, you can use any one of several Lightpipe-to-TDIF converters, including the Soundscape SS8IO-2.

Most of the affordable digital mixers on the market let you use a combination of Lightpipe or TDIF (or AES/EBU even) interfaces via optional I/O cards. Chances are good that you can configure most digital mixer with the I/O cards that will let you use the Mixtreme as your audio interface to the PC.

Soundscape's inclusion of word clock (and superclock, standard) is an indication that they see the Mixtreme fitting into the world of post-production and video where the Tascam DA-x8 is much more common than the ADAT.

The optional Video Sync board ($149) includes a video input and an S/PDIF output and plugs on to the Mixtreme card instead of the S/PDIF option. This allows Mixtreme to sync it's internal sample rate to a video blackburst signal so that there's no sound to picture drift when working with video (for example if using Mixtreme as an audio output device for a video editing system). Drivers supporting the option card are currently being tested and the option appears automatically in the software as an additional item under the Master Clock entry in the Settings menu.

Finally, the MIDI breakout cable will support an (upcoming) MIDI interface. This won't be a standard MIDI interface (there are plenty of very nice ones out there already), but a connection for synchronizing to MTC.

D.S.P. (Don't Squander the Power!)

Far and away the biggest difference between the Mixtreme and other multi-channel digital I/O cards is the Mixtreme mixer application. Calling it an "applet" isn't really doing it justice. The Mixtreme uses the same DSP as the Event Layla and Sonorus StudI/O, but read on to find out about all the cool stuff you can do with the DSP on the Mixtreme.

Before diving in, I need to state the obvious. All DSPs have limits. Even the "digital mixer on a soundcard" products (with several DSPs) have limits. The key to using the Mixtreme is to figure out how to best integrate it into your system. I expect that many Mixtreme owners will use a combination of host-based CPU, Mixtreme-based DSP and perhaps some outboard processing units. The thing to remember is that the Mixtreme is *very* flexible, so it can adapt to *your* needs, not the other way around.

NOTE: The lower right corner of the Mixtreme mixer window displays two values. The "P" value is the percentage of DSP power being used. The "M" value is the percentage of memory being used. Keep an eye on these values as you experiment with various mixer configurations. The DSP required for track inserts (streaming audio to/from your app) is NOT included in these numbers. This will take varying amounts of DSP power depending on your system. You will hear very obvious pops and crackles if you go over the limits, so pay attention when you push beyond 80 or 85% of the resources used.

You can start by creating your own mixer from scratch, or you can load one of the many mixer presets that are installed. It's probably a good idea to play around with the presets to get an idea of the many ways you can configure a mixer before spending the time to roll your own.

To give you a feel for what you can do with the mixer, let's go through the options you have available and I'll provide my take on how to use them. I should note that the Mixtreme manual does a fine job of describing the mixer options. There were a few things that weren't obvious to me at first glance, so take your time and READ the manual. Lots of good info in there if you pay attention. So, let's create a mixer!

When you create a new Mixtreme mixer, you are faced with 128 blank channel slots. You work with the mixer in one of two modes. You are either in Edit mode (changing the mixer configuration), or Control mode (using the mixer). You can switch between the two modes by clicking on a button on the toolbar or simply hitting the "E" key.

In Edit mode, you need to select which "tool" you are going to use. Your choice of tools includes:
Create (add a new mixer element to a channel)
Move (change a mixer element's position in a channel)
Delete (remove a mixer element from a channel)
Mute (disable a mixer element from a channel)
IO Assign (lets you configure the ins and outs of channel)
Info (displays information about a mixer element)

The cursor bitmap changes to reflect the current tool. Let's take a look at using the Create tool since that is where most of the interesting stuff lives.

Once you are in Edit mode, and have selected the Create tool, you simply left click on a channel slot to choose the type of channel you want to create. The Mixtreme supports every combination of mono and stereo inputs and outputs you can think of. I expect most people will either use Stereo In/Stereo Out, or Mono In/Mono-to-Stereo Out. The "Mono to Stereo" output option basically creates a mono channel strip with a pan control for feeding a stereo output (much like the mono channel strips with pan pot on a typical analog console).

TIP: Stereo channels are more efficient than a pair of mono channels. If you plan on applying the same processing to a pair of channels, use a stereo channel rather than a pair of mono ones. This same approach applies to creating a stereo submix of, say backing vocals, in your audio application rather than creating separate mono channels for each backing vocal track.

Now that we have a channel created, we see an input source at the top (one of the Mixtreme inputs) and the output section of the channel, which includes the channel fader, a pan pot, the channel meter (pre-fader), solo and mute button, etc. There is a blank area between the channel input and the channel output. This blank area is where you insert additional mixer elements. If you need a taller channel strip, just resize the window and the insert area gets larger.

The next step is to insert mixer elements in the channel. Signal passes through the mixer elements in top to bottom order, so you have a lot of flexibility in designing the channel strip just the way you want. The standard mixer elements (available in mono and stereo versions) are:

Track (mono, stereo) - inserts a connection to the playback and recording drivers
2-band EQ (mono, stereo) - fully parametric, 2-band EQ
Fader (mono, stereo) - simple volume control
Sample Delay line (mono, stereo) - for aligning audio between channels
Peak Meter (mono, stereo) - small meter w/o numeric feedback
Send Pre/Post (mono, mono to stereo, stereo) – pre/post fader aux sends
Send Pre/Post w/ Equal Power Panning (mono to stereo, stereo) - pre/post fader aux sends

Most of the elements are pretty straight forward and do exactly what you expect they would do. It's worth spending some time talking about the Track and Send elements though, as they are very important to your mixer design.

The Track element is where the recording and playback drivers are connected to the channel. When I read the manual the first time, the fact that the track inserts applied to BOTH recording and playback was lost on me. I was expecting playback only. I should learn to read slower :-) As the manual notes, the ability to place mixer elements in any order lets you do things like process a live input with EQ and record the EQ'd signal. To do this, simply place the Track element after the EQ element. Want to record the straight signal but monitor it with EQ? Place the Track element just below the input (before the EQ) and you'll be recording the raw input. You can even have multiple instances of the same track element in one channel. This allows you to place the playback and record points at different places in the channel. Just mute the one you aren't using.

Having tightly coupled record and playback drivers (via a single track insert) works well enough, but I'd like the flexibility of having the two sets of drivers be more independent. It's certainly easy if you want to use (up to) 16 tracks of audio in your application and have those tracks feed individual channels on the mixer. It's much like using a tape deck with direct outs from a console hardwired to specific tracks on the deck.

I experimented with dedicating a stereo pair of drivers (15/16) for recording and only using 14 channels of playback between Cakewalk and the Mixtreme. I created two mixer channels that were fed by inputs 1 and 2 and routed them to recording drivers 15 and 16 (with no processing). I could monitor the channels in real-time because those two mixer channels were also feeding the stereo output. This worked well for me and also turns out to be a way to digitally record the stereo output of the mixer to hard disk.

The process of creating a mixer from scratch is easy enough, but can be a bit tedious. For instance there is no way to copy channels or groups of elements. I found that creating my own mixer template was the way to go. I basically created a mixer that had lots of (identical) channel strips that were loaded with mixer elements.

Now pay attention (this is important). Muting an element frees up its DSP resources. Bypassing an element does not. So my template mixer had mostly muted elements. I would unmute the ones I needed for any given project, and then save the mixer for that particular song.

I traded several emails with Soundscape about ways to streamline the mixer interface and editing procedures. I'm happy to report that most of my suggestions are already on their "To Do" list. They tell me that putting out solid, bug-free releases, is a higher priority than feature bloat. I'm happy to report that the software was rock solid for me during the review. Given the sad state of most music applications these days, that's saying something.

Get On The Bus, Gus

The Mixtreme has 16 buses. You can use them as 16 mono buses, 8 stereo buses, or any combination of mono and stereo that your project requires. Creative use of the buses is very important to mixer design.

If you left click on the input at the top of a channel strip (in Edit mode using the IO tool), you will see that a channel can be fed by any one of the 16 (hardware) inputs or the 16 buses. If the channel has stereo inputs, the inputs and buses will appear as odd/even pairs. Similarly, the output at the bottom of the channel can be pointed to the 16 outputs or one of the 16 buses. Channels that have stereo outputs will show the outputs and buses as odd/even stereo pairs.

Note that there is no "master" fader or stereo mix bus. By default, channels directly feed outputs 1 and 2. If you want a "master" fader, create a stereo in/stereo out channel that is fed by Bus 1/2. Then assign all your mixer channels to output to Bus 1/2. This new channel is your stereo master. Also note that you can insert mixer elements such as EQ and compression on this master fader channel, just like using master inserts on a console. This same approach can also be used to create subgroups if you wish.

The Send mixer elements are typically used to create effects sends and headphone mixes. Each send element has a destination, just like channels do. Sends can output to physical outputs or to buses. If you route a Send to a physical output, you could have a headphone amp or external effects processor connected to the output. In the case of an external effect unit, connect the effect unit outputs to a pair of Mixtreme inputs and create a stereo effect return channel in your mixer that is fed by those inputs.

Sends that are routed to the Mixtreme's buses are usually used for effect plug-ins. In this case, create an effect channel. Its input will come from a bus (usually stereo). It will output to the "master" bus. Insert the effect plug-in on this channel, and set its wet/dry mix to 100% wet. Now simply add Send elements to the channels that you want to apply the effect to. Set the individual Sends to the same bus as the input for the effect channel. Use the fader on the effect channel as your effect return volume control. You can add EQ or compression to the effect channel for even more control. The only limit to the number of effect channels you can create is the amount of DSP power the effects use.

Multiple Card Support

Multiple cards can be used together. The latest software update (1.04) provides a full set of drivers for each card. I didn't get the opportunity to try multiple Mixtremes (darn!), but here's my understanding of how they work. In essence, you still have two separate mixers, but they share a common user interface (the mixer window). When you create mixer channels in a multi-card setup, you are asked which card the channel applies to. You'll probably want to physically interconnect the cards in some fashion so they can share audio. Rumor has it that multiple cards will be able to talk to each other over the PCI bus in a future software update. Note that multiple cards do NOT require extra IRQs (just one for the original card).

SS8IO-3 Analog to TDIF Converter

The SS8IO-3 is Soundscape's affordable analog-to-TDIF interface solution. The SS8IO-3 is a half rack unit featuring 20 bit converters. It has eight analog inputs and eight analog outputs on the back (RCA, -10dBV). The front panel of the SS8IO-3 has two LEDs per input channel plus buttons/LEDs for changing the clock settings. The SS8IO-3 is powered by a wall wart. The SS8IO-3 is connected to the Mixtreme via the supplied two meter ribbon cable. You might get away with longer ribbon cables if you are careful about routing. If you need a longer cable, you can buy them from Tascam (Warning! They're not cheap!).

I measured the RMS Power level of recorded silence (nothing plugged into an analog input) using WaveLab 2.02. The Mixtreme/8IO-3 combo came in with a -99.6dB reading. That puts it in the same ballpark as similar I/O units. Not as quiet as the pricey gear, but still a bit quieter than 16 bit (which usually measures between -90 and -93dB). Test scores beyond a certain point don't mean a whole to me. To my ears, the converters in the 8IO-3 sound fine and are plenty quiet. Things like mic choice, mic position, and ambient noise in the room make more of a difference (to me) than the differences between converters in this price range.

NOTE: Since the analog I/O is unbalanced, you'll want to take the usual care with unbalanced wiring. Use good quality cable, keep cables short, and keep cabling away from AC power and wall warts. I have a mix of balanced and unbalanced wiring in my studio and generally don't have any noise problems. I really would have preferred 1/4" jacks instead of RCAs though.

Go to the Soundscape web site for product specs and photos of the unit (http://www.soundscape-digital.com/Products/Ss8io-3/).

Optional Mixtreme Plug-Ins

I had the opportunity to use Soundscape's Audio Toolbox ($325 list), the TC Reverb and Dynamizer plug-ins ($599 and $799 list respectively) and the Wave Mechanics Reverb ($349 list).

You get a voucher for 30% off list price for the Audio Toolbox and WM Reverb in the box with the Mixtreme. Additional passwords (for multiple Mixtremes) are available at a 40% discount off list price.

As I mentioned earlier, Soundscape provided me with passwords to unlock all the plug-ins that are currently available for the Mixtreme. This list includes the Soundscape Audio Toolbox, the TC Works Reverb and Dynamizer, and the Wave Mechanics Reverb. More are on the way…

Since effect quality is such a subjective thing, let me just say that all the plug-ins are good values and that the more expensive ones sound better (to me anyway) than the less expensive ones. I suggest that you download the demos and make up your own mind.

Here's some basic info and a screenshot on each plug-in…

Soundscape Audio Toolbox

The Soundscape Audio Toolbox ($325 list) includes three plug-ins: a very cool (but efficient) Dynamics plug-in, a Chorus/Flanger plug-in, and a 2-Tap Delay plug-in. The Chorus/Flange and 2-Tap Delay are useful enough plug-ins, but the real reason for buying the Audio Toolbox (IMO) is the Dynamics plug-in.

As you can see from the screenshot, the Dynamics plug-in provides a Gate, an Expander, three Compressors, and a Limiter (in series). A visual representation of the transfer function is shown to the left of the controls. The three compressors let you dial in increasing compression ratios to build your own "soft knee" compression. Having a gate and an expander is great for removing things like headphone bleed or amp hum (and is way faster than editing out all those bits in your source material). The limiter adds that final level of control if you simply don't want the level to ever exceed X. The individual sections (gate, expander, etc.) can be toggled on and off by clicking on the label above the controls.

My only complaint with the Audio Toolbox is the lack of presets (there are none). This plug-in is definitely worth the money since it can be had for about the price of budget compressor and you get delay and chorus/flange plug-ins along with the dynamics.

TC Works Reverb

The TC Works Reverb is not an inexpensive option ($599), but top shelf outboard reverbs are WAY more expensive. I really liked the TC Reverb. It's supposed to be the same algorithms as the acclaimed M5000 unit. All I know is that it sounds really nice to me. It's also highly tweakable (see screenshot) and comes with a wide variety of presets to use as the starting point for your own masterpieces. One thing to note is that it is a pretty DSP-hungry plug-in. Expect it to take about 30% of a DSP.

TC Works Dynamizer

The Dynamizer is a high-quality, multi-band (3 bands), look-ahead expander/compressor/limiter. I didn't really play with the Dynamizer much. I ran some mixes through it and played around a bit, but I was mostly tracking and overdubbing during the review. Given the large amount of DSP it uses (40% of a DSP), it's more of a mastering tool. It comes with a bunch of presets and has lots of adjustments (see screenshot). It's even more expensive than the TC reverb at $799 list, but this is a serious tool and is way less expensive than the equivalent hardware.

Wave Mechanics Reverb

The Wave Mechanics Reverb ($349 list) is available at a good discount with the Mixtreme. It is also highly tweakable (see screenshot) and sounds pretty nice. I preferred the TC Reverb (at nearly twice the price), but the WM Reverb is a fine value. I wouldn't have any problems using it on my projects. The WM Reverb comes with a ton of presets that are nicely organized (plates, rooms, halls, etc.). This reverb is the most DSP hungry of the bunch, using 42% of the DSP.

Summary

I really liked the Mixtreme. You have to check this card out if you find yourself wanting a TDIF-based digital I/O card that does a whole lot more than just I/O. The software is very solid and easy to use. There really isn't any competition at this point. I think the Mixtreme has carved out its own niche for the time being. It falls nicely between the I/O-only cards and the multi-DSP "digital mixer on a soundcard" offerings.

Although we'd all like to have the latest, greatest PC, it's a fact of life that many of us have to milk older systems while saving our pennies for a new one. The Mixtreme is an affordable way to extend the life of your system by offloading some of the host-based processing to the DSP and still get the multi-IO capabilities you need. Unlike some of the "mixer on a soundcard" offerings, the Mixtreme can run on just about any PC that will run Win9x and has a PCI slot. The mixer application doesn't use a lot of system resources, and runs at a low priority so that your audio application can have the CPU when it needs it.

I traded several emails with Soundscape about where the product is going and it has a very bright future ahead. More native plug-ins have been announced, including offerings from Aphex and a new Dolby Surround plug-in for 4-channel sound (LCRS). One somewhat glaring omission is the lack of automation. I have been assured that full automation (including plug-ins) is a very high priority for the product and will be included in a software update.

I want to thank Chris Wright and Jerry Breiner from Soundscape for patiently answering many rounds of questions via email. They were VERY responsive and knowledgeable folks. Don't forget to check out the Mixtreme info on the Soundscape web site (http://www.soundscape-digital.com). Soundscape also offers a "Users Corner" (newsgroup) on their site for users to help each other out. The only catch is that you must be a registered user to access it.

I really enjoyed my test drive of the Mixtreme. Once I figured how to best utilize the card for my needs (music production), it just quietly got the job done. Can't ask a soundcard to do more than that. My only complaint (for my needs) is that I occasionally ran out of DSP power when trying to run more than a dozen tracks with lots of processing at the same time (such as the TC reverb). To be honest, I bet a two Mixtreme system would really rock, and given the low list price, buying two isn't that big of a stretch. You don't hear a lot of buzz about this card online, but Soundscape has a real winner here if you ask me. Don't be a sheep. Check it out!