What makes different amps produce different sound?

As a basic concept, different technologies and designs and parts, lead to different sonic qualities. While they all reproduce the basic musical signal, they do so with different levels of accuracy, colorations, and transparency, and also that elusive "lifelike" or "musical" quality.

There are many generalizations that are applied to certain kinds of amplifiers, but they are only generalizations, and may or may not apply as much, or at all, to a particular product being auditioned.

There are some indications that an experienced person may expect from any given amplifier design, based upon some knowledge of design/performance characteristics, but it is always better to audition the product because some designers may have done a better job than others. Also, some amps may do better with certain speakers than other amps, and the overall system environment can play a part that is generally called "synergy".

Also, individual listening tastes and preferences will play a part. This is why there is really no overall consensus about any one amp being "the best". It really amounts to what you like "best" in your system context.

I'm sure that others will offer some information on the usual characteristics that we have come to expect from certain technologies and designs. I just wanted to start the conversation out in a general kind of way.

Just for the record, there is a school of thought (and a rather well-credentialed one) that holds that competent amps operating within normal limits do NOT sound different at all. To be more specific, they would argue that if two amps sound different, it will be for one of three reasons:
1) Their levels are not matched.
2) One or both has a frequency response that seriously diverges from flat (as might happen with amps with a high output impedance driving certain loads).
3) One or both is clipping.
Of course, lots of audiophiles reject this, as is their right.

Since every amp sounds different that's probably the reason people reject that idea. There are many people who think nothing sounds different and that there is no reason to buy anything but a Bose Acoustic Wave Machine.

The same equipment in a different room sounds different.

I bought a highly regarded pair of speakers once based on a review. I sold my other speaker to facilitate this change. When the speakers arrived I hooked them up, let them break-in and then sat down to listen. I almost cried since I was a struggling music lover on a very tight budget and these speakers sounded terrible at best. At a loss to know what else to do, I rewired them with some audioquest wire I had left over from some evil experiment I had conducted on an unsuspecting victim.

I hoped for an improvement, but what I got was much more. These speakers actually sounded good. I kept them for some time before moving up to something else.

The point of all this is to say that something as seemingly inoccuous as wire can make a huge difference. When two amps with similar design use different componants they will sound very different.

Consider too, the companies that modify CD players with new caps and resistors. People swear by these mods, or was it that they swear about the mods... I can't remember.

Hi Jeff,

One design in particular, negative feedback has a rather substantial effect on sound... way back, adding massive amounts of negative feedback was a popular way to woo a consumer by drastically reducing THD or Total Harmonic Distortion... In reality, anything below 1% THD is considered (by most) to be inaudible... negative feedback seems to greatly reduce the 3D soundstage abilities of an amp... other features such as high current can sound different based on the power consumption needs of a particular pair of speakers...

There are a multitude of different design philosophies, each seemingly providing the silver bullet to fidelity. Nelson Pass has his own philosophy and mentions a list of amp fads and design issues in his owner's manual for my Aleph 3. The link is www.passlabs.com/pdf/aleph/a3man.pdf. Look under the "Product Philosophy and Design Theory" section. Here's an excerpt:

"We have heard Triodes, Pentodes, Bipolar, VFET, Mosfet, IGBT, Hybrids, THD distortion, IM distortion, TIM distortion, phase distortion, quantization, feedback, nested feedback, no feedback, feed forward, Stasis, harmonic time alignment, high slew, Class AB, Class A, Pure Class A, Class AA, Class A/AB, Class D, Class H, Constant bias, dynamic bias, optical bias, Real Life Bias, Sustained Plateau Bias, big supplies, smart supplies, regulated supplies, separate supplies, switching supplies, dynamic headroom, high current, balanced inputs and balance outputs...Leaving aside the examples of marketing hype, we have a large number of of attempts to improve the sound of amplifiers, each attempting to address a hypothesized flaw in the performance."

A number of these is addressed in his subsequent comments as he describes his design rationale for the Aleph 3.

Tube amplifiers characteristically produce second harmonic distortion whereas transistors like to produce the third harmonic. Musical instruments typically make second harmonic sounds, so the tube amplifier distortion, even if rather high, is often perceived as pleasant or "musical".

Eldartford: Where did you get that information? To my (admittedly limited) knowledge, everything you've said is wrong. Most instruments produce ALL harmonics, but some wind instruments, if I'm not mistaken, produce only ODD harmonics. (Remember that the fundamental tones are themselves odd harmonics!) Also, while it is true that tube amps tend to have more even harmonic distortion than odd (or is it the other way around?), tube tend to have more of BOTH even and odd-order harmonic distortion than typical solid state designs.

Why anyone thinks that added distortion of any kind should make something sound more musical is beyond me.

I think it was Art Dudley who suggested that the color of the face plate had an effect on the sound of amps and preamps.

Some other variables that affect the sound of an amplifier:

Humans are relatively insensitive to even-ordered harmonics and astonishingly large amounts are usually not objectionable. Even ordered harmonics tend to add 'bloom' or 'fatness' (in the bass) and will obscure detail if in large amounts.

Humans find odd-ordered harmonics much more objectionable, in fact higher orders (about the 7th and beyond) are used by the human ear/brain system as loudness cues. Transistor amps with high feedback tend to have very low distortion except for higher/odd ordered harmonics, which is why many transistor amps will have a harder sound. Tube amps, particularly those with low or zero feedback, will have very little high/odd ordered harmonics and will have a softer sound.

Bandwidth plays another role. Generally, due to phase shift/group delay effects, the cutoff frequency of many amps can be audible. For example a 10Hz cutoff will introduce audible artifacts to about 100 Hz, resulting in a bass shyness. Amps that cut off higher, for example 20Hz, will have cutoff artifacts to 200Hz. High frequency cutoff issues generally follow similar rules- to 1/10th the cutoff frequency. A 50KHz cutoff will introduce artifacts down to 5KHz. These are all frequencies that humans can hear very easily! -and is why so many manufacturers strive for wide bandwidth.

There are a number of other influences, but it is getting late and these are good for starters. .

Unsound, I read an article in Stereophile about ten years ago about an experiment with speakers. The listeners were told that different colored grill cloth was used so they would be able to tell the difference between the multiple speakers being tested.

They ran the listening tests for the speakers and the listeners determined which had certain sonic characteristics. There seemed to be somewhat of a consensus as I recall. The ones with the blue grill cloth had the better bass... and so on.

The funny thing about the test was, all the speakers were the same except for the color of the grill cloth. I doubt that these people would qualify as audiophiles. I don't remember who did the test, I think it was a university in the US, but unsound's comments reminded me of the test.

Bomarc...I really don't know where I learned what I posted: it was long long ago, and I know I have seen it many times. See also the comment by ralph@ above.

Sometimes there is confusion about numbering of harmonics. If you do a Fourier analysis of an audio signal, what we call the fundamental is the first harmonic. DC is the zeroith harmonic. When speaking of an AC signal, harmonics are usually numbered as multiples of the fundamental.

Perhaps brass instruments do produce the "unmusical" harmonics. This explains why they sound so strident compaired with strings.