External Summing: Thinking Outside the Box

Gear Autopsy Summing Article

When I think of all the music that I have loved through the years, the songs that have influenced and shaped me as a musician, songwriter, and engineer, there is a clear commonality that I am sure most of you can relate to. Really great and timeless songs and albums don’t sound like musicians and instruments mic’ed up in a studio or on stage. They sound like cohesive pieces of music, singular in nature, almost as if they were discovered rather than created. Enough can be said of the songwriters and performers who have created such masterpieces, but those of us in the audio field know it took great engineers, and thrifty engineering skills to make records that sound so refined. Let’s face it, the gear really had a lot to do with it too, and the methodology behind getting great results from it.

Of course most of the songs I grew up on were recorded through analog consoles, to tape, and mixed back down through those very consoles, hitting tape once again for the final product sent to master. With noise and distortion being cumulative by-products of each stage in this process, great care had to be taken to maintain fidelity and minimize the analog artifacts throughout production. Google “Fleetwood Mac Rumors HF loss” to see how these effects almost ruined one of the greatest records of its era. It’s no surprise that the advent of a new, clean, super-quiet recording medium would turn the heads of the professional audio world and become an instant replacement for tape overnight. Digital recording was crystal clear, almost immeasurably quiet, and what you recorded at the session was the same exact sound you worked with at mix time. Like all things new, it wasn’t without its growing pains, but it was clear this was going to be the new way forward for professional recording. It didn’t take long before computers and software took over completely, and the DAW was born. Digital replaced not only expensive and temperamental tape machines, but the large, expensive, high-maintenance analog mixing desks studios employed as well.

A Real Neve Console - A Luxury Most will Never Know

A Real Neve Console – A Luxury Most will Never Know

 Lost in Translation

Here’s where things got a little sour. In creating a “more perfect” system without the audible noise and distortion of its analog counterpart, a new problem was created. Things were “too clear”, almost harsh, and seemingly more difficult to mix. Sources had a separation that made them much more difficult to put together into those single, cohesive pieces of art we all love. It became apparent that some of the characteristics digital systems did away with actually helped the music come together. So began the quest to bring back the “good parts” of the analog process. For studios that could afford it, the answer was simple: replace the tape machine, but keep the console. Get a great sound going in, record it digitally, and mix on a real desk, like it has been done for years. This is what most of the top studios and engineers are still doing to this day. Most music lovers would agree the results are better than they ever were before (well, they can be). Of course, quality consoles are light-years beyond the budgets of most small studios, and while big multi-channel boards would be great to have come mix-time, most guys really don’t need that many channels at acquisition anyway. A gang of nice quality preamps or channel strips can solve the problem going in, and most DAW mix engines put the capabilities of consoles costing tens of thousands in the hands of all. Unfortunately, many experienced engineers will tell you that something is still missing, and without the console, you don’t get the depth and feel you need to make truly great and memorable productions. There are a few big name engineers proving that great work can be done entirely within the box, but they are widely out-numbered. Still, with the recording business being a tough one, and good consoles costing a fortune, they remain a luxury only high dollar studios can afford.

Get a great sound going in, record it digitally, and mix on a real desk, like it has been done for years. This is what most of the top studios and engineers are still doing to this day.

Greater than the Sum of Its Parts

This is where external summing mixers come into the picture. These are small analog boxes that take multiple streams from your interface’s analog outputs and combine them the old fashioned way, in the analog domain. Since most interfaces feature 8 or 16 individual mono outputs, summing mixers put them to good use by taking these channels from the DAW, and combining (summing) them like a real console does. The resulting stereo mix is sent back to the AD converter where it is “printed”, replacing the tedious process of bouncing to disk. Summing boxes don’t feature channel processing, pan knobs, or faders like consoles do, since these tasks are done within the DAW; they simply do as the name implies and sum the audio, eliminating a major bottleneck of mixing internally. This maintains the advantages of the digital environment such as automation, recall ability, etc., and (in theory) brings back the analog counterpart that’s missing from in-the-box mix-downs. It’s a simple enough concept, and one that’s easily adapted to any workflow. Unfortunately, the benefits of the process are hotly debated. Some guys swear it absolutely works, others will tell you that the latest analog emulation plugins will get you there for less, and still some will swear it’s complete snake-oil and offers no practical advantage whatsoever. Since sound is so subjective there really is no one correct conclusion, but it does seem that most of the engineers speaking out against analog summing are those that have yet to try it. In my experimentation, analog summing has brought a number of advantages that have both changed the way I work and led to improved results. In fact, most of the engineers I know that have tried external summing have come to similar conclusions.

The most obvious reason someone would be inclined to try a summing mixer is to gain a sonic advantage that, according to its proponents, includes a wider and more detailed soundstage, greater depth of fine musical nuance, and a 3D quality that we’re all constantly chasing in-the-box. This really does simplify the merits of working with a hybrid setup, however. Most DAW interfaces available today have very high quality conversion, and chipmakers and circuit designers go to great lengths to make sure this conversion is clean, quiet, and musical. The quality of your DA conversion is going to directly influence the benefit you get from an analog summing workflow. Better converters will obviously yield a greater advantage. Also, since every converter out there will to some extent have a “sound” of its own, it’s important to note that the summing box MUST be incorporated before you begin the mix process. The sound of your DA will influence the decisions you make along the way, and will be there in your final product. In my opinion, this is already an advantage, since in-the-box mix-downs never retain the character of the DA used to create the mix and will always sound different for this reason.

…a more detailed soundstage, greater depth of fine musical nuance, and a 3D quality that we’re all constantly chasing in-the-box.

The Issue of Headroom

There is a little bit of a myth that surrounds digital mixing environment regarding their ability to mimic the analog mixing process. DAW makers will tell you that since their algorithms are taking digital streams and combining them with simple and perfect math, that the resulting audio represents a perfect product that even the best analog desks can’t replicate. Mathematically, they have a point, but when it comes to real ears, and real mix-engineers, it’s impossible to draw the same conclusion. A fundamental difference between digital and analog systems is how they handle levels as they approach (and reach) full-scale output. In an analog system, as levels start to creep up, the circuitry starts to naturally compress and distort – gently at first, and then more aggressively. Those of you who have worked in a real analog environment know this phenomenon well, because it’s often regarded as a tool rather than a limitation of the system. Digital systems are not as graceful. As levels reach full-scale, digital responds with obvious clipping artifacts that are anything but pleasant. Many DAWs use something called floating-point math to compensate by continuously shifting decimals as the digital signal approaches clipping, which emulates usable headroom. This can prevent the nastiest artifacts of digital clipping, but doesn’t really add usable headroom to the mixer. Another technique is to incorporate analog emulations that mimic the way real consoles compress and distort. These can certainly be very cool, but they don’t give you usable headroom. They merely provide compression and distortion as an effect.

So what’s the big deal? Just record clean tracks, keep the levels from reaching full-scale, use compression and limiting to keep things under control and everything will be okay, right? Unfortunately, this isn’t the case in practice. Any mix engineer knows that as you start combining tracks, the level on your master buss starts to creep up. The cumulative levels of your individual tracks start to overwhelm the master buss. In order to bring things back into comfortable territory, your only choices are to bring the faders back down, or compensate for the individual levels through the use of additional processing. This can change the mix entirely, since there is often level dependent processing downstream from these tracks, on groups or even the master buss itself. Aside from that, it is simply a drag to have to do. This is a problem the transition to digital created, and it is one that many people don’t even realize they are working against.

A properly designed summing mixer allows for multiple buffered input channels for summing to a real analog master-buss. The active circuitry provides an optimal load to the analog outputs of the audio device and more headroom than any DA converter in production could ever approach the limits of. The summed output replaces the DAW’s master buss, and removes the key bottleneck from the digital mixing environment. Level creep is no longer a problem because the new analog master buss is immune to the negative effects of level accumulation, and the resulting output is free of the claustrophobic feel that many find in-the-box mixes to be prone to. Of course it’s a phenomenon better understood once you experience it first hand, but a common reaction is that it’s hard to go back once you try it.

Also, keep in mind that there are two types of summing devices on the market: passive and active. While both types can alleviate the issue of limited headroom within the digital mixing environment, a passive approach brings some disadvantages to the table that active devices are immune to. Passive devices are usually less expensive, but are prone to grounding and loading issues, poor crosstalk rejection, increased noise, and other artifacts. Since they feed multiple channels into a passive resistor network, instead of using active receiver amplifiers for each channel, gain is significantly reduced and high-gain microphone preamps are needed to restore the summed audio to useable line level. This limitation of passive summing systems leaves the final mix colored, and with artifacts originating from this extra gain stage. In my opinion, the summing process should remain as clean and transparent as possible to preserve the integrity of the tracks. Color can be added at many stages throughout the mix process and shouldn’t be an unavoidable product of the hybrid workflow. You should consider this before making your hardware selection.

Mathematically, they have a point, but when it comes to real ears, and real mix-engineers, it’s impossible to draw the same conclusion.

Out of the Box and into the D-Box

In my experimentation I have been using a Dangerous D-Box to combine 3 stereo busses (drums, guitars/backing-vocals, and effects), and 2 mono busses (bass and lead vocal) for a total of 8 summed channels. These get sent to the D-Box on respective input pairs 1/2, 3/4, 5/6 (which are internally hard-panned in L-R pairs), and channels 7 and 8 panned center (D-Box has panning controls on the last two inputs for this added flexibility). Just by doing this, and sending the summing output of the D-Box back into the Apollo, I do feel like the resulting mix-downs demonstrate a spaciousness and depth-of-detail advantage when compared to mixes I’ve done in-the-box. This might seem to defy logic somewhat, given that digital mix-downs should theoretically be more accurate, but I attribute it to the smoothness and character of my D/A conversion (which I consider to be very musical) combined with the benefit of not having to constantly compensate for level accumulation. Consequently, I find the effect to be consistent whether I’m using the BLA Apollo’s DA converters, or those of my Lynx Aurora. Both sound comparatively better than the product of the internal mix engine in Studio One (which has been confirmed to be identical to most other DAWs through null tests, ad nauseam). My internal mixes still sound good, but I am definitely hearing an advantage. The benefits seem to become increasingly apparent over time, but are instantly recognizable when trying to go back to mixing through the DAW. It just doesn’t feel or sound the same.

The summing section of the D-box is intentionally very clean and linear in nature (as I am told all of the Dangerous summing boxes are) so that the summing process is pure, and the benefits can be gleaned without additional analog artifacts. Certainly, it can be useful to add color through additional processing while still in the analog domain; this is another capacity in which the hybrid system’s advantages really start to become audible and apparent. By attenuating the summing output of the D-Box, and piping the stereo mix through a line level amp, you can mirror the phase and harmonic response of a real console buss. Modules like Classic API’s VP28 amplifiers are almost legendary in this capacity and can really take things up another notch. Follow that with your favorite buss compressor, equalizer, or other outboard processing gear and the sonic palette becomes almost limitless.

Of course, you can also bring analog outboard into the equation through the stems themselves, without the added complexity of routing out and back in to the DAW multiple times. I have always been a big fan of digital algorithms, and though the latest analog emulation plug-ins are excellent, true analog outboard gear still has a weight and smoothness that plug-ins have yet to replicate. A real VCA buss compressor on the drum group, outboard EQ on vocals, or just running stems through a discrete amp at unity gain, can add some beautiful weight and sheen that algorithms can’t manage. The headroom advantage afforded by the summing mixer makes doing this effortless, and I can say from experience, it will really change the way you think about the place of outboard in today’s small studio. You won’t lose the flexibility of processing within the digital environment, but placing actual hardware into key processing roles throughout your mix can genuinely reduce the amount of overall processing you might need altogether.

The benefits seem to become increasingly apparent over time, but are instantly recognizable when trying to go back to mixing through the DAW. It just doesn’t feel or sound the same.

So What Does All of This Really Add Up To?

Once you start to consider the cumulative effects of these advantages, and put them up against the simple internal mix-down, the benefits really do start to become glaring. Tracks start to fit together more easily, require less processing in the box, and contribute to a solid, more cohesive sounding mix-down. Tracks will fit together more seamlessly while managing to retain their focus and clarity within the mix, something I admittedly have always struggled to achieve. I am also guilty, as I am sure many of you are as well, of leaning too heavily on plug-ins when trying to get that “record” sound in-the-box. I find I am getting closer to the sound in my head, more quickly, and with less processing since moving to a hybrid workflow. In fact, I am finding mixing to be a lot more enjoyable now as well. Whether that’s directly because of the additional analog component provided by the external summing engine, or simply because of a change in the way I approach mixing, I can’t say. Most likely it’s a combination of both. Either way, my perspective is different thanks to external summing and it’s made things fun and fresh again, which is a welcome change.

So what if you don’t see the need for outboard gear, you’re not convinced of the musicality of your DA, or you’re already getting what you consider to be stellar results from within the DAW? Is analog summing really any better than what you’re doing now? Is a hybrid workflow quantifiably better than mixing in-the-box, on paper? That’s not a certainty. It’s more of a hypothesis whose proof is subjective, and therefore you have to do your own experimentation to determine its utility for you. I guarantee that for many, you’ll at least find the change in workflow refreshing which can be a huge benefit on its own. I recently read a blog post boldly claiming that external summing is a fruitless expense, citing renowned engineers like Charles Dye and Dave Pensado as examples of why in-the-box mixing is all you’ll ever need. In my opinion, this is as shortsighted a conclusion as the proposition that you’ll only ever need a single microphone or preamp to make a great recording because somebody else has done it. Great skill and creativity can do great things with modest tools, but that doesn’t diminish the advantage great tools bring. If the sentiment is that there is no magic fix, and no substitute for real experience and skill in the record business, I’d have to agree. Summing out of the box isn’t a magic fix, and you certainly won’t get out what you don’t put in. If you find you’re already getting the most of your method, and want to take your work to the next level, then I’d say external summing is worth consideration. If any of the points I detailed in this article resonate with you then I’d say you have a lot to look forward to when you are ready to give analog summing a try. Also, if you’re finding that inspiration isn’t as easily found as it once was, and you’re looking for a way to liven up the mixing process, I have found just that in the summing section of the D-Box. Whether it’s directly or indirectly responsible for the improvement in my own results, I’m not sure I even care. It’s an integral part of the equation and the clear way forward for me. It will be how I work from now on.


There is 1 comment

Add yours

Post a new comment