Sonic Frontiers SFS-80 Review (Vintage Tube Amplifier)

[ObjectiveSubjectivist]

Looking better than 90% gear nowadays.

 

[Gord]

I have a bit of time each week to try measuring stuff. Since my DAC isn’t so straightforward I can do a series of tests at different voltages and then increase in a stepwise fashion. That said, tubes are a really bad idea when you need lots of power.

I made this chart in a different thread where I show my work for the calculations which sort of shows in both directions why tubes work well for moderate listening levels and why “even though a watt gets you a lot of output, you still want 300-600W juggernaut amplifiers if you are playing music with bombastic peaks.

Speaker	dB at 0.05W (5 vs 15 ft)	dB at 0.5W (5 vs 15 ft)	dB at 5W (5 vs 15 ft)	dB at 50W (5 vs 15 ft)
Revel PerformaBe F328Be	80.3 / 70.8	90.3 / 80.8	100.3 / 90.8	110.3 / 100.8
Focal Aria 906	76.3 / 66.8	86.3 / 76.8	96.3 / 86.8	106.3 / 96.8
Elac Uni-Fi 2.0 UB52	72.3 / 62.8	82.3 / 72.8	92.3 / 82.8	102.3 / 92.8

Peak advertised power handling

Speaker	Peak Advertised Power Handling	In room Peak SPL (5 vs 15 ft)
Revel PerformaBe F328Be	300 watts	118.1 / 108.6
Focal Aria 906	120 watts	110.1 / 100.6
Elac Uni-Fi 2.0 UB52	140 watts	106.8 / 97.3

I don’t know if this actually measures better than a MC275/MC2102. From what I understand the 12ax7/12at7 combo is even better than the 6DJ8. Then the question is how good something like an Audio Research, current production amp can perform.

Definitely! Searching USPTO, I cannot find evidence that they ever trademarked the name and Sega has trademarked the name for their latest Sonic the Hedgehog game.



I’d love to hear a 300B system. It sounds like the power limitations are real, but with a matching high efficiency speaker or appropriate listening level, it’s supposed to be very impressive.

You should put your gear on the test bench too. I have always been told that higher biasing is better too, but I subjectively thought 40 mA was better but I also told myself that it was probably the bias of knowing that these tubes have a finite lifespan and have no reliable replacement.

But at least in this case, the lower bias measures better! The 6550A should easily handle even more power than 50mA. Then you have all the hand waving that EL34’s sound better than 6550a’s if you don’t need power, but EL34’s are typically biased at 40 mA.

I don’t use this tube amp in any of my primary systems but it might be interesting to see if 42 mA, 45 mA, or 47 mA ends up having the “best” results. I just have two datapoints.

No matter how good your 300B setup may be, it would be nice to see if there is a sweet spot for your bias. Certainly, the new current production 300B’s from Western Electric would be interesting to measure.

My speakers are DIY Soundgroup 'Tempests". they're about 98 dbl efficient, so a pretty easy drive. sorry to say the kits aren't available anymore, but the parts might be. They were designed by Jeff Bagby and he's no longer with us, so I feel kinda fond of them. Iwas biasing to around 62 ma. My wife, who has the same setup as me, loaned me her WE 300Bs to break in, so I went on the WE site and found out they could be biased to 100 ma. I went to 75ma and they stated to sing. Previously they were kinda dull sounding and all midrange. I actually thought my Gold Lions had more life. She's since taken them back

 

[Nugget_]

Thank you for your review. I read on ASR for some time now, but i created a account now, since i had to reply to the following, because noone else did.

Remember that my measurements are as much as 9 dB worse than what Amir can get with his carefully designed setup. Still, my 5W SINAD of 74 dB is pretty impressive. It means I'm beating the Lyngdorf TDAI-3400, Anthem MRX520, Pioneer Elite VSX-LX504, and every tube amp tested to date at ASR. I don't know if we'll really get the full 9 dB when measured on a proper setup, but I'm very confident that this unit would break past the mean amplifier SINAD of 78. If this amp can break into the 80 dB range, that really would be saying something about the types of audio gear being passed off today as "audiophile premium products" wouldn't?

The db scale is logarithmic so it doesn't work like you describe it here. With a better measuring setup you wouldn't gain anything here. You can calculate 1 SINAD from 2 different SINADS. You will see that as a rule of thumb, everything above 10db difference between two sources doesn't really change anything compared to the worse measuring source. So based only on SINAD at best measuring level, your measurement rig doesn't add anything to your measurements up until aprox. 100db SINAD.


BTW, the best measuring tube amplifier at least i have ever seen is a unity coupled design based loosely on mcintosh mc60, designed by technical university berlin with the only goal to be the best measuring tube amplifier. They got to 88db at 0.56W, -82db at 9W and -71,5 at 44W.

 

[Redver]

Hi Amir, just wondering what my Conrad Johnson main amp would test like after all these years? A MV60 SE too expensive to post from Australia to yourself, have you ever tested one?

Keep up the fantastic work mate.

Kind regards,

Nick

 

[eeMGee]

Great job again, @GXAlan! You are doing what I didn't dare to try so far: measuring amps with the Cosmos!

As many pointed out in my former reviews, maybe REW should be more accurate for THD+N figures. I personally use :
- REW for 1Khz THD+N and DR.
- Multitone for... Multitone (yup!) and Jitter (the latter not being a thing for amps)
- RMAA PRO for regular tests: FR, Noise floor, SMPTE IMD and Crosstalk.

If Multitone is used, I recommend to display FFT results in dBr - otherwise THD value is not correct (is calculated relative to 0 dBFS, as the SW author has explained).

 

[AdamG]

Pretty soon I can stop doing reviews.... Promoting to home page.

You started this Train rolling. Just sit back and admire the fruits of your inspiration. Snowball effect! Every measurement adds to the collective knowledge library. Great job @GXAlan.

 

[Matias]

@GXAlan if you intend to keep on reviewing amps, you can set up a fundraiser and users here can help buying 4 ohms resistance and a DAC.

Edit: same for @VintageFlanker for resistors.

 

[Gorgonzola]

*Raises hand.

I'm in the "if I wanted solid state sound I'd buy a solid state amp" camp.

(Actually, I did also buy a solid state pre-amp. I like it and my tube preamp).

Beautiful preamp!

Well I think that is essentially reason Sonic Frontiers folded -- just foo many people felt that way.

Subjectively of course, I do find that that my SF Line 1 does add just a tincture of that tube je ne sais quoi. but not much. Some day I'd like to try a nice "tube-like" preamp such as a Conrad Johnson that adds a generous dollop of pleasant 2nd order distortion but financial considerations make it low priority.

 

[Jim Shaw]

Not unlike incense and holy water, the enthusiasm for tube equipment mystifies me. But, of course, it seems harmless, albeit distracting. For me, I expect vacuum tube devices to go the way of whale-oil lamps. Curiosities for certain, but no longer suitable for everyday reading and commerce.

Well, carry on. It's a bit like organizing a St Patrick's Day parade: useless but basically inoffensive. And a distraction from the mundane modern-day devices that run unattended for decades and which we now take for granted.

Would you ride on a 20-story elevator powered by vacuum tubes? Even one with a full-time tech measuring grid biases

 

[RichT]

Good work GXAlan, nice performance, this was really great stuff. Regards

 

[fpitas]

Not unlike incense and holy water, the enthusiasm for tube equipment mystifies me. But, of course, it seems harmless, albeit distracting. For me, I expect vacuum tube devices to go the way of whale-oil lamps. Curiosities for certain, but no longer suitable for everyday reading and commerce.

Well, carry on. It's a bit like organizing a St Patrick's Day parade: useless but basically inoffensive. And a distraction from the mundane modern-day devices that run unattended for decades and which we now take for granted.

Would you ride on a 20-story elevator powered by vacuum tubes? Even one with a full-time tech measuring grid biases

Only if it used 6SN7 dual triodes in cascode.

Yeah, they aren't going away. The modern trend to purposely make them distortion boxes annoys me, but it's not my ears or wallet affected.

 

[JayGilb]

Great review - Thanks for your time and effort.

 

[GXAlan]

So why is this tube amp particularly "good" otoh?

I am not sure if you are asking relative to other tubes or relative to something like a modern solid state amp.

Relative to tubes, this in fact is the highest measuring tube amp that anyone has tested. It’s sort of impressive since the roots of the SFS-80 are in hobbyists who took the Dynaco design with expired patents and tweaked it. Adding to the social interest is the fact that this was from the same company that is known as Anthem. Tubes sound different. Whether that difference is favorable or not is a personal preference.

Relative to solid state, it’s a loaded question. I think my Marantz PM-10 and PM-11s2 are better products. They are better products because I expect them to be more reliable, they don’t need to warm up or heat up the room and have power when I need them.

But I think this is the difference from those German Christmas Pyramids powered by candles versus one that was powered by a battery. Tube amps are decorative objects just as much as they are functional.

So based only on SINAD at best measuring level, your measurement rig doesn't add anything to your measurements up until aprox. 100db SINAD.

Thanks. I was basing that comment off the Marantz PM-90 that Amir tested and then I tested. I get a measurement that is 8-9 dB worse. As you say, it’s hard to estimate what the real difference is because the math and the reality are different. With this tube amp, I don’t see an easy way to actually ground the E1DA to tube amp so I grounded it to the preamp.

BTW, the best measuring tube amplifier at least i have ever seen is a unity coupled design based loosely on mcintosh mc60, designed by technical university berlin with the only goal to be the best measuring tube amplifier. They got to 88db at 0.56W, -82db at 9W and -71,5 at 44W.

Wow! Did they publish the paper anywhere? It would be great to see the tubes and bias that was applied.

@GXAlan if you intend to keep on reviewing amps, you can set up a fundraiser and users here can help buying 4 ohms resistance and a DAC.

Edit: same for @VintageFlanker for resistors.

I am just a pure hobbyist and don’t think I could do a lot of reviews other than measure the collection I have amassed over decades. Just a handful of items left to measure.

What WOULD be helpful is to create a kit and a standardized test protocol for testing line level items. Imagine having a library of a three of four E1DA cosmos/Dx7Pro+ which we could loan out to different parts of the globe. Much easier to ship some cables, and two small DACs.

Logistics aren’t easy to track who is currently borrowing the unit and you probably should have something like a $150 deposit, shipping fee, and rental fee of $20 to borrow the unit but it would be like Lensrentals.com except for a E1DA Cosmos/DX7Pro+ combo and some standardized cables. For example, instead of bare wires to the the XLR, we can built a breakout box with standard binding posts and a grounding lug.

If it works well, adding resistors to the package could be an option. The challenge is figuring out if someone in this readership is willing to take the time to build the database or manually manage it. It’s easier to donate money than time

 

[aschen]

sonic frontiers was a cool company, also remember the parts connection kits and parts. Always wanted to order an assmbelage dac when I was in college.

Sonic frontiers stuff always looked so nice and was semi reasonably priced compared to its competition. You can pick up nice pieces affordably these days if you are into nostalgia.

 

[57gold]

Interesting blast from the past.

I owned a pair of these Power 3 monsters for a while, along with a two box all tube SF preamp, model number escapes me. Play guitar and surrounded by vacuum tube amps and effects, so after hearing my speakers, Aerial 10Ts driven by a Audio Research tube amp, decided to try the SF gear. Ended substituting the Russian small signal tubes with Siemens and the 6550s with KT88s. even had some factory upgraded caps installed, all fine for a while...then one of the power tubes blew taking out some resistors. After shipping back to Canada and many weeks with no music, sold the SF gear and replaced it with a McCormack DNA-500. Turns out the DNA-500 was a better amp for my speakers, better bass control, very dynamic and no loss of mid range sweetness. Still in service decades later.

However, teenagers of all ages were in awe of those Power 3s lit up and making music...

Still have boxes of match quads of the KT88s. Thought I might get a guitar amp that might use them. My friends at Komet Amps did create a great 2 X 6550/KT88 amp, the Silver Cloud. But @80 watts clean, pushed to break up through efficient guitar speakers like Celestions, it would get me thrown out of pretty much any bar other than a death metal club and I stop at 1960-70s rock.

FWIW, they were well built, great packing materials, never got bent in shipping to and from me...weighed a S#!+load!

 

[beefkabob]

2 watts of clean power. Yay!

 

[MattHooper]

Not unlike incense and holy water, the enthusiasm for tube equipment mystifies me. But, of course, it seems harmless, albeit distracting. For me, I expect vacuum tube devices to go the way of whale-oil lamps.

If tube amps were actually going away they would have done so in the 70's and never come back. It's been around 50 years and they are still here, so I think your expectation may be misplaced

Curiosities for certain, but no longer suitable for everyday reading and commerce.

Understandable as your personal point of view. I don't read much about old sports cars. Don't know why anyone likes 'em.

Would you ride on a 20-story elevator powered by vacuum tubes? Even one with a full-time tech measuring grid biases

That's a bit like asking 'would you ride an elevator powered by a Benchmark AHB2 power amp'? Obviously not. Wrong use case.

 

[Nugget_]

Wow! Did they publish the paper anywhere? It would be great to see the tubes and bias that was applied.

It's not online anymore, but downloadable, at this third party website: https://nanopdf.com/download/der-unity-coupled-verstrker-nach-mcintosh-mc_pdf
But it's in german and pretty long, but also with schematics and everything.

 

[GXAlan]

It isn't really. You'll get used to it.

But as you are testing amplifiers, we really need to know what they can achieve in terms of power output.

I am sure I will work my way up there. I think you are a little biased because if you break something, you'll be able to say "oh, it's just ____ that needs to be fixed" Whereas, my electronics repair skills are high-school level soldering/de-soldering, etc.

Sonic frontiers stuff always looked so nice and was semi reasonably priced compared to its competition. You can pick up nice pieces affordably these days if you are into nostalgia.

It looks very nice, but the sheet metal and tube cage metal is pretty thin/flimsy compared to McIntosh. It's a smart move overall because it saves money and doesn't affect the sound quality but it makes shipping the units extra hard because it's so disproportionately heavy and I can see the transformers denting/deforming the sheet metal and tube cage against its own weight.

On the other hand, it also means you can get them for a good deal if there is one local to you -- at least until people start reading this review.

 

[Jim Shaw]

Matt Hooper: Of course...it could always be my imagination.

So mote it be. (That's a good "use case.")