Home Theater Forum and Systems banner
Not open for further replies.
1 - 1 of 1 Posts

· Banned
5,013 Posts
Discussion Starter · #1 ·
miniDSP nanoAVR DL HDMI / Dirac Live Audio Processor Review


nanoAVR DL MSRP: $549

by Wayne Myers


The miniDSP nanoAVR DL is another digital signal processing offering from miniDSP, the Hong Kong-based audio company that specializes in bringing affordable yet high-quality DSP into the hands of the home theater and two-channel masses. Dirac Live is a room correction technology that seems to have caught the attention of room tuners and chasers of sonic perfection in recent years, a market long dominated by Audyssey, with their AVR based products seen by many as the shortest line between speaker setup and great audio. As far as I can tell, Dirac Live is in the process of showing the audio world that said line can be much shorter and reach a more satisfying conclusion, too.

I previously reviewed the miniDSP DDRC-22D, a two channel Dirac Live product, and while I was very impressed with its performance, I had reported that I was not certain the technology was up to the task of preserving the sonic details of the highest quality soundstage and imaging experience. Since then, I have delved quite a bit deeper into the science of soundstage construction and fine-tuning, as well as the ability to judge soundstage and imaging quality, and when I received the nanoAVR DL for review, I was determined to put it to the ultimate sonic tests and see what it was capable of under the most stringent listening conditions I could come up with.

The nanoAVR DL starred in a comparison with Audyssey XT32 back in March at Sonnie Parker's Cedar Creek Cinema High-End Amplifier Evaluation. At that time we chose to focus on an apples-to-apples under-the-hood comparison of core technologies and, given equivalent target curves, we determined that there were no appreciable differences.

For this review, I decided that any comparisons to Audyssey would focus more on an oranges-to-oranges comparison of the end-user experiences common with the two products. While that is not the main purpose of this review, there are certain contrasts between the products that stand out and are worthy of mention.

As was pointed out in the previous comparison, it is no small task determining how to make a true and fair comparison between products as complex as these, especially when referring to something as hard to define as an end-user experience. About all one can do is to relate frustration and satisfaction levels at different points in the experience, how long tasks take, the costs, and the ultimate results achieved, realizing that others may have a different range of experiences with the same products. I have done my best to state pertinent details to make those comparisons as meaningful as possible.


The nanoAVR DL is a small unit with two HDMI inputs and one HDMI output on the rear panel, along with power, Ethernet, and USB connectors. The front panel contains a single control button and 8 LED indicators. Most control functions can be programed into a generic remote using a supplied PC utility.

The sheet metal case is no show piece, but is built to last and do its job for a long time. It contains a DSP clocking data at 48 kHz with 32-bit internal processing depth. Only LPCM audio is handled by the nanoAVR DL, which can receive up to 24 / 192 kHz. All bit depths and rates are resampled to 24 / 48 for processing. No video processing is done within the nanoAVR DL, other than buffering to match the delay of the audio processing.

All the measurements and programming are accomplished by way of the Dirac Live PC-based program. A UMIK-1 calibrated USB microphone can optionally be included with the nanoAVR DL (a $70 adder). The miniDSP version of the control program only works with that model of microphone.

The program allows an impressive degree of flexibility. One can delete and retake individual mic position measurement sets at will. The user also has the ability to customize target curves, Any of the 8 channels can be grouped together to be handled by a single target curve, or split apart for an individual or sub-grouped target curve. Drag and drop GUI allows all this to be done with tremendous ease. I found the interface very intuitive and easy to use, extremely well thought out, and I ran into no bugs or obvious oversights of any kind. Once measurements are taken and target curves set, a single click kicks off the optimization processing which only took a few seconds on my i7-based laptop. Once the filter set is transferred via USB connection to the nanoAVR DL internal memory, you are ready to play music.

With the ability to control the nanoAVR DL functions from its front panel, from the Dirac Live calibration program, from an easily programmed remote, it is easy to get the unit doing what you want it to. I only wished for a readout of gain (Master Volume, or Output Volume), but that is probably just me obsessing over numbers the way I tend to. I left that setting alone once set for the filters in use.

  • Processor: 32-bit floating-point Analog Devices SHARC DSP
  • Input/output resolution: 24 bits
  • Channel count: 8 in, 8 out
  • Input sample rate: 32 - 216 kHz
  • Internal operating sample rate: 48 kHz
  • Output sample rate: 48 kHz
  • HDMI inputs: 2 x HDMI 1.4a selectable from software, front panel or IR remote
  • HDMI Outputs: 1 x HDMI 1.4a output
  • Audio sample rate / Resolution: Input/output resolution: 24-bit integer, 8-channel linear PCM (32-bit internal)
  • Video support: Video pass-through of selected source (i.e. no video processing); 36-bit / 3D support / HDCP embedded keys
  • Storage/Presets: 4 filter sets stored, selectable from front panel, IR remote, or smartphone app
  • Infrared remote: “Learning remote” capabilities (NEC, Philips, Sony, Apple); Controls master volume, mute, HDMI input selection, filter set selection, Dirac Live® filtering enable
  • Smartphone app:Controls master volume, mute, HDMI input selection, filter set selection, Dirac Live® filtering enable supported on Android
  • Power supply: 5 VDC single supply @ 600mA, 2.1 mm center-positive
  • Dimensions (H x W x D): 31 x 161 x 110 mm
Associated Review Equipment
Measurement Methods
  • Asus G74SX Laptop, Intel I7-2670QM @ 2.2 GHz, 16 GB DDR3 Memory, Windows 7 64-bit, Room EQ Wizard, foobar2000, Reaper DAW
  • Digital Audio Workstation, Phenom II x6 1100t @ 3.5 GHz, 16 GB DDR2 Memory, Windows 7 64-bit, Room EQ Wizard, foobar2000, Reaper DAW
  • Media Server, Phenom II x6 1055t @ 2.8 GHz, 8 GB DDR2 Memory, Windows 7 64-bit, Room EQ Wizard, foobar2000, Reaper DAW
  • Roland Quad-Capture Audio Interface
  • M-Audio Fast Track C600 Audio Interface
  • M-Audio Firewire 410 Audio Interface
  • miniDSP UMIK-1 Measurement Microphone, Courtesy miniDSP, Home Theater Shack Sponsor
  • American Recorder SPL-8810 Sound Level Meter
  • Onkyo TX-SR705 Receiver
  • MartinLogan ESL Hybrid Electrostatic Loudspeakers
  • NSM Audio Model 5 Compact 2-Way Loudspeakers
  • Bosch DLR130 Laser Distance Measurer
  • Fluke 77 Multimeter
  • CyberPower PP1100SW Sinewave UPS
Review Goals

While it is not the purpose of this review to focus solely on comparison to Audyssey, it is worthwhile to mention key contrasts. Any comparisons made in this review will focus more on the end-user experience. I will do my best to keep those comparisons accurate and impartial.

In my review of the miniDSP DDRC-22D last year, I stated some hesitation in Dirac Live's ability to deliver and/or preserve ultimate soundstage and imaging qualities. A good deal of time was spent over the past year in pursuit of ultimate soundstage and imaging setup using various kinds of speakers. I have been especially pleased with results achieved with dipoles, specifically the MartinLogan Electro-Motion ESL model, in my listening room. Through the use of carefully placed reflective panels, adjusted reflection timings, and other specific room treatments, I have achieved soundstage with imaging qualities seldom experienced elsewhere, and the ability to make use of specific track segments in judging soundstage quality. This involves matching of rear wave timing from the dipoles to the listening position (LP) with sub-micro-second precision. A major goal of this review was to see how well Dirac Live would behave in handling a finely-tuned soundstage and imaging setup. There was even a curve ball to be thrown at Dirac Live in the process, and I look forward with anticipation to see just how it would be handled.

System Configuration

Just where the nanoAVR DL will fit into a music or home cinema system is a major consideration. Its two inputs will be a limitation for some users. HDMI tends to be quite particular in how components are hooked together, with HDCP setting strict rules, so there is no way to cheat and place the nanoAVR DL both downstream AND upstream of an AVR in a loopback fashion.

This is an area where Audyssey has a distinct edge, so far. Being integrated on the downstream side of the preamp and selection circuitry in AVRs, it can be used along with any input and processing combination built into a given model.

Dirac Live is making inroads into similar products, and another implementation of Dirac Live offered by miniDSP is the DDRC-88A, with 8 analog inputs and outputs, which can work with an AVR that has pre-outs and amplifier inputs, or any system where pre-outs normally feed external amplifiers. In that position, it could even work as part of an ATMOS system.

A common issue with computer audio is the often exasperating process of getting components and programs to behave well together. This was a problem when working with the nanoAVR DL unit at Sonnie's get-together in March, and was again in completing this review. Results will vary immensely depending upon the equipment and applications in use.

A big reason for forums such as this one is so we can commiserate and help each other through such times, so while the prospect can be daunting, it should not be a showstopper for anyone considering a setup like this. The particular hardware and drivers for the HDMI port on my test laptop seemed a bit finicky. So did my Onkyo receiver, a 7-year-old model which performs wonderfully in the audio real, but might be a bit behind the times when it comes to HDMI configuration. The HDMI monitor which I use with the AVR simply had to be unplugged for the laptop to agree to pass audio through the HDMI connection. Oh, well.

Getting the AVR mode to pass properly to the laptop so it would finally let flow the audio bits often required changing modes, turning off and on the AVR, unplugging and replugging cables, etcetera. Audyssey, in comparison, has the advantage skipping most of those particular system configuration funnies. Its plug-in-the-microphone-and-follow-the-sequence simplicity is certainly a powerful factor in its popularity over the years. In fairness though, none of the difficulties I have mentioned were a fault of the nanoAVR DL unit or software, but were rather caused by incompatibilities at the front or back end of my system surrounding it. This could have easily been the case in a system using Audyssey, too,

Getting Dirac Live to work along with the PC and Dirac Live control program was extremely easy. Dirac Live and miniDSP documentation are excellent, as are their online help when I have needed them, and the only hitches I ran into any of their products were silly oversights or carelessness of my own, always quickly resolved. The Dirac Live program itself almost needs no documentation. So I would say the overall complexity of setting up the miniDSP DL and software vs setting up Audyssey is only slightly greater.

I have no direct experience in using the Pro Kit for Audyssey XT32. That must be mentioned at this point, because in order to take the option of moving the Audyssey user experience to the higher level of functionality more equivalent to Dirac Live, use of the Pro Kit is necessary. The complexity level level of mastering that configuration along with the necessary licensing per installation will have to be discovered elsewhere.

Cost comparisons are also difficult to make. Audyssey XT32 is only available in high-end AVRs, and only certain of them have Pro Kit capability, which adds several hundred dollars to the cost of the expensive AVR. Then each installation requires a licensing fee, a one-time expense per AVR. In comparison, the nanoAVR DL may be used with any HDMI-capable system, but big-picture system issues might make it not work for some users where an Audyssey-equipped AVR might. At any rate, the $549 price tag for the nanoAVR with control software, which can work with virtually any AVR on the market, even with many component systems, seems to make full-capability room correction attainable to many users for whom it was not before.

EDIT: An additional point in Audyssey's favor is that there is a huge and very helpful online community that has many years of experience in getting satisfaction with the product. This is a factor not to be taken lightly.

An interesting configuration detail that popped up was the apparent reversal of the surround speakers in the Windows driver for the HDMI output of my laptop. Dirac Live allows speaker positions to be changed for any filter set, so correcting the issue was very easily. This was perfect with my older DVD-Audio player, but was incorrect for my newer Bluray player. Ultimately I ended up with two versions of the final filter set, one for each of the two players.

Configuration issues also surround the use of Room EQ Wizard to verify the results of a Dirac Live filter set in a system. I ended up playing back a prerecorded WAV file of either pink noise or REW’s frequency sweep in the computer at the beginning of the HDMI string, and using another computer altogether running REW to capture the measurement mic input and do the analysis, an advanced technique discussed in the HTS forums. Another question is where to put the microphone in such a case to accurately verify the effectiveness of Dirac Live? The pink noise source combined with a moving-mic-method RTA measurement gives the most accurate quick frequency response view, but is devoid of impulse response and timing data. That along with a single point measurement at the LP where the first Dirac Live sweep was taken for an impulse and timing verification measurement is the best I can suggest.

Soundstage and Imaging Test

My first setup measurements for Dirac Live involved a single mic position setup pattern, to be discussed in detail later. Once the calibration was complete, the viewing power of Dirac Live becomes apparent. With a click and or drag, any area of the measured or corrected waveforms can be zoomed in on. You can zoom in even further, and a double click at any point takes you back to the previous level of zoom. Dirac Live allows one to view before and after views of both the frequency response and the impulse diagram.

I was anxious to see how this had turned out because of the curve ball I had in store for Dirac Live. The dipole front left and front right electrostatic speakers had strong rear wave reflections being directed at the LP, and the timing of those reflections relative to each other and to the front wave timing was crucial to the soundstage results I had received.

The fine-tuning of the soundstage and imaging had been done while listening to two kinds of tracks. The song Roscoe by Midlake is a dense but simple mix with four or five male voices and several guitars closely spaced across the soundstage. Maximum image clarity and separation was the goal with this track. The second test case involved the B-52s song Ain't It A Shame. An echo of Cindy Wilson's vocals was mixed with so it would occupy an empty area of the soundstage, and rather than being a point source, it is mixed with a certain width. When properly tuned, the image of those echoes coalesced with density and seemed to have an almost solid outline reminiscent of early animation characters.

I had been listening to the soundstage for week, and while it seem almost perfect in many respects, it had its shortcomings. At higher frequencies, the sweet spot became narrow almost to the point of nonexistence, and I knew there would be fine adjustments to be made in the coming days, perhaps even a complete setup from scratch.

How well did Dirac Live fare in the situation? The hand-tuning of those reflections involved at one point adjustments of nuts on machine screws in the amount of one twelfth of a turn at a time , then listening for that particular echo change from broad vague boundaries to tightly-defined with very distinct boundaries, all from a single timing adjustment of under one-quarter of a microsecond. I say this not so you will question my sanity, but so you will appreciate the precision with which Dirac Live would have to work with the impulse timing to not be disruptive of the soundstage experience I had already achieved.

The fine points of that soundstage were preserved to perfection. I am pleased to report that the integrity of that little echo blob was completely unchanged. Another echo blob I had to check on was at the beginning of Weird Fishes by Radiohead, with the snare drum somehow casting a little echo just behind my right shoulder. Again, Dirac Live preserved it perfectly. Several densely packed mix tracks suddenly showed clear separation between sonic images that had slightly overlapped before.

I found myself in conflict. The dedicated Audio Tweaker in me was already setup for another round of adjustments to improve the soundstage at high frequencies, but the end user in me was hearing sonic perfection with the polishing effect of super-tight frequency response matching from top to bottom by Dirac Live. Was this the end of an era? Rather than tweak over and over in hopes of reaching perfection, I could get close and then let Dirac Live carry me across the finish line, with no apparent sacrifice in sonic quality.

The final hurdle was cleared when I saw that Dirac Live had performed its impulse correction magic upon not just the first direct wave impulses but also on the second reflected wave impulses. It had knocked my curve ball out of the park. I was truly impressed.

At that point, I did quite a bit of listening, and I have to say that the final polish brought to the soundstage and imaging went far beyond what I had hoped for. Image sharpness and separation were well beyond what I had achieved hand. I am of the belief that good room correction can take you from wherever you are to a somewhat better place. The better the place you start from, the closer to Sonic Nirvana you might end up. I had never been sure if a room correction tool could handle those final refinements with the appropriate finesse. At that point in the evaluation it appeared that Dirac Live could do so.

What put Dirac Live over the top for me was listening to Devin Townsend's Ki album. A long time fan of his music, I have often felt that his heavier track mixes could be a little too dense for their own good at times, bordering on sonically indecipherable at others. What Dirac Live had done was like going through the sonic landscape with a fine-toothed comb, sharpening and separating all of the images and teasing out into the open minute details that rarely saw the light of day. My opinion of Devin Townsend's mixes was transformed from thinking them overcrowded to admiring them as gorgeous complex soundscapes begging to be delivered by the right listening environment, somewhat intolerant of a typical speaker setup.

Finally, there was a distinct three-dimensionality to images and sounds resulting from the careful timing of reflections, along with a strong sense of depth acuity that is rare in a soundstage. All this was perfectly preserved by the Dirac Live.

Odyssey is yet to be put to the test with this kind of soundstage to see how well it can preserve its qualities. I will report on that test when I am able to complete it.

Target Curves

I gained a whole new appreciation for the level of capability provided for the user to refine and work with target curves with Dirac Live. Audio channels can easily be linked together with a simple drag-and-drop so that a single target curve applies to all in a group. After a few adjustments, any individual channel can be removed from that group and then have an independent target curve for further refinement if desired. The endpoints of the range of correction can also be changed with a drag-and-drop. I played with this quite a bit, and see it being a real strength for Dirac Live.

Remember that each target curve must be separately saved. At one point I was editing a target curve text file directly then saving to load it into Dirac Live.

The importance of being able to easily manipulate target curves is well worth pointing out here. I spend a bit of time optimizing my favorite. I went for a slight bass boost, just a couple dB at 20 hurts, for a nice solid bottom end. This was flat by about 120 hZ, as it seemed that boosting the band between 120 and 500 by any amount simply muddied the mix, with some masking starting to set in and drown out detail in the higher ranges. Above that, I favored a flat response except for allowing a single broad peak of a couple dB at about 4 kHz to bring out detail. That one little area of emphasis ended up being like a detail magnifying glass for many instruments and sounds, all without sounding overly bright.

Here is where Dirac Live truly shines over stock Audyssey XT or XT32. Audyssey users sometimes complain about not being able to get satisfactory results with Audyssey, and the target curve limitations are the first thing I think of. So much can be accomplished with a few small tweaks, and personal preference is such a powerful factor. The two target curves supplied are only slightly different. Many Audyssey users resort to additional EQ to enhance the final result.

Again, not having worked with the Audyssey Pro Kit, I do not know its limits of capability or the ease of use in modifying and applying custom target curves. As stated, it is only usable with a few of the most expensive AVRs, so that level of capability is not available to most Audyssey users anyway. Having worked with Dirac Live's target curve system and experienced what it can do for your sound, I now consider it an indispensable element of a proper DRC system, and would not undertake a room correction implementation without it.

EDIT: How could I forget to mention the ability to save measurement / filter / channel configuration sets! Saving, reloading, and modifying at will and with ease, building on a core of several previous measurements and adding a few others at different points in the listening area, having four different filter / config sets in the nanoAVR DL to switch among, these are powerful features to consider. The Audyssey Pro Kit also allows one to save and reload filter sets; otherwise, the filter set captive in the AVR at the end of an Audyssey calibration will be lost when the next calibration is saved, although a few Audyssey XT32 equipped network AVR models that are not Pro Kit ready do allow saving and restoring the filter set over the network.

Bass Management and Other Complexities

While my experiences with bass and subwoofer management with Dirac Live are not extensive, I was very pleased both with its ability to control room modes and its ability to integrate the sub nicely with my other speakers. The subwoofer was set off to the side, making integration with the mains a challenge for Dirac Live. It was able to do so easily and smoothly with only minimal narrow notches at a couple of points and no mode peaks it could not easily control.

Starting with an 80 Hz crossover, I could easily hear certain bass runs move toward or away from the subwoofer. To reduce localization, I changed the AVR crossover point to 60 Hz, then changed the Dirac Live target curve for the subwoofer, also very easy. With these mods, there was no localization detectable.

Dirac Live's mixed-mode filter strategy seems a major strength in its design. Using minimum phase filters for control of low frequency modes allows much stronger correction than using IR filters through the bass range. It is common for Audyssey users to have to use additional hardware equalization devices to pre-correct larger bass peaks, even with XT32 capability. In fact, this was an area where Sonnie Parker was an early expert and his sharing of those experiences online led to the foundation of Home Theater Shack.

Playing with AVR subwoofer distance settings to optimize subwoofer integration after Audyssey has been applied is another valuable skill users have needed to hone. All that along with Audyssey not always giving the same crossover points and sometimes having to rerun Audyssey because of changes in distance settings. Every Audyssey run has to be followed by a check of distance and crossover values, knowing that mysterious changes can occur for no apparent reason and with sonically destructive results.

None of this is necessary with Dirac Live. The settings as you set them in your AVR remains the same and Dirac Live works with them, refining and preserving timing with extreme precision as previously mentioned. I ended up adjusting my surround speaker levels after applying Dirac Live. That and the subwoofer crossover change to eliminate localization were the only times I had to go back to AVR settings, neither a result of Dirac Live's operation. The interaction between Audyssey and AVR settings, while seeming to be a strength, has always been a bit perplexing to me. I much prefer Dirac Live's separation from the AVR settings, and find it to be easier to understand and work with.

There were usually a few narrow notches in the frequency response curve with Dirac Live, but they were always narrow enough that I was not concerned about them. The correction algorithms have wisely been programmed to not work too hard at eliminating those notches, as they are nearly impossible to hear even if you go looking for them with sine waves.

Another important contrast between Dirac Live and Audyssey is Dirac Live's forgiving ability to re-do individual measurements. An interruption with Audyssey forces you to restart the process. Also with Dirac Live, previous measurements can be re-used, with any individual measurements deleted from the set to allow for new ones. I made good use of that capability. These also are areas where the Audyssey Pro Kit helps, although I do not know to what degree.

All in all, my experience of getting from here to there is about as straight a line with Dirac Live as I can imagine it being with any room correction product, much straighter than with some I have experienced.

Two little wishes:

1. A gain (Master Volume, or Output Volume) readout in the Dirac Live Control Program and in the Android control app. If you can change it, you ought to be able to read what it is.

EDIT: A recent update to the Dirac Live calibration program includes the numerical readout of the Output Volume gain setting I was hoping for. Perfect! This is also soon to be added to the miniDSP smartphone remote control apps.

2. Some gain. Autocorrection filtering almost always makes you back off your gain a little to avoid distortion. The level and volume controls for the nanoAVR DL are all loss-oriented. Some systems might not be able to get to desired volume levels with gain cuts when more severe correction is needed.

EDIT: A closer look at the gain question since publishing the review has led me to change my stance somewhat. The product manual suggests keeping the Master Volume (Output Volume) setting as close to zero (no change in signal level) as possible without distortion. When a DSP is used for equalization of any kind, one is never quite sure what frequencies will be amplified by how much and when, and I had initially backed off the nanoAVR DL Master Volume far more than was needed, as I know at least one other user has done who complained about not having enough gain to get to desire volume levels.

The internal gain structure and headroom of a DSP device is always a little bit of a mystery to the user, and as it turns out, the filter set I ended up with allowed me to run with a Master Volume setting of 0 dB with no obvious clipping. To be a bit conservative it was finally left at -6 dB, which is still almost 10 dB higher than I was running it before.

So I retract the request. There are always possible situations where adding a component to a system will change the gain structure enough that additional gain is needed somewhere, and it is not necessarily correct to consider it a design fault in the component added.

Surround Performance

My 5.1 surround setup made use of a pair of small bookshelf speakers for the surrounds and a phantom center channel. I favor a phantom center channel because the front soundstage is then a continuous soundstage from left to right, more natural and cohesive than having a center channel speaker. I also subscribe to the philosophy that if there is ever another person in the room with me, they could not possibly care about good sound as much as I do, and if they do I am going to let them sit in the center seat anyway, so the center channel speaker and anchored center image source for other listening positions is of little importance to me.

I listened to Porcupine Tree's Deadwing album from 5.1 DVD-Audio, along with the 2013 5.1 remix of Yes's classic Close To The Edge and Pink Floyd's The River. Even with the vastly different speaker types between mains and surrounds, Dirac Live correction had them working very seamlessly together. Synthesizer sounds on a couple of Deadwing tracks moved around the circle from surround to surround and stayed very sharp and distinct all the way around with no obvious change in tonality while moving from speaker to speaker.

It was the strength of the image clarity and distinctness of the apparent position of that image in the room that impressed me more than the frequency response consistency. I will venture that it is in no small part a result of their impulse response correction that image clarity and positioning can be handled so consistently between different speaker types as it was.

My surround speakers were located straight to the left and right of the listening chair so they would always have a view of the setup mic (explanation below). I would like to find a way to get equal performance with them placed more behind the LP, allowing for a more complete surround circle on appropriate mixes.

Setup Mic Pattern

I have always been a bit of a rebel when it comes to mic setup patterns for auto correction programs. My first experiences with Audyssey led me to recommending either a single position or a tightly-spaced symmetrical pattern around the sitting position.

When I reviewed the DDRC-22D, my impressions were that the single mic position gave better image clarity and soundstage performance. Flavio at Dirac Research suggested that I try a widely-spaced pattern, even wider than the widest I had tried in my testing. This seemed to defy logic, but I promised to do so.

A recent experiment has led me to use a hardwood board against the seat back and hang the measurement mic tip actually just touching that boared. There are no reflections from the chair back and the impulse response is much cleaner. Frequency response is quite smooth, although not quite accurate, so a pink noise moving-mic measurement provides a reference. Once an inverse of the difference is added to the target curve, accurate correction can be achieved. Allowing Dirac Live to correct for speaker and room and not have to clean up chair reflections seemed to put its performance over the top.

To complete the testing, though, I completed a three-part calibration following the prescribed method. Starting with a single mic position hovering over the chair at the LP (no board in place), then adding four more, then four more, and generating filter sets at each level. Listening to those filter sets, they all worked very well, and gave much better frequency response than my more complex approach, with its required manual offsetting of the target curve. I will even say that the final version, with all nine measurements, sounded the best. But the precise preservation of the finest details of the soundstage was simply not there. The impact and the three-dimensional POP were missing. Bear in mind, this is picking at extremely fine detail, it is not a negative statement in any way whatsoever about the product.

So I combined the two approaches. The first three measurements were done with the mic tip touching the board leaning against the back of the chair, as before. Then the board was removed and the next 6 measurements were widely spaced, all at least 2 feet away from the chair surface and pretty much randomly placed, as suggested.

The result was that all of the fine soundstage detail and image clarity and separation were preserved in their entirety, along with the depth and depth acuity, as before, but with the benefit of an excellent frequency response for all speakers, including the surround speakers, with no need for extra measurements or offsets to the target curve. The performance could not have been more ideal.

I do not know what magical potions were programmed into Dirac Live to allow it to take such widely spaced information and preserve the image clarity and soundstage information so nicely for the listening position, but they have done so. Again, my little trick with the hard surface for the mic to avoid having to deal with any reflections in the first measurements is going above and beyond, but for absolute perfectionist soundstage preservation, it somehow relaxes the task for Dirac Live and lets it focus where it needs to.

My recommendations for mic setup pattern for home theater use would be to follow miniDSP / Dirac Research's suggestions exactly as stated. The first point should be right at the center of the head location for the best imaging and soundstage timing information. The other locations should be widely spaced around the listening position, even extending to seats to the left and right if desired. That is as far as I took it, and it worked extremely well.

I would suggest the same for two-channel users, except those extreme chasers of soundstage clarity who might benefit from taking the first measurement or two with that hard surface so the chairs reflections are avoided.

Here is another interesting contrast with Audyssey. To get good soundstage and imaging clarity, the Audyssey mic pattern must be a single mic or a very carefully spaced symmetrical pattern right around the area of the head of the listener. With Dirac Live, the setup pattern requirements are incredibly relaxed once the first measurement(s) are done. (I will try it with just one measurement against the board some time soon.) The need to carefully measure fractions of inches between setup pattern points to ensure a symmetrical pattern is simply not needed with Dirac Live. That tedious process alone had pretty much stopped me from using Audyssey.

With Dirac Live one would still have the option of the widely spaced pattern for home theater, extending over more seats, but I would still caution against going too far with them, as frequency response compromises are inevitable. It starts to get out of hand. The main listener is always going to be the one caring most about the sound. I will always recommend giving the main LP top priority in this way.

Is 24/48 Enough?

Every product design involves trade-offs. The nanoAVR DL includes a lot of capability for the price. Deciding upon a single sample rate for internal processing helps keep RAM and hardware requirements under control, as filters need only be stored for operation at the one processing frequency.

In my listening I never felt the 24-bit / 48 kHz limitation hold me back in any way. For those who insist upon higher bit rates, the full version of the Dirac Live supports 24 bit / 192 kHz audio. I will be reviewing that product in the near future.

The Audio Tweaker's Dilema

Being an audio system tweaker at heart, one who enjoys setting up by hand and takes a little pride in the result being as good as or better than what could be achieved with auto correction, it is a little hard for me to bring myself to admit that without the nanoAVR DL, I might not have achieved the level of image clarity and separation and soundstage perfection that I did. Dirac Live is built upon principles that allow it to go further than some DRC technologies, to take giant steps instead of regular-sized steps in the "mother may I" of the room correction gain.

The Audio Tweaker's Dilemma is How far do I go? How close am I to perfection, and how much more work will give me how much more benefit? Will I get there with this particular setup, or have to back off and start over and take another day to get beyond where I am now? Do I break down and let some automatic technology finish the job or stick to my guns and insist on tweaking to the end? For some it is a matter of pride, of being invested in the idea of doing it by hand. That applies to me a little, but in the end the results win out, and here I am admitting that I will probably never implement a setup of my own again without Dirac Live as my polishing tool. It is simply that good, and there is no downside or reason not to.

Even a well-matched pair of speakers is never perfectly matched, and one can tweak them and move them again for hours or days on end trying to find the best imaging from that position where the frequency response matches the best. The nanoAVR DL can make the job a whole lot easier. And no matter how close you think you are already when you activate Dirac Live, you will probably realize in a few seconds that it just got better.

Audio Tweakers beware. If your hand tuning is a matter of pride for you, and you would be crushed to discover that you could not get it as good without Dirac Live as with it, then you had better never try it. But if like me you are simply goibg for the best sound you can get, do yourself a favor and get an implementation of Dirac Live in your toolbox. I guarantee you will not regret it.


I do not often rave about products that I review. There is almost always some combination of pros and cons that need to be positioned for a given user or set of needs.

In the case of the miniDSP nanoAVR DL, and the use of Dirac Live in general for room correction and speaker fine tuning, I simply have to say my experience reviewing these products has been one of the most positive I have had to date. This is simply a phenomenal product, and, where it fits one's system configuration needs, I can think of no reason it should not be at or near the top of any serious listener's purchase list, if they do not have one already.

The miniDSDP nanoAVR DL is an excellent product for the home theater user, and will serve the surround music listener as well. If you are just getting started with room correction, there is no easier way to do it than with Dirac Live. If you want to add a little polish to an already high-performing system, it can do that very nicely. If you don't think you will get any benefit from a product like this, you are probably wrong. My experience has shown that it is all upside with no downside when properly applied, is easy to use, and is an incredible value. Buy one.

1 - 1 of 1 Posts
Not open for further replies.