GoldenEar Technology Triton One.R loudspeaker Measurements

Sidebar 3: Measurements

I used DRA Labs' MLSSA system and a calibrated DPA 4006 microphone to measure the GoldenEar Triton One.R's frequency response in the farfield, and an Earthworks QTC-40 mike for the nearfield responses. I removed the loudspeaker's front and side grilles for the nearfield measurements but left them in place for the farfield measurements (footnote 1).

The Triton One.R measured very similarly to the original Triton One. As with the original, GoldenEar specifies the One.R's sensitivity as 92dB/W/m. My estimate was 91.2dB(B)/2.83V/m, which is both within experimental error of the specification and identical to the earlier speaker's sensitivity. The One.R's impedance is specified as being "compatible with 8 ohms." The solid trace in fig.1 reveals that the impedance magnitude ranges between 3 and 6 ohms for much of the audioband, with a minimum value of 3.1 ohms between 290Hz and 390Hz. Like the original One, the One.R's use of a passive high-pass filter with a low corner frequency means that the electrical phase angle becomes increasingly capacitive below that frequency. Although the impedance magnitude rapidly increases below 100Hz, mitigating the effect of that phase angle, there is still a combination of 4 ohms and –48° at 100Hz, which will require a good 4 ohm–rated amplifier to drive the speaker to acceptably high levels.


Fig.1 GoldenEar Triton One.R, electrical impedance (solid) and phase (dashed) (2 ohms/vertical div.).

The impedance traces are free from small discontinuities that would imply the presence of panel resonances. When I investigated the cabinet's vibrational behavior with a plastic-tape accelerometer, the only resonant modes I could find were at 452Hz on the sidewalls (fig.2) and another at 609Hz on the back panel. These modes are probably too low in level and too high in frequency to affect sound quality.


Fig.2 GoldenEar Triton One.R, cumulative spectral-decay plot calculated from output of accelerometer fastened to center of sidewall level with lower midrange unit (MLS driving voltage to speaker, 7.55V; measurement bandwidth, 2kHz).

Like the earlier GoldenEar speaker, the response of the One.R's midrange units extends almost to 100Hz, with a steep rolloff below that frequency (fig.3, blue trace). The powered woofers' nearfield response (fig.3, green trace) rolls off sharply above 80Hz, without any peaks evident in their upper-frequency output. The woofers have their minimum-motion notch at 27Hz, with then a steep low-frequency rollout. Because of their large surface area and correspondingly small motion, I measured the passive radiators' nearfield response with an accelerometer rather than a microphone in order to minimize acoustic crosstalk from the woofers. Their output (red trace) peaks between 20Hz and 30Hz, implying excellent low-frequency extension, and like that of the woofers, their low-frequency rollout is also very steep, close to 6th-order, 36dB/octave.


Fig.3 GoldenEar Triton One.R, anechoic response on tweeter axis at 50", averaged across 30° horizontal window and corrected for microphone response, with the nearfield responses of the midrange units (blue), woofers (green), and passive radiators (red) respectively plotted below 300Hz, 450Hz, and 125Hz.

Higher in frequency in fig.3, the response of the midrange units and tweeter on the tweeter axis is impressively flat, though like the original Triton One, the top octave is elevated. Looking at the One.R's lateral-dispersion plot (fig.4), which is normalized to the tweeter-axis response, it appears that the speaker's output declines rapidly to its sides above 10kHz. This will tend to produce a flat top-octave balance in small to medium-size rooms. The contour lines in this graph are evenly spaced in the midrange and low treble, which implies stable stereo imaging. In the vertical plane (fig.5), the GoldenEar's on-axis balance is maintained over a relatively wide listening window, which is a good thing when you consider that the tweeter is a high 40.5" from the floor. A suckout does starts to develop in the upper crossover region 15° above and below the tweeter axis, but these are unrealistic listening axes.


Fig.4 GoldenEar Triton One.R, lateral response family at 50", normalized to response on tweeter axis, from back to front: differences in response 90–5° off axis, reference response, differences in response 5–90° off axis.


Fig.5 GoldenEar Triton One.R, vertical response family at 50", normalized to response on tweeter axis, from back to front: differences in response 15–5° above axis, reference response, differences in response 5–15° below axis.

Looking at the Triton One.R's behavior in the time domain, the loudspeaker's step response on the tweeter axis (fig.6) indicates that the tweeter and midrange units are connected in positive acoustic polarity, the woofers in negative polarity. (I confirmed that this was the case by looking at the individual outputs.) More importantly, the decay of the tweeter's step blends smoothly with the start of the midrange unit's step and the decay of that driver's step blends smoothly with the start of the woofers' step. The bulky grille and grille frame are responsible for a strong reflection of the tweeter's output 500µs after the initial arrival. This reflection results in ripples above 2kHz in the on-axis response and some low-level, top-octave hash in the GoldenEar's cumulative spectral-decay plot (fig.7). This plot is very clean at lower frequencies, however.


Fig.6 GoldenEar Triton One.R, step response on tweeter axis at 50" (5ms time window, 30kHz bandwidth).


Fig.7 GoldenEar Triton One.R, cumulative spectral-decay plot on tweeter axis at 50" (0.15ms risetime).

Like its predecessor, GoldenEar's Triton One.R offers excellent measured performance.—John Atkinson

Footnote 1: Despite what Kal wrote, the grilles can be removed—with difficulty—although removing them may invalidate the warranty.—Ed.
GoldenEar Technology
PO 141
Stevenson, MD 21153
(410) 998-9134

Bogolu Haranath's picture

Speakers to beat for $7k :-) .........

Bogolu Haranath's picture

My be a better value for the money compared to many passive bookshelf/stand-mount speakers and some floor-standers selling between $5k and $10k :-) ........

JRT's picture

The price difference could offset some of the cost of a better performing separate low frequency subsystem.

Bogolu Haranath's picture

Agreed ......... Several bookshelf/stand-mount speakers and some floor-standers, augmented by powered subwoofer(s), under $5k, could be a better value :-) ........

Bogolu Haranath's picture

Heaven forbid ...... KEF LS-50s, augmented by a couple of powered subwoofers, under $5k, could be a better value :-) ........

JRT's picture

The KEF LS-50 small satellite monitor is just a tweeter and coaxial midrange which runs out of breath in the upper bass (in lower midrange at higher SPL), and would be better behaved with the addition of a couple of well behaved midwoofers (eg. pairs of Peerless HDS P830869 eight inch Nomex, $59.44/each in quantities of four or more at Parts-Express) underneath for better performance in low triple digit and high double digit frequencies, above the crossover to the separate low frequency subsystem.

Bogolu Haranath's picture

KEF LS-50s with just a couple of powered subwoofers with a crossover point set at 150 Hz, may work just fine, in a small to mid-size listening room ........ See Stereophile measurements, including JA1's in-room frequency response measurements taken from listening position :-) ........

dc_bruce's picture

I'm running a pair of LS-50s with a REL Q400e subwoofer. In a modest-sized room, this combination delivers good results with all kinds of music at satisfying, if not bone crushing loudness. The little KEF is quite strong to 50 hz., and the REL's have a gentle slope low-pass filter. I'm quite sure that the speaker under review will generate higher undistorted SPLs. The question is: do you need that capability in your room? I tuned my setup with a real time analyzer to get the smoothest transition and best bass. The room adds about a 3dB boost at around 40 Hz.

Bogolu Haranath's picture

dc_bruce, could you repost your comments about using LS-50s with subwoofer? ...... I got your reply, but that reply somehow didn't show-up in this comments section ...... Your comments could be useful for answering some of the questions JRT (see below) and others may have, about using LS-50s with subwoofers ...... Thanks in advance :-) .....

JRT's picture

There are some exceptions, but most subs are not well behaved at higher frequencies, so must be rolled off more steeply if the low pass corner is pushed a little higher. That increases rate of change of phase with respect to frequency, and so too group delay (group delay is defined as being the negative derivative of phase with respect to frequency, the negative slope of the curve, the negative rate of change of phase with respect to frequency).

Using a pair of medium size woofers better behaved in their passband and in nearby stop band allows a somewhat higher crossover frequency to the smaller midwoofer, with more shallow slopes in the blend region, allowing the woofers to contribute more of the volume-velocity in the low side of the baffle step, improving blend and reducing interference from floor reflection in that range.

For a driver operating on infinite baffle, in pistonic range, at constant SPL with respect to frequency (flat frequency response), there is an inverse square relationship between swept volume and frequency, 100 times the excursion at 1/10 frequency. Crossing to a pair of woofers that quadruple the radiating area of the pistonic diaphragms reduces excursion requirement similarly for the needed swept volume, for the needed volume-velocity, for the needed SPL.

Bogolu Haranath's picture

dc_bruce (see above) is using REL subwoofers with the LS-50s ....... He says, he is happy with the results ....... I got his reply but, somehow his reply didn't show up in this comments section ...... I asked him to repost his comments ........ His comments may answer some of the questions you may have :-) .......

dc_bruce's picture
Bogolu Haranath's picture

Regarding group delay ......... See, 'Precedence effect' (Haas effect) in Wikipedia :-) ........

JRT's picture

Precedence effect is largely about signal doublets, with leading half of the doublet taking precedence in perception, depending much on time interval spacing between doublets. An example of a doublet would be direct radiation in combination with a delayed arrival reflection.

Group delay is about one portion of frequency spectrum being delayed relative to another portion of spectrum, for example a low frequency fundamental and low order harmonics delayed relative to higher order harmonics/overtones.

Bogolu Haranath's picture

Not exactly ........ Did you read the articles in Wikipedia? ........ Precedence effect also applies to multi-way loudspeakers, where the frequency is divided between multiple drivers ........ Also see JA1's time domain response measurements :-) ........

Bogolu Haranath's picture

Precedence effect also helps to create a good, stable and phantom center image, with just two speakers, if both speakers are properly positioned :-) .......

Bogolu Haranath's picture

Also see, 'group delay and phase delay' in Wikipedia :-) ........

Bogolu Haranath's picture

According to the Wikipedia article 'group delay and phase delay', ear is most sensitive to group delay from 1 kHz to 4 kHz :-) .........

Bogolu Haranath's picture

The notorious BBC dip is located from 1 kHz to 4 kHz :-) ........

jimtavegia's picture

The one speaker my wife says she loved and misses were the Polk Audio Model 10's and I do as well. They were the first passive radiator model we owned. It is amazing how many design differences there are in the speaker world and many of them work to provide excellent performance.

Ortofan's picture

... a speaker with a passive radiator.
The KEF Q550, Q750 and Q950 models use a pair of passive radiators.
Also, REL uses passive radiators in some of their subwoofers.

JoeinNC's picture

Oh, man. Thanks for reminding me of Polk 10's. I never owned a pair myself, but a friend of a friend did. I didn't like the dude at all, but even so, I had to give him credit for having a great sounding system and a pretty awesome record collection.

ednazarko's picture

I've enjoyed a couple of different speakers with passive radiators. I've got a pair of Aon 3 stand mounts (as my rear channels in my AV system, and another set as my printing/framing room system), and the thing that I really love is how radiating that bass energy makes the sound more enveloping. If you look at the history of Sandy's speaker life, he did a lot of speakers with passive radiators, along with a number of dipole speakers, which if placed correctly give the same effect.

I have another system I listen to a lot (in my photo studio and editing area) that achieves the same effect with a bit different design - Gradient Revolutions. The woofer is open air, which gets that wrap around feel, and with cardioid mid and tweeters. Less sensitive to placement (unless you aren't using a sub and really need the room reinforcement for better bass) than any other speakers I've ever owned.

I've had a number of other speakers come and go, and the more traditional designs all sounded flat to me. Less engaging, shallower sound field.Not sure what I'm going to do if my Gradients die...

Anton's picture

Does Herb like these?

helomech's picture

...every time I've auditioned these and the original Triton 1s, their peaky treble was incredibly fatiguing. The accolades for these models is somewhat baffling, no pun intended. I have to presume they're designed for the over 60 crowd.

Bogolu Haranath's picture

Did you by any chance audition the new Revel Performa F226BE floor-standers? ........ They are about the same price, $7,000/pair ....... They are a totally passive design ......... They use Beryllium dome tweeters :-) ........

helomech's picture

and found them far better balanced than the GE Tritons. I'd take a pair of the old F208s over any of the Triton series. I imagine the new Be series is even better.

Bogolu Haranath's picture

May I ask, were they using tube or SS gear with the GE Triton speakers? ........ Just curious :-) ........

JRT's picture
Bogolu_Haranath wrote:

May I ask, were they using tube or SS gear with the GE Triton speakers?

Here is the link:


Digital sources:
Oppo Digital UDP-105 universal disc player, Baetis Prodigy-X PC-based music server running JRiver Media Center v25 and Roon, exaSound e38 Mark II D/A processor. QNAP TVS-873 NAS.

Audio Research MP1, Parasound P7.

Power amplifiers:
Benchmark AHB2, Hegel C53, Classé Sigma Mono, Parasound Halo A 31.

Bogolu Haranath's picture

I know what KR used for this review ....... My question was for helomech ........ See above :-) ........

Bogolu Haranath's picture

JA1 also used SS amps with GE Triton reference speakers ........ Interestingly, JA1's in-room frequency response measurements of KEF Reference 5 rolls off above 3 Khz, while GE triton reference stays flat to 10 Khz :-) .......

Bogolu Haranath's picture

Revel Ultima Salon2 speakers also use Beryllium dome tweeters ........ JA1 and LG in-room FR measurements show mild roll-off above 7 Khz :-) ..........

Bogolu Haranath's picture

Revel Studio2s also use Beryllium dome tweeters ....... KR has reviewed them ........ KR also recently bought Studio2s for his surround sound system :-) ........

helomech's picture

I'm not sure which models but they were SS and the amp was most likely the 452.

ednazarko's picture

Used to have the original Triton Twos, in a terribly challenging listening space. 30 foot tall ceilings at one end, 20 at the other, open at one side, with a balcony/bridge across the back area. When I had the Twos I initially kept the subwoofer I'd had with my previous speakers. Then I dialed up the powered sub... and when alien spacecraft flew into a scene, the entire house vibrated.. just like you see on the screen. (I have a GoldenEar center, the SuperSat 60C, and for the rears, Aon 3.) For video, the setup was breathtaking.

Then I upgraded to the original T1 for the front. The power of the onboard powered subs was crazy, I had the settings way lower than expected. Then, we moved to a smaller place with low ceilings, and we've got the T1 sub amps set to about one third power. Alien spacecraft rattle the china, wine glasses, and windows.

But in two channel mode, that's the real test. These speakers create eerie presence on really well recorded music. When we have guests eating in the nearby dining room and are playing background music in the family room, it always happens... guests will startle, get up and walk into the family room like they're sneaking into a performance space. And then stand there, transfixed, but occasionally looking up, left and right. The comments are always "she/he materialized in the room... I was trying to figure out how that happened." There's definitely depth, but not symphony hall depth, more like recording studio depth.

As to the comment about the upper range being overdone. I can still hear the scanning in the CRT monitors used in some home stores. On a hearing test recently, the doctor congratulated me for taking good care of my ears. (I'm in my 60s, he was early 30s. I was an MC5, Cactus, Ted Nugent, Detroit Dogs etc fan. Lucky genes.) I don't find these to be peaky. I can tell what size studs are in a ride cymbal, which needs that top end to be true and right. And by comparison, I do find my LSX powered speakers to be a bit peaky (easy to adjust.)

What I love most about the GoldenEar speakers is that they create a sound field. A LOT of the auditory energy comes from those passive radiators. I performed in orchestras, jazz combos, rock bands. I crave a sound that if I get in close feels like I'm in the middle of the performance. The GoldenEar get me there. I often find myself standing very close in, between the speakers, and it feels like all those live performances.

helomech's picture

A 6db peak over the midband at ~12kHz. Sorry, but that's definitely the smoking gun and why I couldn't tolerate them for longer than five minutes. Like your doctor, I'm in my 30s and definitely can't hear as well as I did in my teens despite taking every precaution to protect my ears. It only takes one loud concert to cause permanent damage. The fact is we all lose hearing ability to some degree as we age, regardless of exposure.

After my third audition of the Tritons, I honestly began to wonder if there was something wrong with my ears. How could a speaker that's earned so much praise among the reviewers sound so intolerable? Well, at least my girlfriend felt the same way, so that was a big clue it wasn't just me. Subsequent conversations with local audiophiles, most of whom are aged somewhere between 30 and 50 confirmed my suspicion. Each and every one of those who had heard them shares my opinion. I later see JA's measurements and sure enough, there it is - a piercing treble response with a steep rise. I'd bet anyone who can still hear well (well being the key word) at >10kHz will be fatigued by these speakers when listening on-axis.

Bogolu Haranath's picture

You are right about the elevated high treble FR ....... That elevated treble starts at about 6 kHz and extends up to 20 kHz ....... There is about +6 db elevated FR at 20 kHz ........ Also, there is kinda 'saw-tooth' appearance to the FR in this region, which could be due to treble distortion ....... Hi-Fi News also reviewed the GE Triton Reference speakers with measurements ....... See their lab report and look at their graphs 1 and 2 :-) .......

Archguy's picture

People are saying these speakers are strident.

Bogolu Haranath's picture

Stereophile could consider adding Grover Neville (Inner/Fidelity, AudioStream) to the reviewing team ...... He is a 'millennial' ......... Most likely he has his hearing well preserved .......... Just a thought :-) ........

helomech's picture

What's the average reviewer age for this publication? Must be at least 55 these days.

avanti1960's picture

away from the wall is 2-feet added to your typical starting point? An important consideration for people thinking about them.
I personally marvel at the technology of these speakers and wish more manufacturers would offer powered bass (or sub bass) drivers.
My one minor complaint during a few extended sessions is that they have a definite character of their own (in the midrange / upper midrange) that is often noticeable, especially with jazz horn sections. Gives the sound a somewhat muted flatness. Overall a grand achievement that sound noticeably than the "ones".

Kal Rubinson's picture

How far away from the wall is 2-feet added to your typical starting point?

The distance from the front wall to the front baffle, normally 6', was then 8' for the GEs.

Bogolu Haranath's picture

May be KR could also review the Boenicke Audio W13 speakers? ......... They have 'wood cone' bass-mid drivers and DSP controlled powered bass drivers ....... HR favorably reported about them in his 2018 Munich show report :-) ........

Kal Rubinson's picture

Interesting but not immediately appealing because there is little acoustical (as opposed to mechanical) attention at the website. OTOH, they say they have a NYC dealer, so I will try to take look/listen.

Ortofan's picture

... Tony K in FLA about these great speakers by Sandy G in NYC.

Bogolu Haranath's picture

Correction ........ Great 'half-price' speakers :-) .........