Axiom Home Page Forums General Discussion Stereo INSANELY BRIGHT?

Starting the thread isn't the problem. Naming it in all caps "INSANELY BRIGHT" was what caused the ruckus. (likewise on the soundandvision board)

Littleb,

Don't sweat it about starting this thread. We have been down this road before on this forum on many topics. I look at it this way...it's an opportunity for all of us to hear each other's opinions and share our own without antagonism. In many instances in this and other forums, my perspectives have been challenged and often broadened.

Mark

Cooper,

Belatedly... Let's be a bit more precise here...

In reply to:

---

Sure, those curves are flat because they go into a non-reactive test load. And speakers measure awfully flat when tested at a fixed volume (1W ?). However, impedance curves are far from flat, on any speaker. So what would happen if I took a fixed voltage and fed actual music to a speaker who's impedance varies from 4 ohms to 20 ohms (rated with a nominal impedance of 8 ohms)? Well the high impedances are pretty easy to deal with linearly, but as the impedance drops, the speakers demand much more power. To perform linearly, the amp must provide exactly double the power from 8 ohms to 4 ohms given a constant voltage. And quickly.

---

So far, the argument is accurate.

In reply to:

---

But wait a minute. Notice that the curves for the different resistive loads, while individually flat, are not at the same absolute level. 2 watts at 4 ohms will virtually always produce less than double the power than 1 watt at 8 ohms; the magnitude of the difference depends on how well-equipped the amp is to handle high-current / low-impedance. Okay so there's maybe a few-decibel difference at 1W. How about comparing 10 watts @ 8ohm to 20 watts @ 4ohm? What happens if a peak in the music demands instantaneous power of 300 watts for a tiny fraction of a second?

---

If the power (or voltage) output of a solid-state amp decreases by as much as a few dB within audio frequencies when driving a 4-ohm load, the amp should be considered defective (unless it is a tube amp with a very low damping factor, i.e., high output impedance). This aspect is covered by the output impedance measurements over the frequency range.

The vast majority of today's solid-state amps have output impedances of well below 0.1 ohm, which stays low and ruler-flat up to (at least) 20kHz, where it slowly starts to rise. When driving a 4-ohm load, this translates to a power loss of less than ~0.1dB, NOT "a few dB" loss. And this is maintained until the maximum available power output (for the given load impedance) is approached, where the sonic degradation and its differences between amps becomes very obvious anyway.

In reply to:

---

Additionally, Stereophile also tests amps into a "dummy loudspeaker load," which has a typical loudspeaker impedance curve. Amps always produce the same general freq. response curve, dipping where impedance drops, with the degree of interaction being dependent partly on the amount of amp output impedance. The slope of the loudspeaker impedance curve also has an effect on its load to the amp.

---

Yes, and as stated above, those "dips" are only ~0.1dB deep or less. Even if one can reliably hear a 0.1dB dip in frequency response (highly unlikely), this does NOT provide a physical explanation for the alleged "brighter" sound of the Yamaha receivers on many different speakers (with wildly different impedance curves from each other). If you have any logical idea here, please tell me...

In reply to:

---

Back to transients. This is partly why a square wave is often used in testing amps. When there's a rapid change in the voltage signal, the amp has to quickly supply the current as well. Does the amp just go straight to that voltage/current, especially at high wattages? Depends on the amp. It may overshoot the increase to the top of the wave, then undershoot the drop to the bottom.

---

Again, if a solid-state amp exhibits an overt overshooting/ringing under a resistive or resistive-inductive (speaker) load, the amp should be considered defective today, period. Under a realistic load conditions (resistive/inductive combined with some capacitative components), the vast majority of current solid-state amps reproduce "text-book" square waves.

In reply to:

---

Tube amps round off these square waves substantially, leading to a softer, warmer sound that many people appreciate or prefer.

---

That is simply because, in many tube amps, the frequency response starts to roll off at 10-15kHz. No magic here -- and some tube amps have a definitive sonic character precisely because of this and several other reasons (whether it is subjectively good or bad). Anyway, my initial post deals with modern solid-state amps.

In reply to:

---

Regarding distortion. Decades ago, amp manufacturers engaged in a THD war. Solid state was in, and so was negative feedback. Problem is that extreme use of negative feedback can have a very detrimental impact on sound quality. Sure, the 1W frequency response looks great though! Some amps run with little or no feedback, though the middle road today is to simply use it in moderation.

---

I am not aware of a modern solid-state amp that runs with "little or no" NFB, say, less than a total of 10dB feedback (please show me if you know any such commercial designs). The open-loop characteristics of solid-state op-amps and power stages have come a long way in the last 3 decades, so that today's amps achieve, with moderate levels of NFB, a very low THD+N and output impedance over and well beyond audio frequencies, while maintaining a robust stability and load tolerance.

Bottomline: Your "technical" arguments still do not provide a good explanation for the alleged sonic characters of solid-state amps.


curtis,

Yes, I had instantaneous A/B compared, in several occasions, two current models of Yamaha receivers (the flag-ship RX-Z1 and a $1000-ish model, of which I don't remember the model number) in the direct stereo mode, with various other receivers including the Pioneer, Denon, H/K, Onkyo, and Marantz. After carefully ensuring that every processing is defeated and the listening level is within +/- 0.5dB, I did not reliably hear a sonic difference among them. This wasn't even a blind test -- I knew what I was listening to and was looking for the alleged, published "characters."

I personally suspect that, if any real sonic difference exists among these receivers, it resides in the digital-domain algorithms and/or the ADCs/DACs. I believe that it is fairly hard to hear the sonic difference among modern solid-state amps (or the amp portion of the receivers), although I do not want to claim or prove that all SS amps do or should sound identical. To my ears, they certainly do not provide a strong enough sonic signature to affect/modify the timbral characters of loudspeakers, which are, of course, highly audible in most cases.

In reply to:

---

Do you know if listening in STEREO mode on the 45TX (not DIRECT) applies any types of equalization changes to the music? I ask this because I notice a slightly different sound in DIRECT vs STEREO. And no, I did not have any of the tone controls or MCACC eq. turned on while we were listening in STEREO mode that night.

---

If you are using analog inputs from a CD player, Stereo mode is most likely still re-digitizing the signals and performing digital crossover work before converting back to analog. "Stereo" mode can still heed your speaker settings and hi-pass filter your mains while providing a low-passed signal to the sub.

Many amplifiers don't even have a direct mode: they are always digitizing your analog inputs, no matter what. Last I checked, ALL Yamaha receivers digitize all inputs. One generation back, all HK receivers did it: I emailed the company. Fortunately HK ended up giving a true bypass feature for the current generation.

Many people are unaware of the huge sonic penalty of an extra analog->digital->analog step. My Yamaha CD player has very good DACs, and I lose all the benefit if I were to to go Yamaha DAC -> Receiever ADC -> digital processing -> Receiver DAC. While I do have an optical output, I believe the Yamaha DACs are much better than the DACs on most receivers below $1500. This is a rough estimate of course.

-Cooper

Might that explain the reason why the sound is perceptively better when a one of my CDs is played through my 3 month old DVD player vs. my 18 year old Pioneer 5 CD player?

Nice and astute response to my rant, Sushi. I was trying to illustrate two points: 1) there are measurable differences between all amps, and 2) there are lots of measurements we take which are deceiving or simply absent.

You are quite correct that solid state amps typically have a very small reaction to the "dummy speaker" load in Stereophile. This is where I will again stress that it's done into roughly 1 watt with a simulated nominal impedance of 8 ohms. Probably almost any amp can cope with a tiny dip down to 4 ohms / 2 watts. But where we separate the men (big amps) from the boys (50wpc stuff) is when dealing with extremely high powers.

So if you take a 50wpc amp and a 300wpc amp, being equal except for power, we couldn't easily tell the difference at very moderate volume levels. Big amps are all about handling a symphonic clash or a cannon shot or some other dynamic peak. Turn up the volume and your 50wpc will sound quite loud during the song, but the big bang will fall short even if it's not clipping.

But I also believe that as volume levels increase, those dummy speaker results will get exacerbated. Why? Because that impedance linearity fades away under higher power. I own "the most powerful receiver Harman Kardon made in 30 years," the HK 3470. It puts out 100wpc into 8 ohms, but something like 130wpc into 4 ohms. A really good 100wpc amp would actually put out almost 200wpc into 4 ohms. The discrepency means that my amp gets progressively less linear, with respect to impedance, as power output increases.

In other words, if the music draws 100 watts for the 8+ ohm parts of the clash, then it's going to want to draw 200 watts for any sound frequencies that have 4 ohm impedance. But it won't be able to on my amp, which means during that burst of sound my amp will be several decibels lower at the frequencies where my speaker impedance dips. (130 W is a far cry from 200 W)

Some amps have an "instantaneous" power draw statistic, but those last for an extremely short amount of time during high current situations. So if the echo of the cannon blast continues for a second, you'll be relying on those continuous power ratings of 100wpc and 130wpc. This is why having a lot of overhead on power amps is so important: it helps you deal with tremendous musical dynamics and sound levels without losing neutrality or impact.

The bottom line is that most statistics rely on continuous signals, while the real challenge to an amp has everything to do with dynamics. I'm speaking in gross terms and I am no electrical engineer.

-Cooper

Mark,

The STEREO mode on the 45TX leaves all digital-domain processing turned on (obviously except for the matrix surround processing). For example, the MCACC level/delay/EQ adjustments for the front L/R will be used, and the high-bit/high-sampling feature will stay on if you have turned it on. Also, in my experience, the STEREO mode seems to play it 3-4 dB louder than the DIRECT mode.

In reply to:

---

But I also believe that as volume levels increase, those dummy speaker results will get exacerbated. Why? Because that impedance linearity fades away under higher power. I own "the most powerful receiver Harman Kardon made in 30 years," the HK 3470. It puts out 100wpc into 8 ohms, but something like 130wpc into 4 ohms. A really good 100wpc amp would actually put out almost 200wpc into 4 ohms. The discrepency means that my amp gets progressively less linear, with respect to impedance, as power output increases.

---

Cooper,

We need to separate two distinct issues here: (1) impedance linearity of maximum available power, and (2) power-level dependency of frequency responses under realistic loads. We should not mix these up together.

(1) Yes, most moderately-priced receivers/amps cannot provide a linear increase of maximum power when the load impedance is reduced. This is simply because there are limits in maximum current draw in these amps, as well as limits in power-rail voltage regulation under high-current situations. Also, as you said, there are definitive differences even among the amps/receivers with similar maximum stationary power outputs as to how well they handle transient peak power demands (so-called power researve or "crest factor"). These differences are certainly measurable, and can be audible when one tries to drive an insensitive/inefficient pair of speakers (such as the Magnepans) in a very voluminous room. But when you are driving speakers of decent sensitivity (like the Axioms) at moderate volume levels in a moderate-sized room, the amp will hardly put out more than a few 10's of watts in 99.9% of music passages. My point is, these max power-related differences CANNOT render the "sonic character" of the amp, which supposedly is present constantly through the music.

(2) These limitations in current draw and power-supply regulations seen in many moderately priced amps DO NOT result in a modulation of the frequency response until the actual power output approaches the maximum available power. For example, the frequency response of your H/K 3470 under a realistic speaker load will be virtually flat (except for the tiny dips due to the finite output impedance as explained earlier) when driven at 1W, 10W or 100W into 4 ohms (until it approaches the 130W limit). The frequency response DOES NOT dip more when it is putting out 100W, as compared to the response at 1W.


So, I do agree with you that the dynamic power handling measurably differ from amp to amp, including the amount of transient power reserve as compared with the amp's sustained maximum power, and the gracefulness of distortions when the maximum power is reached. These attribute can be an important factor under certain situations where you do actually encounter a three-digit power demand frequently enough so that the difference is audible.

However, my point remains that these attributes related to the maximum available power output CANNOT explain the alleged "sonic signature" of the amps/receivers.

In reply to:

---

We need to separate two distinct issues here: (1) impedance linearity of maximum available power, and (2) power-level dependency of frequency responses under realistic loads.

---

The impedance of any speaker varies dramatically with frequency, regardless of the power output. It becomes a stronger factor as power levels increase.

In reply to:

---

the frequency response of your H/K 3470 under a realistic speaker load will be virtually flat (except for the tiny dips due to the finite output impedance as explained earlier) when driven at 1W, 10W or 100W into 4 ohms

---

That is a good point. However, I don't believe it's as clear cut as you're indicating. I have to fall back on my argument about transients. I think this is where the differences get revealed, because even if you're not sustaining 100wpc, music is very dynamic. Every single cymbal clash and kick-drum whack will instantaneosly demand a huge rate of current draw. You can calculate exactly how many farads of energy are stored up in capacitors, and how long it could supply energy at a steady voltage, but how does that current draw affect the nature of the sound, and is there a kind of hesitance in your wall current supplying the sudden jolt?

One very well-established, but counterintuitive, observation in high-end equipment design is the effect power supplies have on the sound of a component. I'm not just talking about amps; even preamps and digital sources are affected. The more "overhead" the power supply has, the less constrained the sound. This is one of the reason high end components weigh so much. It's oh-so-easy to measure continuous current draw situations, but with dynamics... well, remember the wonderful world of derivatives? If the current goes from 1 amp to 100 amps in .001 seconds, what does the current graph look like? A straight line?

This goes back to the square wave. Many amps do not actually produce great-looking square waves. It has to do with amplifier bandwidth: how high the frequencies go before the response tapers off. On a $9000 100wpc Linn Klimax Twin, Stereophile measured a 3dB rolloff at 58kHz. I will quote from the article: "This curtailed ultrasonic response also rounds off the leading edges of a 10kHz squarewave, but the waveform is refreshingly free from overshoot or ringing." The graph indicates this very obviously, and it also has an interesting curve when dropping down to the low part of the signal.

I'll stop here, as I can no longer make a strong assertion about exactly what causes the frequency-specific variation. But most measurements vary greatly with frequency, such as channel separation (by 10s of decibels), noise floor (often by 10s of decibels), and other things. Given the fact that plenty of people on Audioreview will comment on the "brightness" of a $400 Yamaha amp or the "warmth" of a Harman Kardon, I am inclined to believe there is some cause. But I'm pretty much out of steam on this argument, so maybe we can call it a draw?

-Cooper

In reply to:

---

But I'm pretty much out of steam on this argument, so maybe we can call it a draw?

---

Hehe... Cooper, this is only starting to get interesting.

In reply to:

---

The impedance of any speaker varies dramatically with frequency, regardless of the power output. It becomes a stronger factor as power levels increase.

---

I don't know what exactly you meant by "a stronger factor." But as you stated yourself, the speaker impedance curve essentially stays the same regardless of the power output. It does NOT suddenly become more important from amp's perspective above certain power output levels, as I explained in detail above.

In reply to:

---

That is a good point. However, I don't believe it's as clear cut as you're indicating. I have to fall back on my argument about transients.

---

It is indeed as clear cut as I indicated -- there is nothing hidden or mystical here. The hard fact is, if the amp can maintain an essentially flat frequency/phase response at continuous 100W output with a real-world load, then it will maintain the same frequency/phase response through the highly dynamic music playback, AS LONG AS the transient power peaks do not exceed 100W. In other words, the problem of power-supply regulations, dynamic current draw etc. are already covered here when measuring at steady-state 100W output, as a worst-case scenario so to speak.

Now, if the dynamic power demand overtly and frequently exceeds 100W (or whatever maximum available), that is an entirely different story. But I strongly doubt that people (including the "expert" magazine reviewers) routinely put the SS amp into that kind of situation during a typical subjective listening test using typical music materials. In the vast majority of home audio settings, I would predict that your ears start to hurt way before the amp starts to exhibit stresses. As an example, my Adcom power amp (rated 325Wpc into 4 ohms) has nifty instantaneous distortion indicators; I have NEVER managed to turn them on even when listening to highy dynamic music/soundtrack materials at insane (basically intolerable) volume levels in my 5000 cu. ft. room.

In reply to:

---

It's oh-so-easy to measure continuous current draw situations, but with dynamics... well, remember the wonderful world of derivatives? If the current goes from 1 amp to 100 amps in .001 seconds, what does the current graph look like? A straight line?

---

The dynamic behavior of an amp is (unlike that of loudspeakers) in fact quite simple, again as long as you do not swing the amp beyond the maximum continuously available power. All you need to fully describe the amp's transient/dynamic behavior is the (complex) transfer function across the entire audio frequencies, measured at near-maximum power outputs. EVERY transient/dynamic behavior will be covered by this transfer function, including your example if the amp is actually capable of doing it. [incidentally, not many amps can ever do it in the first place, because 100A into 4 ohms and 8 ohms correspond to 40kW and 80kW, respectively.] An important fact here is, the transfer function of an amp can already be fully determined by obtaining simple power/phase response curves over the entire audio frequency range. There are many ways to do this, including the classical frequency sweeps, pink noises, square waves, and more fashionable impulse and step responses, but these are in fact 100% equivalent to each other when measuring amps (again, they can derive totally different results when it comes to loudspeakers, because they produce so much time-domain aberrancies).

In reply to:

---

One very well-established, but counterintuitive, observation in high-end equipment design is the effect power supplies have on the sound of a component. I'm not just talking about amps; even preamps and digital sources are affected. The more "overhead" the power supply has, the less constrained the sound. This is one of the reason high end components weigh so much.

---

I do NOT think that observation is "well-established" outside the audiophile marketing machinery for expensive audio gears. I fully agree that the more expensive the equipment is, the bigger, more over-spec'ed power supply it tends to boast -- this is important because they have to justify the higher cost in some ways. I, for one, love to look at the beautifully built innards of some of those expensive gears, which alone would hugely boost the pride of ownership for me. But I am NOT aware of any reliable (subjective or objective) reports convincingly demonstrating that these over-spec'ed parts actually produce a better sound quality. Let us not forget that ALL audiophile magazines operate within the realm of the huge marketing machinery (in fact, they ARE perhaps the most important part of the whole marketing mechanisms).

In reply to:

---

On a $9000 100wpc Linn Klimax Twin, Stereophile measured a 3dB rolloff at 58kHz. I will quote from the article: "This curtailed ultrasonic response also rounds off the leading edges of a 10kHz squarewave, but the waveform is refreshingly free from overshoot or ringing." The graph indicates this very obviously, and it also has an interesting curve when dropping down to the low part of the signal.

---

This is an entirely expected, "text-book" example of beautiful squarewave response. Of course, if the amp's high-frequency pole is at around 60kHz, it will NOT perfectly reproduce a 10kHz squarewave; its 7th harmonics is already down significantly. These sort of things only reinforce my assertion above: when it comes to electronics, all these seemingly different measurements are in fact functionally equivalent. Indeed, I am NOT aware of any modern SS amps/receivers that behave badly or unexpectedly on squarewaves (if squarewave overshoots or rings, there must be a peak(s) in the frequency response, somewhere in the treble-ultrasonic region, which I've never seen these days).

In reply to:

---

I'll stop here, as I can no longer make a strong assertion about exactly what causes the frequency-specific variation. But most measurements vary greatly with frequency, such as channel separation (by 10s of decibels), noise floor (often by 10s of decibels), and other things.

---

Yes, these measurements (especially noise floor) can vary wildly across audio frequencies. But, unless the amp is defective, it is a variation from, say, -120dB to -90dB -- it remains essentially inaudible in most modern SS amps.

In reply to:

---

Given the fact that plenty of people on Audioreview will comment on the "brightness" of a $400 Yamaha amp or the "warmth" of a Harman Kardon, I am inclined to believe there is some cause.

---

Now, we are finally back to the "faith" domain. I fully respect your faith on these anecdotal reports. However, my own version of faith here, faced with this same assortment of reports, is that the listening impressions of many of these people might well be already influenced and pre-conditioned by the very widespread notion of the "Yamaha brightness." I simply cannot convince myself with these reports, especially given that I myself do not hear the alleged brightness on the Yamahas, and because I do not see any (measurable) physical foundation for the Yamaha's brightness.