Rogue Audio RP-7 line preamplifier

Stereophile 02/2019

Herb Reichert

You know I'm a lucky guy. I maintain two separate audio reviewing systems.
The core component of my beloved, daily-driver desktop system is a Mytek Brooklyn DAC-preamp-headphone amp. Through this system I play high-resolution files and Internet sources (Tidal, Qobuz, Netflix, and YouTube). One of the Brooklyn's two line-level inputs delivers NPR news and baseball from my Kenwood KT-990D FM/AM tuner. I mostly use this system with headphones, but currently, the Brooklyn's line-out feeds a pair of Bel Canto Design's compact e.One REF600M monoblocks driving the shelf-mounted Dynaudio Excite X14 speakers I use to play movies and videos.

My bulkier, more elaborate floor system employs a modified Home Depot rack as a totem and a variety of moderately priced cables as fetish objects. This system is not anchored by a newfangled, multipurpose device like the Mytek Brooklyn; instead, its core component is an old-school, line-level preamplifier I rely on for selecting sources, adjusting volume, and, most important, setting the entire system's gain, tone, and temper.

The sound quality of every source component is affected and established by the sound quality of this single active device. Likewise, the sound character of every amp and speaker I connect to it. That is a huge audio system responsibility—so huge that most audiophiles, especially those whose only source component is a computer, are delighted to abandon separate preamps.

In my floor system I prefer to experience the texture and vivid intimacy of tubed gear—such as Rogue's RH-5 headphone amp/preamp driving their own Stereo 100 amplifier (100Wpc, KT120 tubes); or PrimaLuna's ProLogue Premium preamp driving their own ProLogue Premium amp (35Wpc, EL34 tubes). Occasionally I go all solid-state, with Pass Laboratories' HPA-1 preamp and headphone amp, and either Pass's XA25 amp or the Bel Canto e.One REF600Ms.

But what I like best of all is a tubed preamp driving a solid-state power amp. Lately, I've spent a lot of time with a PrimaLuna preamp feeding the solid-state Pass XA25. The PrimaLuna's tubes add a touch of feminine flush and glow, generating deeper reproductions of the sounds of recording venues, more distinct voices, and richer timbres—while the supertransparent Pass XA25 grips the speakers more firmly than the ProLogue Premium amp. This unusual hybrid system of tubed and solid-state separates gives Harbeth's Monitor 30.2 speakers satisfying balances of yin/yang, masculine/feminine, disegno/colorito.

Unfortunately, when I use the PrimaLuna ProLogue Premium preamp with my newly beloved First Watt SIT-3 amplifier, the sound becomes a little too flushed with second-harmonic distortion, resulting in a tiny amount of dark murkiness that's most noticeable in orchestral climaxes.

This slight dullness is caused not only by the SIT-3's unusually high (for a solid-state amp) component of second-harmonic distortion (it has no feedback), but also by its unusually low gain of 11.5dB (most power amps have at least 25dB).

The PrimaLuna has only 12dB of gain, which means that it and the SIT-3 provide a total gain of only 23.5dB—which leaves me about 10dB shy of a well-managed, sparkly-dynamic condition. If I use low-sensitivity speakers—eg, Harbeth P3ESRs or Magnepan .7s—this problem is exacerbated.

That was all before the Rogue RP-7 preamplifier ($4995) entered my system. In single-ended mode, the RP-7 sports 14dB of gain—only slightly more than the PrimaLuna ProLogue Premium. But unlike the PrimaLuna, the RP-7 is also a fully balanced design, in which mode it delivers 20dB of gain. That's great—but the SIT-3 has only unbalanced input jacks (RCA). However, because the RP-7 is fully balanced, I could run the HoloAudio Spring "Kitsuné Tuned Edition" Level 3 DAC (which has no volume control) into one of the RP-7's balanced inputs and pick up another 3dB of gain. Which worked quite nicely.

Best of all, the RP-7 sounded drier than the PrimaLuna. It didn't flood the First Watt with second harmonic. Like Rogue's Stereo 100 power amp, the RP-7 barely sounded as if it was running on tubes.


One of my favorite things about the Rogue RP-7 is that I can easily read its azure-tinted OLED display from 9' away. Rogue's RH-5 preamp and headphone amplifier, which I reviewed in November 2017, used a less-geezer-friendly VFD display.

The RH-5 measures 15" wide by 4" high by 13.5" deep and weighs 19 lb. The RP-7 is slightly wider, higher, and deeper—18.5" by 4.5" by 17"—and heavier, at 30 lb. It looks and feels sturdy and substantial, and, as with Rogue's Stereo 100 amplifier, the style of its steel case and thick aluminum faceplate is unaffected and timeless.

The RP-7 uses four 12AU7 dual-triode tubes and has 17 (!) separate power supplies—including an individual regulated filament supply for each tube. According to Rogue's founder and chief engineer, Mark O'Brien, the RP-7 uses military-spec circuit boards with heavy copper traces, populated with oil-filled Mundorf capacitors, Vishay resistors, and Vishay HEXFRED diodes. All of which are genuinely expensive, audiophile-certified bits.

Although bigger and wider than the RH-5's, the RP-7's front panel has the same number of buttons and knobs. The big knob at far left is the Balance control, which, late at night in the dark, I always mistake for the Volume control—which, counterintuitively, is the big knob at far right. The volume and balance knobs are stepped in increments of 0.5dB, but on the remote, balance operates in 1dB increments down to –15dB, then jumps to –90dB.

This permits fine adjustment where it's most typically needed, yet facilitates the extinguishing of the channel without going through all the intermediate steps. When the Balance control is used, the OLED display shows the level of each channel separately.

Each of the nine small, round buttons strung in a row across the bottom of the front panel is topped by a little blue status LED, to make it easy to recognize and assess from afar. Close up, their tiny labels are almost legible—from left to right, Display, Proc Loop, Line 1, Line 2, Line 3, Bal 1, Bal 2, Mono (I love Mono buttons!), Power—but it doesn't matter: The display tells you, in large characters, which input is selected.

I almost never touched the RP-7's front-panel buttons—they're too high-quality and positive in their actions. I prefer the remote's cheesy plastic "lazy buttons." Unfortunately, if you want to extinguish the RP-7's bright display, you have to get up and directly push the Display button.

I'd have been happier had this important feature been included on the remote. Likewise, though there's no Mute button on the front panel, the remote has one: when I hit it, a little red LED just left of the Volume knob lit up.

The chief difference between a headphone amp that includes a preamp and a preamp that includes a headphone amp is easy to spot. A proper headphone amp, such as Rogue's excellent RH-5, has separate, three-pin left and right XLR jacks, plus a four-pin XLR jack, for balanced headphone listening, and two ¼" (6.35mm) jacks for single-ended headphones.

The RP-7 has only a single ¼" phone jack, and no balanced options. The owner's manual explains: "For performance reasons the headphone jack does not include a switch to break the audio signal. To mute the output from the preamp either mute the preamp or turn off your power amp."

At the left of its rear panel, the RH-5 had an optional RCA phono input. In contrast, the RP-7 has line-level inputs only. At the left of its rear panel are eight black XLR jacks: two pairs of balanced (MOSFET-buffered) outputs and two pairs of balanced inputs. These dark XLRs are followed by a regiment of bright gold RCA jacks: two unbalanced outputs, a unity gain input, a processor loop, a fixed output, and three line-level inputs.

Happily, this arrangement gives me up to five line-level sources. Very nice. Stacked at far right are the main power rocker, a fuse bay, and a three-pronged IEC power-cord inlet.
The RP-7 includes a 30-second soft start with auto mute, to go easy on the tubes and suppress turn-on transients.


Listening with the First Watt amps

The HoloAudio Spring DAC driving the Rogue RP-7 driving the First Watt SIT-3 power amp driving the 93dB-sensitive DeVore Fidelity Orangutan O/93 speakers created a superbly balanced and invigorating—nay, intoxicating—system that didn't sound like tubes or solid-state. But it did reproduce, with extraordinary weight and saturated tones, my latest favorite piano album: Alexander Melnikov performing Debussy's Préludes, Book 2, then joined by Olga Pashchenko in Debussy's own four-hands arrangement of La Mer, all performed on an 1885 Érard piano like the one Debussy used when composing the Préludes (24-bit/96kHz FLAC, Harmonia Mundi/Qobuz). This exotic album is all about the sound of a vintage piano. Its purpose is to show the listener how richly toned and expressive these old, less tightly strung instruments can sound. The playing of Melnikov and Pashchenko isn't lacking in feeling or poetic insight, but the intent of this recording is to present what Debussy must have heard while working on these compositions. Through the RP-7, note attacks were arousing. Note decays were extended. Pedaling was psychedelic. Tone was full-spectrum and saturated.

After my experiences with the First Watt SIT-3, which has only single-ended inputs (RCA), I felt a need to try the RP-7's balanced XLR outputs. I connected another First Watt amp, the J2 (25Wpc into 8 ohms, $4000), which accepts balanced inputs and is a pure made-in-heaven match for Harbeth's Monitor 30.2 speakers ($6495/pair).

When John Atkinson measured the M30.2s for my review of them in the April 2018 issue, he estimated their sensitivity at 87.3dB(B)/2.83V/m—which, he said, "is usefully higher than the specified 85dB." The M30.2's impedance remains above 6 ohms throughout the audioband. As a result, the Harbeths sound rich, lively, and articulate with the First Watt J2. Most important, the Rogue RP-7 kept a cool, steady hand on the J2. This combo of preamp, amp, and speakers made beautiful sounds: saturated tone, never too hard or too soft, never too warm, never stressed or compressed. My only dilemma was which music I should listen to next. Pleasurable.

Now the sound of Melnikov and Pashchenko playing Debussy reminded me of aged French brandy: It had a sensuous, mood-enhancing flavor that lingered, inspiring reverie. Suave and sophisticated.

The bass frequencies on Pat Metheny and Charlie Haden's Beyond the Missouri Sky (Short Stories) (24/96 FLAC, Verve/Qobuz) sounded slightly digital-artificial, but the RP-7 plus J2 made the Harbeth M30.2s sound deeper and fuller in the lower octaves than I'd thought possible. Haden's bass work was just-right tight, and his instrument was almost full height. The added reverb surrounding Metheny's guitar fashioned a dreamy, intoxicating mix with pulsing illusions of a vast dark space. This combination of gear and recording generated the kind of pure beauty of sound that, for me, makes being an audiophile a rewarding enterprise.

In fact, this combo of tubed preamp and solid-state amp sounded so enjoyably sophisticated that I'm compelled to include one more supreme listening experience. Anyone who thinks digital is more truthful or lifelike than analog has never owned a Koetsu Rosewood Signature Platinum moving-coil cartridge, and has definitely never experienced the vivid analog reality of the Purcell Consort, under the direction of Grayston Burgess, in Now Make We Merthe: Medieval Carols (LP, Argo ZRG 526). I know this 1967 LP intimately; it showcases stunning three-dimensional sound in the service of great but humble human art and seasonal merriment. I've heard it through countless systems, some with preamps costing six times the RP-7's price, but none has put me in closer, more intimate touch with these ancient Christmas songs sung by the Boys of All Saints, Margaret Street. With the Tavish Design Adagio phono preamp added to the system, I played this record on a wet, cold December night, my apartment lit only by beeswax candles. I wish you could have been there to sense the purity of space and tone. This delightful mix of Rogue tubes, Koetsu magic, and First Watt silicon put my mind near a Christmas choir singing in a stone church. I know this system is not inexpensive, but to my taste, it played up there with the very best I've heard.

Listening with the Bel Canto Design monoblocks

Connecting the RP-7 to Bel Canto's e.One REF600Ms showed me how low in so-called tube coloration this Rogue actually was. The REF600M's input impedance is 200k ohms in balanced mode, 100k ohms in single-ended. Therefore, it mated comfortably with the RP-7's MOSFET-buffered output impedance of 50 ohms, balanced. When I used the RP-7 and REF600Ms to play the Debussy piano recording, I was surprised at how dark and nontransparent it sounded—not thick or slow, just slightly dark-gray opaque. I doubt this was caused by the Rogue, which all along had sounded cool-air crisp and water-clear with every other amp in the house. I was expecting the Bel Cantos to produce more light and air, more of a sense of openness—but I got less. To its credit and my pleasure, the Rogue–Bel Canto pairing did an excellent job of reproducing the weight and impact of acoustic pianos and pipe organs.

Listening with the Rogue Stereo 100

But Rogue's best preamp didn't need to pass those other tests. It needed only to do a perfect job of driving Rogue's own 100Wpc power amplifier, the Stereo 100. Which it did.The first song I played, via the Harbeth M30.2s, was "I'm So Glad," from Skip James's well-recorded (ca 1964) comeback album, Blues from the Delta (16/44.1 FLAC, Vanguard/Qobuz). It sounded so good that it jumped out and smacked me down, and made me feel a little bad for praising those other amps as much as I had. That first track with the Stereo 100 was an instant wow moment. This preamp-amp-speaker ensemble sounded so startlingly present, so absolutely clear, so completely undistorted, that I sat there laughing and shaking my head. What's that corny old audiophile line? "Veils were lifted!" In this case, it felt like all the veils were gone. James's ethereal voice, his guitar and piano, were laid bare, blatantly there.

You say you want a revolution
Well, you know
We all want to change the world

Next I played The Beatles (aka the White Album), and remembered what had led me to this Rogue combo in the first place. While reviewing Harbeth's Monitor 30.2, I'd realized that this medium-size stand-mounted model would make a perfect, neutrally balanced reference speaker. But I needed just the right amp—one with more horsepower than my Line Magnetic, PrimaLuna, First Watt, or Pass Labs could provide. Now, while listening through the RP-7 to Giles Martin's remix of The Beatles in this 50th Anniversary Edition (2 LPs, Apple B0028831-01), I felt completely validated in my decision to choose Rogue's Stereo 100. The sound was transparent, insightful, and extraordinarily alive. I had almost no criticisms of it.

Take this brother
may it serve you well

Listening through headphones If you came of age with the Beatles, as I did, I humbly suggest that you listen to "Revolution 9," from The Beatles, from start to finish, through audiophile-quality headphones—say, Sony's overachieving MDR-Z1Rs, Audeze's lively LCD-Xes, HiFiMan's state-of-the-art, planar-magnetic Susvaras, or JPS Labs' revelatory Abyss AB-1266 Phis. Any of those revealing transducers will show you everything your big speakers don't.

The RP-7's headphone amp and Giles Martin's remix showed me gobs of what I'd never known this incomparable classic contained—but the Rogue's headphone output also tainted "Revolution 9" with rolled-off high frequencies that made it sound dull, and with more noise than I find acceptable. The noise seemed like a 120Hz ground-plane problem. It was worst with the 103dB-sensitive LCD-Xes, least noticeable with the 83dB-insensitive HiFiMan Susvaras.


Rogue Audio's RP-7 line preamplifier is definitely a Herb Product—how could it not be? It's built like a farm tractor, and looks unpretentious and timeless. It sounded remarkably uncolored—to its great credit, the RP-7 walked the narrow path between tube sound and transistor sound. It was dynamic. It delivered exceptional transparency along with extraordinary weight and body. It played with a relaxed, organic tone enhanced by only the slightest touch of burnished tube glow. Alert and water clear are the most apt descriptors. My new reference.


Dane techniczne

Description: Line-level hybrid (tubed/MOSFET) preamplifier and headphone amplifier. Tube complement: four 12AU7/ECC82. Frequency response: 1Hz–100kHz, ±1dB. Line-level inputs: 3 unbalanced (RCA), 2 balanced (XLR). Variable outputs: 2 unbalanced (RCA), 2 balanced (XLR). Maximum gain, line stage: 14dB unbalanced, 20dB balanced. Output impedance: 50 ohms (1kHz).

Dimensions: 18.5" (470mm) W by 4.5" (115mm) H by 17" (430mm) D. Weight: 30 lb (13.6kg) net, 45 lb (20.4kg) shipping.

System odniesienia

Analog Sources: Dr. Feickert Analogue Blackbird turntable, Jelco 850L tonearm; AMG Giro G9 turntable & 9W Turbo tonearm; AMG Teatro, EMT TSD 75, Etsuro Urushi Cobalt Blue, Hana SL, Koetsu Rosewood Signature Platinum, My Sonic Lab Ultra Eminent Ex moving-coil cartridges; Kenwood KT-990D FM/AM tuner.
Digital Sources: Mac mini computer running Audirvana Plus 3.2; HoloAudio Spring "Kitsuné Tuned Edition" Level 3, Mytek Brooklyn DACs; JVC X-Z1010 CD player (used as transport).
Preamplification: Tavish Design Adagio phono preamplifier; Pass Laboratories HPA-1, PrimaLuna ProLogue Premium line-level preamplifiers.
Power Amplifiers: Bel Canto Design e.One REF600M (monoblocks), First Watt J2 & SIT3, Pass Laboratories XA25, PrimaLuna ProLogue Premium, Rogue Audio Stereo 100.
Loudspeakers: DeVore Fidelity Orangutan O/93, Dynaudio Excite X14, Harbeth Monitor 30.2 & P3ESR.
Headphones: Sony MDR-Z1R, Audeze LCD-X, HiFiMan Susvara, JPS Labs Abyss AB-1266 Phi.
Cables: Digital: AudioQuest Cinnamon (USB), Kimber Kable D60 Data Flex Studio (coax). Interconnect: AudioQuest Cinnamon, Auditorium 23, Triode Wire Labs Spirit, Wireworld Silver Eclipse 7. Speaker: Auditorium 23. AC: AudioQuest Tornado, Triode Wire Labs American–Digital & Seven-Plus.
Accessories: AudioQuest Niagara 1000 power conditioner; Harmonic Resolution Systems M3X-1719-AMG GR LF isolation platform; Sound Anchor speaker stands; Acoustical Systems SmarTractor, Dr. Feickert Analogue cartridge-alignment protractors; Musical Surroundings Fozgometer azimuth meter.


John Atkinson

I measured the Rogue RP-7 using my Audio Precision SYS2722 system (see the January 2008 "As We See It"). The RP-7's volume control operates in 0.5dB steps, with the maximum level indicated as "192" on the front-panel display. All of its outputs—balanced, single-ended, and headphone—preserved absolute polarity (ie, were non-inverting). The balanced input impedance measured a very high 174k ohms at 20Hz and 1kHz, dropping very slightly to 162k ohms at 20kHz; the unbalanced input impedance was lower, at 24.5k ohms at 20Hz and 1kHz, and 21k ohms at 20kHz. The RP-7's maximum voltage gain into 100k ohms was 15.33dB for both balanced input to balanced output and unbalanced input to unbalanced output. Measured at the headphone output, the maximum gain was a very high 30.2dB into 100k ohms, which is probably why HR noted some noise issues with headphones.

The Rogue's output impedance is specified as 50 ohms. However, I found that this varied with both output and frequency. Measured at the headphone output, the impedance was 1.6 ohms at 20Hz, 0.8 ohm at 1kHz, and 1.2 ohms at 20kHz. The unbalanced output measured a very low 0.8 ohm at 20kHz and 3.5 ohms at 1kHz, but rose to 2.44k ohms at 20Hz, this presumably due to a DC-blocking capacitor in series with the output. The respective balanced line-output impedances were 1.4 ohms, 12.4 ohms, and 5.64k ohms. As a result of the increased output impedance at low frequencies, the RP-7's response drastically rolled off below 400Hz into 600 ohms, reaching –3dB at 200Hz (fig.1, cyan and magenta traces). The RP-7 should not be used with power amplifiers having an input impedance much below 30k ohms. Into 100k ohms, however, the RP-7 offers a flat, wide-bandwidth response with an upper –3dB frequency of 170kHz (fig.1, blue and red traces). Fig.1 was taken with balanced operation and the volume control set to "192." Commendably, the response didn't change with the setting of the volume control or with unbalanced operation. The frequency response at the headphone output was also flat and extended, though the –3dB point dropped to a still-high 120kHz.

RP7 rys1

Fig.1 Rogue RP-7, balanced output, volume control at max, frequency response at 1V into: 100k ohms (left channel blue, right red), 600 ohms (left cyan, right magenta) (1dB/vertical div.).

Channel separation was similar for balanced and unbalanced operation, and somewhat asymmetrical, at 80dB L–R at 2kHz and 97dB R–L, these respectively decreasing to 60 and 78dB at the top of the audioband. The unweighted, wideband signal/noise ratio, measured at the line outputs with the inputs shorted to ground but the volume control set to its maximum, was only moderate, at 68.3dB ref. 1V output. This improved to 76.1dB, when the measurement bandwidth was restricted to 22Hz–22kHz, and to 80dB with an A-weighting filter switched into circuit. Fig.2 shows spectral analyses of the Rogue's low-frequency noise floor while its balanced output drove a 1kHz tone at 1V into 100k ohms. The random noise-floor components lie at –100dB, with the primary noise contribution from power-supply–related spuriae.

219RP7 fig2

Fig.2 Rogue RP-7, balanced output, volume control at max, spectrum of 1kHz sinewave, DC–1kHz, at 1V into 100k ohms (linear frequency scale) (10dB/vertical div.).

Rogue specifies the RP-7's maximum output voltage as 18V balanced and 9V unbalanced. Figs. 3 and 4 respectively show how the percentages of THD+noise in the balanced and unbalanced outputs vary with voltage. With clipping defined as when the THD+N reaches 1%, the RP-7 clips at 24V balanced and 12V unbalanced. These very high clipping voltages were maintained into the punishing 600 ohm load. The fact that the traces in these two graphs slope downward from left to right below 1–2V means that the distortion actually lies below the noise floor at these lower output levels. When I measured the percentage of THD+N against frequency at 1V into 100k ohms, it didn't change with frequency (not shown).

219RP7 fig3

Fig.3 Rogue RP-7, balanced output, distortion (%) vs 1kHz output voltage into 100k ohms.

219RP7 fig4

Fig.4 Rogue RP-7, unbalanced output, distortion (%) vs 1kHz output voltage into 100k ohms.


 As with Rogue's RH-5 preamplifier and headphone amplifier, which Herb Reichert reviewed in "November 2017, the distortion is predominantly the subjectively benign second harmonic at a low –76dB (0.015%, fig.5). This graph was taken from the single-ended output; with the balanced output, the second harmonic lay at –80dB (0.01%). Fig.5 was taken with a low-frequency fundamental; when I repeated the test with a 1kHz signal, while the second harmonic was still the highest in level, there were now some higher-frequency enharmonic tones present (fig.6). Fig.6 was taken with the volume control set to "192." Reducing the setting to –20dB ("152") and increasing the input level by 20dB dropped the levels of these spurious tones by the same 20dB (fig.7), which suggests that the circuitry is picking up these tones ahead of the volume control. It turned out that this problem was fixed with v2.2 of the RP-7's firmware. Repeating the spectral analysis with the new firmware eliminated the spurious tones (fig.8).

219RP7 fig5

Fig.5 Rogue RP-7, unbalanced output, volume control at max, spectrum of 50Hz sinewave, DC–1kHz, at 2V into 100k ohms (linear frequency scale).

219RP7 fig6

Fig.6 Rogue RP-7, unbalanced output, volume control at max, spectrum of 1kHz sinewave, DC–1kHz, at 2V into 100k ohms (linear frequency scale).

219RP7 fig7

Fig.7 Rogue RP-7, unbalanced output, v2.1 firmware, volume control at –20dB, spectrum of 1kHz sinewave, DC–1kHz, at 2V into 100k ohms (linear frequency scale).

219RP7 fig8

Fig.8 Rogue RP-7, unbalanced output, v2.2 firmware, volume control at –20dB, spectrum of 1kHz sinewave, DC–1kHz, at 2V into 100k ohms (linear frequency scale).

Keeping the volume control at "152," I looked at the spectrum of the RP-7's single-ended output while it reproduced an equal mix of 19 and 2kHz tones at 1V into 100k ohms, again with the new firmware (fig.9). The high-order intermodulation products are very low in level and the second-order difference product at 1kHz lies at –70dB (0.03%). The picture was similar from the balanced output, though the difference product dropped to –66dB (0.05%).
219RP7 fig9

Fig.9 Rogue RP-7, unbalanced output, v2.2 firmware, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 1V peak into 100k ohms (linear frequency scale).

As long as it's driving a power amplifier with a high input impedance, Rogue's RP-7 offers generally respectable measured performance. Although HR auditioned the RP-7 with the earlier v2.1 firmware, he used sufficiently low volume control settings that the spurious high-frequency tones would not have had audible consequences.