DSPeaker Anti-Mode X4

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#1
Part 1

Background

My comments will not be able to assess all the X4's capabilities. My current system uses it strictly as a DSP equalizer for a pair of stereo speakers. I'm not using the subwoofer or crossover functions since I don't use subs. I'm not using the headphone function since my Benchmark HPA4 is the cat's meow in that department. I'm not using the DAC or analog/digital preamp functions since the combination of my Benchmark DAC3 HGC and HPA4 fill that bill extraordinarily well for me. Thus, all I'm talking about are the plain vanilla equalization functions.

I had my X4 in hand from Walter at Underwood Hi Fi (Underwood Wally) a mere two days after giving him my credit card information. Thank you, Walter. Now, I have some reason to believe that some hardware changes to the unit will occur in the next production run. But as far as I can determine, the only changes to be made in later units will implement a better USB 1.1 workaround, something which is not important to me.

If you, too, decide you'd want to purchase an X4, check out Underwood Audio's Audiogon ad for the X4 here. Also, check for sales on this thing by contacting Walter at Underwood. Oftentimes, the X4 is on sale for less than the price shown in the Audiogon ad. I got mine for $3,750 delivered as part of Underwood's Halloween sale. Now there is another sale advertising an even lower price.

You can also order the unit from most international markets, albeit probably at a higher price, directly from DSPeaker's VLSI Solutions Webstore. One advantage of ordering directly from the Webstore is that this will undoubtedly assure you of getting a unit from the latest production.

Some may wonder if there is still really a market for separate audio-box equalizers. Many computer audio server programs incorporate equalizer functions. Many units aimed at the home theater market, even AV receivers, incorporate some sort of fairly sophisticated proprietary or third-part (e.g., Audyssey) equalization functionality. Add to that the aversion of many hard-core audiophiles to any sort of intentional signal processing and we do see that the market for such an audio box is certainly more limited today than it once was. Still . . . there are reasons for such a unit, as I'll discuss.

Equalizers I've Owned and Used

The DSPeaker Anti-Mode X4 is the latest in the series of anti-mode products from DSPeaker. It is the third DSPeaker DSP equalization device I've owned and used in my audio systems, the prior ones being both the 2012 and 2013 editions of the DSPeaker Anti-Mode 2.0 Dual Core units.

I've written a bit about my experiences with equalizing audio systems here in Tom's Corner a couple of times before, first in 2010 in "Back in the Fold—of the Equalized" and "To Equalize or Not to Equalize: That Is the Question," and then in 2016 in "ART EQ355 Equalizer Added to Stirling LS3/6 + AudioKinesis Swarm System." As you can tell from those discussions, I'm a veteran of using electronic equalization to modify the frequency response of home audio systems and am quite familiar with many of the issues surrounding the use of such equalization.

As far as I can remember, here, in rough chronological order, is a list of the audio equalizers I've owned and used in my home audio systems over the years since the late 1990s, in addition to old-fashioned bass and treble controls on various preamps in the 1990s and before:

  • Cello Palette Preamp
  • Z-Systems rdp-1
  • Legacy Steradian (for Legacy Whisper speakers)
  • Rives PARC
  • Rane DEQ-60L
  • TacT RCS 2.2XP AAA stock
  • TacT RCS 2.2XP with fully Maui Mods
  • Audient ASP231
  • DSPeaker Anti-Mode 2.0 Dual Core 2012 model
  • DSPeaker Anti-Mode 2.0 Dual Core 2013 model with both stock and two different after-market power supplies
  • Behringer DCX2496 + DEQ2496
  • ART EQ355
  • RoomPerfect (in Lyngdorf TDAI-2170)
  • Z-Systems rdq-1
  • DSPeaker X4

Throughout the decade of the 1990s, I also operated a professional grade 24-channel mixing board as a member and then leader of a large local church's technical production crew. I used this to manually equalize by ear the live house sound during church services and concerts as well as for recording the live program. Each channel had parametric and shelf filters available to alter response. The response of each microphone in use was separately equalized through this mixer. We also used a 31-band Klark-Technic equalizer to provide a "floor" equalization for the live sound.

The Long Gestation Period

This product has had an unusually long gestation period. It was announced no later than 2014 or 2015, has been in beta testing by a small group of home users since 2016, and just this year has begun to be offered to the more general public. Supplies are still quite limited and other than manufacturer-direct purchase, I know of only one USA dealer, Underwood Audio, who has stock on hand, much less any dealer who can demonstrate or loan one to you. It has been in action at many audio shows in the USA and Europe over the past couple of years, however, so some may have seen or heard it in action.

I find this unusual in that DSPeaker already produced (and still sells) quite capable equalization products in the form of its prior Anti-Mode devices. This is not a blank-slate product. The same equalization theory is used here as in the earlier devices with the primary changes (other than better sonic performance!) seeming to be a lot more flexibility in terms of inputs and outputs as well as making the unit more able to deal with subwoofed systems. Frankly, most of the added flexibility features here (headphone amp excluded) were part of the software/hardware on offer by the TacT RCS 2.2XP AAA I owned more than a decade ago. Other current products, such as some from DEQX and Lyngdorf, also offer many of the other features of the X4.

But I gather that two of the things which make this equalizer quite unusual if not unique are (1) its ability to set up the high- and low-pass crossovers for a subwoofered system by analysis of distortion from the mains and subs at various possible crossover points and slopes, computing the lowest distortion configuration and (2) its inclusion of separate equalization and head-related-transfer-function programming (what the manufacturer calls "auralization") for headphone use. Neither one of these functions is of much interest to me at present since my current system doesn't use subwoofers and since my current headphone amp, the Benchmark HPA4, is superb without such bells and whistles. Other users may find these features invaluable, however. Many use subwoofed systems and I know that many users of electronic equalization are looking for products which allow subwoofer integration into the overall equalization scheme.

[Continued in Part 2]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#2
Part 2
Why So Many Equalizers?

One can legitimately ask: if I think electronic equalization of a home audio system is a net plus, why have I purchased and later sold so many? Some of the reasons are detailed in my 2010 article "To Equalize or Not to Equalize: That Is the Question." Basically, like many audiophiles, I've been looking for the Holy Grail of equalizers: a unit that is easy to use, while also having all the necessary flexibility for dealing with different home-audio configurations and equalization challenges, and—perhaps most importantly to me—does not have any sonic downsides to its use. Until now, this Holy Grail has eluded my grasp.

But as important as the sonics are, I don't want to downplay the importance of convenience, either. As I've "matured," this is an ever-more-important factor to me for audio equipment. With all equalization devices, for best results, if you move the speakers, the listening position, or even make any changes in furniture arrangement or decor, you have to re-measure and readjust the equalization based on those new measurements. Most equalizers are a bit or more of a pain to get adjusted even once. The pain repeats every time you make any changes in your room set up. Thus, I've also looked for an equalization device which makes it as easy as possible to run new measurements and adjust the EQ settings.

Here's a rundown on the pluses and problems I encountered with prior equalizers:

Traditional analog bass and treble tone controls: Easy to use, but not very flexible. They do not deal with room-mode-created bass peaks at all. In most, the turnover frequency for bass is too high and treble too low, meaning that their use messed up the midrange balance. An exception to the turnover frequency problem was the tone controls on the AR Amplifier/Receiver from the early 1970s. Also, tone controls are not really sonically transparent, as was evident when you flipped the switch taking them out of the circuit, a feature present on many later preamps with tone controls.

Cello Palette Preamp (about $6,500 when new in the 1990s): Together with the even more expensive ($25k!) Cello Audio Palette, this Cello unit was probably the best of the manually adjustable analog tone control units. The six adjustable bands were cannily chosen for maximal effectiveness in dealing with frequency balance problems in program material. Again, however, the unit really couldn't deal well with the listening room's bass modes. Cello also made an Audio Cart (see bottom right picture at this link) which was a stunning-looking clear acrylic supporting cart on casters which could put the controls of the Palette right by your hand in the listening chair, an arrangement even better than remote control (not available) would have been. The sound was fairly transparent for the time, with the main sonic shortcoming being a slightly increased noise floor when the tone controls were in the circuit.

Z-Systems rdp-1 (about $5,200 new in the 1990s): This is the one equalizer I truly wish I'd never sold. Some are still available occasionally on ebay for under $1,000, less than 1/5 the original 1990s asking price. For those willing to do manual measurements with a separate modern measuring system such as the OmniMic V2, this is the way to save serious cash and still get a hyper-transparent audio equalization device. There is a steep learning curve for adjusting this, however, and the necessity of using a separate system to do the measurements makes adjusting it, even with the remote control which came with it (and most units I see for sale today don't come with a functioning remote control) less than convenient. It was really inconvenient back in the days before computerized stand-alone measurement systems with real-time display of the results. That was one reason I sold my rdp-1. Then there is the issue of what shape your target curve should be for best sound. It does allow saving many preset EQ curves, however, so if you can keep track of which curve suits which music, with considerable experience adjusting by ear, this can eventually become a very handy device. The Z-systems rdq-1 (which I currently own) is basically the same device without the preamp functions, without remote control, and without the ability to individually EQ left and right channels. But all the rdq-1 units allowed output of up to 24/96, while many of the rdp-1 units were limited to 16/48 or 20/48 (the rdp-1 unit I owned had all the factory upgrades and thus did process at 24/96). Even with today's measurement equipment, I'd give it at best a 5 for convenience even though sonically it is a 9 or 10, close to top-of-the-heap. My sonic comparison of the Cello Palette Preamp and the Z-Systems rdp-1 from 19 years ago when both were fresh in my experience are available at the Audio Review site here (see the third review by Tom).

Legacy Steradian (included within the 1990s $13,500 price of the Legacy Audio Whisper speakers): This was a specialty EQ box for use with the early version of the Legacy Audio Whisper loudspeakers I owned. My year 2000 Stereo Times review of those speakers, including comparative comments of the Steradian EQ with the Z-Systems rdp-1, are available here. It was very convenient for dialing in the bass range sound of the Legacy Whisper speakers, but it is not a general purpose audio equalizer. Part of why it worked so well with the Whispers was the composite dipole nature of the bass from those speakers. Since those speakers only activated bass room modes to a minimal extent, there really wasn't much to correct in the bass other than to extend the low frequency limit and add any amount of upper bass warmth you cared to dial in.

Rives PARC (about $3,500 new around 2001): This was the first equalizer I used with my Harbeth M40 speakers, around 2001 or 2002. Whereas the Legacy Whispers needed little EQ in the "concrete bunker" basement listening room I used at the time, the M40s desperately needed it. This totally analog unit had three bands of parametric equalization per channel and was limited to adjustments below 350 Hz. But it did the trick as transparently as anything else I'd heard up to that time, just about as well as the Z-Systems rdp-1. Like the Z and most other equalizers of the time, however, you needed to use a separate measuring device to set it up. Even with such a device (the manufacturer used an early laptop to set it up in my room when I first purchased it), setting it by ear gave subjectively superior results. Here is where my long experience with using parametric EQ both at the church mixing board and with the Z-Systems device came in hand. No sonic downsides, but just not very convenient and totally limited to the bass region.

Rane DEQ-60L (about $1,100 new in the early 2000s): This was an early digital equalizer in the form of a 30-band-per-channel graphic equalizer. Its patented Perfect-Q technology really does create an equalizer where the pattern arrangement of the sliders corresponds to the frequency response alterations fed to your speakers. WYSIWYG control, in other words. While it may not be obvious to the EQ-uninitiated, this was and still is as rare as hen's teeth with graphic equalizers. The adjacent bands (and usually even the bands two bands away) interact considerably with each other so that you have to measure the results as you go along in order to determine what kind of response alteration you are really dialing into your system. Perfect-Q is a great concept. I only wish the sonic results were as good. Now, I must say that some users (even those with Harbeth M40 speakers) seem oblivious to any of the sonic downsides I heard from this thing, holding the unit in their lap while making adjustments and claiming sonic Nirvana from it. For me, the A/D and D/A conversions at 24/48, while they should have been fairly transparent, seemed anything but, with considerable flattening of images and staging and considerable high frequency glare apparent. I used this primarily with my Linkwitz Orion speakers, but also a bit with my original Harbeth M40s.

[Continued in Part 3]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#3
Part 3

Tact RCS 2.2XP AAA (both stock and Fully Maui Modded were about $5,500 in 2006 and 2007—the modifier either bought stock units at a discount from TacT or took your stock unit in trade, added his mods, and was thus able to sell them for about the same price as stock): This is another unit, which in the fully Maui-Modded version, I really wish I'd kept for the long term. The 2006 stock unit I first owned sounded a bit flat in staging and bright in aspect compared to my Bryston BP-26 analog preamp I also then owned. The full Maui Mods treatment on a later 2007 sample I also owned erased those sonic downsides (see my "Back in the Fold—of the Equalized" discussion), creating a truly state-of-the-art digital preamp with exceptional flexibility with a full complement of then-relevant inputs and outputs. Of course HDMI and USB interfaces were missing, given the design date. The software was complicated, but I mastered it just fine, even giving many suggestions to the designer, Boz, which he incorporated into later versions of the software. However, the software remained a bit flakey, often requiring restarts to cure problems and you had to know the software well to know when it was not operating properly. And unlike the Lyngdorf products, the application of EQ in the TacT was never automatic. The Lyngdorf and TacT software seemed polar opposites, in fact, with the TacT supremely adjustable manually with no automatic functions at all, while the Lyngdorf RoomPerfect was maximally automatic with little manual futzing capability. Then, seemingly suddenly, Boz disappeared from the scene, as did Anthony Padilla, the designer of the Maui Mods, and TacT owners were left supportless with no further product development on offer. Still, when it worked—and I could always get it to work—the Fully Maui Modded TacT RCS 2.2XP AAA sounded great. It was very flexible, offering full digital and analog preamp functions, a flexible two-way crossover, separate control for one or two subwoofers, and a throw-in Ambiophonics generator. It's built-in measurement program never worked well, however, being especially flakey in operation. Thus, I still had to rely on external measurements via my Liberty Audio Praxis to see how the system response was really being altered by the TacT.

Audient ASP231 (about $1,400 new when last sold in the mid-2000s): Probably the finest sounding analog 31-band 1/3-octave stereo graphic equalizer ever made. Nice looking, too. This unit won a sonic pro-audio shoot-out against many other graphic EQ units. I totally understand why. It was almost as sonically transparent as the 3-band Rives PARC, but with obviously far greater EQ flexibility. The downside was that this thing was a royal pain to adjust. I owned this at the same time as the Rane DEQ-60L and while the Audient sounded better, getting the EQ adjusted took hours instead of the minutes needed for the WYSIWYG Rane. The adjacent and second to adjacent bands interacted A LOT, necessitating seemingly endless tweaking of the sliders to get the response just so. Ultimately, it was just too much bother, even for me.

DSPeaker Anti-Mode 2.0 Dual Core 2012 and 2013 models (around $900 these days new): I owned both versions, the differences primarily being revisions of the unit's input sensitivity and output levels. The need for this is a clue to one of the sonic shortcomings of the unit: it always walked a fine line between input overload, dynamic headroom, and background noise, especially when used with analog input and/or analog outputs. When used strictly with digital input and digital output, it was pretty fine. Still, even at this sonic best, it suffered from what sounded like early transistoritis, a bright brittleness overlaid on the music, together with foreshortened stage depth. This could be ameliorated to some extent by adding one of the aftermarket ($100 to $500 extra) power supplies. However, when using the automatic equalization plus manual parametric EQ, the level of background hiss would rise to what I felt were unacceptable levels even when used via its digital ins and outs. Also, the digital ins and outs were limited to toslink, with coax not available and there was only one toslink input and output. Still, its automatic equalization worked extremely well, was superbly easy to implement, and was also superbly easy to adjust via remote control from program to program from the comfort and discernment available at my listening seat. This ease of adjustment, together with the great automatic EQ and the promise of more inputs and outputs and enhanced internal power supply was what made me sign up for the new X4.

Behringer DCX2496 + DEQ2496 (about $600 the pair new these days): I used this duo to control the response and crossover of the Sanders 10C speakers I owned. The pair provided great flexibility in terms of adjusting the crossover point and slopes between the bass cone and electrostatic upper range driver, as well as adjusting the fine points of the response. Strictly digital domain devices, these provided astounding control of response, with both parametric and 1/3-octave graphic controls. The learning curve is steep, primarily because of all the functionality available through the multitude of multi-function buttons and knobs, but once mastered, the functionality was extremely high. The DEQ2496 unit, when combined with even the $50 Behringer microphone also offered automatic adjustment of response to a user-chosen target curve. Very neat, especially for the low price. Unfortunately, the quiescent background hiss level produced by the units was extraordinarily high, clearly audible immediately upon entering the listening room many feet from the speakers. In addition, there was a bright glaze added to everything and the sonic stage was foreshortened. Allowing my Fully Maui-Modded TacT RCS 2.2XP AAA to handle all these adjustments for the Sanders speakers was sonically a huge improvement. Sanders later abandoned the Behringer crossover for a dBX model used in the later 10E.

ART EQ355 (about $200 new these days): My experience with this is detailed in Tom's Corner at "ART EQ355 Equalizer Added to Stirling LS3/6 + AudioKinesis Swarm System." While I found the analog ART EQ355 to be as sonically transparent as the Audient ASP231 when the ART was merely used to adjust a small range around 4 kHz, when I later used it to adjust the full-range response of speakers by moving many sliders out of centered position, the old problem with analog graphic equalizers reared its ugly head. With most such units, the more analog pots the signal passes through in a graphic equalizer, the less sonically transparent the equalizer becomes. The Audient seemed basically immune to this effect, but not so the ART. In addition, to use the unit to adjust full range response was fairly tedious since I needed to use an external system to measure and then adjust each slider. To its credit, the ART's graphic bands did not interact with each other to nearly the extent of the Audient's so adjusting it was not so arduous. Still, it did not reach the level of convenience I seek.

RoomPerfect in Lyngdorf TDAI-2170 (about $4,500 new with all the options): The Lyngdorf TDAI-2170 is a Swiss Army Knife of digital componentry. All its very flexible functionality works extremely well, too. With one exception. At least as implemented in this particular Lyngdorf unit, RoomPerfect has a glaring sonic flaw. Literally. RoomPerfect, no matter how I worked with it, automatically produced an excess of response level in the area from 2 kHz to about 5 kHz, about two to five dB too much level, which at this frequency is a huge, glaring error. I could not listen to the RoomPerfect equalizer for more than a few minutes. That's why I tried the ART. RoomPerfect is very easy to implement and the Lyngdorf unit comes with a measuring microphone and mike stand. I even ordered a replacement measuring microphone, thinking mine must be defective, but the new one yielded virtually the same results. RoomPerfect automatically measures your room and automatically implements a corrected response. Unfortunately, in the TDAI-2170, there are no manual adjustments available which can fully counteract the automatic boost in the presence/lower highs range. I see that the newer TDAI-3400 allows manual editing of the automatic RoomPerfect equalization.

[Continued in Part 4]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#4
Part 4

Idiosyncrasies

That brings me full circle, back to the DSPeaker X4. Before describing how the unit essentially works so very well, I will deal with the, to my mind, relatively unimportant ways in which the unit could still be improved as to its basic equalization function. I acknowledge that these are minor issues in that they don't significantly impede my goals of ease of set up, sonic transparency, sonic excellence in automatically adjusting bass response, and ease of manual adjustment of response.

First, in the before-and-after frequency response graphs, the "before" response is shown as a colored line and the "after" response is shown as a black line. In the key to which is which that appears on each graph, however, the "after" line is shown as appearing in a pale pastel color, clearly incorrect. This surely could be dealt with via a firmware change. Show a colored line within the light square and a black line within the other light square, instead of a light-colored square.

Second, nowhere in the manual does it suggest that the automatic correction can operate full range. However, full range is clearly a choice. And, in fact, as suggested by Walter at Underwood Hi Fi, my dealer, most users have found the full-range correction choice to sound best. In my experience, it truly does sound best, and not by a tiny margin, either. The user's manual needs to mention and indeed highlight this.

Third, the first time measurements are made, the unit automatically populates five or six presets with its suggested response changes. These suggested presets vary in terms of the range affected and the degree of low frequency compensation added back in to subjectively compensate for the lack of overall bass weight which the removal of bass peaks can cause. But once you change those settings for each preset, there is no way to recall or reset the originally generated automatic presets. To get them back, you would have to first perform a factory reset and then run the measurement sequence again. It's best to jot down the parameters of each preset so that you can recall them manually if you wish. This is especially so since the unit's auto presets all sound very natural indeed, just with bass weights which vary a bit from each other. I would hope that this could be corrected via a firmware update.

Fourth, the "after" graphs of response do not change if you manually adjust the preset Profiles. What is shown on the "after" graph seems to remain the auto-generated "after" response as measured during set up. It would be helpful for the unit to compute and show the "after" response with manual corrections input, even if those actually are not measured responses, just calculated "after" responses. I would hope that this issue could also be dealt with via a firmware update. (In defense of the current arrangement, however, note that you can generate the actual response of any manual setting just by using the Measurement Only function of the Anti-Mode software which allows you to take measurements without altering the computed corrections.)

Fifth, the back of the remote control needs to be redesigned. I very much appreciate the high-quality heft and sturdiness of the all-metal remote, its great button feel, minimal buttons, and sure-fire operation. However, the backside of the remote should be smooth. The five round-head screws that stick up from the case are unsightly. In addition, the fact that the screws stick up can easily cause you to mistake the backside screws for the raised frontside control buttons in a darkened room, especially the three screws which are arranged in a triangle on the back side behind the frontside control buttons. Since the remote does not light up (it would be nice if it did!), you need to rely on tactile cues in a darkened room. Redesign the back so the screwheads are flat and rebated so that they are flush with the backside.

Sixth, upgrading the firmware is a bit more difficult than it should be. The only method of doing this is by downloading the firmware update file into the root directory (that is, not within a file folder) of a USB stick. The USB stick must first be formatted to the FAT or FAT32 formats, which is a bit unusual these days. Also, you'd best use a USB stick of 32 GB or smaller. Despite my using programs which were able to format a larger 64GB stick into FAT32 format, I could not get that USB stick to work with the X4. Once I loaded the firmware update file onto a FAT32 formatted 16 GB stick, the update proceeded smoothly, taking only a couple of minutes or less. Something else to note: new firmware will not load unless the file name of the new firmware is FIRMWARE.X4. If it has a date or any other characters in the file name, it probably will not load. For example, new firmware sent to me with the name FIRMWARE_22_Nov_2018.X4 would not load until I renamed it FIRMWARE.X4.

Seventh, this is the one matter that may cause some a bit of pause and this may not be correctable without a significant redesign. But I've decided to put it here rather than make a larger fuss about it since, to me, in my system, it does not significantly get in the way of using the unit to correct the frequency response as I want it to be. Probably to avoid the problems of overload encountered in the earlier DSPeaker Anti-Mode 2.0, the design seems to adjust the overall midrange and volume level of mids and highs downward as any bass compensation is applied. This makes it more difficult to directly compare various levels of bass compensation since the overall volume changes as you make changes in the bass compensation level. But as long as you boost the volume level up by the amount of bass you are adding in dB, comparisons are possible and once you have the subjective volume correct again, it is a matter of only seconds before you can determine whether the new bass level is more or less satisfying to you. Note that, for reasons I don't understand, this idiosyncrasy does not affect the bass boost or tilt (Quick Tone) controls; applying those does not affect the level of mids and highs.

That's it. Nothing earth shattering. Despite these niggles, the X4 works EXTREMELY well for its intended purposes as an equalizer.

[Continued in Part 5]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#5
Part 5

The Set Up

My latest equipment rack is the short 28" Salamander Archetype with solid cherry wood shelves. I've added two additional shelves to the stock three, giving me a total of five shelves. This allows me to place all my electronics on this one short rack between the two speakers. Keeping the rack short—below tweeter height—helps keep the center stage imaging rock solid steady and generally helps with soundstaging, in my experience. I've placed the X4 on the center (third) shelf, with a large, heavy sand-filled Bright Star Little Rock platform beneath the equalizer to further isolate the equalizer from its vibrational environment.

IMG_6355 50%.jpg

I'm using the stock power cord directly into one of the quad of outlets fed by one of the two dedicated 20-amp circuits I use in my audio room. This circuit feeds all the front-end electronics; the other dedicated circuit feeds the amps.

Wiring is via one-meter Benchmark coaxial digital cables which feed the X4 digital signals from my Auralic Aries G2 streamer and my Oppo UDP-205 into two of the X4's coaxial digital inputs. Another meter of Benchmark coaxial digital cable feeds the X4's coaxial digital output to a coaxial digital input of my Benchmark DAC3 HGC. Neither the analog inputs nor analog outputs of the X4 are used; that means that neither the X4's A/D nor D/A converters are being used.

The system volume is controlled in the analog domain by the switching relay volume control circuit of my Benchmark HPA4 line/headphone amp. The digital volume controls of both the X4 and the DAC3 HGC are bypassed by setting them at their maximum unity gain positions. The available digital volume controls of the Oppo and Auralic also are not engaged.

All non-soldered connections in the signal path are treated with Caig Deoxit Gold G100L, the brush-on variety. All unused RCA, XLR, HDMI, USB, and ethernet jacks in the system are capped with Cardas or AudioQuest caps. Each electronic component has an Electronic Visionary Systems Ground Enhancer attached to a grounding point on that component. I damp the top of the X4 chassis with an issue of The Absolute Sound magazine. I also use auxiliary platforms under the two source components: a Bright Star Little Rock with three Bright Star IsoNodes beneath that to support my Oppo UDP-205; and a Mapleshade 4" maple platform supported by four Mapleshade Isoblocks to support the Auralic G2 streamer. Copies of TAS are used to damp the tops of all component chasses except for the Auralic G2, whose own isolating bearing feet are critically matched to the weight of that unit alone—nothing should be placed atop it.

[Continued in Part 6]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#6
Part 6

Measuring & Automatically Equalizing

The most difficult part of this process is the measuring and it is no more difficult than setting up the microphone. I have it on good authority that the proper way to set up the provided measurement microphone (a cylinder-like affair, similar to the measurement mics provided with the TacT, Behringer, and Lyngdorf products—narrower at the tip than at the rear) is to point the microphone straight up at the ceiling. This is not the way I've taken measurements with such microphones in the past, but this pointed-straight-up method apparently works best here. At least I'm not going to argue about it, given the results it yields.

As usual when measuring my system, I left the listening chair and all the other furniture in place as they usually are when listening. I leave the room while the test tones are playing. I've found with much experience at this sort of thing that there is nowhere I can stand, sit, or kneel elsewhere in this small room that does not affect the measurements more than if I just leave.

A boom microphone stand, XLR microphone cable of adequate length, and the measurement microphone all come in the box with the X4. The mike stand is okay, being very easy to adjust and adequately tighten. The one trick about it is that since it is quite lightweight compared to most boom stands, it is easy to overbalance the relatively short mike stand legs when you have the microphone and its cable boomed up and forward of the legs. To prevent the stand from tipping over you just have to make sure that one of the boom stand's feet is jutting out right under the boom. Having the boom project out between two legs of the stand will tend to get you into balance trouble. Of course, you could always weigh down the rear leg(s) with something, but that makes maneuvering the mike stand into exact position trickier.

My listening room is a converted bedroom and is thus pretty small. It's a rectangle 161 inches long, 132 inches wide, and 103 inches high. Given this room's size and shape, I have found that the listener and speaker positions computed by the Rule of Thirds 29% Version tool to work very well with the M40.2s and the other speakers I've had in this room. With the room's 132" Main Wall width and 161" Side Wall length, that puts the center of the front baffle of the Harbeths 38 9/32" from the side walls and 46 11/16" from the wall behind the speakers. The listening position is 94 11/16 from the Main Wall behind the speakers, or 48" from the plane of the speakers. The speakers and listener form an equilateral triangle of about 55 7/16" on a side. This is near-field listening as I and most others define it. It keeps my head more than five feet away from the wall behind the listening position, which is important, in my experience, for the best spatial presentation. Bass room modes, while certainly both measurable and audible, are less than with most other arrangements and the spatial presentation is the best I've heard in this room.

The microphone capsule (the part at the very front of the microphone) should be placed halfway between where your two ears are when you are seated in the listening position. In my case, that is centered between the side walls, 66 inches from each, 94 11/16 inches from the wall behind the speakers, and 39.25" above the carpet.

IMG_6350 50%.jpg

Since I've done this sort of thing many times before in this room, but have never pointed the measuring microphone straight up before, getting the microphone into position and hooked up took about an hour. The next time will take only 10 – 15 minutes since I'll be familiar with the way things should be arranged.

Once the microphone is in position with one end of the XLR cable plugged into the back end of the measuring microphone, you are ready to perform measurements. Wait until there is no one else home, and turn off noise sources like TVs, fans, washers, dryers, HVAC, etc. It also helps if you pick a time of day when there is not too much noise intruding on your listening room. In my location, I waited for the weekend when there was less train, plane, and vehicle traffic in my area; but you may not live where such is ever a problem noise-wise.

Once you believe it will be relatively quiet for awhile, you plug the other end of the XLR cable into the front-panel (VERY convenient) combo headphone/XLR jack of the X4. Presto, the X4 automatically goes into measurement mode, starting to play pink noise through the speakers. You can use the remote control or the front panel knob to the right of the display to advance the level of sound you hear until the cursor is just to the left of the Good marker on the display. You don't want to exceed the Good level.

Then, with remote control in hand, you head for the listening room's exit door. As you exit, click the enter button on the remote, and the measurement test tones will start within a few seconds, giving you just enough time to shut the listening room door behind you before they start. Tiptoe away and do something else quietly for a few minutes.

At least when the X4 is just measuring a stereo pair of speakers, the test tones only take five to ten minutes to complete. Once the tones stop, you can re-enter the room, unplug the microphone cable from the front panel of the X4, move the mike stand out of the way, and you are basically done!

The first time you make a measurement, the X4 will automatically populate six Profiles you can pick from to listen to the automatic equalization it has performed. You choose among these Profiles by tapping the A, B, or C keys on the remote. In my case, the first profile, Profile A, is the best sounding for most program material. Here are the before and after equalization plots the X4 constructed for this Profile A with the X4's measurement smoothing function applied. These are unretouched iPhone X photos of the X4 screen display. Before is in color, after is the black line; the first photo shows response from below 20 Hz to 500 Hz, the second one shows the full audio range response. The horizontal lines on the graphs are 6 dB apart:

IMG_6357 50%.jpg

IMG_6358 50%.jpg

As you can see, in the case of my Harbeth M40.2s as I have them set up in this room, there was not much to correct. I knew that by the way they sounded, which is fantastic even before equalization. If you look back at my original review of these, you will see that I said that I thought there was an emphasis of about 4 dB centered in the 80 Hz region and that turned out to be an accurate by-ear assessment. The X4 also knocked down a bit of excess level in the 1 kHz region, a problem REG has mentioned many times that the M40.1 speakers have. A bit was also shaved off around 4 kHz and a bit around 300 Hz. The bit of relaxation (much less prominent than in the earlier M40 and M40.1) through the presence range centered around 2 kHz is unaffected.

You can also see why I don't need subwoofers. In my room these speakers are at full level down to at least 30 Hz and still at basically the 1kHz level down to about 21 Hz. Full range bass without subs.

Note that the stated goal of the DSPeaker X4's equalization is to provide a corrected response which has a bit of emphasis in the lower frequencies. As you can see, the range below about 150 Hz is up about 3 to 4 dB compared to the 1 kHz reference level.

Another part of the X4's equalization operating theory (like that of the earlier DSPeaker Dual Core units) is that equalization is applied only to reduce peaks, never to fill in dips. Thus, the well-known-to-me room-induced dip around 50 Hz, which is visible in the response of all the speakers I've had placed similarly in this room, is left as a dip with no attempt to fill in the response. This, to my ears, is the best-sounding way to equalize audio systems.

[Continued in Part 7]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#7
Part 7

Adjusting to Taste

Adjusting the baseline automatically equalized response is also superbly easy with the X4. You may just prefer a generally different bass balance than the automatic program thinks is correct. Or, you may want to compensate for particular program material which has more or less bass or treble level than most other recordings.

There are several ways to do this and the manual covers this part quite well. However, I'll also describe the three ways I find easiest.

First, don't bother tweaking the automatic settings at all. They are awfully good, at least if you do your measurements correctly and actually listen from the spot and under the conditions which were present when the measurements were taken. For example, results will sound best if you listen with the listening room door shut if it was shut when the measurements were taken.

But, most will want to fiddle a bit, if for no other reason than you can and doing so is so easy. So, second, you can change the amount of bass compensation the automatic system is applying. From the remote, press the Enter button, then select the Anti-Mode menu, then select Compensation. Compensation is adjustable from 0 to +10 dB, with 10 being the bassiest. Adjust the Compensation level to taste with the remote control from your listening seat. You're done.

If you do decide to tweak the automatic settings, one way to decide what sounds best to you as your preferred baseline equalization is to listen to some pink noise played back at a healthy volume. I suggest the high-quality YouTube pink noise track available here. Compare the overall weight of the bass end of the noise when the Anti-Mode correction is active with the weight when it is bypassed (by a long press on the bass clef key). The correct setting is that which gives comparable weight, but which stops short of being so much that you start to hear a bass "tone" or "tones" within the pink noise of the type you probably will hear when the Anti-Mode correction is eliminated. If you are having trouble zeroing in on this by ear, with anti-mode correction on, try running the bass compensation control up to +10 dB and listen for the bass "tone" that will surely appear then. Once you identify this type of tone, reduce the bass compensation until the tone just disappears. That will probably be the correct level. In my case, this correct level matched the automatic setting the X4 picked for me. Using pink noise also helps in another way. I tend to think that getting rid of excess midbass "unmasks" the lower treble and thus having an easy-to-adjust bass compensation feature like the X4 allows you to dial back in enough smooth/non-peaky bass weight to keep the lower highs in balance. Listening to pink noise is a great way to do this. You can dial in enough bass weight to keep the lower highs from seeming to stick out.

Third, if you want a more customized adjustment, you can use the Quick Tone controls to (1) increase or decrease bass, (2) "tilt" the entire spectrum (a la the old Quad "Tilt" control on their preamps) toward bass or treble, or (3) increase or decrease treble. To pick any of these three options, first quickly touch and release the bass clef key on the remote. Then touch the A key for bass adjustment, B key for tilt, or C key for treble adjustment. Then the two top buttons on the left side of the remote adjust the response in the selected manner. Again, you do this while listening from your listening seat. You're done. It takes longer to read that description than to do it.

Since the manual doesn't show a screenshot of this, here's a picture of the screen showing applying a +2dB Tilt; with this setting the bass range is elevated by 2 dB and the treble range is decreased by 2 dB (the MAX word just means that I have the X4 volume control set at its maximum rather than at any lower level; as I mentioned above, I bypass the X4's digital volume control be setting it at MAX):

IMG_6371 50%.jpg

There is a further adjustment possible for the Quick Tone controls. You can go into the menu and adjust the turnover frequency for bass and treble boost and cut. Adjusting the turnover frequency can be used to make the bass and treble shifts more obvious or less obvious. Adjusting the turnover frequencies can also allow more of the middle frequencies to be affected by your adjustments, or to make sure that the mids stay unaffected regardless of what you do with the low bass and high treble.

At any point you can compare the effect of the equalization you have applied with bypassing equalization by long-pressing (holding down) the bass clef key. The display will then show an eighth note with a line through it to indicate that the effect of the X4 on your music has been bypassed. To restore the equalization to your settings, just quick-tap the bass clef key again.

You can also measure the results of the equalization you have incorporated without changing any of the recommended profiles. The Anti-Mode menu allows you to run the test tones for the sole purpose of measurement. Of course, if you also have a stand-alone measurement system such as the Omnimic V2 I use, measuring the response that way is almost as easy, although it does require you to bring a computer into play to display the measurement results.

Much finer control over the frequency response is also available via nine parametric band adjustments. The manual warns against overuse of such in attempting to achieve a measured ruler-smooth response. I agree. Unless you are very sure that a particular frequency response wrinkle you see in your measurements is part of the inherent speaker response and not a room effect, in my experience it is usually best to leave narrow-band anomalies alone. Attempting to fix them often leads to worse, not better, subjective results.

To determine whether a given response wrinkle is inherent to the speaker, I'd suggest using a stand-alone measuring system that gives you a real-time frequency response display (like Omnimic V2) and then move the measuring microphone around as you watch the real-time response display on your computer screen. If the response wrinkle stays fairly constant no matter where you place the microphone (near or far, up or down, left or right of the intended listening axis), you can be fairly sure that response wrinkle is inherent to the speaker rather than room induced. If inherent to the speaker, that response wrinkle can usefully be attacked via parametric equalization, either via a cut or boost at the frequency band in question. The rule against filling in dips doesn't apply to such problems because you are not trying to fill in an infinite null caused by room dimensions. Still, peaks are more important to eliminate than dips since peaks in response are inherently more audible than dips of similar magnitude and bandwidth.

As you can tell, my experience leads me to agree with most of the design choices DSPeaker has made for equalization. That is a primary reason I wanted the X4. I strongly suspected that if it could conquer the electronic colorations of the prior Dual Core, I'd really have something which could render room-induced coloration a non-issue, as well as something which could easily deal with common frequency-response colorations one encounters from one recording to another.

One other tweak I should mention: the Headroom sub-menu within the Sound menu. This has nothing to do with headphone listening. It refers to the amount of signal headroom available for processing before overloading the X4's output. Contrary to the case with both versions of the DSPeaker Anti-Mode 2.0 Dual Core I owned, I have never been able to produce gross distortion from the X4 due to digital overload. However, at least in my application, the Normal setting for Headroom seems to have a combination of the best sonics (primarily audible in the macro dynamics area) and highest volume available from the digital output. The Minimum setting has higher volume but doesn't sound quite as clean and unrestrained, while the More and Maximum Headroom settings seem to overly restrain dynamics and limit maximum volume a bit much to satisfy the gain structuring built into my Benchmark electronics. Your mileage may well vary since most systems are not gain structured in the way the Benchmark electronics are.

Finally, I think the best practice is to turn off functions you aren't using if they are not automatically defeated. One such function in my case is the headphone drive. Since I'm not using it, I have it disabled. No sense having the power supply available to drive the headphone amp if it is unused.

[Concluded in Part 8]
 
Last edited:

tmallin

WBF Technical Expert
May 19, 2010
287
5
18
66
Chicagoland
#8
Part 8

The Sound: Easy Equalization With No Downsides!

That title basically says it all. At least as I use it (coax digital in to coax digital out, with all the set up steps and tweaks applied as mentioned above), I hear absolutely no ill effects from applying equalization via the X4. There is no added noise, no added coloration other than what one intends to apply via the equalization controls, no added distortion, no dynamic restraint, no additional coloration, and no adverse effects on imaging or soundstaging. Using the X4's bypass function sounds the same as physically bypassing the unit; mere insertion of the unit into the signal path makes no sonic difference I can hear. And this is in the context of a system whose Benchmark electronics are about as clean and low-noise as they come.

Getting the desired equalization is, as I hope I've conveyed, about as easy as it gets. Profile A gives me just about what I want.

Oh, occasionally I add a bit more bass since I tend to like a rich sound. To do that I may bump up the Compensation control from the baseline of +1 to at most +2 dB. For especially bright recordings, I may knock down the treble a dB or two, but nothing much more than that. I do suppose that eventually I will construct a custom profile to deal with the peaky microphones used in the Mercury recordings I otherwise love. From long experience, I know just about what kind of parametric filter parameters to apply to flatten the high frequency response of those Golden Age recordings.

But for most material, the automatic adjustments the X4 computed for Profile A are just fine. The bit overly warm midbass is cleaned up and damped and the low bass is thus better exposed. My Harbeth M40.2s now more clearly sound like they have the full 20 Hz bass extension measured in my room. Electric and acoustic bass lines are much easier to follow on rock and jazz recordings.

Distortion is yet further reduced since both the amps and speakers are less stressed because the bass drive is reduced by more than 3 dB (a three dB reduction means power is halved) in the midbass frequencies which figure heavily in a lot of program material. The speakers/system will play significantly louder without overloading the room or stressing the speakers/amps if I want to go there.

The soundstage seems a bit more expanded in all dimensions and the imaging a bit firmer on that stage.

The reduction in level around 1 kHz eliminates a bit of "shoutiness" in the mids which I was not overly sensitive to, but which I can clearly recognize now that I hear what the speakers sound like when this excess is absent. Yes, this makes an improvement to speakers whose sound is already ravishing.

There is a general further clarifying of the sound without adding any high frequency edge or glare. At least some of this undoubtedly results from the elimination of midbass peaks which are notorious for covering up or obscuring clarity and detail higher up in the spectrum. This is one reason many reject the full, naturally rich sound of speakers like Harbeths; they want that added detail and value speakers with "tighter" bass in order to get that clarity. But unlike tight-bass speakers, with the X4 the presentation can still be as warm as you care to make it, just with smoother midbass which doesn't obscure clarity and detail further up the frequency range nearly as much as does a midbass full of peaks.

Bottom Line

Yes, the DSPeaker X4 is an expensive equalizer. But, as is apparent from my above descriptions of other equalizers I've owned, I've paid more for less in terms of sonic transparency and convenience.

It may well be a much higher value if you plan to take advantage of one or more of its other functions, which include the ability to act as a full-function analog and digital preamp, a digital crossover for one or two subs and your main speakers, a DAC, and a headphone amp with somewhat sophisticated DSP-powered head-related-transfer function settings for headphone listening.

Are these changes worth the asking price? Well, if I didn't know the X4 existed, I would just have lived with the truly excellent sound I was getting without any equalization from the Harbeth M40.2s in my set up. Or, I may eventually have tried inserting my Z-Systems rdq1 and played with getting the equalization just so—a many hours over many days process, as I know from prior experience.

But in terms of subjective results just as an equalizer for a pair of stereo speakers, the X4 both sounds better and is easier to adjust for best sound than any other equalizer I have ever used. I hear absolutely no sonic downsides related to its insertion as an equalizer in the digital signal path and I hear many sonic advantages to so using it.

As you can see from the graphs, there wasn't much to correct with my Harbeth M40s in this set up, particularly for one who, like me, likes a rich, purring midbass and isn't very sensitive to the slight "shoutiness" around 1 kHz. But, the changes wrought in a total of about 75 minutes of set up and measurement are all to the good, beyond anything I've ever achieved with analog or digital equalizers before. Gilding the lily, yes. But life is too short to worry about the price when the result is this beautiful, this easily achievable, and still relatively affordable to me. Yes, for me, the Holy Grail of electronic equalization has truly arrived!
 
Last edited:
Likes: DonH50

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. A place where audiophiles and audio companies discuss existing and new audio products, music servers, music streamers and computer audio, digital to audio convertors (DACS), turntables, phono stages, cartridges, reel to reel, speakers, headphones, tube amplifiers and solid state amplification. Founded in 2010 What's Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals we enjoy learning about new things and meeting new people and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing