No such thing as the perfect amp.

[pauln]

No. Accurately reproducing what is on the recording is not the same as accurately reproducing the live event. The two CAN produce the same results ONLY if the recording correctly captures the live event.

A simple example is:

Accurate reproduction of the recording means that if the recording is bright, then the sound coming from the HiFi system should be equally bright.

Accurate reproduction of the live even means that if the recording is brighter than the live event, then the HiFi will have to be warmer to compensate for the failure in the recording.

So what do you do if the recording isn't bright? - change your "warm" hifi? Why not just have a neutral, accurate amplifier with tone controls?

 

[Macspur]

No. Accurately reproducing what is on the recording is not the same as accurately reproducing the live event. The two CAN produce the same results ONLY if the recording correctly captures the live event.

A simple example is:

Accurate reproduction of the recording means that if the recording is bright, then the sound coming from the HiFi system should be equally bright.

Accurate reproduction of the live even means that if the recording is brighter than the live event, then the HiFi will have to be warmer to compensate for the failure in the recording.

So what do you do if the recording isn't bright? - change your "warm" hifi? Why not just have a neutral, accurate amplifier with tone controls?

Precisely!

 

[CnoEvil]

No. Accurately reproducing what is on the recording is not the same as accurately reproducing the live event. The two CAN produce the same results ONLY if the recording correctly captures the live event.

A simple example is:

Accurate reproduction of the recording means that if the recording is bright, then the sound coming from the HiFi system should be equally bright.

Accurate reproduction of the live even means that if the recording is brighter than the live event, then the HiFi will have to be warmer to compensate for the failure in the recording.

So what do you do if the recording isn't bright? - change your "warm" hifi? Why not just have a neutral, accurate amplifier with tone controls?

Precisely!

Show off!

 

[steve_1979]

Speakers on the other hand all have relatively high levels of distortion and don't even get close to being 'audibly perfect'.

This is why they all sound so different to one another.

IMO. Distortion is only one of the many reasons why speakers sound different eg. (to name but a few)

- Overall design (ported, sealed etc)

- Rigidity / inertness of cabinet

- Isolation of sensitive components

- Handling of sound reflections inside cabinet

- Type of tweeter (metal, ribbon, silk dome)

- Quality / size of woofer

- Arrangement of transducers

- Type of crossover

- Whether a dipole design

etc etc.

All of those things that you have listed here introduce various different types of distortion. That's what makes them sound different.

 

[davedotco]

If recording engineers always got it right, there would be no bad recordings, and no "loudness war".....certain labels and certain recording engineers do much better than others eg. I like Chesky, Telarc, Twisted records and Harmonia Mundi

FWIW I use un-miked stuff as my real world reference, especially piano, violin, trumpet and soprano....and if you go to enough concerts, you get a personal benchmark on how it should sound. It helped that I played in a school orchestra and my mother was a classically trained singer.

IMO. It's easy to over-analyze the whole thing...and even harder to explain to someone who comes at this from a completely different perspective.

Loudness wars is down to post production/mastering and nothing to do with the recording.

I agree that there are good and not so good recording engineers. The not so good ones get less work, charge less and therefore work on lower budget productions. I'm talking here about recording live gigs rather than recording in a studio which is a whole different thing and something I don't have any first hand experience of.

It's also important to realise (as I'm sure most do) that any live gig where the sound goes out over a PA system is essentially recorded, amplified, then broadcast through speakers. Generally in the region of 10 tonnes of speaker at your average rock gig.

Cno. Being able to hear real instruments, up close and personal, is a huge advantage in evaluating the sound of hifi systems, being able to hear them in your own environment is even better. Walking around the studio, miking up, screening and the rest is always pretty instructive, particularly when you move from the studio floor to the control room. Just listening to the piano tuner doing his job is pretty educational.

Pauln. Recording a live gig invariably involves miking the performers and their instruments on stage, well before these signals get anywhere near the PA system. If the band are disciplined and do not insist on outrageous levels on stage it is much the same as a 'one take' recording in a studio, just a little more 'bleed'.

 

[CnoEvil]

All of these things that you have listed here can introduce various different types of distortion.

I'm not sure it's as simple as that.....but I'm far from a speaker designer.

 

[steve_1979]

Amplifiers sound different for many reasons, including the robustness and design of the power supply, or whether they use valves, or how well they implement Class D or Class B.

3. Amplifiers deliberately 'voiced' to sound different to the competition or conform to a house sound.

4. Amplifiers with poor transfer characteristics, such as output impedence, so that their amplitude and phase response reflects that of the speaker.

Amplifiers that sound different only do so because they have high enough levels of distortion for it to be audible.

There are many reasons for why they can have high enough levels of distortion for it to be audible. Some of these reason have already been suggested by yourselves such as the design of the power supply, or whether they use valves, or how well they implement Class D or Class B, or they can be deliberately 'voiced' to sound different etc etc.

I totally agree with you both that all of those things you listed can effect the way an amplifier sounds. But they only effect the way an amplifier sounds by introducing distortion to the amplified signal.

By keeping the levels of distortion low enough for them not to be audible an amplifier can be 'audiably perfect' (maybe I should have used the phrase 'audibly transparent' instead).

 

[steve_1979]

All of these things that you have listed here can introduce various different types of distortion.

I'm not sure it's as simple as that.....but I'm far from a speaker designer.

If the things that you listed don't alter the sound by introducing distortion then how do they alter the sound?

 

[pauln]

Pauln. Recording a live gig invariably involves miking the performers and their instruments on stage, well before these signals get anywhere near the PA system.

Do you really think I was suggesting that there's a mic in front of each PA wing?

 

[CnoEvil]

By keeping the levels of distortion low enough for them not to be audible an amplifier can be 'audiably perfect' (maybe I should have used the phrase 'audibly transparent' instead).

This makes more sense to me.....though it's still a problematic area.

 

[CnoEvil]

All of these things that you have listed here can introduce various different types of distortion.

I'm not sure it's as simple as that.....but I'm far from a speaker designer.

If the things that you listed don't alter the sound by introducing distortion then how do they alter the sound?

That may take an engineer to answer.

For example, I know that different types of tweeter sound different, but how their distortion levels compare is more than I can tell you.

Different types of ports sound different (Ported vs TL).....again, i don't know how the distortion levels compare.

Are colourations the same as distortions?

How do you know which distortions are audible (your earlier point) and which are not......more questions than answers I'm afraid.

To what extent does frequency response / sensitivity / impedance, effect sound, and how much of that is distortion?

 

[oldric_naubhoff]

is there a perfect amp? for me it would have to meet a few criteria:

- no global negative feedback: I want to be sure that what distortion characteristics I see for a sinewave I'll be getting for music too

- vanishingly low distortion at lowest power levels (up to 1 Watt) with gently rising THD with power before clipping - a.k.a. a natural power transfer: most commonly seen is situation where an amp will exhibit lowering of THD with power (due to use of NFB) which is totally useless when you realise that most of your listening drifts below 1Wpc

- even THD with relation to frequency: most amps would have rising THD with rising frequency for any given power level, even from as low as 1kHz. it's not uncommon that an amp which boasts THD levels lower than 0,05% @ full throttle would have some 0.5% THD in treble region. it's possible because the power measurements are usually done @1kHz

- would have ruler flat frequency response extending faaaar beyond audible range (like -3dB point @ around 300kHz or more). this ensures the amp will have a high enough slewing rate to prevent the amp from exhibiting transient saturation and distorting high frequencies. it's very common observation that "bright" sounding amps would be the ones which frequency response drops off the cliff right beyond 20kHz mark

- would be a perfect (or near perfect) current source down to 2 Ohm load. which means with every impedance halving the power should double (or close to that). that will ensure the amp will be able to efficiently drive everything on this planet. even ESLs will pose no problem. there's one proviso though; I'm taking into consideration only the true IEC RMS powers and how they relate. because it wouldn't be uncommon for some manufacturers to measure power output @ 2 Ohms and then simply divide the result by 2 for 4 Ohms and divide it by 2 again for 8 Ohms for the spec sheet. where in fact the results for 8 and 4 Oms would be much higher according to IEC standards. this is cheating (although not entirely telling lies)

it's hard for most amps to meet all those criteria (especially the first one) but I actually found a manufacturer of audio gear (making mostly pre and power amps) which amps meet all of them. never heard the amps but I wouldn't hesitate to buy one blind because from studying the performance graphs I know they must sound simply glorious. and they are class A BTW

 

[CnoEvil]

.....from studying the performance graphs I know they must sound simply glorious. and they are class A BTW

Sounds like my sort of amp. :bounce:

 

[steve_1979]

By keeping the levels of distortion low enough for them not to be audible an amplifier can be 'audiably perfect' (maybe I should have used the phrase 'audibly transparent' instead).

This makes more sense to me.....though it's still a problematic area.

A bad choice of word on my behalf.

Just read 'audiably transparent' instead of 'audiably perfect'.

 

[steve_1979]

Are colourations the same as distortions?

Yes.

Different types of ports sound different (Ported vs TL).....again, i don't know how the distortion levels compare.

Different enclosure types will cause different phase distortions and amplitude variations in the frequency responce.

How do you know which distortions are audible (your earlier point) and which are not

Some types of distortion are more noticable than others (for example phase distortion is generally more noticable than small amplitude variations in the frequency responce).

But what's actually audible can vary from person to person because some people have better hearing than others.

To what extent does frequency response / sensitivity / impedance, effect sound, and how much of that is distortion?

Any variations from a perfectly flat frequency responce is a type of distortion.

But sensitivity and impedance are just measurements of efficiency and resistance so they are not a type of distortion. Although a speaker with too low a sensitivity and/or impedance can cause distortion in the amplifier if it's underpowered.

 

[plastic penguin]

Are colourations the same as distortions?

Yes.

Different types of ports sound different (Ported vs TL).....again, i don't know how the distortion levels compare.

Different enclosure types will cause different phase distortions and amplitude variations in the frequency responce.

How do you know which distortions are audible (your earlier point) and which are not

Some types of distortion are more noticable than others (for example phase distortion is generally more noticable than small amplitude variations in the frequency responce).

But what's actually audible can vary from person to person because some people have better hearing than others.

To what extent does frequency response / sensitivity / impedance, effect sound, and how much of that is distortion?

Any variations from a perfectly flat frequency responce is a type of distortion.

But sensitivity and impedance are just measurements of efficiency and resistance so they are not a type of distortion. Although a speaker with too low a sensitivity and/or impedance can cause distortion in the amplifier if it's underpowered.

Steve, you seem like a learned character. If you walked into a house (could be painful) and heard a decent set-up for the first time could you immediately tell if there is distortion or colouration?

Does this all mean that a warm set-up like Marantz pearl Lite with Dynaudio speakers would be distorted?

 

[Macspur]

is there a perfect amp? for me it would have to meet a few criteria:

- no global negative feedback: I want to be sure that what distortion characteristics I see for a sinewave I'll be getting for music too

- vanishingly low distortion at lowest power levels (up to 1 Watt) with gently rising THD with power before clipping - a.k.a. a natural power transfer: most commonly seen is situation where an amp will exhibit lowering of THD with power (due to use of NFB) which is totally useless when you realise that most of your listening drifts below 1Wpc

- even THD with relation to frequency: most amps would have rising THD with rising frequency for any given power level, even from as low as 1kHz. it's not uncommon that an amp which boasts THD levels lower than 0,05% @ full throttle would have some 0.5% THD in treble region. it's possible because the power measurements are usually done @1kHz

- would have ruler flat frequency response extending faaaar beyond audible range (like -3dB point @ around 300kHz or more). this ensures the amp will have a high enough slewing rate to prevent the amp from exhibiting transient saturation and distorting high frequencies. it's very common observation that "bright" sounding amps would be the ones which frequency response drops off the cliff right beyond 20kHz mark

- would be a perfect (or near perfect) current source down to 2 Ohm load. which means with every impedance halving the power should double (or close to that). that will ensure the amp will be able to efficiently drive everything on this planet. even ESLs will pose no problem. there's one proviso though; I'm taking into consideration only the true IEC RMS powers and how they relate. because it wouldn't be uncommon for some manufacturers to measure power output @ 2 Ohms and then simply divide the result by 2 for 4 Ohms and divide it by 2 again for 8 Ohms for the spec sheet. where in fact the results for 8 and 4 Oms would be much higher according to IEC standards. this is cheating (although not entirely telling lies)

it's hard for most amps to meet all those criteria (especially the first one) but I actually found a manufacturer of audio gear (making mostly pre and power amps) which amps meet all of them. never heard the amps but I wouldn't hesitate to buy one blind because from studying the performance graphs I know they must sound simply glorious. and they are class A BTW

I'm intrigued... so what brand are these amps?

Mac

 

[lindsayt]

Amplifiers, digital sources and cables can all be 'audibly perfect' to the human ear.

But 'audibly perfect' doesn't mean that there isn't any distortion there at all. It just means that the levels of distortion are so low that it is impossible to hear them.

As far as I understand there are only two reasons why some amplifiers sound different to each other.

1. Because they have high enough levels of distortion for it to be audible.

2. Because they are being overdriven into clipping (which often happens at a lower volume level than many people realise).

OK Steve, we can make a game of this.

You nominate a digital source or amplifier that you think is audibly perfect. Then I see if I can come up with an alternative source or amplifier that sounds better in certain respects and better overall with a certain type of music or a certain type of speakers. It's then up to you if you then go on to actually compare your audibly perfect nomination against my challengers.

 

[oldric_naubhoff]

I'm intrigued... so what brand are these amps?

Mac

since you ask. it's not a secret. AM Audio of Italy (surprisingly enough most examples of audio engineering philosophy that strikes the right chords in me comes from Italy. maybe there's something right about the air in northern Italy? )

there's a section on the website called "about us - reviews" where there's a lot of scans attached. the ones most interesting are from magazine "Audio Review" (I wish I could could speak Italian, I would then surely subscribe to the mag. this may be one of the most comprehensive hi-fi mags available, with extensive tech section for every review). unfortunately the scans are not of the best quality so you'll be missing on some of the fine print but if you know what you're looking at you'll be able to decipher them.

this scan is from the review of their new flagship pre-power (less powerful models should be available soon) and it's a good starting point as it's better resolution scan. this pre - power represents very solid engineering IMO as the very good measured performance would suggest, especially when you take into account the fact that there's no negative feedback to make up for the design deficiencies to make the amp measure acceptably. but ironically the less powerful models already available offer even better measured performance in some areas than this one. but they wouldn't be offering 300W RMS into 8 Ohms in pure class A! (not that I ever needed that much power for my typical listening levels).

 

[Overdose]

But it would largely put an end to box swapping in the never ending quest for 'better'.

i don't have a dog in this fight, but I'm interested to know what you have against "box swapping". Sales sustain the industry. With fewer sales, retailers and manufacturers will go to the wall. Would that be a good thing?

I don't have anything against box swapping. If that's what you want to do, then fine, but it would be a more honest situation if people realised that by doing so would not neccessarily gain them any sonic benefits, however much they believe that to be the case.

ok, that wasn't my point, but it doesn't matter.

I know it doesn't matter (to me anyway), but the hifi industry is in decline, certainly in this country and manufacturers are going to have to move with the times or go under. Moving with the times is progress and largely, a good thing.

 

[CnoEvil]

but they wouldn't be offering 300W RMS into 8 Ohms in pure class A! (not that I ever needed that much power for my typical listening levels).

:O

Does it weigh a quarter of a ton and reach 300 deg C?

 

[oldric_naubhoff]

but they wouldn't be offering 300W RMS into 8 Ohms in pure class A! (not that I ever needed that much power for my typical listening levels).

:O

Does it weigh a quarter of a ton and reach 300 deg C?

yeah, a nearly quarter of a tonne (for the pair)

don't know anything bout the temperature though so you may be right look at it that way; you could easily get rid of a cooker and prepare your meals while listening to music!

 

[CnoEvil]

but they wouldn't be offering 300W RMS into 8 Ohms in pure class A! (not that I ever needed that much power for my typical listening levels).

:O

Does it weigh a quarter of a ton and reach 300 deg C?

yeah, a nearly quarter of a tonne (for the pair)

don't know anything bout the temperature though so you may be right look at it that way; you could easily get rid of a cooker and prepare your meals while listening to music!

The lecky bill would be eye watering......but at least the sound would help take the mind off it (for a while).

 

[gregvet]

I dunno, the class a integrated looks interesting too. 4,600 Euros, specs as below

http://www.amaudio.it/eng/page83.html

No way of hearing in the UK unfortunately from what I can see tho.

 

[oldric_naubhoff]

I dunno, the class a integrated looks interesting too. 4,600 Euros, specs as below

http://www.amaudio.it/eng/page83.html

No way of hearing in the UK unfortunately from what I can see tho.

they seem to be doing factory only kind of distribution so even if you lived in Italy unless you were based near Milan you'd find it hard to have a listen. but the upside is where else would you find a brand new 40 watter class A power amp for under 3k Euro?