A listening test of two power amps

"@surdo  you have to measure voltage at the outputs using test tones."

Ah, yes, that makes sense. Thanks

Here’s the switching program that I did in the graphical audio scripting language "puredata". You can jump between up to 3 different audio files and a quick equal-power crossfade of 5 milliseconds is applied when the switch occurs, so no click. It only works on Wav or Aiff files, not flac. If anyone wants to try it, please let me know and I’ll give more details.

I prefer to keep that unknown, that’s why I was vague about it. I think it’s better to keep the test as blind as possible, so the emphasis is on which recording represents live performance and live sound more accurately.

I couldn’t disagree more. It’s the music and its reproduction that matters, not the device that is rendering it.

But as I inferred in the 1st message, I will say what the amps are after a time, and hopefully after people share their thoughts on the two recordings. If the amps are known, people might play "guess the amp" and apply any prejudices they might have against either in relation to the recordings. I don’t have the advantage of not knowing what the amps are, unfortunately, but I do believe that both setups have their merits.

If no one cares to respond, that’s fine. I’m slowly coming to my own conclusions and thought I’d share the test on this forum.

Ha ha, yes indeed! 😀

Ah, ok. How would you suggest? In any case, I think I got very close. Here are the stats on the two files:

$ sox e.flac -n stats
Overall Left Right
DC offset -0.000000 -0.000000 -0.000000
Min level -0.368343 -0.368343 -0.368343
Max level 0.444594 0.444594 0.444594
Pk lev dB -7.04 -7.04 -7.04
RMS lev dB -24.57 -24.57 -24.57
RMS Pk dB -16.77 -16.77 -16.77
RMS Tr dB -inf -inf -inf
Crest factor - 7.52 7.52
Flat factor 0.00 0.00 0.00
Pk count 2 2 2
Bit-depth 23/24 23/24 23/24
Num samples 60.1M
Length s 313.128
Scale max 1.000000
Window s 0.050

and

$ sox t.flac -n stats
Overall Left Right
DC offset -0.000000 -0.000000 -0.000000
Min level -0.448870 -0.448870 -0.448870
Max level 0.400301 0.400301 0.400301
Pk lev dB -6.96 -6.96 -6.96
RMS lev dB -24.07 -24.07 -24.07
RMS Pk dB -16.72 -16.72 -16.72
RMS Tr dB -67.62 -67.62 -67.62
Crest factor - 7.17 7.17
Flat factor 0.00 0.00 0.00
Pk count 2 2 2
Bit-depth 23/24 23/24 23/24
Num samples 60.1M
Length s 313.128
Scale max 1.000000
Window s 0.050

So if the RMS peak value is the best measure to judge the maximum loudness, that of e.flac is -16.77 dB. The overall RMS peak value for t.flac is -16.72 dB.

I’m no psychoacoustics expert, but that looks pretty close.

 

I didn’t realise that this would be such a contentious post! :-) Starting with the most recent:

"What is the "real instrument" reference? Do you have an orchestra or maybe a jazz combo around? Maybe you live in an apartment above Carnegie Hall. In any case it’s personal taste that drives the "reference" myth as all recorded music is an illusion filtered through our addled brains."

Yes, the reference is the sound of real unamplified musical instruments, played by musicians. In my case, I know well the sound of a piano. I played violin (still play a bit) and have been around quite a few double basses. I also played drums for a number of years and I still play other instruments. And actually, in the recordings, it’s the snare drum and the cymbals that have helped me detect realism. So those are my references. Anyone else that’s heard an instrument played unamplified can use that as a reference too. When you hear a recording, it might (and often does) sound too bright, for example, and that might be the fault of the stereo that’s playing back the sound or the way the instrument was recorded (with mics) or equalised. And yes, that’s a big issued when you’re trying to compare gear - one piece of equipment might be rendering the rotten recording perfectly, when another smooths out the rough edges. But the listener can tell whether or not the sound appears "natural" through his or her experience of listening to real instruments. I’m sure most people on this forum have listened to instruments played with no amplification.

I don’t think it has anything to do with personal taste. A real instrument sounds like a real instrument.

"Which recording produces the most realistic rendering determined by what ? Not the device that rendered it ? but by ?? I am easily confused not being rude but I am in need of little more guidance in order to participate please. You have provided a recording engineer’s stats and it is obvious you know what they mean."

As above, but re, the stats: I just put those there because someone mentioned that my method of level matching wouldn’t work. I don’t doubt that there’s a better way, but I printed the stats to show that my recordings were fairly evenly matched, at the least with any difference being imperceptible, I think. The peak of any audio file is the digital sample of the file with the greatest amplitude and the highest sample amplitude possible in a digital system is 0 dB, when using the logarithmic decibel scale of intensity. Other values are negative and the more negative the value, the less amplitude it has. The trouble with these peaks is that they occur rapidly, so fast that our ears can fail to perceive their intensity. RMS stands for root mean square and this is a type of average of adjacent values (so those amplitude values that occur around the same time). These "RMS" values relate much more closely to our perception of "loudness" (the human perception of the acoustic property, intensity). So since the two recordings had RMS peak values -16.77 dB and -16.72 dB (0.05 dB "louder"), the difference is so low, that I believe the loudness of the two recording is reasonably well matched.

"I’m not aiming to deride your efforts, OP, just sharing objectively: by limiting the samples to 2 and stating each sample is from a different amp, you’ve set the stage for expectation bias in responses. Repeated measures can make for a challenging experimental design online."

I agree. But that’s all I had time to do and I didn’t think anyone here would have time or inclination to listen to more, let alone 2 recordings;

"Much a do about nothing IMHO...sorry."

Don’t feel bad about it, humble one. That’s ok.

"What an incredible amount of trouble to go to to make a comparison between two amplifiers that does not necessarily align with real world sq and whether or not one sounds better to the listener or not."

The test and editing only took an hour. Writing and responding here is taking longer. Although I’m enjoying it.

"I would venture to say few here have the desire to dive into the minutiae the way you seem to enjoy. So which amp sounds better to you? Don’t over analyze it..."

That’s true, but it would surprise me that people don’t put this effort (and concentration) into their own gear selections and configurations, especially with such large sums of money involved. I’ll get to my opinion soon.

"I can’t read the flac files and don’t care to load the software to enable it. Put them on YouTube for a bigger audience."

If you’d like me to put a CD format wav file online, please let me know.

"Tracks actually loaded easily via copy/paste into a url search field on my Mac. Listening through Sennheiser headphones and a very decent dongle Dac. e.flac seems to have more body, realism relative to the room of the performance. t.flac gets more detail, but seems a bit colder in tonality. Level match seems close enough to render the important qualities to me? But maybe try a couple more…"

Thanks for listening @riccitone! And since we have a listener, I’ll reveal the amps:

A pair of Quad 303 amps used to bi-amp the monitors and a Sunfire Signature 600 (early version) bi-wiring the speakers with the "voltage source" speaker outputs attached to the main driver and the "current source" outs connected to the tweeter.

What I did to listen, was to load the two files into two VLC media players. I have a way on my computer that i can shuffle the two players, with one on top of the other, so that I don’t know which is on top after shuffling. I then separate them, so they can be played, and conceal the title bar to make them anonymous until satisfied enough to check. I did this a whole lot of times.

My conclusions were very similar to yours and the first thing that struck me was the full body-ness of e.flac. I also paid a lot of attention to the snare drum in the opening passage and that appeared to have the airiness of a real drum in e.flac despite t,flac sounding a tad cooler. I also found the bass a little clearer in t.flac and I think the piano masks it more in e.flac. I have to admit it’s really difficult though. I think both sound good. And after some repetitions of listening and revealing the file name, I didn’t always reach the same conclusion. But mostly did.

Which amp corresponds to which file?

e.flac is the Quads
t.flac is the Sunfire

Without going through the rigmarole of setting up a mic and listening to the recording, I think I hear the same tonal differences and instrument separations when sitting in front of the system, listening in stereo. I think there, the Sunfire has the advantage of slightly better imaging and I suspect that is due to its slightly cooler sound and with the higher frequencies (usually easier to locate) taking predominance. But the thing with the snare (and cymbals) sounding more real and airy on the Quad still confuses me. i.e. why wouldn’t this help with imaging? Perhaps my two "identical" 303s aren’t so identical, there might be a mismatch. I have tried using just one of course. It didn’t sound so dynamic and I didn’t at the time compare it with the Sunfire.

"GREAT version of U.M.M.G btw 👍🏼👍🏼👍🏼"

Well spotted! I didn’t give the full name of the track and the musicians because I think there might be copyright issues. But it’s a Brazilian trio residing in New York. Recorded two years ago and released on vinyl last year. The pianist was 83 when the recording was made and he was one of the founders of the Brazilian jazz trio movement in the sixties. A cult figure. Pioneer of samba-funk too.

I think there have been other replies since I started writing this message.

"Haven’t listened to the files yet, but hope to later today."

Great, thanks

"I find your wiring configuration of the Sunfire amp confusing. To my understanding, current drive would have the most effect on the lower frequencies. The tweeters should not need for much current regulation or drive. Have you tried the Sunfire with the current/voltage switched. I think that might change the sound presentation significantly."

Yes, I see you what you mean. On the back of the amp, the suggestion for bi-wiring is to use the current source on the tweeter and mid-driver and voltage source on the sub. Since the 12" driver on the monitors handles the mids as well as the lower frequencies, I should probably swap it around. I guess there's nothing much to warm up in the sound coming from the tweeter alone. Thanks for the suggestion, will try soon.

"surdo; You actually played it well. This forum has it's share of haters and naming your gear will usually open you up to a kick in the nads from at least one or two of them. You're still getting a good taste of that even without brands. I for one am ashamed that your first visit has been so contentious."

No worries. I'm having fun and I'm here to learn. I also occasionally frequent bicycle forums. They're far worse. People here will have to try harder to get me wound up. :-)

"You are your own enemy in this over-technical quest"

I think it was pretty simple test and reasonably quick to do, as I mentioned above

"Answer is obvious or it isn’t. It’s preference, not better."

Sorry, but once again, I don’t think it’s preference. I’ve listened to many systems, good and bad, and I’m never every fully convinced that there’s a musician sitting somewhere to my front, playing an instrument. But some systems approach this ideal better than others. They reproduce the reality of an instrumental sound and its performance, better, through their rendering of timbre, spatial qualities, etc. (although the recording itself also plays a huge part in this). So that’s what I’m looking to do in this, yes, experimental test, choosing to concentrate on the tone and timbre of the instruments rather than the spatial aspect.

"Forget the earphones, listen to the speakers in the space."

I of course do this all the time, as we all do. The problem though is in making a clear comparison of two systems when you need to switch amps on and off and mess around with cables in between sittings. My listening memory is not good enough to make this type of comparison when the two systems are of a comparable quality. For those that can do this with confidence, good for them.

At some point, on the weekend, hopefully, I’ll use some software to devise a switch, where I can jump, with a single click of the mouse, between one recording and another - like you can toggle two solo buttons on a mixing desk with two hands. As it is, I have to stop one recording on a certain player, go to another player, and switch that on. It’s not as awkward as messing around with cables, but it could be bettered. So the test will be more technical still, but I think more advantageous.

"1. Match level by multimeter instead of SPL. Get a 60 Hz signal and check the output on either speaker. You can get really accurate this way.

2. Consider recording AT the speaker connections. You have to be careful to keep the peak voltages ~ 1V but this would let you get a noise-free recording of what the amp is actually offering the speakers. Then you can do some very interesting analysis such as output by frequency, compression by frequency, etc.

This option also makes it easier to listen to any differences directly as opposed to via another microphone, though the max signal to noise will be different."

That’s brilliant erik_squires, thanks for your input!

Am I understanding correctly that you’re talking about getting the signal to be recorded by attaching a cable to the terminals on the the speaker? ie. connect positive and negative terminals to the wire and shield respectively of a short unbalanced cable (one cable for each speaker)?

I think I’d have to test with the speakers wired traditionally - ie. not bi-wired or bi-amped, correct?

I know very little about electronics, but do have a meter, soldering iron, etc. What sort of voltage might I expect at "normal" (not too loud) listening levels, seated just a bit more than two meters away from each speaker? They’re 8 ohm speakers.

"All amps of low distortion, low output impedance and flat response will sound indistinguishable operated below clipping. Decades ago Peter Walker of Quad did a test using RTR tape copies of classical master tapes between two Quad amps - a transistor and a tube amp. Speakers used were the then new Quad 63’s. The listening audience were British HiFi press. Nobody could tell the two amps apart!"

 

jasonbourne71 I’ve heard about that experiment and if it’s really the case with my two amps, I’ll be very happy. Two good amps!

"It's am interesting quest, certainly do not be deterred from pursuing your methodology. I do find fault with the logic of searching for what sounds most like live instruments, as if there is some absolute. And there lies the fallacy in the premise.

If we can agree that all devices color the sound with distortion, I believe that some of us find some distortions consonant with the fabric of the music, or "pleasurable" if you prefer, and others find different distortion spectra to be closer to the sound of real music. What works for you will not work for me. And this is supported by the wide variety of amplification topologies that sell well in the marketplace. Like finding a mate, there are some faults that one can live with and some that one cannot. Thank goodness we dont all find fault in the same things."

viridian Thanks for your message. I don't believe its a fallacy that one can be able to distinguish an excellent reproduction from another. It's just that some people are better at it than others. An excellent recording engineer will be somewhere way up on the scale of this ability, especially when they are working closely with musicians to get the sound that the musician wishes (as they should). An excellent musician will also know and even an excellent audiophile that keeps regular contact with live acoustic performance of music. I think it's our ears that get estranged from the absolute sound of instruments because of our distance from them and because we mostly hear instruments played in amplified settings and via reproduction - and in the case of the listener of recorded music near-exclusively, the ears can be corrupted in feedback loop of "taste" - with his or her preferred sound redefining mentally and incrementally what they believe an instrument should sound like. Certainly, every musician has his or her playing technique/embouchure, etc. which affects the sound as does the sound of the instrument itself and the room, but those familiar with the instruments, room acoustics and with the particular player where possible, playing live unamplified, will be able to get closer to an absolute recognition of natural live performance.

So when someone above joked asking if I lived above Carnegie Hall, that would really great and would help me as listener a lot. I have to rely more on memory. But if we want to be good listeners, we do need to know what it is that we're listening to. And I'm not saying what amplifier.

It's interesting, I sent the links of the two recordings to a nephew who is also a drummer, but has younger ears than mine. He heard the opposite to me in terms of body and fullness but agreed with me when it came to the cymbals, choosing e.flac as more realistic in this sense.

I'm still reserving my final decision on which amp I consider more natural until I can write this little program that lets me switch instantly from one recording to another. I'm also really keen to try eric_squires suggestion.

"If you attempt to record off the speaker terminals though you have to be sure you keep your voltages low, probably 2V or less at all times."

erik_squires Thanks Erik. Yes, I won't use one of my good recorders for this or the laptop input, but I have a cheap Behringer dongle interface that I can risk. Would it matter if I connect the wires at the amplifier end, as long as the speakers are attached? I can keep the wires shorter that way and won't have to shift heavy speakers around.

Thank you very much budjoe for taking the time to listen to the recordings and for the detailed account. Yes, it’s a very good point about environmental sound affecting the recording and the listener’s ability to distinguish between the two. There was perhaps a bit more wind in the first recording too along with those angry birds at the end. I think the Tascam and the valve mic had been on for over half an hour and the second test was done shortly after the first. So probably not much of an imbalance there. But good point, also.

jji666. I agree!

Thanks too, erik_squires. I think it’s a great idea to do this test bi-passing the speakers and consequently any unwanted ambient noise and irregular room interactions. I’m hoping I can do this later today or on Wednesday. I’ll likely do the recording at the amp end of the speaker wire. Some recent surgical work has left me a bit shy of shifting equipment!

Back with a new test as per erik_squires’s suggestion

This time the tannoys high and low frequency terminals were jumper-ed and there was no bi-amping or bi-wiring done.

Three recordings were made of three pieces of music with two amps. One amp has two speaker out options, so that’s why there’s three recordings. The three amp configurations are name "j", "y" and "p". The three recordings are of a sparse piano track with a change in dynamic in the middle: "piano". A samba jazz trio: "trio". And a voice with orchestra track (great production but the master recording could be better, i think): "voice". So the 9 edited recordings are:

https://escuta.org/webtmp/j-trio.flac

https://escuta.org/webtmp/y-trio.flac

https://escuta.org/webtmp/p-trio.flac

https://escuta.org/webtmp/j-piano.flac

https://escuta.org/webtmp/y-piano.flac

https://escuta.org/webtmp/p-piano.flac

https://escuta.org/webtmp/j-voice.flac

https://escuta.org/webtmp/y-voice.flac

https://escuta.org/webtmp/p-voice.flac

Steps.

1. A 60Hz sine tone was recorded at approximately -1.0dB

2. The tone was played with each amp setup with the voltage of the amp adjusted with the preamp to measure 0.5V. Fine adjustments of the voltage were made with the volume control of the digital source (Logitech Media Server)

3. For each amp, the three musics were played at the adjusted amplitude.

4. Recordings were made at 48kHz with a Behringer UCA222 interface on a laptop running the DAW Ardour.

5. Nine edited ecordings exported as 48kHz 24bit Flac files.

If anyone has time to listen and give some feedback, please do!

 

Listening to these now, I think I’ve found an unpleasant problem and not just a difference, in one of the amps....

The problem is in the recordings j-piano.flac and y-piano.flac which are recordings of the Sunfire’s "Voltage source" and "Current source" speaker outs respectively. The piano, in its quieter passages, produces a distinct distortion, especially on certain notes. The recording p-piano.flac, that of the Quad 303, produces no such distortion.

I have a theory on what the problem is:

The Sunfire’s speaker outputs produce a low level, but audible on headphones, 60Hz hum (or a harmonic of 60Hz). The Quad, while perhaps more noisy (with hiss) has no 60Hz tone. I tested both the Sunfire’s unbalanced inputs with the Quad 34 preamp and the balanced inputs with the sound coming from a mixing desk. Both produce the hum and the amp produces a hum even with no inputs connected and with the laptop, that’s monitoring the terminals, running on batteries. I’ve tried disconnecting the Sunfire’s earth and reversing the pins, but the hum is unchanged.

My theory is that when the piano is played softly, certain notes seem to interact with the 60Hz hum, producing what seems like distortion, so if the hum can be removed, perhaps that distortion will disappear.

Is some kind of a mains filter likely to solve the problem or is there something that can be done to the Sunfire to make it less susceptible to this 60Hz hum?

And yes, I do hear the distortion through the speakers. I had noticed this sound before on occasions, but since I was using the Sunfire in my studio, I always suspected the speakers (a pair of JBL 4312A monitors), which have had a long hard life.

 

"I played both the normalized and originals thru 2 different playback apps and my observations hold across both the original and normalized thru both apps. If you want the full analysis values from the originals and the normalized files, I will send them in a direct message. Let me know."

Thanks a lot budjoe. I really appreciate your comments. Yes, please send the analysis. I’m interested to see exactly what you’re looking at.

"Were listening to these files through the speakers or headphones? Possible you were hearing issues with the headphones and possibly being overdriven?"

I don’t think the phones were overdriven. I hear this sound through the speakers and on headphones (AKG K501 and K246). What I don’t hear on the speakers is the 60Hz buzz. This I only hear when "tapping" the speaker output of the amp and listening on headphones, either directly or on a recording. Although that doesn’t necessarily mean that the sound is not there - there’s a bit of other noise here always.

The more I listen to the sound, the more it sounds like a type of ring-modulation of the piano and I wonder if some sort of cross-modulation is occurring. If this is actually the case, there would be additional side bands appearing in the analysis and not necessarily a distortion.

Listening now to the 3 piano files on phones and at the same volume, p-piano.flac does not have this ring-modulation sound in the first half of the recording, but y and j certainly do.

The original recording is of tracks 8 and 9 ("Var. V. Lento" and "Var. VI Poco movendo") of Alessandro Simonetto’s album "Erik Satie: Works for Piano". I streamed them into a DAW from Qobuz and used this recording for the three amp tests.

budjoe, I normalised the peak of the j-piano and p-piano recordings to -0.3dB (which is what you did?). The above is Ardour’s spectrum analysis of the 1st note of the two recordings. The j-piano recording seems to have a lot more noise. The p-piano part seems to have stronger high frequency transients between 1 and 5kHz.

Also, just above the 1st orange band above 100Hz, there’s what seems like a pulsing partial that’s not present in the p recording. This was evident before normalising both recordings and in which there was less apparent noise but still very weak upper partials (between 1 and 5kHz) in the j recording.

 

@budjoe Thanks again for your help with this. Yes, the last set of recordings was done with the same cables, speakers (Tannoy), preamp (Quad 34) and recording device. Only the power amps were changed.

I wonder if normalising the peak values is the right way to go. I imagine that the two amps will potentially render peaks quite differently. And if that’s the case, wouldn’t that produce audio mismatched for loudness and with the background noise exaggerated in the recording with the lower peak?

I’ve gone back to my Tannoy/Quad34 recordings and normalised the files, constraining the RMS to -5.0dB. There was no clipping and the two files sound reasonably equal in loudness to me in headphones.

Yes the Quad recording has noise and a thin, grainy sounding hum, but the hum of the Sunfire is louder and deeper, to my ears at least. On the Quad the hum is louder in the right channel. On the Sunfire the hum is louder in the left but more balanced.

All the same...

More important to me than the noise, is the distortion of the piano sound in the Sunfire recording. Are you (or anyone else) able to hear this distortion?

Here are the two files normalised for loudness:

https://escuta.org/webtmp/QuadPiano.flac

https://escuta.org/webtmp/SunfireVoltagePiano.flac

Cheers,