What makes a good amp?

Onyxia

Established Member
Joined
Mar 26, 2005
Messages
131
Reaction score
3
Points
24
Question is in the title. For video, it's very easy to discern personally speaking. There are many elements to consider and can be easily prioritised in order of importance. For audio, it's a bit more difficult for me. What should one be looking for in a good amp? Power? I don't really enjoy my ears being blown off, rather listen to average volume high quality music.

Speakers may range from B&W 685,684, CM1 and CM8. Regardless of price, what are the innards that make the difference and matter most? I understand a lot of this is subjective, on paper what should perform well?
 
Power? I don't really enjoy my ears being blown off, rather listen to average volume high quality music.

Definitely.
You need sufficient power to provide good dynamics.
Hence the power supply (and weight) can be a good indicator, imo.
 
Good design, i.e. simplest is often better. Top grade (audiophile status) components. Big PSU in terms of current delivery and smoothing i.e low e.s.r. Discrete components (Transistors not op-amps or ic's. Tripath ic's being the exception).
 
IMHO audio is much more subjective. Good, well respected amps generally do not exhibit neutral sound, but have a small amount of tonal lift. They often change their tonal balance at different volumes, as many operate at class A at low levels and class AB or D at higher levels. As AB & D have inherently higher levels of distortion, how an amplifier handles itself in this mode often decides its character.

Having impulse tested a few back in the 90's when I worked as an audio designer, Most consumer amps had a warm (Lifted) bass and a little roll off at the high frequency end - usually as a result of an ultrasonic filter. We built an amp with dead flat response from less than 10Hz to over 200KHz using a switched mode PSU and some meaty output devices. Honestly ? It sounded "OK", but improved massively when we warmed it up a bit and actually increased the distortion at higher levels!

Valve amps are another subject entirely. The most critical component is the output transformer, imparting the frequency response (Never flat) and also the graceful overload (Magnetic saturation being much nicer than electronic clipping) Of course, good, well matched valves are needed to drive the thing in the first place!

It there was one amp design that worked perfectly, we would all buy the same one! Speakers do not present a static load to the amp, so the way the amp drives different speakers will also change. Hence, certain amps suit certain speakers.

In short, great sounding amps are well designed and use quality components. But, and its a big but, the speaker imparts much more character upon the sound than the amp driving it.
 
IMHO audio is much more subjective. Good, well respected amps generally do not exhibit neutral sound, but have a small amount of tonal lift. They often change their tonal balance at different volumes, as many operate at class A at low levels and class AB or D at higher levels. As AB & D have inherently higher levels of distortion, how an amplifier handles itself in this mode often decides its character.

Having impulse tested a few back in the 90's when I worked as an audio designer, Most consumer amps had a warm (Lifted) bass and a little roll off at the high frequency end - usually as a result of an ultrasonic filter. We built an amp with dead flat response from less than 10Hz to over 200KHz using a switched mode PSU and some meaty output devices. Honestly ? It sounded "OK", but improved massively when we warmed it up a bit and actually increased the distortion at higher levels!

Valve amps are another subject entirely. The most critical component is the output transformer, imparting the frequency response (Never flat) and also the graceful overload (Magnetic saturation being much nicer than electronic clipping) Of course, good, well matched valves are needed to drive the thing in the first place!

It there was one amp design that worked perfectly, we would all buy the same one! Speakers do not present a static load to the amp, so the way the amp drives different speakers will also change. Hence, certain amps suit certain speakers.

In short, great sounding amps are well designed and use quality components. But, and its a big but, the speaker imparts much more character upon the sound than the amp driving it.

Best post I've read for a while :smashin:

I've also noticed that the older you get, the more you actually want it warm, even though failing ears hear the top end less and less.
I haven't quite worked that one out yet :confused:

But, and its a big but, the speaker imparts much more character upon the sound than the amp driving it

I particularly agree with this point, but it seems to be contrary to many people's opinions :confused:
 
Hi Noiseboy,

Thanks again for your post - very informative. I remember reading that a warm amp usually has higher IMD so I assume a neutral one will have lower values?

Thanks,
Simon
 
Last edited:
In technical terms, amps have distortion of less than 0.5% while speakers probably chuck out 3-6% and more. For this reason alone, speakers will always impart more of their tonality than the electronics driving them.

I suppose the hierachy of better amp then better speakers might be true, but I remain to be convinced.
 
Another point is that there is often poor correlation between how an amplifier measures,and how it sounds or is appreciated by the person listening to it.

Single ended triodes often measure appallingly,but can sound stunning,and those of us old enough to remember will recall the amazingly low distortion figures of many high powered amps from the 80's,which sounded dreadful,thanks to lashings of global negative feedback.

There are many factors influencing the sound quality,but prioritising them in some sort of meaningful order is difficult,and you also need to factor in the synergy or lack of,between amplifier and speaker,which can be very significant indeed.
 
Another point is that there is often poor correlation between how an amplifier measures,and how it sounds or is appreciated by the person listening to it.

Single ended triodes often measure appallingly,but can sound stunning,and those of us old enough to remember will recall the amazingly low distortion figures of many high powered amps from the 80's,which sounded dreadful,thanks to lashings of global negative feedback.

There are many factors influencing the sound quality,but prioritising them in some sort of meaningful order is difficult,and you also need to factor in the synergy or lack of,between amplifier and speaker,which can be very significant indeed.

Good point about the feedback circuit. A fact of life if you use op-amps as part of the design as they effectively have infinite gain. We tried all sorts of tricks like putting the EQ in the feedback loop, but at the end of the day, a well designed discreet amp or one using potted IC modules usually sounded better.

Power supplies are another area of design that often gets over looked, or just plain wrong!

A big meaty supply with huge capacitors would at first glance seem to be the best approach, but what happens when the transients deplete the charge in the capacitor ? You have to wait for the next power cycle from the bridge rectifier to recharge it, leading to ripple. This leads to poor bass response, so the designer bungs in bigger capacitors... The end result is huge power consumption on start up and a bigger transformer creating more electromagnetic noise.

A better approach is a large switched mode power supply. This recharges the rail more than a thousand times faster than a conventional supply giving a "faster" bass response. Current designs are ultra low ripple and throw out very little spurious RF. Most professional amps designs now use switched mode, and IMHO, consumer amps should follow. The manufacturers will have to throw in a big chunk of steel to make the thing weigh more, becuase we all associate weight with quality (!) but more importantly to reduce resonance.

...And of course, the amplifer should sit on its own, not under the CD or Blu Ray player or if a single stack is needed, right at the top so the heat is disipated.
 
Power supplies are another area of design that often gets over looked, or just plain wrong!

A big meaty supply with huge capacitors would at first glance seem to be the best approach, but what happens when the transients deplete the charge in the capacitor ? You have to wait for the next power cycle from the bridge rectifier to recharge it, leading to ripple. This leads to poor bass response, so the designer bungs in bigger capacitors... The end result is huge power consumption on start up and a bigger transformer creating more electromagnetic noise.

A better approach is a large switched mode power supply. This recharges the rail more than a thousand times faster than a conventional supply giving a "faster" bass response. Current designs are ultra low ripple and throw out very little spurious RF. Most professional amps designs now use switched mode, and IMHO, consumer amps should follow. The manufacturers will have to throw in a big chunk of steel to make the thing weigh more, becuase we all associate weight with quality (!) but more importantly to reduce resonance.

...And of course, the amplifer should sit on its own, not under the CD or Blu Ray player or if a single stack is needed, right at the top so the heat is disipated.

Exactly,and it's interesting how the high end has grasped that,with Chord,Linn and others moving to switch mode PSUs,and both of those were among the first in audio circles to do so.

P.S....You'd love the PSU section of my old Krells.....coke can sized caps,and 2kW transformers.
No mechanical noise,and some industrial electrical screening but huge weight as well.
 

The latest video from AVForums

Is 4K Blu-ray Worth It?
Subscribe to our YouTube channel
Back
Top Bottom