Is Swan Song Audio's New Cygnet Phono Preamplifier Worth Its Weight In Copper?

Swan Song Audio (www.swansongaudio.com), newly arrived in the audiophile electronics market will make its show debut at October’s Rocky Mountain Audio Festival 2014. Earlier this year Anthony Crocker (who I think at this time is Swan Song Audio), emailed and offered for review his Cygnet Phono Preamplifier.

The photos on the company’s website displayed an orderly circuit board inside that uses surface mount technology and attention to sometimes fanciful detail outside including using as a power “on” indicator an LED illuminated 7mm Amethyst gemstone.

But more attractive I found was Swan Song’s "Philosophy" page. It comes across as genuine. The phonopreamplifier primer page contains much useful information about gain, noise, loading and the RIAA curve. It's definitely worth your while to read.

Stereophile requires for a product review a minimum of five domestic American dealers, but analogplanet.com has no such requirement. We’re open to reviewing products from start-ups and new companies but we do so with the warning that while products from new companies are often attractively priced, the resale value often will be lower than that of established brands and if the company folds, lower yet—unless you happened upon a short lived product that achieves legendary status but you’ll first win the lottery.

The Cygnet is a compact, polished and oxidation-proofed copper chassis design featuring an outboard copper chassis power supply connected to the main unit via a multi-pin umbilical cord. The copper is of course handsome, but it’s used because of its shielding effectiveness—5X that of steel and twice that of aluminum. (The photo stacks the two boxes for display purposes only. The two should be separated).

Custom Tailored to Meet Your Needs

The Cygnet’s base price is $2100. Add $300 for Sowter 9570Z 1:10 step-up transformers. Add another $300 for the Special Edition version that moves loading, RIAA and balance controls to the front panel, though loading options are reduced from 8 in the standard version (via internally mounted jumpers) to 4.

Because the Cygnet uses socketed op-amps, Swan Song also offers a selection of op-amp upgrades at various price points. They’ll sell you a Shunyata Venom 3 S power cord for $75.00 and silver colored screws. You can even customize the “on” indicator gemstone. Seriously.

Unusual and Unique Operating Features

Making the reasonable assumption that cartridges used with the Cygnet can range from 1.mV to 6mV and that amplifier input sensitivity can vary from .25 to 2 volts (a wide range), Crocker’s design offers an unusual 30 different gain options to assure optimum matching. The point of this is to allow you you utilize the full range of the preamp or integrated amp’s volume control.

Each of the unit’s 3 gain stages can be adjusted to low, medium or high gain settings, to optimize both cartridge gain and what the amplifier “sees” for a total of 36 steps (32-61dB) in the standard version and 52dB to 81dB in the step-up transformer equipped version.

In other words, with the standard version you can run both MM and MC cartridges but the step up transformer version is for MC cartridges only.

Once gain is optimized you can adjust channel balance via a +3dB variable gain control but the 0.1% tolerance resistors generally assures that you won’t need to make this adjustment.

Even more unusual and perhaps questionable is the end user (that’s you) ability to adjust the RIAA filter. While the 500Hz bass turnover frequency is fixed, the 2122Hz treble turnover filter can be adjusted from 1879Hz to 2283Hz, while the 50Hz filter is user adjustable from 37Hz to 60Hz. Obviously the lower and higher filters will have greater effect on low bass and treble response. However, since these are standardized, why would you allow user-adjustability here? Read on!

Setting Up the Cygnet

I requested a review sample that included the $300 transformer option. Otherwise the Cygnet was standard, for a total price of $2400. While the multiple gain stage settings might at first appear daunting, Swan Song’s instructions, which come on a USB stick, includes a less than easy to follow Excel spreadsheet. The instructions make it sound simple enough: you open the spread sheet, enter the two numbers (cartridge output voltage and amplifier input sensitivity) and locate the highlighted suggested gain settings. Those are in green in the photo below. Once you open up the unit, what this means might become clearer but it’s been made too complicated here in my opinion.

For each of the three stages there are two sets of pins (positions 1 and 2). What’s highlighted means, in plain English: Stage 1, jumpers are used on both position 1 and 2. Stage 2, jumper only on position 2. Stage 3 jumper only on position 1. I think the spread sheet could be made more user friendly but if you follow the prose in the directions and look at the manual, the spread sheet and at the circuit board it will be clearer.

Properly set, the Cygnet will provide the optimal gain for the cartridge and for the amplifier’s input sensitivity. The Cygnet’s adjustability in this regard is unique in my experience. Virtually all phono preamps provide cartridge output gain settings but not overall output gain settings. Swan Song also includes final level trim pots that acts as a balance control but under most circumstances they will not need adjusting.

Five internal dip switches used individually or in pairs set loading and then there are the very unusual RIAA adjustments that act as tone controls. Increase or decrease gain above 2122Hz and add or attenuate the top end.

The manual has an unfortunate typo:

“The first black shaft, on the left when facing the Cygnet, is the upper midrange RIAA setting. Adjusting this to the right will increase the gain of all frequencies above 2122hz. This will help the system to sound more open with more treble response. The first black shaft is the lower bass RIAA setting…..”

That is less of an issue than the concept of using the RIAA as a tone control, which I’m less than comfortable with and the fact that there’s no click or detent or other indicator showing where the 2122Hz and 50Hz settings are on the two potentiometer shafts. So if you get a Cygnet and decide to “play” with these, first carefully mark the starting position!

Once you’ve confirmed your settings and screwed on the top plate, cable connections—no matter how large the plugs or thick the wire—— are facilitated via the use of recessed high quality Neutrik RCA jacks. All internal signal path connections are via .999% silver wire.

By the way, the claimed specs for this unit (see website), particularly in terms of noise and distortion are impressive regardless of price.

Sound

Analogplanet.com readers overwhelmingly voted in a blind test among moderately priced phono preamps for the $1100 Lehmann Black Cube SE II and I concurred. It produced both warmth and detail plus excellent spatiality. The easiest way to describe the Swan Song Cygnet’s sound would be take a Black Cube SEII, expand laterally somewhat the soundstage, maintain much of the Cube’s wonderful warmth but in order to increase tonal neutrality carve away the warm part that made its character obvious, increase transient speed and clarity and improve transparency.

And where does that bring you? Next door to the similarly priced LKV Research Phono 2 SB. These two similarly priced products do not sound identical or look identical but they operate on a level playing field in that both start with ultra-low noise floors.

One offers balanced operation and balanced and SE inputs and outputs (LKV) one doesn’t, one is op-amp based (Swan Song) and one is based upon discrete transistors. Both designers advocate well in the instruction manuals and information blurbs for their choice, which is as it should be.

Timbrally, the LKV is somewhat leaner and more “analytical” and the Swan Song is somewhat warmer; the LKV slightly “faster” the Swan Song lingers longer on the note.

ORG’s outstanding double 45rpm reissue of Peter, Paul And Mary’s eponymous debut album (Warner Brothers WS 1449/ORG 069) arrived during the evaluation (this one and In the Wind were mastered from the original tape not a copy and they sound it!). The gold label original issued May of 1962 is a sonic gem engineered by Bill Schwartau. It places the men L/R (Paul Stookey usually on the right channel, Peter Yarrow on the left) and Mary center stage. All are closely miked and sound as if they are in your room as opposed to you being in the studio, which is how Kingston Trio albums were miked—plenty or reverb, somewhat distant vocals often bunched in the center. Play this album with headphones and the perspective is almost too close!

Make no mistake: this was more a #1 pop album that sold more than 2 million copies than a “purist” folk album, with PP&M and Milt Okun curating great tunes and delivering them with chill producing exuberance and stunning 3 part harmonies. Yet PP&M delivered the material with sincerity sufficient to produce emotional resonance with the material. They epitomized clean cut “coffee house” and that is from where they came.

I first evaluated the reissue with the Lyra Etna cartridge into the Swan Song configured for a .5mV cartridge and a 1 volt sensitivity amplifier. Load was set to 100 ohms.

Without reference to the original pressing I thought the reissue produced greater three-dimensionality, more low level detail, better textured and cleaner transients and a timbral balance that sounded remarkably similar to the original as I remembered it. The guitars were far better separated in space and more cleanly delineated than I’d ever heard them. I also thought I heard greater transparency and detail slightly at the expense of a pleasing warmth heard on the original but after a full play I decided the reissue betters the original by a wide margin.

Then I put on the original and I ended up coming to the same conclusion. The original sounded more similar than different to how I remembered it sounding through my considerably more expensive reference Ypsilon VPS100, which is transformer, tube and LCR based.

All of this says a great deal about the Swan Song’s resolution, tonal balance and dynamic performance, though once I listened through the Ypsilon the differences were obvious in both micro and macro dynamics and in terms of ultimate “see through” transparency and overall stage expansiveness.

. But was it a more than 10X price differential? Well yes in the context of my system and listening experience but I have to admit I listened happily to the Swan Song Cygnet for well more than a week without switching back to my reference and wasn’t made overtly aware that I was listening to a product not in the price league of the rest of the system.

Conclusion

The Swan Song Cygnet is an attractive and distinctive looking, well-made, compact phono preamplifier that as a few quirks best dealt with cautiously. I wouldn’t mess with the RIAA settings for instance, unless I knew precisely how to return to “ground 0” before twiddling either knob. The Cygnet’s adjustability in terms of gain and sensitivity matching are also unique and very useful in some circumstances. While the engraved script on the cover’s underside has a “home-made” quality, the surface mount board components and high quality parts used throughout indicate a manufacturer serious about high technology. Incidentally, hopefully today, audiophiles have dropped their prejudices against SMT (surface mount technology) and the use of op amps for gain. When carefully selected and matched they perform well as this phono preamp, the Black Cube SE II and the battery powered $5000 ASR Basis Exclusive I reviewed way back in 2003 that’s extensively updated and considerably more expensive today. It was the first high performance op-amp based phono preamp I’d heard.

So now I’ve reviewed on analogplanet two “factory direct” similarly priced phono preamps priced at around $2500 give or take a few hundred. The standard $2100 Swan Song Cygnet should have sufficient gain for all but the lowest output MC cartridges and based on the transformer equipped one reviewed here, it should be a sweet-sounding option.

Which would I choose if given the choice between the LKV Research and the Swan Song? You shouldn’t care what I’d choose. And I’m not sure which I’d choose. Were my system balanced throughout I might lean towards the LKV. Otherwise I might lean the other way. The LKV offers greater transparency tending towards the analytical, the Swan Song offers greater richness and smoothness tending towards warmth but both get out of the way once your ears adjust. Neither produced gross colorations.

Designer Crocker supplies the Cygnet in a really nice “road case”. He tests and listens to each unit for a considerable time before shipping to the customer.

Having spent a considerable amount of time (months) listening to the Cygnet in a variety of systems I can understand why he does that. He’s created a really find sounding phono preamplifier. Whatever Mr. Crocker’s background, it’s obvious that the Cygnet was designed by someone who listens and not by a technocrat who only measures. He’s produced a phono preamp that has excellent claimed measurements and sounds as good as it’s said to measure.

Being mindful of the cautions to be considered and benefits to be derived when buying from a new company, I’d think once you’ve seen and heard the $2100 Cygnet you’d agree with my conclusion that buying one is a chance definitely worth taking.

COMPANY INFO
Swan Song Audio
2852 East 85th Street
Tulsa, OK 74137
info@swansongaudio.com

COMMENTS
Ortofan's picture

Would you really take a chance on this unit rather than a Parasound JC-3+, or a Lehmann Decade, or a Sim Moon 310LP or a Luxman E-200...?

Claimed RIAA matching is 2% - is that the best they can do?
More info needed on the "Nuetrik" [sic] connectors being used.
Is there a subsonic filter?

Only 47K loading for fixed coil cartridges?
Some have flatter frequency response when loaded with ~33K or up to 100K.

Michael Fremer's picture
Some might. It sounds very good and is well-made. Sorry I spelled Neutrik incorrectly. What info do you need? Neutrik NF2D-B-2 RCA D-Series Panel-Mount Jack. Very few phono preamps offer options other than 47K for MM cartridges, though of course some do better with other loads. But why single out Swan Song? The RIAA matching spec needs further elucidation from the manufacturer and I hope he logs in and explains....
SwanSong's picture

Thank you
First and foremost, thank you Michael for a complete and thoughtful review. I have always found your reviews to be insightful and thought provoking. Below are some specific responses. I will be happy to answer any further questions.

Opamps
SSA uses opamps to allow the advanced “tweaker” to change to sound of the Cygnet. The opa227 is the standard opamp due to its bandwidth of 8mhz low noise levels. Nearly 20 opamps were tried and considered but none provide the black levels that allow the notes to linger in space for the price of the opa227. If money is not an issue, the opa627 is the favored opamp and is available upon request. As with all upgrades, SSA only charges the additional cost of parts, ordering and any extra labor.

Gain
The Cygnet was designed and tested using a .5mv cartridge. Noise levels are low enough to support the gain required to use a .5mv cartridge without transformers. Transformers are provided at cost as a service to our customers.

Documentation
Comments taken and well said. We will be working on clarifying the documentation and creating a series of videos.

RIAA curve
It is not necessary for the listener to make any adjustments. SSA will set the gain and the RIAA curve. However, a phono preamp does not operate in isolation. The ability to adjust the 50 and 2122hz points allows for some compensation in the rest of the signal chain. The unit provided to Michael was within .1db of the RIAA or within 1%. At SSA we state 2% because we look at the frequency response from 20 to 15Khz not just the 3 points. Due to interactions between the internal components, to achieve a better overall response these points may have to drift from their theoretical values. Please note the traditional measure is +/-dbs. At 500hz a +/-.2db actual versus theoretical RIAA difference is actually 8% (.2db / 2.6db).

Loading
Any loading can be provided. The standard provides 8 levels from 69 to 47K. The transformer option provides 8 to 1K. The 1K limit is imposed by the transformers.

Neutrik
Neutrik have been making connectors for the professional market for a number of years. Their RCA connectors are recessed and therefore help to protect your cables from getting knocked loose. http://www.neutrik.com/en/video/phono-rca/phono-rca-chassis-connectors/

About SSA
SSA is a boutique company focused on making high end audio accessible in a manner that enhances the listeners décor as well as their listening experience. We are focused applying experiences and knowledge obtained in R&D for high tech military electronics to the audio world. R&D is a heavy focus. Every part and every mm of track is scrutinized. Cost only comes into play at the very end in order to set a price. At SSA, first and foremost is how does the item perform and sound. No product is brought to market without months or trials and real world usage. When you purchase a product from SSA you can be assured that in every step of the way it was agonized over and cared for by someone who cared, not a machine. We are well funded to the point that making your listening experience better, not sales, is our primary concern.

Again. Thank you Michael for this opportunity. I am looking forward to the RMAF. Both the Cygnet solid state and tube version will be present. There will be a series of high end LP demonstrations and AB phono comparisons.

Ortofan's picture

..but is the reproduction any more (or less) faithful to the sound from the master tape than with those other phono preamps in the same price range?

The "Nuetrik" error I cited was not yours – it’s spelled that way on the swansongaudio website. Oversights such as that make me wonder what they might have overlooked in the design process.

IMO, allowing the user to fiddle with the turnover frequencies of the playback equalization just seems wrong – unless the intention is to provide eq curves for various pre-RIAA recordings. Swan Song makes a big fuss out of this adjustability feature, but having a wider range of fixed-coil cartridge loading adjustments would better suit my requirements. Otherwise, if you want such a tone adjustment facility, then get a Cello Audio Palette (or the AR Limited version) or one of the McIntosh preamps or integrated amps with multi-band tone controls.

Reading through their website, there seems to be too much emphasis on the chichi and not the substance. The reference to “female and high wealth consumers” evokes the image of Mr. Hugenetworth showing off his latest hi-fi acquisition to his trophy wife, whose focus is solely upon the bejeweled power indicator. It reminds me, coincidentally, of the Aston-Martin Cygnet, which is little more that an overpriced, overembellished Scion iQ.

Of course, that could just be the skeptical technocrat in me speaking.

Michael Fremer's picture
"Oversights such as that make me wonder what they might have overlooked in the design process." When a typo elicits that kind of comment, the part of you that's talking is obvious to me. It seems to be that the focus of the design is not the bling. The bling is simply a nice addition. i agree though that the adjustable RIAA is controversial.
Ortofan's picture

At my first job out of college, the president of the company would sit in on presentations made by the junior staff members. Although his expertise was in marketing, and his technical background was rather limited, he could detect spelling and math errors like a savant. Anyone whose presentation contained such errors would be tossed out of the meeting unceremoniously. The assumption was that the entire proposal was prepared with the same lack of care and attention to detail as was the presentation. That experience was a formative one and has stuck with me through the years.

On the subject of bling, Sony used to copper plate the interiors of the chassis on some of its ES products to a functional effect. Is there really some non-aesthetic benefit to putting copper on the exterior, as well?

Michael Fremer's picture
IMO you are throwing the baby out with the bathwater. If a presentation was sloppy and rife with errors i agree, but if there is but ONE typo in an entire presentation associated with a well-produced and well thought out product like this (whether or not you agree with all of its implementations) then I'd say you are being overly harsh. As for copper, one could turn it around and ask Sony the question: "Why cheap out on the copper and just plate instead of solid copper"?
Ortofan's picture

…without seeing a schematic. Otherwise, it’s difficult to make that determination. All you can do is go by whatever info is on the website.

Indeed it could be construed as equivalent to throwing the baby out with the bathwater. OTOH, if time was so heavy on their hands that they were able to install an amethyst jewel in the power indicator, you’d think that they might have already gone to the trouble of proofreading their website content. The jury is still out on the precision of the RIAA equalization. The Swan Song website shows only 2%, whereas, for example, the Liberty Audio preamp you reviewed recently is specified at 0.1%.

Good question why Sony only chose to copper plate instead of using solid copper. Maybe they compared both approaches and found that just a layer of plating was sufficient. Maybe the difficulty of working with solid material was the deciding factor. Maybe the price point that they needed to meet was the limitation. Do you have a contact at Sony who might be able to obtain the answer? Marantz also uses copper plating in some of its products. Do you have an e-mail address for Ken Ishiwata?

Michael Fremer's picture
The manufacturer's comment re RIAA. You continue to make a BFD out of the transposition of the letters "e" and "u" in the word Neutrik. Having received for review hundreds of products many of which don't come with directions, many of which come with inadequate directions, for you to make such an issue over the misspelling of "Neutrik" is laughable. But I'll leave the characterization of your fetish to readers. Yes, I have Ken's email address. I am friends with Ken. I can ask him but you're not getting his email address you can be sure. Perhaps in his response he'll misspell something and who knows what that could lead to?
Ortofan's picture

I did read the manufacturer’s comment about the RIAA tolerance, but, without the results of independent test measurements, their claim remains unverified.

It’s not so much that the spelling error is such a “BFD”, rather it’s that any error can make one (or at least me) question the validity of all the other information as presented. Was the spelling error just a one-off mistake or is it symptomatic of the way in which the entire organization conducts itself – only close enough for government work, as they say? How can one be certain without questioning it? The tone of response by the manufacturer speaks volumes, as well. He could have simply stated that we overlooked it and will fix it, rather than belittling the inquiry or the inquirer.

Maybe products with poor documentation are the norm in your experience, but it doesn’t have to be that way. If you continue to accept products with poor or incomplete support, then you tacitly encourage such behavior. Again, faulty documentation may bespeak other systemic shortcomings with the way the product itself was conceived and produced.

I’m thrilled to hear that you are friends with Mr. Ishiwata and would never dare to risk the possibility of somehow compromising that relationship by asking you to reveal his contact information to me – that was never my intention. My assumption was that he would be in the best position to answer the question about the application of copper in Marantz products. My expectation was that, based upon your professional experience, you would know how to get in touch with him and obtain that answer.

Michael Fremer's picture
Ken Ishiwata said:

1) Well, you know very well copper is better conductor than steel, thus, this point of view it’s an advantage. Unfortunately copper is softer compare to steel(Mechanical strength), thus, I was in favour of using steel with thick copper plating on top. I’m sure, purely sound point of view, pure copper chassis made big difference in sound quality!!!

2) I’m sure SONY reason is purely cost and manufacturing and their copper plating is rather thin…. One additional advantage of copper is reducing eddy current, thus, smoother mid high and high frequency sound reproduction!!

Ortofan's picture

Thanks to MF for the prompt follow up and to KI for the response.

Typically, the copper plating would be applied for the purpose of shielding/screening. However, it’s interesting to find out that the side effect of reducing the induced currents in the chassis/enclosure accrues discernible audible benefits, as well.

This would appear to be another factor to take into consideration when evaluating the acquisition of new equipment.

audiof001's picture

Kudos for allowing some of us to tweak the design with op amps of our choice. - there are many audio lovers out there who would complain that they couldn't if the SSA was a closed system. On a side note, I did enjoy the discussion of gain, RIAA and loading on the SSA site, though a bit of copy editing would help make the designers message clearer and more compelling.

curiousperson's picture

What a beautiful amp. Personally I enjoyed the website and found it much better designed and easy to navigate than most. Ortofan reminds me of the person who sees a Boxster and says "it's not a real Porsche if it's not a 911." Of course they drive a five-year old Accord and have only read about Porsches. Hard to believe that you have listened to all the amps you mention at length. In your next to last sentence I think you meant than instead of that, but we all make errors. Lighten up.

Ortofan's picture

Between a Boxster and a 911, I’d take a Cayman. However, a Cayenne would better suit my everyday needs. Then, again, you can get the same vehicle with less bling as a VW Touareg. If you’re not seeking to make a public statement of your wealth, what’s wrong with a five-year-old Accord – or a Saab, for that matter?

I haven’t listened to all the units I mentioned, but I don’t evaluate products professionally, and I long ago concluded that the Borbeley designed preamp I use pretty much sees off anything else. Even the Leach design (upgraded with Rohm transistors) I used before that was hard to beat. If you expect your equipment to make some sort of fashion statement, that’s an entirely different subject.

Ortofan's picture

Between a Boxster and a 911, I’d take a Cayman. However, a Cayenne would better suit my everyday needs. Then, again, you can get the same vehicle with less bling as a VW Touareg. If you’re not seeking to make a public statement of your wealth, what’s wrong with a five-year-old Accord – or a Saab, for that matter?

I haven’t listened to all the units I mentioned, but I don’t evaluate products professionally, and I long ago concluded that the Borbeley designed preamp I use pretty much sees off anything else. Even the Leach design (upgraded with Rohm transistors) I used before that was hard to beat. If you expect your equipment to make some sort of fashion statement, that’s an entirely different subject.

curiousperson's picture

Ortofan. With all due respect, for someone who has not listened to the items mentioned you seem very opinionated and almost hostile for some reason. I would completely agree with your wealth statement and highly believe in value which doesn't always mean cheap. I do value education, careful thinking, and character. The cheap shot on the typo was not needed, says something about you, as MF pointed out, and was amusing given your own typo. Careful about throwing stones. By the way my son has a five-year old Accord (2010) and it is a very nice ride. As a professor I'm far from wealthy, but have no problem with the success of others. Some people actually buy things for the performance, not to make a statement, but I think you are giving yourself away again. I almost agree with your Cayenne vs. Touareg remark, however, where I live you can't give away a used VW while the resale on the Cayenne is very high and probably wouldn't cost much more in the long run. A loaded Touareg is not cheap and I would suggest a CPO Cayenne although I remain a manual transmission guy. Just chill.

Ortofan's picture

…since you brought it up twice, let me say that neglecting to express my appreciation for taking the bait and, thus, helping to prove my point, could be considered inconsiderate of me. Having said that, and under other circumstances, an edit function for this forum could prove useful.

Devil Doc's picture

And for $2400, it seems like a steal. If I was in the market for a new phono-stage this would be on my list.

SwanSong's picture

Thank you Ortofan for the spell checking. Given the thousands of words on the website and the arduous task of starting up an audio company, we may have missed a word here and there. Perhaps I should post a bounty?

Thank you curiousperson for your reality check. You sound like you may have a lot of experience with high end audio equipment and German cars.

Audiof001, you sound like the tweaker part of the target market that appreciates good design. Here are some of the challenges faced by the Cygnet:

Low noise
Musical reproduction begins with clean power. Dirty power results in dirty music. This dirtiness can come from poor filtering or noise from sources such as a light bulb. The ground up design for the power supply resulted in the .00000025 volts of noise on the DC power going to the preamp. This allows for a signal to noise ratio of 153db before additional power filtering at the opamps.

However, clean power and a good circuit won’t help if outside interference is amplified by the phono preamplifier. Given than the superior conductivity and shielding properties of copper are well documented, a solid copper chassis was selected for the Cygnet. What most people don’t know is that copper is extremely difficult to work with. Because it is non-ferrous, it cannot be cut with a plasma cutter. Copper’s reflective nature means it cannot be cut with a laser, nor cut with a water jet. The only option is milling. Due to the small bit sizes required to mill the tight tolerances, fabrication shops will not work with copper at these specifications. Therefore, I had to develop fabrication processes and skills in house. I personally CNC mill and CNC engrave all chassis (silkscreen on copper seems a shame). I have consulted with master machinists and engineers from Ford Motor Company and American Airlines. The end result was there are very few people with the skill set required to mill copper. This is why you don’t see copper used more often. Even if, as Devil Doc stated, copper is beautiful to look at. As an aside try and purchase copper screws. I work with a couple suppliers in China, which was a challenge all in itself

Finally, the most common way to remove distortion is through feedback. This is accomplished by taking a portion of the output signal, inverting it, and using it to cancel out common noise. The problem with this approach is that it also reduces dynamics. The Cygnet uses NO feedback.

Component distortion
The more components put in the signal path, the greater the likely hood for distortion. One to the biggest contributors to signal distortion is the use of coupling caps to remove the DC offset created by amplification. The Cygnet has NO coupling caps in the signal path and a DC offset of about 10mv or less. The manner in that this is accomplished is not commonly done and therefore this is one item I will not discuss in more detail.

Speaking of capacitors, the type of capacitor used in the RC filters for the RIAA curve will affect the sound. I use WIMA and Vishay caps. If you wish to change them out you may. But because the exact value of capacitor may not be available in your brand of choice, you can use the RIAA adjustment pots to compensate.

I could continue to wax on poetically, instead I invite you to come to the RMAF where I will give a detailed discussion on designing a phono preamp from the ground up. Besides I am off to enjoy some LPs on the Cygnet Tube phone preamp.

Ortofan's picture

...how do you reduce the open-loop gain (and increase the bandwidth) of the op-amps? With NO coupling caps in the signal path, how much DC gets passed through the phono cartridge or the step-up transformer? Did you ever try a copper plated chassis instead of one fabricated from solid copper?

SwanSong's picture

Answers to Ortofan
Bandwidth of the OPA227 is not an issue as the gains of each stage are recommended to be kept around 40 or less (The first stage can be run to 60 but the voltage levels are so low slew rate is not an issue). If you want more than 8mhz bandwidth, move to the opa627 at 16mhz. For transient speeds slew rate is a more important factor. At the low voltages of a phono preamplifier, you have to have a high-end system to hear the differences. Remember the phono preamplifier is but one link in the chain. If the preamplifier and amplifier cannot handle the transient speeds, then the speed of the phono preamplifier does not matter. Using the 227, I have run the Cygnet at 2mhz with no waveform distortion, albeit a very small waveform due to the RIAA filters.

DC offsets are caused by DC in amplification circuits and input bias currents. I have not seen any DC offset coming from any of my cartridges (I have 4) or through the setup transformers. Coupling caps are used due to the amplification of the signal. A properly designed circuit can minimize their usage. That said, check your amplifier it has plenty of coupling caps. At the higher voltages in the amplifier they due less damage then they would in a phono preamplifier.

The effectiveness of the copper shielding is directly related to its thickness. Copper plating would mean that the shielding is limited to the metal that was plated. Yes, I tried it and no it is not the same. I have special equipment to measure very low voltage levels.

The very first public Cygnet
The first owner of the Cygnet has been watching the blog and asked if he could join in the discussion to provide his feedback. I won’t provide his actual identity but perhaps we could just call him “Al”, like in the Paul Simon song. However, first a little background might help.

Origins
The Cygnet started life as Tube Hybrid design that I was undertaking as an engineering exercise, it was never intended to go public. “Al” and I were exchanging emails and he expressed an interest. We pushed the limits in many areas and the end result sounded good but the typical tube noise levels were higher than I desired.
Al was interested in a good solid-state phono preamp so I took a break to create a ground up design that allowed us to explore different technologies. With build release number 4, a Cygnet was born that not only met his needs but demands were being made to go public. My arrangements with Al are that as I develop new products, he provides suggestions in return for a board and bill of materials.

Versioning
Each build release is a combination of several versions and can incorporate 20 to 100 changes. Each new build must contribute significant performance or usability gains. For example, build release number 2 was a through hole design and was ok in its own right. Noise levels were not up to par so out came build release number 3. Number 3 was a move into SMD and with some through holes components. Better but as an engineering exercise number 3 was not where I wanted it to be. Number 4 is a full up SMD design with advanced changes to the power supply and an active ground plane. I feel the PSRR, SNR/THD performance of number 4 is on par with the best in the market. Michael reviewed build release number 5. Number 5 has better performance, more flexibility and a smaller board size than number 4. Number 5 also moved to a copper chassis and is the result of various parties asking for release number 4 to go public. The current build release is number 6. To explain number 6 the Cygnet tube should be discussed.

Cygnet Tube
Once the Cygnet build release 4 was in the “listening” months, the Cygnet Tube version was begun as an undertaking to complete the original engineering exercise. The Tube version is much more complex 4 layer board with 3 separate power supplies. This complexity led to changes that were ported over to Cygnet solid-state version. These changes are such that the casual observer would not notice. But the changes led to increased performance, a greater range on the output balance controls, an increased sharing of parts between the two models, and new parts on the market meant the RIAA adjustment pots are set to the middle when set to the standard RIAA curve. Currently, I have been unable to locate a small low noise pot with a single detent in the center of the pot, sorry Michael.

About “Al”
Al works as an engineer developing hi-tech components for the military. He has several patents and over 25 years of experience. He saw the birth or microprocessors and was an early adapter of surface mount technology. The speeds he typically deals with are in the megahertz range so transient speeds are nothing new to him. He has no financial stake in SSA and can provide an independent evaluation of the Cygnet.

Ortofan's picture

The open loop gain of the op-amps is not invariant with frequency . Using “NO” feedback, how do you make it so?

What is the input bias current specification (upper limit) for your unit?
How do you maintain that limit in production?

The photos show gaps in the enclosure that would permit the ingress of RF energy. Likewise, there is no evidence of metal fingers or other gasket material to reduce seam aperture leakage. How are you sealing between the panels of your enclosure?

ahannan's picture

It’s “Al”, and really it is Al. I waited until the mortar shells stopped flying (hopefully). Michael, I am honored to hear such a positive review for this pre-amp. Tony brought me back into the Audiophile world a couple years ago after a hiatus into Home Theater. A little background first. I was the high school kid back in ’76 that cruised with my brothers to Musicraft, Pacific Stereo, & Playback, and tortured the salesmen for hours switching in various pieces of equipment and speakers. It was at Pacific Stereo where I fell in love with the AR9 speakers. I went off to college and spent my summer earnings on those AR9s as they went on close-out at half price, and then got a degree in Electrical Engineering. I've been working for a company designing instrumentation and test equipment at the circuit level including microprocessors for the past 33 years. So, fast forward about 30 years. Well I still had the AR9s, in a HT setup, and they just didn't have that sparkle they used to (and yes I re-foamed all the low end drivers). Tony comes by and hooks up a little tube amp to the high end drivers and says all you need is a real amp, the speakers are fine. The sparkle was back. And, then he says records are still better, you only need 2 channels. I knew I lost something for the convenience of CDs back then, but I was fully engaged in the lazy digital world. I still had the old turntable and records. I knew someday they would be valuable, but I never expected I’d be playing them again. So I pull out the old Dual 1249 with a Micro Acoustics 530MP (high school budget), but nothing has a pre-amp anymore. So Tony says try one on these rounded red pre-amps with a tube that starts with a “B” off eBay and make these modifications. Well I went though that thing and practically upgraded everything. I even found they left a track off (it was on the schematic) in the voltage multiplier circuit and was bulging the caps. I hope I was the only lucky one to receive that. I might misspell from time to time, but I do check out my designs thoroughly (sorry, cheap shot). My biggest problem with it was it was a single ended power supply and it used coupling caps. The case wasn't big enough to fit a good set of film caps in it. Tony was off working a tube version pre-amp and I said let’s make a solid state one with no coupling caps. So Tony started designing it with select leaded components because those were the audiophile’s choice. So I started being a springboard for demystifying some of the snake oil out there and pointed him to some of the really good new stuff in surface mount. Tony was doing all the real work, I just have the posh position of being a consultant (for a free set of boards of course). I have to admit he had some circuits that were even new to me. Once it was designed, it was scrutinized, and tweaked before layout, and then scrutinized and tweaked some more. There were many op-amp options and I actually selected 3 different op-amps to go in my copy (none were in the final version). Tony picked the 227. When I didn't like what I had picked I went with the faster version of the 227, the 228. The 627 is an excellent choice too, but I couldn't see spending that much in op-amps right off. I can always selectively roll them in later. In the mean time Tony talked me into upgrading the turntable to a Rega P7 with a Dynavector 10X5 as an entry level turntable. I guess I can’t hide behind the college budget excuse anymore.

So enough of the history. When I first got it wired up, ran some waveforms through, and I checked out the offsets. It was under 10mV on the outputs. Success. When I went to tune it, I was having some trouble setting the RIAA, but I thought I had it. I hooked it up and it wasn't good. Tony already had his built and was in nirvana. Then I got Tony on the horn, and we got it set right. Then when I hooked it up it was nothing short of amazing. But you don’t need me to tell you that, there are better spokesmen for that. But what I can say is this design left no corners cut unlike the pre-amp I mentioned above. All the best parts, in a no hold design, put together as best as one can, but more importantly it had to sound the best too. That’s when the engineer walks away and the audiophile takes over. And that is what Tony did.

Oh, and do play with the RIAA pots after marking the position. A little shift can make a significant difference. After all, it’s all about the music.

Al

-------------------------------------------------------------------------
Rega P7 TT w/ Dynavector 10x5, Swan Song Cygnet release build 4, Passive Stepped Volume Control, Vincent SP-T100 Mono blocks, AR9 speakers (plus a modified Benchmark USB-Pre DAC & a modified Sony NS999ES SACD)

SwanSong's picture

When I talk about feedback I am talking about the typical audio use of the word in the meaning of global feedback directly injected into the signal chain. The Cygnet is configured in a direct coupled non inverting configuration. The gain of the opamp does affect its frequency response. With the opa227 at a gain of 40 the upper limit is 200Khz. The 228 is around 3 Mhz. "Al" runs a 228 in his version. I run both 227s and 627s in various units. With any of these chips, I have had no problem with my tube and solid state amplifiers. I am purposefully not discussing the specifics of my opamp configurations as that is part of the months of work that I had to undertake. If you purchase a unit and decide to roll opamps, it would be best to consult me on different choices.

Input bias current is controlled via balancing the input resistors to balance the bias current. Incorrectly configured the input bias current will cause a DC offset. Given the 3 stages, there are offsetting biases based on the loading resistor and gain levels chosen. There are plenty of online videos from TI on this matter. Again, this is an area I had to spend weeks of testing and design to overcome and is part of my intellectual capital.

The "gaps" you are seeing are caused by the round bends that result from the thickness of the metal. The flanges are wide enough to correct this issue and provide adequate sealing. I have run the unit sitting on top of live AC lines powering 30 watts of light and have had no ill affects. My passive preamp, that I use in my personal system, sits on top of this light box and also has no ill affects. Looking at the back of the Benchmark DAC 1, their corner gaps are about the same as mine.

I think at this point we are running a bit far afield of this purpose of this forum and more into the DIY realm. Are there specific concerns that you have concerning performance and how the Cygnet will work with your current system configuration? Or are you looking for assistance to build your own phono preamp? If the latter, there are plenty of kits available on the market. They will save you months of work. And are much more cost effective unless you are located in a dedicated R&D facility.

The Cygnet has over 1000 development hours. If you are looking to start a business be prepared to spend 1-2 years of dedicated time and a minimum of $30K just to start. For that kind of money one can purchase an Ypsilon. For myself, the Cygnet was the outcome of an engineering exercise.

Ortofan's picture

…perhaps you’re right in being of the opinion that this discussion has moved beyond the scope of this forum. Based upon the fact that MF regularly analyzes the recordings he reviews in extraordinary detail, my presumption was that the same thoroughness could be applied to the playback equipment, as well.

Reviews in the Audio Critic magazine used to include circuit analysis and measurements in addition to listening tests. Those evaluations often showed that even the famous names made obvious errors in their designs and that higher prices did not always correlate with higher performance.

I still consider the Apt preamp and power amp to represent an epitome of design and presentation. Of course the GenRad look did not appeal to everyone. However, the manuals were extraordinarily comprehensive and, as I recall, a schematic was attached to the bottom of each unit. There were no secrets withheld or any attempts made to shroud the designs in some mystic aura.

As a final point, your website states that “components should be built to last a life time.” Would you therefore be prepared to guaranty your products for an extended period of time, such as the 20-year warranty offered by Bryston?

Thank you for the time and effort you’ve taken to respond to my questions.

SwanSong's picture

If I gave away my schematics and board layouts then what would I have to sell? Furthermore, even if I gave you the schematic (which I won’t) you could not execute it. The Cygnet utilizes an active ground plane that takes skill and time to implement. Part selection, and wire placement are all critical. You would need to be able solder SMD components on both sides of the board. You would need to know the particular components to buy and from whom. The picture from Michael only shows a part of the story. Finally, the design uses concepts not used in the typical audio components. Therefore, you might change something thinking it is not needed and the whole thing goes up in smoke. I looked at both the preamps you mentioned and they are light years behind the Cygnet. Even the power supply from your preamp could not appropriately power the Cygnet. The Cygnet is not like any other design. The circuitry and board layout comes from outside the audio world. That’s why the numbers are so good. Also remember, Michael reviewed the base unit. Not the more advanced unit.

As to your other point, that is what Michael said at the beginning of his article. As a boutique manufacturer I build to order. Like getting a custom suit you pay for the fitting and basic design and everything else is time and materials. You want a ruby for your gemstone then the cost is the cost of the ruby. You want the chassis out of sterling silver then the additional cost is the cost of the silver. Once you place an order, there are a variety of options that open up. Including lifetime warranties that extend to your children once you are dead, provided I am still alive or SwanSong is still in business and I am dead. The difference between Bryston and SSA, is my extended warranties also include upgrading parts. But first you need to make a purchase. . . . Which like a tailor also means I need to determine what are your needs and goals. We may talk and I may tell you that until you upgrade the rest of your signal path you are wasting your money. The Cygnet is not for everyone. Nor do I want it to be. It is something special.

Littlezilla's picture

Wow, I bet Ortofan is a blast at parties. With that said, he did ask a question of interest. MF, please provide a quick comparison of the Cygnet and JC3+.

Note: This email is a draft and is not intended to be considered a final product. All errors should be corrected with caution as they may have been intentionally placed for entertainment purposes.

Michael Fremer's picture
Had I both here I could do a direct comparison but I don't. I think were they both here for a back to back, the Cygnet would be warmer and more "velvety" sounding without giving up detail but I can't really be 100% sure.

I don't understand Ortofan's treatment of Mr. Crocker and his product. He didn't "examine" any other phono preamp reviewed here with equal skepticism/hostility. I don't understand it.

I'll say this: I am not an audio designer and don't claim to be. I understand the basic concepts and that's it. My job is to listen and report. I leave the design criticism to those better equipped to do so.

Ortofan's picture

Perhaps my reaction to the Cygnet preamp can be characterized as too negative. On the one hand, the Swan Song website states that they “set out to build a product that is not about hype and fashion.” On the other hand, references to “high wealth consumers” and a power indicator fitted with one’s choice of precious jewels suggest (to me) that the target market is subscribers to the Robb Report rather than members of the IEEE.

Over the years I’ve encountered numerous examples of under-engineered electronics, often dressed up in over-engineered enclosures. The Cygnet struck me as just another example of that breed. The designer’s reluctance to demonstrate otherwise and continually refer to the product as “something special” only serves to reinforce that opinion.

Consider the Audio Research LP1 phono preamp – roughly the same price as the Cygnet and produced by a company with a long track record. It appears to come packaged in a sturdy, attractive enclosure and appears to be well constructed. Yet look at the test report published in HiFi News. There’s a greater than 1dB overall difference in frequency response, with the range from 20Hz to 1kHz shelved down by about .35db and the range above about 8kHz shelved up by about 0.8db. Was this done intentionally or is it the result of a manufacturing error? Maybe it was made that way because it sounds “good”, but it’s not accurate. This is but one example of a pretty box containing electronics of somewhat questionable performance.

Perhaps everyone else is satisfied so long as a given product sounds sufficiently “warm” or “velvety” or “detailed” or whatever. Instead, I prefer to be certain that there is also a properly cooked steak behind the sizzle.

SwanSong's picture

Can I use your disclaimer :)

SwanSong's picture

Part of the velvet sound is from the copper chassis and part is from the Burr Brown chipset. When the Cygnet was is in a steel chassis it was a little harsher like traditional solid state. I won't go back to a steel chassis but the I can swap opamps provided they are single opamps. I use singles so I can put some distance between the two channels. If speed is more important then the opa228 can be had for about $12 more (I may start keeping a stock on hand). The opa627 will raise the price about $120 plus shipping so lets say $130. It is a $25 chip versus a $5 for the 227. There are 6 opamps. You can always change the opamps yourself. They are socketed.

Littlezilla's picture

Mickey/SSA,

Thanks for the additional insight into the Cygnet sound and how it compares to the JC3+, equipment with which I'm familiar. If the Cygnet compares favorably to the JC3+, that's noteworthy. For me, the ultimate test is in system/in room performance and reliability vs measurements, etc (although they have their place). I look forward to listening to the Cygnet at RMAF.

SwanSong's picture

Great! I look forward to seeing you. I will have two competitors phono preamps and switch boxes for on the fly a/b comparisons. I will also have a Opa627/buf634 headphone buffer with a standard size headphone jack to allow for headphone listiening. The Cygnet can drive headphones but the others cannot. The speakers will be Zu Druids IV/Rel Storm 3 and the amplifier a fully restored MC225 and a MC7150. The table is an Oracle pairs, SME309 arm and Soundsmith Aida cartridge. I will be bringing LPs. If you have something in particular let me know and I will see if I have it. I am a bit picky about cleaning LPs.

SwanSong's picture

I see comments now and then about a "true RIAA curve". Can a true RIAA curve ever be reproduced given that most rooms have a bass affect, most speakers are not flat, most amps have some reaction with the speakers, and all cables have some sort of inductance? Yes, the phono preamplifier should reproduce close to the RIAA curve. However, it is doubtful that one can accurately reproduce that curve in their room. As an example, perhaps the AR LP1 is shelved high to offset the bass affect that most people have in their room? Ultimately, if the listener prefers a particular sound, who is to say they are wrong? Should one be forced to listen to a component they do not like because someone says this component is more accurate? I think that listening to music should be an enjoyable experience, even if the reproduction is a bit customized to your taste. Just something to consider..... In the end this is a question that has long been debated.

Ortofan's picture

Do you want a precisely de-emphasized version of the signal from the disc or some editorialized version? In the case of the ARC LP1, the specification is within plus or minus 0.3db – which the sample unit under test failed to meet. Testing a second unit could have provided some clues and if there was any comment from ARC, it wasn’t published.

I don’t agree with the concept of trying to have the flaws in one component compensate for the flaws in another – the two wrongs making a right method of assembling a system. It’s agreed that most speakers are not flat and interact with the room to a greater or lesser degree – often as a function of dispersion patterns and bass loading configurations. Corrections can be accomplished with a fractional octave equalizer or the newer DSP-based devices.

Amplifiers can, and should, be designed to be stable with any load and be immune to the effects of reflected signals or stored energy from the speaker. Sufficiently low output impedance/high damping factor and non-intrusive protection circuits help achieve this goal. Borbeley/Hafler accomplished this objective over three decades ago. Locating the power amplifier(s) near the speakers will minimize any reactive side effects from the cables. Bud Fried used to cynically suggest that people bothered by this issue should solder the amplifier output terminals directly to the speaker input terminals.

There’s nothing to stop you from choosing a set up that sounds “best” to you – but don’t necessarily kid yourself that it is the most true to the source. Maybe you like a bit of added second harmonic distortion with your music. Maybe the non-linear output impedance of your vacuum tube amplifier creates a pleasant loudness curve effect. Maybe you’re over age 60, can’t hear anything much over 10kHz and like the rising high frequency response of a particular moving-coil phono cartridge. Does anyone recall that Doug Sax once identified the Stanton 881S moving-magnet cartridge as providing disc reproduction closest to that of the microphone feed – closer than that of any moving-coil cartridge?

Switching topics, I doubt that I’ll be able to attend the RMAF, but I’m curious if you would have any of the recordings I use to evaluate systems? First is the “Famous Duets” recording by Carlo Bergonzi and Dietrich Fischer-Dieskau on Orfeo. Second is any of the Liszt piano recordings by Earl Wild on Etcetera. Third is the live in concert at Carnegie Hall recording by Robert Merrill and Richard Tucker on London. I’ve heard Bergonzi, Merrill and Wild perform in person and these particular recordings make it straightforward for me to judge whether a given system can (in my opinion) recreate a reasonable facsimile of a live event.

SwanSong's picture

How a system is voiced is an excellent question. My collection is over 1300 LPs and ranges from 1953 to 2104. I listen to both LPs and Maxi Singles. I have standards such as The Planets, Firebird, and Steve Winwood. I don’t try to recreate a live event as there are to many other factors and most live music cannot correctly deal with the venue (I do attend many live performance each year). I am more interested in reproducing what the engineer hears in the recording studio. Below are some of the artists I used to voice the Cygnet.

Orchestral
Enoch Light (Command Records), Mancini, Mantovani, Xavier Cugat,Yo Yo Ma

Vocalist
Dean Martin, Doris Day, Engelbert Humperdink, Frank Sinatra, Julie London, Tony Bennett, Vikki Carr, Vic Damone

Blues
Bonnie Rait, Diana Krall, Eric Clapton, Melody Gardot, Ray Charles

Country
Johnny Cash, Willie Nelson

R&B
Anita Baker, George Benson, Jerry Butler, Marvin Gaye, Phoebe Snow, Sade

Jazz
Bob James, Chic Corea, Chuck Mangione, Doc Severinsen, Grover Washington Jr, Herb Alpert, Keith Jarrett, Joe Sample, Maynard Fergusun, Miles Davis, Wes Montgomery

Pop
Adele, Air Supply, Al Stewart, Barry Manilow, Belinda Carlisle, Billy Joel, Billy Ocean, Bob Segar, Bon Jovi, Bruce Hornsby And The Range, Carly Simon, Cher, Chicago, Dire Straits, Eagles, Elton John, Fleetwood Mac, George Michael, Hall and Oates, INXS, Journey, Linda Ronstadt, Lionel Richie, Madonna, Melisa Manchester, Santana, Simon And Garfunkel, Steely Dan, Steve Winwood, Stevie Nicks, Wham,

Rock
AC/DC, Billy Idol, Evanescence, Jimi Hendrix, Led Zeppelin, Queens Ryche, Rolling Stones, White Snake

As an aside, AR could have given the correct specification based on their measurements. It could have been measured at only the 3 RIAA points. Or depending on the tubes used, the internal capacitance could have shifted the RIAA filters. This is an issue that I faced when designing the Cygnet Tube. I use opamps as buffers to keep the tubes from interacting with the RIAA filters. I also use solid-state regulation to keep the operating point of the tube from shifting. I have voiced it and tested it with a variety of new and NOS tubes and have found the buffers and regulators to be sufficient to keep the RIAA filters from shifting. In any case, this is all speculation since I was not present at the time of the original measurements.

Ortofan's picture

Among all the artists you mentioned, the only one I've heard perform in person (and non-amplified) is Yo-Yo Ma. Regarding other references for live sound, I live near a college town where there are classical concerts each week during the school year. In addition, I attend several symphonic concerts annually and a nearby concert hall features chamber music concerts throughout the summer. I've studied and play the piano and have one in my listening room.

The idea of “voicing” a piece of electronics is anathema to me. You’re not building a piano or crafting an acoustic guitar. By somehow trying to “voice” a device, aren't you either adding distortion or, worse, subtracting something from the original signal. If you’re meddling with the frequency response, isn't that introducing some false coloration? Shouldn't the electronics, at least, be as self-effacing as possible?

When you’re evaluating your products, are you making blind, level-matched comparisons? If so, maybe you should take Alan Shaw (of Harbeth) up on his challenge.
To my knowledge, no one has yet claimed the prize.

SwanSong's picture

Each type of resistor, opamp, transistor, tube, capacitor, inductor, internal wiring, connectors, etc has its own sound. When these components are put together they can “color” the sound. Next the interaction between a system component and its connecting components must be tested. Interactions can further change the sound as resistance, capacitance, and inductance can add unintended filters. Voicing a component refers to listening to the various combinations and making sure that the “sound” is accurate based on the designers intentions. This is how a cold, neutral, or warm sound is developed. In the case of the Cygnet, Michael has described the sound as velvety and detailed. There are plenty of articles on the web and tests of resistors, capacitors, and opamps if you want to learn more. But in short, no just putting a set of electronic components together will not ensure that the sound is transparent. This is why Michael felt it important to state that I listen to my creations.

To accurately A/B test the listener should be able to immediately switch between components or ideally have the left side of one phono preamp and the right side of another playing at the same time. I have built the equipment to do this. A little more thought was required for low level signals. Yes, I have done this and will do so at the RMAF. I also have a “control group” of listeners who do not know what has changed. I do not tell them there has been a change. They evaluate and give me their thoughts.

But in the end, the question of what is truth arises. Each listener will have their own preferences. Each user will be subject to their own experiences that color their truth. I have been in many fine concert halls. In short they to not all sound the same. Even changes in altitude and humidity change the way an instrument sounds. Individual experiences lead to a need for many manufacturers who develop a sound and capabilities that match the listener’s needs. Reviewer subjective comments then can help the listener discern the sound of the component. Numbers tell only part of the story.

Ortofan's picture

Please, have either yourself, or one of your "control group" of listeners, take Alan Shaw’s challenge and let us know who prevails.

Michael Fremer's picture
Alan Shaw's challenge is as ridiculous as his entire post about vinyl. He purports "scientificism" but doesn't say what his turntable was, or his cartridge or his phono preamp and then he runs off at the mouth about digital and blah blah blah.
SwanSong's picture

Either I was not clear enough in my response or you have selective reading. I clearly stated that I have instantaneous A/B switching. I level match my components. I have several other manufacturers phono preamps. They do not sound like each other or mine. I also stated this will be demonstrated at the RMAF.

Ortofan's picture

...in Mr. Shaw's presence and with his switching equipment? If not, you can't yet claim to have met the challenge and succeeded. Further, were you just doing A/B switching or were you performing an A/B/X comparison?

Michael Fremer's picture
Are a total waste of time and effort. They produce stupid results because using them to judge audio gear or to "prove" that cables all sound the same etc. is a farce.
Ortofan's picture

Are you suggesting that the experimental method contains some flaw?
If so, what would that be?

Michael Fremer's picture
The experimental methodology known as A/B/X testing as applied to audio results in stupid results. Therefore I think it's a stupid methodology when applied to audio. I was challenged to such a test at an AES by someone who asserted that all amplifiers that measure the same sound the same. I took the challenge. He arranged the test at a national AES. I got 5 of 5 identifications correct. John Atkinson got 4 of 5 correct. The overall results were that audio engineers at the AES could not identify what they were hearing even though one amp was a VTL 300 tube amp and one was a Crown DC-300 solid state amp. These amps do not measure the same and certainly do not sound the same! Yet audio engineers were not able to distinguish one from the other in the test. Why? Unless one is experienced at such tests confusion reigns, often borne of anxiety. A/B/X audio tests generally produce CONFUSION and stupid results. I did very well in another blind test at Harman International's speaker testing facility so I am good at this because I am well-experienced at it. But being well-experienced means I can listen at home over long term and get a much better idea of how something sounds. I am fully capable of filtering out my "prejudices" and don't prefer gold face plates to silver ones (and all of the rest of the bullshit written that attempts to discount actually LISTENING to something over a long period of time rather than in quick A/B/X tests that sow nothing but confusion. BTW: in that AES test, because my result was at odds with the "average" clueless engineer's result, I was declared a "lucky coin" and my result was tossed! So the guy challenged me, I passed his challenge and my result was tossed and I was declared a "lucky coin" so that this guy could continue to grind his axe. So, yes, A/B/X testing for audio gear is STUPID. It leads to STUPID results. Music requires long term listening. Assessing the sound quality of a piece of audio gear (or a cable) requires a long term listening approach. To sum it up: if a methodology results in STUPID results, the methodology is STUPID. The same goes in many cases for measurements. Vinyl does not measure as well as CD in great part because the measurements are based on the old parameters. Based on those CD "wins" particularly in terms of flat response from 20Hz to 20kHz. Unfortunately it doesn't sound very good and especially didn't in 1984. Yet those lead around by measurements instead of their ears insisted it sounded better because it measured better. Later measurement techniques discovered the issued: ringing and pre-ringing filters, jitter, etc. The EARS RULE. And the brain is not as stupid as some claim in the "measurements are king" world. Believe me, we survived because of our innate abilities not because we did "A/B/X" tests on tigers stalking us.
Ortofan's picture

Based upon the experience you related with the AES amplifier comparison, one can understand how you might be disabused of the validity of A/B/X testing. Apparently the expectation of the audio engineers at the AES, who it is said could not identify what they were hearing, was that no one else would be able to distinguish between the two amplifiers, either. What was “stupid” was not the test itself but rather the decision to discard the results from MF, simply because they appeared to be an outlier. It seems too obvious to suggest the test should have been repeated, either to confirm the acuity of MF’s discrimination abilities or to ascertain if there could have been some fault with the test apparatus that aided in the identification of A or B. Also, although the switching is instantaneous, there need be no constraint on the length of time that the test subject can spend listening to A or B or X. Frankly, I tend to doubt that no one would have been able to distinguish between two amplifiers as disparate as the Crown and the VTL 300. It also seems plausible that some measurable parameter could account for a perceived audible difference between those two units.

Drawing comparisons between digital discs, analog discs and speakers into the discussion isn’t particularly relevant since the product under consideration is an amplifier.

Back on track, the manufacturer asserts that some aspects which define the “sound” of his product defy measurement and can only be discerned by a “listening panel.” At what point does a given parameter become immeasurable and detectable only by listening? In fact, instead of trying to specify a suite of tests, one of which might correlate to any specific audible characteristic, let’s go back to the Hafler SWDT test and the Carver transfer function nulls. Hafler was able to get his XL-280 amp to null with a “straight wire” down to about -70dB. Perhaps coincidentally, Carver was able to achieve a null (or match) of about -70dB between his solid-state amplifier and one using vacuum tubes. It was subsequently reported that listeners at Stereophile were unable to distinguish between that pair of amplifiers.

The question for this (or any) manufacturer is, as arbiter of accuracy, how well does your product perform in the SWDT test? Likewise, how close is the null between the amplifiers being compared? If the answer is -70db or greater, can the “listening panel” still reliably identify either of those devices in an A/B/X evaluation?

SwanSong's picture

Ortofan, from the beginning you have made clear your detest for what you don't understand or does not fit within your limited framework. I know of others who have not posted due to your toxicity. No matter how much effort I spend on educating you, you will not change your opinions. This is a shame because your attitude will discourage other manufacturers from contributing to this forum. I think at this time I have been more the patient with you and it is time you move on and make room for those who have something to contribute. I think I can safely say I know how to wire a switch. Please move on, I will no longer reply to your insults.

Michael Fremer's picture
To stop taking his bait. Never before has any piece of gear reviewed here or in Stereophile that I can recollect been subjected to such skeptical scrutiny nor has any designer of what is obviously a legitimately designed and engineered product been so abused. I don't know what is Ortofans's game but I don't like it.
SwanSong's picture

Thank you Michael. If there are legitimate questions and a desire to understand I am happy to reply and to take the opportunity to expound upon related issues so that my learnings can be shared. Building a ground up design takes a lot of work. Add to that the starting of a business. . . .

I really appreciate that Michael agreed to review my product. I really appreciated the positive and constructive comments. I also appreciated "Al" providing some color. His experiences have reinforced my belief that it is not a single component that makes a good system. It is the sum of the components and the signal chain must start with a good source. Once music is lost it cannot be recovered later in the chain.

I am looking forward to Denver and am working on a good show for the attendees. Along with getting ready for Denver, I am also moving my operations. I will have a much larger listening room, development room, machine shop, and assembly facilities.

X