Jump to content

MACKIE ONYX 400F (audio interface)

Recommended Posts

  • Members

Mackie Onyx 400F – Prologue


Short form: Mackie’s Onyx 400F is a cross-platform FireWire audio interface that can also work in standalone mode, with bundled sequencing software and a mixer applet. Currently, my main interfacing needs are handled by the Creamware SCOPE system, which I use for its dual ADAT light pipe, SPDIF, and MIDI I/O. It also has two channels of analog I/O, and while these are okay, they are on a card located inside my computer…so there are, shall we say, some inherent limitations.


For analog recording I’ve been using the preamps and converters in my Panasonic DA7 digital mixer, which are definitely better than average, but not “boutique quality.” As it looks like I’ll be doing some more acoustic recording in the future, including an on-location piano recording with a FireWire-friendly laptop, the Onyx 400F definitely piqued my interest. What better way to find out about it than by doing a Pro Review?


The hype on Mackie’s web page for the Onyx 400F starts with “There are plenty of FireWire audio interfaces on the market today.” Well, no doubt about that, so points for honesty. But then they make their case as to why you should choose theirs: Really good mic preamps, rugged construction, good mixer software applet, etc. They even throw in a bundled version of Tracktion 2, which is actually a pretty good deal (hint: laptops love it). The package list price is $899.99, considerably above the “cheap ’n’ cheerful” category but well below the stratospheric “well, maybe when I get a gold record” level.


Click on the attachment to see the Onyx 400F front panel.


What I want to find out is how easily it installs, how it sounds with my mics, and whether the mixer application is flexible and easy enough to use that it’s an asset to the package. I’m also going to see if I can crack it open and check on the quality level (guess I’ll void the warranty, eh?). So all in all, this should be a fun Pro Review.


For those of you not familiar with how the Pro Review concept works, it’s simple: This is a forum, so everyone’s invited to participate in this “open source” review. If you have questions or comments, step right up and voice them. I’ve also invited Mackie to participate, and we’re fortunate that Dan Steinberg, Mackie’s Recording Product Manager, will be around to answer questions and offer insights on the product as required.


Here’s the link to the Onyx 400F “landing page,” which gives the basic specs, features, system requirements, and the like so I don’t have to waste bandwidth repeating them here:




All right, time to open the box and start the review!

Link to post
Share on other sites
  • Replies 598
  • Created
  • Last Reply

Top Posters In This Topic

  • Members

The Onyx 400F is packaged well in a sturdy cardboard box, covered with a cardboard wrap. Even UPS couldn’t dent it. You get:


 The Onyx 400F itself

 Two detachable line cords (one for US and one for the European connectors used in Germany and other countries)

 A loooooong 6-pin FireWire cable (didn’t measure it, but I’d guess 15 feet) Printed manual written in Mackie’s inimitable, breezy-yet-comprehensive-with-a-dash-of-humor style

 CD-ROM with Windows drivers for ASIO/GSIF/WDM (the Mac doesn’t need drivers), control panel software, and a full – not demo – version of Tracktion 2 software. There’s also a PDF version of the manual on the CD-ROM for those of you who appreciate a format with a search function.


The manual is online, so you can check it out here:



What you don’t get is a 6-pin to 4-pin FireWire adapter. If you’re planning on using the Onyx 400F with a laptop that has a 4-pin connector, visit your local computer or office supply store and buy the appropriate adapter.


Of course, even before powering it up or plugging it in, the first thing I did – which I do with any software or hardware that runs software – was to check the CD-ROM for a read me. And yes, there was one on how to update the firmware and software.So next was a trip to Mackie web site to check for updated drivers. Sure enough, V1.05 was there. After a mercifully small 897K download, I was ready to go.


Note to Mac fans: The Onyx 400F communicates with the Mac using OS-level FireWire drives, so you have run System 10.3.9 or higher. Mackie also notes that the Onyx 400F is Tiger-compatible, but as my Mac hasn’t gotten past 10.2.8, I can’t vouch for that one way or the other. Guess I really do need to update…


I decided to test the Onyx 400F with my dual Athlon Windows XP Pro machine. It meets the system requirements (XP with SP1, 256MB RAM, Athlon XP/Pentium 4/Celeron), and I felt this might be a somewhat tougher test because FireWire isn’t native to the machine, but comes courtesy of a cheapo 3-port card with the TI chip set. I also figured it would be a good test to use it with my Digital Audio Wave laptop, which has a FireWire port…but I gotta get the 4-pin adapter first.


And now, the “smoke test”: Time to plug it in and see what happens.

Link to post
Share on other sites
  • Members

Even though FireWire devices are supposed to be hot-swappable, a recent article by PC wizard Martin Walker in Sound on Sound magazine questioned whether that’s always true. Rather than fry anything, I figured I’d play it safe and power everything down before plugging in the Onyx. If someone from Mackie would like to comment on whether this is unnecessarily paranoid or not, I’d appreciate it.


Like so many other hardware+software driver combos that run on Windows, you install some of the software first, so that when you do plug in the hardware, Windows will recognize it rather than assume it’s some generic plug-and-play device and want to install its own drivers. That held true here.


When I turned on the Onyx 400F, though, there was about three seconds of fear – nothing happened! Then the little LEDs started dancing, the “FireWire present” LED turned on – success. Next, the New Hardware wizard appeared on the computer screen. There was a slight discrepancy compared to the manual; it said the wizard would ask if I wanted to check the web for updates, but my computer started right in on the second wizard screen so I just picked up from there. Installation was totally non-eventful, which is always a nice thing.


So was it really there? I checked System Properties, and there it was, in Device Manager under “Sound, Video, and Game Controllers.” (Click on the attachment to see where the device gets installed under Windows.) As shown in the attachment, there was also the always-reassuring “This device is working properly” message. Things were off to a promising start.


Next up: Time to run some programs with it, and get familiar with the settings and console software…and see what kind of latency I could get away with.

Link to post
Share on other sites
  • Members

The console software is a collaboration between Echo and Mackie, with Echo contributing the FireWire technology. It has five mixer pages, which control the ins and outs, as well as provide a zero-latency mix if you want to bypass your host software. We’ll get into the mixer later, as it’s an important part of the package; but first, let’s set up the Onyx to match the computer system, and run some audio to make sure it actually works.


The system characteristics are edited on the Settings page. Click on the attachment to see this page, which includes the controls described in the rest of this post. The big knob at the top (ah, the recurring Mackie “big knob” theme) chooses among the sample rates: 44.1, 48, 88.2, 96, 176.41, and 192kHz. Do I care whether it has 192kHz? Well, not really, even though that’s a big deal for this particular interface. What I care about much more is that it can do 88.2 as well as 96kHz, as that seems to sample rate convert more elegantly to a 44.1kHz CD.


This is also where you select the Clock source. It can be internal, or come from S/PDIF or the hardware word clock connection on the rear panel.


The DSP Mixer switch enables or disables the mixer (when disabled, the mixer pages are grayed out). If you’re monitoring through the host, you might as well leave this off. But if you need zero-latency monitoring, it had better be on.


Another switch determines which outputs (1/2 or 7/8) get “mirrored” to the headphone outs. When switched to 7/8, then you can have separate headphone and control room mixes; with 1/2, both get the same signal. We’ll cover the available outputs shortly. For now, as we just want to make sure the thing works, so we’ll mirror outs 1/2 and use headphones. I always find it best when testing gear to go from the most basic to the most complex situations.


There’s also a S/PDIF format switch that chooses between Professional and Consumer. According to the manual, this relates to the subcodes used in the signal, not the data itself. I’ve had other devices with this option, and set it to Consumer unless things didn’t work, in which case I set it to Professional. One of them would always work. When using a passive AES/EBU to SMPTE adapter, Professional seems to be the way to go. Whatever; just flip it until things work – besides, you have a 50% chance of getting it right the first time!


The final editable option is a buffer samples slider, which goes from 32 samples all the way up to 2,048 samples.


An info box toward the bottom shows the software, firmware, and driver versions. I used the latest ones available as of this writing (1.05 software, 1.04 firmware, 1.05 drivers). In preparation for testing with audio, I set the sample rate to 44.1kHz, with a 256 buffers – a setting that works well with the Creamware interface and doesn't tax the computer too much, so I figured it would work here as well.

Link to post
Share on other sites
  • Members

I loaded Propellerhead’s Reason, and set the audio page preferences to ASIO Onyx F Series. The 256 samples translated to about 5ms of latency. Good.


I pressed Play, and yup, it played back. It’s then that I noticed just how good the headphone amps are – quiet, clean, and very well-behaved. This is definitely not one of those boxes that stuffs in a cheap amp chip for phones for quickie monitoring; these are nice. I don’t know if they’re Class A or not (I doubt it, because the unit doesn’t get very warm!) but they’re a cut above.


Also note there are two headphone amp outs, each with an individual volume control. The +48V phantom button is located next to the headphone 2 jack; I don’t know if this is intentional or not, but if you plug into this jack, it provides a natural barrier that helps inhibit hitting the +48V switch accidentally. Click on the attachment to see a closeup of the headphone section, and you’ll see what I mean about the +48V switch.


I thought I’d push the ASIO latency a bit further, so I tried 128 samples. Of course in the grand scheme of things, this really tests the computer’s speed and the complexity of the host software project more than the interface, but really well-written drivers can sometimes make a computer behave better than expected. The Onxy 400F and Reason passed the “ASIO 128 samples test” without problems.


So hey, how about 64 samples…nope, the audio started to break up, as did 96 samples in complex passages. I then tried Wavelab, a program that doesn’t place much demand on a CPU when just playing back files, and it worked fine with 64 samples.


Next, I loaded a really complex project in Sonar: Lots of MIDI and digital audio, Reason 3.0 rewired into it with a big Reason Drum Refill loaded, and East-West Colossus loaded as a virtual instrument. 256 samples didn’t cut it, as there were lots of dropouts. Just in case using Sonar 5’s 64-bit engine was screwing things up, I disabled it; same thing. So, I bumped the number of samples up to 512, and everything worked perfectly.


Just out of curiosity, I checked out using the Creamware interface and while 256 samples didn’t break up as much, it was still unuseable, whereas 512 samples worked fine. Bottom line: With ASIO, the Onyx worked about as well as my computer would let it, and within a hair of performing as well as an internal PCI card (which should theoretically give a slight performance advantage compared to going through FireWire, all other things being equal). Additional bottom line: I’m really glad I’m getting a new computer soon to replace my vintage 2002 model!

Link to post
Share on other sites
  • Members

It's getting toward the end of the day, time to call it quits for now. I've started doing some testing with Sonar and the WDM drivers, with mixed results to say the least...I can't get good results without really high latency settings. However, I'm not a huge fan of WDM - I've been using ASIO with Sonar once that option became available - so I could be doing something wrong within Sonar itself. I'll look into this further and report back tomorrow.


Please remember, this is an "open source" review. Got any questions? Comments? Complaints? Whatever? Chime in!


Note to Mackie: And you have some homework, too...let me know your opinions about the viability of hot-plugging FireWire connections, and also, whether you've done any special mojo to the headphone amps.

Link to post
Share on other sites
  • Members

Hi there everyone, Dan Steinberg here, product manager for Mackie recording products.


I've been a longtime lurker on Craig's old forum on the musicplayer boards, so I feel like I know a bunch of you already! I'm honored to help take part in this review and answer any questions anyone may have. So here we go.......


"let me know your opinions about the viability of hot-plugging FireWire connections"


I have definitely done it all the time with my Mac powerbook G4, including times when it’s been the only thing on the chain as well as times when it’s been on the chain with an external hard drive. So, for my setup at least, hot plugging has been just fine. HOWEVER, that only speaks for that model of powerbook. I have heard tales of some PC laptop or desktops where the manufacturer did not spec the Firewire chip to handle hot plugging that well, so I suspect how "safe" it is depends on the computer. It’s always supposed to be OK, and it is for our unit and my powerbook I tried device, but that’s really as far as we can “guarantee”.


"and also, whether you've done any special mojo to the headphone amps."


Glad you like them. I'd say that the "mojo" is that we’ve had a longstanding expertise in making clean and loud headphone amps in our mixers for a long time, and were able to call on that expertise, same as with the pres. That’s one of the benefits of having an audio interface come from a company with a long pedigree in analog circuitry, we sometimes have more tricks in our bag to call on than some other manufacturers. Our competitors in this product category all make some great stuff, but some of them started out making software, MIDI interfaces, etc, so although they make some awesome prodeucts, they sometimes are still learning about how to do great sounding analog.


Craig, to address a couple of other points I saw you raise:


"A loooooong 6-pin FireWire cable"


This was no accident. Since studio setups come in all shapes and sizes, it’s entirely likely that your 400F might be mounted somewhere that’s way more than 6 or 10 feet from where your laptop or desktop computer sits. I didn’t want to have someone start setting everything up and then find the cord didn’t reach, that would put a damper on one’s day for sure!


Including a 6 to 4 pin Firewire adapter


Funny you should bring this up, we actually did just finish sourcing a 4 to 6 pin adapter to include in the 400F’s forthcoming big brother, the 1200F, and since we were able to get a good price on them, we will probably be able to afford to start including them in newly built 400Fs as well. So in the long term life of the product, the 400Fs will probably start including those like you had wished. For anyone who needs one in the meantime, I have found that online sources tend to have much better prices than brick and mortar retailers, on 4-6 pin cables and adapters.


For example, www.monoprice.com has a great rep among home theater nuts trying to find good prices on DVI cables, and they can get you a 15 foot long 4 to 6 pin cable for just a few dollars, the same cable would probably be $30 at radio shack or Best Buy. Here’s the link




"next was a trip to Mackie web site to check for updated drivers"


Thanks for reminding users to always check the wesbite for updated materials, we'll rev the CD-ROM at some point to bring it up to date. Users will also want to do this for the bundled Tracktion 2, as we came out with a newer version of T2 since the disc went to press. When you register the version that’s on the CD-ROM, you'll then have an online my.mackie.com account where the newer version will be there for the downloading.


Buffer Sizes


It's true, the 400F can get nice and low and with a powerful enough machine, you should be able to run at 64 samples for very low latency. Craig, I noticed that you mentioned your desktop PC had 256 megs of RAM, I'd be interested to see if bumping it up to 512 megs might allow you to achieve some lower latencies than you got. I have always found that Windows XP (like Mac OSX) seems to run best with at least 512 megs of RAM, and this would hold even truer for audio where lots of RAM "breathing room" is always a good thing.


As Craig said, feel free to post any questions or comments you have, we're really excited about this product and are very excited to have it undergoing a Pro Review!

Link to post
Share on other sites
  • Members

Great comments Dan, I like the extra info like where to get a good price on FireWire cables and adapters!!


A couple comments on your comments...


Just in case it could be misunderstood, my mention of the long cable was not a complaint. In fact this is the first time that an included FireWire cable was able to reach from the back of my computer to a convenient place to put the interface. So maybe it added a couple more bucks to the cost of the package, but I didn't have to run out to Best Buy and pay some ripoff cable cost just to get the pro review off the ground.


Also, the 256MB reference was to the system requirements, not my computer. It has 1GB of RAM, like all good music computer desktop machines should :)


It's actually still a pretty capable machine, it's a dual processor AMD with Tyan motherboard spec'ed by the legendary Pete Leoni. But the 2002 processors are a little slow by today's standards...I'm going to be going dual core before the end of the year, that should be a nice little boost!


Once again thanks for your participation. It's great to be able to get answers to questions "from the source."

Link to post
Share on other sites
  • Members

Craig - great job as usual!


Craig or Dan (this is probably more of a Dan question - Dan: I appreciate any response, but "company speak" tends to wash over me, so I would appreciate as un-company-speak a response as possible :) ):


How would you compare the 400F with the 800 and new 1200F?


I was very impressed with the 1200F's intro at AES.


Are the differences between these different models merely in the realm of features, OR does the 1200F benefit from further R&D and include better-sounding components? Will the 1200F actually sound better than the 400, or will they sound the same?




Peace, Love, and Brittanylips

Link to post
Share on other sites
  • Members

- "company speak" tends to wash over me, so I would appreciate as un-company-speak a response as possible ):


The Onyx 400F is such a good interface that it will wash your car, do your taxes, and improve your love life.....wait a minute, you said you DIDN'T want hype, oops


- How would you compare the 400F with the 800R and new 1200F?


I'll try to make some brief comparisons here. The three pieces are really quite different from each other, and are each suited for a different type of customer. When you know what you want to do with your gear, which of those three pieces to get will probably be pretty clear.


800R: Pretty "apples and organges" to the 400F and 1200F because it's not actually a computer audio interface, it has no USB or Firewire connectors on it. This unit is an 8 channel A/D converter, with 8 Onyx mic Pres and 8 line level inputs, and analog line level and 8 channel AES, SPDIF and ADAT digital out. Great as a line or mic expander for an existing audio interface (or hard disc recorder, or 2" reel to reel, or portastudio) but it won't get audio into a computer on it's own.


1200F: This is the "big brother" to the 400F and gives you MORE, of well, everything, and will separate you from more of your money for the privilege :)


30 in, 30 out channel to/from computer, as opposed to the 400F's 10X10 channel count

12 mic pres instead of 4

12 analog inputs instead of 8

built in control room fucntions such as talkback and monitor A/B switching

4 channel headphone amp with individual audio streams to each jack

2X2 MIDI instead of 400F's 1X1

16 channels of ADAT I/O (hmm, an Onyx 800R would be perfect for those)


So as you can see, whereas the 400F is great for a guy recording up to 10 tracks at a time, the 1200F takes it further, since you can actually track 30 tracks at a time and also get fancier with headphone mixes at the same time. So, which unit to get really depends on what kinds of work you think you'll be doing.


But really, what you want to do is buy both. The kids can go to community college instead of one of them fancy out of state schools, tell them Dan said it was ok.


Are the differences between these different models merely in the realm of features, OR does the 1200F benefit from further R&D and include better-sounding components? -


I would say mainly features. The two units actually share a very similar development platform. Same Firewire chip, same co-developers (the fine folks at echo) and when the 1200F comes out, they will even share the same control panel that auto-senses which unit you have connected and configures accordingly.


- Will the 1200F actually sound better than the 400, or will they sound the same? -


I checked with ther appropriate propellorheads and they had this to say:


The 400f and 1200f use the same converters, same input and output circuitry so they should sound the same. However, due to the loads of circuitry inside

of the 1200f, the noise floor is about 1-2 dB higher than the 400f, so the 400f is a bit quieter (measurable, not audibly). From a specs standpoint,

everything should be within a dB or two of the 400f, in some cases better, in some cases not. So in summary:


Same "sound"

Same performance


Also, the 400F and the 1200F use the same Texas Instruments DSP chip, so both units' standalone DSP mixers have the same ultra-neutral sound when summing signals together. The 400F web page has some good info on how the DSP chip is pretty cutting edge and actually gives the 400F a better sounding digital mixer than a lot of "actual" digital mixers out there.


Thanks very much Mr. Brittanylips!

Link to post
Share on other sites
  • Members

Okay, I had a chance to test the WDM drivers with Sonar and Project5, and the results are definitely not impressive. The latency with Sonar has to be set very high to get satisfactory results, as there was significant breaking up with latencies under 150 ms - even with a really simple project that just played back some loops (and dinged the CPU meter a mere 5% or so).


For a bigger project, I had to bump the latency up to 200 ms. Matters were much better with Project5; I got satisfactory results with as little as 40-50 ms of latency, but to put that in perspective, that's still not acceptable for a comfortable real-time work with a host program.


I was wondering if maybe the newer versions of those programs (Sonar5 and Project5 V2) were the issue, and that perhaps there was a change in the way they handled WDM. So I tried both with my Creamware card's WDM drivers and got 5 ms latency easily with decent-sized projects, and 10 ms with behemoth projects.


I tried a zillion different things with Sonar and the Onyx control panel, changing bit justification, disabling dual processor operation, everything I could think of...no go. I think that either the WDM drivers are not mature, or need optimization. I believe they are true kernel streaming types, not "wrappers" for plain-vanilla MME drivers, so I don’t know why this is an issue.


Interestingly, in Martin Walker's Sound on Sound review of the Echo FireWire interface, he noted the same level of excellent ASIO performance that I found, and the same questionable WDM performance. I have an inquiry in to Mackie about this, and expect that Dan will have an answer soon.

Link to post
Share on other sites
  • Members

And while we’re waiting for a definitive answer, it made me wonder just how important WDM support is. Granted, Adobe Audition requires WDM, but it seems just about every pro application for Windows is switching over to ASIO (there are rumors that Audition will as well). Even Sony/Sonic Foundry, a stalwart supporter of Windows and DirectX, has made the move to ASIO and VST. And of course, Mac people don’t have to worry about this at all.


I do know that Sonar’s WDM support goes deep into the OS to get the high level of performance. Since Sonar started supporting ASIO, though, I virtually never use WDM with Sonar. Actually, make that “never.” There are so many ASIO-based programs in my system that seems to be the best way to go. Programmer friends seem split over whether ASIO or WDM is better, as they feel there are “band-aids” present in both of them.


Let’s travel into the Speculation Zone as well. Why would Cakewalk, which was really on the WDM bandwagon, support ASIO? The obvious reason is that they wanted to open up the application and be more inclusive, much like how they supported VST along with DirectX – now they’re even producing an instrument (Dimension Pro) that supports Apple’s AU spec.


But maybe they’re also cautious about cozying up too closely to WDM for other reasons. I’ve heard totally unverified rumors that Windows Vista will introduce yet another sound protocol that’s not related to DirectX or WDM. I would assume Cakewalk has a pretty good pipeline into what’s happening at Microsoft, and as a developer, would be aware of any upcoming changes before the press or general public would. Maybe ASIO is a convenient safety valve so they don’t get caught in the trap that happened when Sonar was first introduced and supported only WDM, but WDM hardware support was scarce.


If indeed the industry is moving to ASIO for reasons related to WDM being a lame-duck protocol, then that would explain why the Onyx 400F has a solid ASIO implementation and lame WDM. The WDM works okay for Windows Media Player, DVD Playback operation, and similar stereo/consumer-oriented apps, it’s just not up to pro specs or operation for use with a multi-track host like Sonar.


Let’s see what shakes out…tomorrow we’ll look at other aspects of the Onyx 400F.

Link to post
Share on other sites
  • Members



Thanks for the helpful, direct, and dare I say, Craig-like response!


I’m already set with an RME FireFace 800 at home (competitive with the 1200F), so for me, the 400F would be something I would consider - seriously consider - for a smaller interface to take around with my laptop if I didn’t want to lug around the FireFace. (I am also considering an Apogee mini-me).


In any case, it is nice to know that the difference between the 1200F and 400F is in features rather than quality, so it is not as though Mackie is slumming on the sound of the 400F.


But really, what you want to do is buy both. The kids can go to community college instead of one of them fancy out of state schools, tell them Dan said it was ok.


You mean Mackie doesn’t offer a “need-blind” admission policy (like some of them fancy out of state schools) with a sliding scale for customers who want to buy 2 interfaces for the price of 1? You should do that - some guy named Greg told me to tell you it was OK.


Thanks very much Mr. Brittanylips!


Thank you! But “Mr.” – are you sure?


Also, and most importantly, what’s a guy with a name like “Steinberg” doing working for a company called “Mackie”?


-Peace, Love, and Brittanylips

Link to post
Share on other sites
  • Members

So, here's some info to speak to Craig's experience with Sonar and the WDM drivers:


-We did some more testing with Sonar 5 and found the same issues with latency items, using the WDM driver, so it's not a local issue with Craig. Sonar, compared to other WDM programs like Adobe Audition, works with the WDM drivers on a deep, "windows bypassing level", that our WDM drivers are not currently optimized to deal with, thus the issues Craig mentioned.


Since Sonar DOES however, work with the 400F's ASIO driver very well, we chose not to devote a lot of engineering resources to working through this Sonar/WDM issue. Instead, we officially recommend that any 400F/Sonar users simply use the ASIO driver, not the WDM driver.


Having said that, we want to be very responsive to customers' needs, so if anyone here knows of a reason that we do not, about why that is not a practical method of working, please let us know. If there is an important reason why you must use Sonar with WDM drivers and not ASIO, post about it here.


-The 400F WDM driver works fine for system level stuff like system sounds, itunes, windows media player, web browsers, video games, etc. That was actually our main intention for the WDM drivers, apps that had to use it, as opposed to apps like Sonar where ASIO is a valid option.


I am hoping that speaks to the issues Craig raised. If not, please let me know and I'll be ahppy to weigh in further, I'll be monitoring the thread over the weekend.



Link to post
Share on other sites
  • Members

Thanks much, Dan. I'll be interested to see if anyone pipes up in support of WDM with Sonar. Some people claim the program is more solid with WDM, but I suspect that's more a function of the drivers that were written for it. Certainly when using the Creamware drivers, there is no discernible performance difference between ASIO and WDM.


I'll be adding more posts over the weekend, so Dan, keep your laptop on "standby."

Link to post
Share on other sites
  • Members

Originally posted by Anderton

The DSP Mixer switch enables or disables the mixer (when disabled, the mixer pages are grayed out). If you’re monitoring through the host, you might as well leave this off.
But if you need zero-latency monitoring, it had better be on.



Near zero latency.


One man's silly little millisecond or two is another man's aeon. :D


Personal bugaboo, here. Sorry for nitpicking.



Great review so far. I was really hoping to find a thorough write-up on this unit somewhere and I can't think of a better writer or format for the job!





I would add one thing, since Craig noted that his laptop -- like my own -- required a 4 pin FW connector. Like the Mackie, the MOTU 828mkII I bought last year came with a long, beautiful 6-to-6 that's been of no good to me.


If I'd known that I could have shopped the internet and eased the pain of paying something like $25 for a too-short connector from my local micro-chain computer store (same cable/brand was a nasty $40 at CompUSA!)


Get a long enough cable -- the added flexibility will seem heavensent -- and help reduce strain on connectors (particularly of concern to us laptop users whose connector ports are often mounted directly on the mobo).

Link to post
Share on other sites
  • Members

Most people don’t know that in my print reviews, I usually take things apart but there’s never enough space to go into details…I only write something if the construction is noteworthy, either for being good or shoddy. But hey, I have all the space in the world and I can’t resist taking things apart…


So 18 Phillips head screws later, I had a naked Onyx 400F in front of me. First thing I noticed is that the FireWire interface is a daughterboard that plugs into the main board via two multipin headers (not those !#$&$@ ribbon cables that are always so hard to get back in), secured by three screws with lockwashers that screw into standoffs (nice…thank you). There’s no way the daughterboard will vibrate out of place. This is the Echo Digital Audio part of the show: Their FireWorks board includes three main chips, among these a huge honkin’ Texas Instruments DSP chip. If FireWire gear always has to have this special mojo circuitry in it, I bet that definitely adds a few bucks on to the price.


The power supply is another drop-in module. Whenever I take a unit apart I’m always a little paranoid about brushing against the incoming AC line, but no worries with this one: All the AC connections are covered with a thick insulating plastic sleeve, and the two AC lines terminate immediately in a connector that again manages to keep live wires away from the outside world.


It’s also clear that a lot of effort went into making sure any power supply switching noise didn’t get into the audio lines. The power supply is located at the opposite end from the mic pres – if it was any further away, it would need to be outside the case. There are plenty of caps and inductors, and decent heat sinks; and while I didn’t remove the power supply module, it looks like there’s a ground plane in the circuit board area beneath it.


The power supply outputs go through heavy gauge wires, and like the AC line ins, terminate in a connector. Bottom line: If I had to substitute a new power supply, I bet I could open the case, replace the power supply, and close it back up in under 10 minutes…probably less, actually.


The rest of the board has the expected surface-mounted chips along with capacitors, resistors, a couple local voltage regulators, and the like. All jacks except the FireWire and MIDI connectors are secured to the rear or front panels via screws or nuts, so you can plug and unplug audio connections without worrying about stressing out the circuit board to which they connect.


However, the controls are not secured, and do the usual “terminals-mount-on-circuit-board, shaft-sticks-through-hole-in-panel” thing. I was concerned about this, because the classy brushed aluminum knobs (with setscrews, no less!) extend far out from the panel. But the pot bushing goes right up to the front panel, so there’s very little “play” when you push down hard on the knob. It would take a lot of downward pressure to snap the shafts off; I pushed down to what I judged to be what the most ham-fisted user could possibly do, and they survived just fine.


Similarly, the front panel LEDs (there are four, four-LED meters) are secure – even if you push hard on the LEDs, you’re not going to pop them out the back of the panel because they’re retained by a plastic shell that anchors into the circuit board.


The jacks appear to be rugged and well-constructed, which is a good thing – I’d hate to have to replace one of those suckers, although it is doable. One of the clever design features is that the front panel can be removed fairly easily, so if you remove all the jack nuts and XLR+1/4” combo jack screws, you can pretty much pull out the board if you need to work on it.


And speaking of jacks, one of the truly great features is that the ground sleeves on the 1/4” jacks are plastic, and physically insulated from the front and rear metal panels. This means the ground connections are made internally to the unit, at the jacks themselves; therefore there’s a great degree of control over how grounding is handled, rather than just grounding to the panel. Proper grounding can have a huge effect on reducing noise and hash, and this is definitely an example of proper grounding technique.


Finally, it’s worth mentioning that the entire casing is made of metal, which helps give a feeling of security. Overall, I would rate the construction as high level stuff, and there are indeed some serious analog chops on display here.

Link to post
Share on other sites
  • Members

Originally posted by Anderton

Thanks much, Dan. I'll be interested to see if anyone pipes up in support of WDM with Sonar. Some people claim the program is more solid with WDM, but I suspect that's more a function of the drivers that were written for it. Certainly when using the Creamware drivers, there is no discernible performance difference between ASIO and WDM.

I'll be adding more posts over the weekend, so Dan, keep your laptop on "standby."


When I got my MOTU 828mkII I was pretty trepidatious about MOTU's level and quality of support for WDM drivers, since I had the perception (and nothing has really shaken that) that MOTU has always been a more Mac-centric outfit.


Oddly, the WDM drivers have apparently mostly worked just fine, allowing me to run the i/o buffers at 128 samples (min of 96 presented glitching) and Sonar's 'mix-latency' setting at 2.9 ms. (It should be noted that I do not do huge projects... seldom going over 20 audio tracks... although I have let myself slip into using more and more plugs and v-synths.)


Odder still, perhaps, I have never been able to get the ASIO drivers to even show up in Windows -- although twice, now, the MOTU driver installation prg has said it installed them. (IOW, I did a reinstall to see if I couldn't get them to show up. Nope.)



At any rate, the 828's seemingly been fine with the WDMs in Sonar 2.2, 4, and 5. (Although I still have not given S5 a real workout; I've been busy with small acoustic recordings as I build a catalog of informal performances of all my songs... preparing my 'legacy' as it were. Other people build science buildings at their alma mater -- I'm sticking all my stuff on Archive.org. It's my era of diminished expectations. Besides, my alma mater already has a science building.)

Link to post
Share on other sites
  • Members



Just as an aside...when EQ asked if I wanted to review the 2408, I said I was pretty much using Windows, so they should probably get someone who lived and breathed Mac to use it. MOTU stepped in and said they were hoping I would review it with Windows because they wanted to change the perception they were a Mac-only company. I figured what they hey, and was pleasantly surprised that it worked very well under Windows, as well as Mac.


However, I'm surprised ASIO doesn't show up...given a choice between having ASIO not show up or WDM not show up, I'd be happier if WDM was the no-show.

Link to post
Share on other sites
  • Members

First of all, let me say thank you. There have been very few (perhaps zero?) in depth reviews of this unit, and I couldn't be happier you've decided to undertake this.


At a shootout on another forum, one user reported fantastic AD conversion but very thin, brittle, DA conversion. I've been looking very seriously at the Mackie, but bad DA would be a deal breaker for me. What good are great preamps and AD if you can't trust the sound coming from your monitors and headphones? All I've heard so far is one man's opinion, so I'd be very happy to hear another.


I'd also like to hear about latency within tracktion 2 and, if possible, Cubase.


Thanks so much,



Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...