Columns Retired Columns & Blogs |
This is the only time a butt upstages a piece of equipment. It appears Melody Gardot has put in some gym time on the stair master
Enter the full-function Soulution 727 preamplifier ($74,975; footnote 1), whose optional MC/MM phono section ($11,975) will be evaluated in a future issue (footnote 2). Because Soulution claims that the 727 "sets benchmarks in terms of noise, phase errors, common mode rejection and distortion," one would hope that there's far more than 62lb of classy casework and an easy-to-handle lightweight remote to account for its price.
The 727 is the top-level preamplifier in Soulution's three-tiered lineup of Series 7, Series 5, and "entry-level" Series 3. Its front panel is intentionally simple: three large, slightly raised Power, Mute, and "Prog" (program) buttons on the left; a display in the middle, whose readout is clearly legible from a distance of 12', and a large rotary knob on the right that handles volume, input selection, display brightness, and programming/configuration adjustment.
The rear panel, whose layout is almost as elegant as the front's, includes a 15-amp mains input, two pairs (one for each channel) of balanced outputs, one pair of RCA outputs, three pairs of balanced inputs, and two pairs of RCA inputs. Add in an RS-232 port; two "LINK 3-Series" sockets for use with Series 3 units; link-in, firmware, and link-out ports; and a model-identifying panel that includes the serial number, and you know what's on the surface. Settings and configuration choices include balance, phase, ground lift, start volume, maximum volume, and input selection/deactivation/naming.
Beneath the surface
To get the lowdown on the company and the technological underpinnings of the 727, I Zoomed with company owner and chief designer Cyrill Hammer, 58. During our 55-minute chat, he discussed Soulution's origins.
Soulution is a subsidiary of Spemot (for Special Motor), a company that was founded in Dulliken, Switzerland, at the close of WWII. Its initial focus was on manufacturing electric motors for everything from power tools to air conditioners. Motor-driven household appliances, eg, a specialty ice cream machine currently used in Michelin three-star (and more) restaurants, helped put it on the map.
In 1997, Hammer's father bought Spemot. When he died shortly thereafter, son Cyrill and his partners assumed control. Not long after that, one of the partners, who was an audiophile, suggested that they diversify into audio. Five years later, the German engineer they hired designed the Soulution 710 amplifier. Its debut at the 2006 High End Munich put the company on the map.
"When we started, our very first idea was to make a 'very good value for the money' product," Hammer said. "But once we saw how much effort we needed to devote to building it, and figured in the cost of componentry and manufacture, we realized that our projected cost would most likely not work.
"At the same time, we realized that our product really was something very special. It was different from all the products that were on the market back then. So, we decided, okay, let's aim for the High End, which back then was not as crazy high priced as it is today. For example, our 710 amplifier cost something like 20,000 when it launched (footnote 3). But due to different euro exchange rates and price increases, today a 7 Series power amplifier costs something like $60,000 or $70,000."
Once Soulution launched the 7 Series, its distribution partners advocated for a less-expensive series that would appeal to clients who could not afford 7 Series prices. As a result, Soulution launched the 5 Series, whose prices were half those of the 7 Series, in 2012.
"When we introduced the 5 Series, we expected that they would sell in higher numbers than the 7 Series products," Hammer said. "This was in fact the case in the first year after the market introduction. But then, sales of 7 Series products picked up again and the 5 Series did not grow as fast as we expected. Due to its lower price point, it was an easier and quicker decision for new clients to go for a Soulution product. However, after one or two years, most of these clients upgraded to the 7 Series. While we did attract clients who could not afford the 7 Series products, thus enlarging the client base, to a large extent it was the same client segment who bought the 5 Series and then upgraded to 7 Series products.
"The 3 Series, which we began in 2017, is another story. These products cost approximately 25% of the 7 Series and do attract a different client segment."
The 727's history and technology
The Soulution 727, which replaces the 725, is the third top-level preamplifier the company has produced. From the first drawings, it took seven years to bring the 727 to market (footnote 4). Motivating its design were desires to reduce noise, enlarge bandwidth, reduce phase error in the audio band, and improve the balanced input's common-mode rejection. While reducing distortion was also a factor, Hammer said, "The distortion level of the 725 was already so crazy low that we knew it would be very difficult, or perhaps impossible, to reduce it even further."
When asked to tell Stereophile readers what he considered special about the 727's circuit design, whose dual-mono layout uses separate circuit boards for each channel, Hammer was far more forthcoming than most designers I've interviewed. Due to the comprehensive nature of his explication, it seems best to simply step out of the way and share his words en toto (footnote 5). In several places, I've augmented Hammer's information with bracketed information I drew from the product manual (footnote 6).
"At the heart of the 727 preamplifier are four special building blocks," he said. "The first one is the completely redesigned input stage, which receives and buffers the signal. Then, the volume control is important in the pre and output stages. Everywhere, everything is fed by a good power supply.
"I'll start with the full balanced input stage, which is derived from instrumentation amplifiers. [Soulution's instrumentation amplifiers are designed for 20MHz bandwidth, noise density <160dBV/√Hz, and optimal common mode rejection >105dB.] The input stage's structure allows us to optimize common mode rejection so that noise, which is common on the input signal, can be actively rejected by the input stage. The standard design used for these kinds of balanced inputs is a so-called difference amplifier. But the very simple and easy design of this difference amplifier requires a tradeoff: either you get the best common-mode rejection and add a lot of noise, or you optimize for low noise but lose common-mode rejection. You cannot have both with the difference amplifier's simple structure.
"Instead, we use an instrumentation amplifier which is built from four op-amps with a configuration of resistors. This requires a lot more components, but we were a lot happier with both common mode rejection and noise. Then, to further bring down the noise, we use eight of these structures working in parallel. Ultimately this requires 32 op-amps. It's a huge effort to do this, but it pays off because it's extremely important to have the first stage as noise-free as possible. Our op-amps are very high grade and have close to no distortion, add no additional noise, are extremely wide bandwidth. Theoretically, you can push them to 40MHz of operating bandwidth, which is really phenomenal.
"The next difference from the 725 is in the volume control [which uses relay-switched, high-precision, metal-foil resistors]. We had to push its impedance level down roughly by a factor of 10. This was a challenge because it required much more current to drive it. But the input stage was able to drive it just fine. [Each channel's volume control has its own resistor network. Volume is adjustable in 1dB increments over the span of 80 steps. The 727 also uses a secondary, IC-based volume control that alters volume without producing any audible signal; once volume is set, this secondary volume control is bypassed.]
"Next, we wanted to have signal leave the unit via a class-A push-pull output stage without any distortion or additional noise. This led us to completely redesign the output stage with discrete transistors to make it compatible with the 20MHz (3dB) bandwidth we wanted to have. We also used a feedback loop, controlled by an op-amp, to ensure wide bandwidth and low distortion.
"Our power supply is also really special and completely different from what came before. I think it is also completely diffeent from what I believe everybody else in the industry is doing. We call it a 'distributed power supply,' which requires several gauges of power filtering and regulating. At the first step, we have a switched-mode power supply (SMPS) which converts mains voltage to DC voltage. We go from 120V in the US to 24V DC. From there, we produce all the different voltages we need for the unit. The display controller needs maybe 5V and 3V, but the analog stages require a highly efficient switching DC regulator to bring the voltage down to 18V.
"Then we have a so-called low drop regulator, a linear power supply that is extremely noise-free and highly precise, to control the voltage we want to supply to the amplifier and distribute to the whole board. One huge layer of the board is devoted solely to that single voltage power supply; this allows power to be distributed with close to no loss. Then, next to each power sink (eg, an op-amp or a transistor that is sinking current) there is another regulator, next to the pins just two or three millimeters away, that performs the final regulation. It is ultralow-noise and extremely close, which creates the shortest signal path possible from the power supply to the device it powers."
You go your way, and I go mine
Some designers tune their designs first; after they've got the sound where they like it, they tweak things as necessary to achieve the best possible measurements. Others optimize measurements first and sound thereafter; yet others focus mainly on sound. And then there are companies like Linn that, once they achieve measurements that they consider ideal, trust that the sound is optimized and forgo tuning. What process does Soulution follow?
"We begin with our engineering goals," Hammer said. "Once we move forward, we measure to see if we've met our goals. Usually, if the unit does not perform to the specs we want, we go back to the drawing board without listening. Only when we achieve the desired specs do we begin listening.
"We never say, 'Okay, let's put a capacitor here or do this to that to make the treble less hard or improve the midrange.' We do a lot of listening, but we do not tune according to what we hear. If we don't like what we hear, we ask what could be wrong with the design. A discussion process follows where I ask where something may be wrong that we did not see. Sometimes we find it right away, and sometimes it takes months to understand.
"To give you an example, when we were building prototypes for the new power amplifier that may be in production by the time your review goes to print, the measurements were not necessarily better than for its predecessor. They were a little bit better in some areas and not so in other areas. But it sounded phenomenalunbelievably better than what we had before. So, we scratched our heads and asked what we had done to make such a difference. It took us maybe six months to understand what design choice was responsible and optimize further.
"There are a lot of strong beliefs in the high-end industry. Some believe you must have big transformers or separate outboard power supplies to have the best design. Negative feedback is a bad thing, switch-mode power supplies are evil, etc. We don't care about these orthodoxies. Whenever we think negative feedback will make an improvement, we use it. Regardless of industry belief, if we think the best option for a power supply is switch-mode, we use it. We go with what works best for our design. It may not be the best option for someone else's design, but if it's best for ours, we use it."
Footnote 2: According to the manual, "With its adjustable gain (40dB to 80dB at 1kHz) and termination impedance (20 ohms to 1k ohm, 47k ohms, 0pF to 70pF) the phono input can be adapted very precisely to your pickup system. The use of very fast amplifier stages (3dB at 2MHz) ensures highly accurate reproduction of the RIAA equalization curve (frequency and phase response)."
Footnote 3: When I asked Hammer what he thought the 710 might cost if it had first launched toward the end of 2024 rather than in 2006, he estimated between 40,000 and 50,000.
Footnote 4: One reason it took so long is that several other products were being designed simultaneously.
Footnote 5: As is always the case when I interview people for whom English is a second language, this explication has been edited for grammar.
Footnote 6: The manual is distributed in print form to 727 owners and available online at soulution-audio.com/downloads.
This is the only time a butt upstages a piece of equipment. It appears Melody Gardot has put in some gym time on the stair master
As a 70 year old man I am quite confident I would never be able to hear an audible difference between this piece and a Benchmark LA4. For those who can, good for you. Enjoy it while you can.
Do we know the age of the reviewers here and do they share with us every year the results of their hearing test by an audiologist? Or every 6 months? Not to mention their personal preference, as like professional wine tasters give very different ratings as well.
Don't be so sure. Unless you have hearing damage, I'm quite confident you'll be able to hear the difference.
jason
Who needs a preamp worth this when everything sources or dacs etc have their own volume controls these days with enough gain to "feed direct" into poweramps with one less set of interconnects.
All it serves to do is to add more distortion/colorations into the signal path and have "glitz value" for $75k. It does have a phono stage??
Cheers George
"This preamp really has no sound of its own."
When the ancient Greeks were finally able to create realistic true-to-life statues, the reaction of everyone was not "OH! how great" but "OH! how boring" and sculptors immediately went back to creating unrealistic statues.
If audio were purely aesthetic and subjective like art, this would be an apt analogy.
But it’s not.
Sculpture as an art is more akin to music, not audio playback— which is science and technology based.
The goal of high *fidelity* audio is realism, and accuracy plays a definitive part in that. Measurements would be totally irrelevant if the goal wasn’t accuracy.
So measurements are vital in audio. They give an objective, reproducible, highly detailed, and science based description of what a product will sound like.
I can’t imagine anyone wanting a deliberately unrealistic sound signature of playback equipment that is intended to faithfully reproduce music.
"This preamp really has no sound of its own."
And for those that believe that, the second set of interconnects I suppose don't either
Wow that's like inventing "perpetual motion"
Cheers George
I would like to add that I am a fan of extreme engineering. This was a large fraction of my work as a researcher before I retired. There is value in attempting to expand the limits of a given technology without regard to cost. Since we have an economic system that creates a relatively large pool of people for whom $100,000 represents pocket change better that they spend their spare change on luxury goods rather than buying politicians. A Rolex watch is a status symbol rather than a superior instrument for keeping track of time. The average human being is not going to miss their train by relying on a Timex watch. The person with the Rolex is not going to be more on time.
The Times/Rolex analogy isn’t a great one TBH.
You assume that a watch’s only function is to tell the time, and that the most accurate watch is “the best watch”. Chronographs have multiple time keeping functions besides just telling the time, for example.
And many watches are sought for their durability, waterproofness, and reliability. If you want something mission critical, it ain’t worth the metal it’s made out of if it stops working completely.
I think there’s legitimate reasons why the watch chosen for the moon landing was an Omega and not a Timex. To a certain point you get what you pay for.
I have a Rolex, and to me it is a beautiful, beautifully made watch. Picking it up, you can feel and see the quality right away, There is no Timex like it in the world. I enjoy wearing it, and for me it justifies the price differential over a Timex.
But if there were some sort of apocalypse or if I were stranded on a desert island, the watch I’d most want to be wearing is my G-Shock. Solar powered, mulitfunctioned, shock and waterproof, and radio controlled automatic synchronization to atomic clocks around the world, it’s more technologically advanced and rugged than either Rolexes or Timexes.
They all tell time accurately. But that’s far from the whole story
Thank you Jason for a very enjoyable read of something… I will never have. I can’t even imagine what it’s like to behold a unit like this - with or without it being plugged in! Like sw23 alludes, this is undoubtedly a fine piece of jewellery that only certain folk will ever enjoy. However, that’s why I love coming to Stereophile to read reviews like this. It’s the second-hand experience; the audible passion and appreciation of such craftpersonship that makes the reading compelling. Thank you as well for the notes on Ernst Schlader, I haven’t stopped listening since!
I’m with Sw23 on this one. I’ve waiting for my Benchmark HPA4 to arrive so I can plug it into my AHB2s and would like to think the journey will stop there. Hopefully my ears will convince me!
It appreciates the work that went into creating the product.
And JVS needs some appreciation too- That was classy.
Truly appreciated.
jason
First of all, thank you. it's great to hear from you.
While I'm certain you will not be traveling across the border to buy Canadian goods, there are other reasons to make the trek. If you can find your way across, I'd be glad to let you hear the 727, and the Relentless, and compare them to your Benchmark.
To everyone else: Chris and I go way back. Well, not as far back as I go.
jason
Good review, JS. But ... "the maximum gain for the balanced and unbalanced inputs to the balanced outputs was a little lower than the specified 4dB, at 3.7dB" --- that's a 7.15% deficit from spec, and it makes me wonder just how "precision" this device really is. If anything, I would have expected a box like this to be conservatively specified (e.g., 4dB spec = 5dB or 6dB actual). I have $200 Chinese boxes that are far more accurate at meeting their published specification.
This one, or another - priced tens of thousands less, no less - which JVS characterized as "clear and transparent and truthful as can be."
He went on to conclude that it "could very well be the keeper, the component that delivers a lifetime of joy and pleasure. It is one of those rare products whose excellence is proclaimed with every note. It is more than an object of beauty; it reaches into the emotional core of musical experience and opens a window onto ultimate truth."