RETRO-MEASURE: Pioneer SX-880 receiver (1978). And the amplifier "THD Wars" of the '70s?

Back in the 1970's, Pioneer among various Japanese brands like Sansui, Sony, Yamaha, and Technics brought popular, budget-conscious audio gear to the masses. With the prices of solid state components dropping, these amplifiers powered many college parties and brought music into the homes of many audiophiles of that generation.

Check out this ad from back in the day featuring the Pioneer SX-780 (45Wpc):

Nice to see that the idea of "high-fidelity" featured prominently rather than some nebulous idea of "high end" audio. Today, let's measure and discuss the sound of the last in the Pioneer SX-800 series of receivers, the Pioneer SX-880 featuring up to 60Wpc into 8Ω. The SX-880 came out in 1978 with a retail price around US$425. Thanks to a large part to central government mismanagement... err... inflation, this would be about US$1700 2020 dollars which would buy a rather advanced multichannel receiver like the Denon AVR-X4500H these days.

It's a great looking ~27lb/12kg unit that's very well built. It's got the classic look with silver knobs and switches, power meters, AM/FM signal strength meter, and an FM tuning meter to help with precise radio channel reception. Of course, all this is made even more classic with the amber illumination and the wood veneer up top and sides.

There is a nice selection of functions available including the "loudness" switch for when you're listening at low levels to emphasize bass and treble, separate bass and treble EQ knobs, and of course a balance control. I noticed that the knobs were a little noisy due to age but otherwise the functions all worked and the FM receiver sounded great.

After wiping a bit of dust off, this particular unit literally looks "mint" on the outside. The wood grain veneer appears immaculate. It was originally bought in the late 1970's (~1979) by a previous owner from Inner Sound in Portland, Oregon with a label on the back I presume of the store's previous address and phone number (here's their current website).

There is a normal assortment of RCA inputs including of course phono in. I'll be using the AUX input for my testing today since I'm mostly interested in the quality of the amplification.

Notice back in the '70-'80s before ubiquitous remote controls, many receivers also had convenience AC outlets, typically switched with the main unit so you can plug in your turntable for example and everything turns on when the receiver is activated. Speaker cable terminals are the spring compression variety for bare wires with a convenient push release button. I suppose 12AWG thick wires were not popular back then, at best I was able to comfortably squeeze 14AWG wire into those holes without fraying the strands. I'm glad that most decent amps these days use universal binding posts with the convenience of banana plugs.

Since I didn't want to open this up to poke around (borrowed after all), visual inspection from the outside looks good. The components I can see appear to be clean with no evidence of bulging capacitors or any unsightly burn marks.

Based on the information on HiFiEngine and Pioneer's manual, this is a 60Wpc power amp continuous into 8Ω "with no more than 0.05% total harmonic distortion", is capable of driving 4Ω loads, and has an advertised damping factor of 30. The detailed specs in the manual also lists THD at 1W into 8Ω as 0.03%.

I. Basic Characteristics

Okay, let's start with the usual basic parameters. At full volume the voltage gain from this integrated amp is up around the +43-44dB range:

Voltage gain: Left   = +43.5dB
                       Right = +44.1dB

As you can see, there's an imbalance between the channels of 0.6dB which is higher than what I'd like to see. I actually noticed this when listening to the receiver before it got on my test bench with the stereo image a little shifted to the right side; thankfully with the channel balance knob, I was able to subjectively get the balance mid-line. We'll talk more about the listening down below.

Given the age of this machine, let's jump ahead to look at the square wave to compare the 2 channels at 2V into 4Ω to see if there are issues:

Hmmm, looks like indeed the left channel is weaker than the right by 0.67dB at around 2V (varies between 0.4-0.7dB depending on how high I turn the gain up). Furthermore, it looks like the left channel is symmetrically "tilted" down suggesting some mild low frequency attenuation or phase shift. Given what I see here then, I'll focus on measurements of the right channel which is likely representative of the true performance of this receiver back in the day.

Given how good the right channel square wave looks, the device should have a well extended frequency response.

Next, let's look at the damping factor:

Remember that I'm measuring this into 4Ω. Overall it looks good with a 7-point average damping factor of 18.4 from 20Hz to 20kHz (into 8Ω the specs say 30).

As alluded to above with the square wave, I am expecting a well extended frequency response and with the damping factor at around 20 into 4Ω, control of the Sony SS-H1600 (8Ω nominal) bookshelf I use as my comparison speaker should be quite good:

As you can see, there is indeed a channel imbalance between the left and right with the right one louder (this is with 1V into 4Ω).

With a decent damping factor (compared to something like the Pass ACA 1.1 with damping factor of only 3), frequency response into the Sony's reactive load is well controlled.

Here's a peek at the phase curve - nice and flat, nothing surprising...

Since this is a stereo receiver, I checked the cross-talk using a simple 300Hz/4kHz tone in either the right or left channel and found an average of -63dB channel separation at 1W into 4Ω; certainly this is good enough for music playback. Remember, back in the '70s to early '80s you had vinyl, cassette tapes, reel-to-reel, and FM radio as stereo sources so even approaching 50dB separation would be excellent already in the home using the best gear and best recordings.

II. Single-Tone Harmonic Distortion and Noise

As usual, let's start with a look at harmonic distortion vs. frequency at 2V/1W into 4Ω using REW's frequency step function. I'm using the right channel:

The cursor is at 1kHz and we can see that THD+N is around -71dB/0.028% at that point. This would be in line with the Pioneer spec of <0.03% at 1W into 8Ω (we're using a more challenging 4Ω load here of course). The harmonic distortion does show some frequency dependence with a gradual increase in the higher frequencies which is not uncommon. Notice that below 600Hz, the 3rd harmonic is highest and above that, the 2nd harmonic dominates although not leading by a huge amount.

Noise floor (brown) looks good hovering around -100dB below fundamental.

Let's have a peek then at 1kHz harmonic distortion across various power levels to see the FFT details with both channels driven (again, we'll look at the better-performing right channel):

If we scan the results in that matrix of FFTs, we see that the receiver is very well behaved. THD+N remains better than -60dB (<0.1%) all the way to 20V/100W into 4Ω. Clearly by 23V/132W, the device has started clipping with a large jump in distortions and noise.

If we graph those numbers as THD+N vs. voltage output for our distortion-power curve with both channels driven:

That's very good, it looks like the amp is able to deliver good quality <0.1% THD+N reproduction up until it starts clipping (very quickly I might add!) at just over 22V or 121W. Remember, it's rated as 60Wpc into 8Ω, so it's great to see that the power supply is able to indeed deliver twice the watts into half the load.

III. Multi-Tone Testing: Intermodulation Distortion and Triple-Tone TD+N

Alright, let's now have a look at some of the intermodulation tones at 2V and 10V into 4Ω of the right channel, again, both channels driven with the caveat that the left seems to be sub-optimal:

Overall, not bad at all. IMD results were excellent at the 2V level around -80dB across the various IMD signals, and at 10V we're seeing a larger variation from -80 to -60dB, depending on the test variant used.

Here's the "transient intermodulation" (TIM) distortion test result for this vintage receiver; again both channels driven:

Again, very good. At a 2V output level, there are essentially no distortion sidebands on either side of the 12kHz sine tone. By 10V, we start seeing the anomalies but at low levels around -89dB which is not bad, but not as good as the 2011 Onkyo TX-NR1009 receiver as a more modern consumer-level product. Remember, these are late '70s-era transistors.

Finally, here's the Triple-Tone Total Distortion and Noise graph at my standard 2Vrms level into 4Ω:

I used the channel balance control so they're both ~2V output into 4Ω. As noted before, the left channel is not functioning at the same quality as the right. Based on what I have seen through these measurements, I suspect what I'm finding with the right channel is likely representative of the receiver's true performance (about -74dB) and that the left channel has deteriorated over the years (TD+N reduced to -64dB).

IV. Wideband Noise

Typically, I show the 1kHz square wave tracing here but given my concern with channel imbalance, I've already demonstrated the anomaly above (I).

The wideband FFT is not of high resolution but good enough to make sure we don't see any major issue with excess noise in frequencies >100kHz.

Looks fine up to 600kHz as one would expect with good Class AB amplifiers, no significant noise components unlike Class D amps discussed before.

V. Impressions and Conclusions

From an objective performance perspective, the results are certainly very respectable after all these decades!

Remember that this device is 42 years old. I was in grade 1 when this machine was built. As such, I suppose it's forgivable that the owner of this receiver did not notice the <1dB channel imbalance and higher distortion of the left side. For years, this receiver has been used as an amplifier/FM radio for casual music and TV playback in an asymmetrical living room.

Using the data from the right channel which I believe is representative of the true quality of this device back in the day, here's my summary AMOAR Score:

If we look at each component of the score, we don't see any major issues. In fact, if you compare this to the 2011 Onkyo TX-NR1009 receiver, you'll see that in fact many of the results are similar. This old Pioneer has slightly higher damping factor than the Onkyo. The triple-tone distortion factor are within 1dB of each other. The much newer Onkyo is capable of more power before hitting 0.1%THD+N and of course has all kinds of digital features, capable of multichannel playback, various surround decoding, and DSP capabilities. While features have been added, objectively it would be hard to argue that sound quality as an amplifier has changed substantially between these two consumer-oriented receivers.

Over the years, there has been talk about Slew-Induced Distortion (SID), specifically the type we call Transient Intermodulation Distortion (TIM/TID) also known as Dynamic Intermodulation Distortion (DIM). This was said to be a problem in the past with slow transistors. However, by 1978 with this amplifier, I'm not seeing any problems with the TIM test signal (similar to a 1975 Tektronix recipe using digital signal generation but more challenging with 1kHz square, 12kHz sine and 96kHz bandwidth). In fact, the amplifier I've seen showing the highest TIM was the Pass ACA 1.1 compared to other devices at 2V into 4Ω. We know that the ACA utilizes low feedback and has a commensurate low damping factor. There has been discussion recently by Bruno Putzeys that TIM is reduced by negative feedback - not the other way around as per audiophile lore about feedback being "bad" for TIM. As usual, be careful with all kinds of audiophile myths out there unless "fact checked".

Subjective listening of this old Pioneer I felt was consistent with the objective results. On my system playing music off Roon to the Oppo UDP-205 as network DAC to Emotiva XSP-1 preamp to the Pioneer SX-880 amp to the Paradigm Signature S8 v.3 speakers (I turned the subs and DSP off), I was able to hear the slight channel imbalance as noted in Part I above before putting this on the test bench. This was remedied with a ~15° turn of the balance knob to bias the left channel. By the way, the "mono" switch is convenient for this - just flick it to "mono" and turn the balance knob to center the sonic image.

While I know from the objective results after testing that the left channel had more distortion than the right, I honestly was not able to hear a difference while enjoying music (other than the channel imbalance). Remember, when we listen to music at normal levels at home, most of the time the amplifier doesn't need to produce more than 1W unless you have extremely inefficient speakers.

Perhaps it's just the bias of knowing that this machine hails from the '70s, I could not help but have the thought run through my mind that what I was hearing was basically the epitome of the "solid state hi-fi sound". Tube-amp lovers will call this "clinical", and "lean" sounding (like most solid state, right?). Not exactly a great match for '80s recordings like the first press of OMD's The Best of OMD (1988) with synth tracks like "Enola Gay" or "Souvenir" sounding a bit more sizzly than I prefer. I think some dramatic audiophiles may suggest that this kind of sound would "have me running out the room". No need to be so histrionic, my friends! :-)

"Regular Pleasures" from Patricia Barber's Verse demonstrated that this amplifier has no issues with the low end. Subjectively I thought the bass wasn't as tight as my reference Hypex NC252MP with volume pushed up; it's hard to know unless I have the other amp running in parallel with A/B switching. With the balance control tuned in, soundstage was excellent and Ms. Barber sounded like she was singing in the room seemingly a bit deeper than the 10' sitting position from my couch to the plane of the speakers. Natural instruments like the muted trumpet on the track sounded realistic and "popped" out nicely with a sense of space around it.

More classic female vocals like Sheila Jordan's "I'm a Fool to Want You" (Portrait of Sheila Jordan) extended that impression of this being a "hi-fi" amplifier playing what it's fed without adding its own editorialized "color". Wonderful details on this sparsely recorded 1963 track rendered cleanly although one can easily tell that this was not a modern digital recording (certainly no need for hi-res!). In particular, I appreciated the lack of any excess sibilance in Ms. Jordan's voice. Likewise, the unforgettable Louis Armstrong's "St. James Infirmary" (Satchmo Plays King Oliver) was simply divine - a track like that just subjectively sounded "right" played through vintage gear. :-)

Modern recordings like Ramin Djawadi's Westworld: Music from the HBO Series, Season 1 (2016) came through nicely as well. Expansive soundstage on "Nitro Heist", appropriate bass "weight" and balance with the highs.

Very enjoyable 3 evenings of listening to this amp; I didn't experience any fatigue listening for 2-3 hours each time.

Bottom line: The vintage Pioneer SX-880 appears to be a good performer - a reminder of the high-fidelity one can get from gear at least dating back to the late 1970s. While objectively I can measure that the left channel is not as clean on this unit, at normal listening levels provided that I compensate for the channel imbalance, I didn't think the sound was significantly compromised. Overall, the sound was clean and akin to the previously-measured Onkyo TX-NR1009. Honestly, I would hate to blind-test this against something like the objectively better Hypex NC252MP that I use as reference! I bet it would be very difficult at normal playback volumes; perhaps the only thing I would notice is if I pushed the volume up, the higher noise level with the Pioneer might be evident.

I will suggest to the owner that he tracks down the cause of the left channel imbalance (audible) and distortion (much less audible). It's worth cleaning out the innards a bit, use some Deoxit on the knobs and switches, inspect the old electrolytic capacitors closely. I think it goes without saying that when a machine is as old as I am, it would not be surprising to find some wear-and-tear compared to when it was young :-).

Unless you're buying used at a real bargain, make sure to get things checked out thoroughly.

BTW I think it's good to take the opportunity to have a listen to some of the vintage gear once awhile. Just like my "retro-measures" of the 1994 Sony LasterDisc as CD player or the old 2002 Lynx PCI audio card, it helps us maintain perspective on subjective sound quality correlated to objective results. Certainly digital/computer technology has changed the way we access audio in very substantial ways. However, despite all the dramatic "Best sound ever!" headlines every few months on the front covers of magazines, one might be surprised at actually how little difference there has been over the decades with mature products like the sound of power amplifiers at normal listening levels.

--------------------

Every once awhile on discussion forums, I run into comments like this one referring to stories from the past used to suggest that measurements / specifications are somehow of little use when it comes to sound quality:

"It is about much more than distortion. The lowest-distortion-possible race of the '70s created many amplifiers with vanishing levels of distortion, which nonetheless sounded like crap."

To be honest, I don't remember much of the '70s because I was only a kid back then. Stereophile talked about "The THD Wars" of the "1960s and early 1970s" for example. Interestingly, it's hard to find much information about such a "war" on distortion when I dug around. While it has been documented that the consumer receiver brands were battling over market share, I'm just not sure "THD" specs were a major selling point. In fact, looking at a few ads from the 1970s through the archives of Stereo Review, there did seem to be the "power wars" with devices like the 160Wpc Pioneer SX-1250 among other receivers like the 120Wpc Kenwood KR-9400. However the lowest THD measurement used in advertising I saw from the era was this one from Yamaha in 1979:

Okay, 0.02% THD (-74dB) is pretty low but certainly not unrealistic for what they consider the "leading" number. I don't know how good this receiver is and even if there is a bit of an exaggeration in the spec, it can't be all that off the mark, right? Many receivers over the decades would achieve this without great fanfare. As expected, I see significantly better results from my Emotiva monoblock and Hypex Class D amps. So where are numbers like "0.0001% distortion" in the 1970's I was led to believe existed? It appears the answer is nowhere.

I do remember the 1980's and later with false claims like grossly exaggerated advertised power numbers such as "1000W PMPO" from little boomboxes as I was "coming of age". Stuff like this still quite common to see:

But that's not serious hi-fi gear though.

It wasn't actually until the 1980's that we saw amps like the Yamaha MX-10000 with advertised "0.0005%" THD (-106dB) released in 1987. Or the Onkyo M-510 Grand Integra with "0.005%" THD (-86dB) released around 1984. IMO these amps still look fantastic and I think the meticulous construction of those devices remains enviable. Whether those THD numbers hold up to modern scrutiny is unclear as the numbers lack context. Is that "0.0005%" THD at 1kHz? At what output power/voltage? Into what load (for all we know it could be a 16Ω resistor)? Up to what harmonic did they measure (typically I report up to the standard 9th harmonic)? Remember that it's nice to also know the noise level, so even with pristine THD results posted, the THD+N might be significantly higher.

From what I've seen, perhaps there's no need to complain so much about the '70s and the supposed "THD Wars". It all seems rather tame from what I can tell, and nothing like snake oil hype with ridiculous cables, fuses, audiophile ethernet switches, unsubstantiated USB tweaks and such in the "high end" today and in the last few decades...

I agree with the forum comment quoted above though that sound quality is indeed "more than distortion". There certainly are perceptual preferences as well which are psychological and idiosyncratic to each of us as alluded to in the Pass ACA post a few weeks back.

Having said this, I have yet to find a component that genuinely measured well but "sounds like crap"! I've asked Internet audiophile denizens about such a beast for years. But like the Loch Ness Monster, Yeti and Sasquatch, I have yet to see plausible evidence of these creatures. Recently, there was a claimed sighting regarding the Topping D30 which was refuted by Greg Dunn.

Such is the nature of many (even most) audiophile claims and beliefs. The moment you ask for clarification, look for concrete examples, or test something for yourself, you quickly realize that much of it had either been poorly remembered at best, or worse... mythical yet dogmatically embedded into the subjectivist audiophile "culture".

Safety and health to you and yours. Enjoy the music as we enter the latter half of August.

PS: I just saw this video from Paul McGowan of PS Audio as I was finishing off the write-up... I'm a bit concerned about his generalization that today's amps sound significantly better, and calling brands like Pioneer, Kenwood "trash" and "dreadful sounding things" (~2:20). Nonsense! IMO, he's very wrong if he's applying this generalization to higher-fidelity products and likely just speaking as a salesman perpetuating the typical beliefs, promoting the idea of "sounding musical" - whatever that is.