• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Stereo Amplifiers >$500 (Not Integrated)

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#1
Anyone know any great stereo amplifiers that measure well that are close to $1k (maybe a little more)?

Seems hard to find as most reviews are just subjective opinions backed by nothing.

The stereo amplifiers measured by Amir so far all seem to be not so great compared to the great amplifier options we have for headphones!
 

Blumlein 88

Major Contributor
Joined
Feb 23, 2016
Messages
4,693
Likes
3,117
#2
The soon to come March Audio amps might be worth looking into if you aren't in a hurry. Listed specs are very good, and Amir will likely test one in time.

Would you consider 2nd hand? What speakers are you going to be using? How much power are you looking to find?
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#3
Why not an integrated?

Since the late 1980s, 'integrated' amplifiers have essentially been power amplifiers with high gain stages and passive front ends when operated in direct mode. That is no different to a power amp with a level pot. What you gain of course is a stack of useful inputs and flexibility.

Performance doesn't have to suffer at all. There's an enormous range of very high quality amplifiers on the secondary market.
 

JJB70

Addicted to Fun and Learning
Joined
Aug 17, 2018
Messages
684
Likes
1,043
Location
Milton Keynes, England
#4
I think there is a lingering perception that separate pre-amp and power amp combinations are better than integrated amplifiers, possibly driven by manufacturers and hi-fi reviewers. In most cases a good integrated amplifier is more than good enough. On the second hand market if you avoid some of the real cult and/or collectible models it is amazing just what you can get for $500, if buying new you and taking $500 as price to pay rather than SRP then you can still get some very good new amplifiers from the major players.
 

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#5
I want a dedicated amplifier and not integrated because I am using it for my desktop nearfield stereo bookshelf pair of speakers and I already have a DAC and I want to use balanced out as the grounds noise leakage I am dealing with is very troublesome (balanced out seems to get rid of it).

Also those older designs probably don't measure as well as the solid state Class D derivatives like a Hypex NCore or something (this is just a guess?)?
 

jasonq997

Member
Patreon Donor
Joined
Nov 4, 2018
Messages
54
Likes
60
#6
Why not an integrated?

Since the late 1980s, 'integrated' amplifiers have essentially been power amplifiers with high gain stages and passive front ends when operated in direct mode. That is no different to a power amp with a level pot. What you gain of course is a stack of useful inputs and flexibility.

Performance doesn't have to suffer at all. There's an enormous range of very high quality amplifiers on the secondary market.
I would be curious to hear what you think represents a "top 5" for good used integrated amps.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#7
I would be curious to hear what you think represents a "top 5" for good used integrated amps.
It would be very difficult to narrow down to even a top 50 or 100. There are simply so many quality, integrated amplifiers out there made in the last 40+ years.

The OP appears to have a noise issue which apparently is 'solved' by going balanced. Using an input topology to solve a problem instead of curing the issue is a band aid approach IMO.

Also those older designs probably don't measure as well as the solid state Class D derivatives like a Hypex NCore or something
These hyped class D amplifier designers are very clever by extolling what they see as virtues and conveniently ignoring, dismissing, or failing to specify parameters that have for decades defined the very essence of high fidelity amplification.

The focus seems to be on biggish power numbers, but power at rated distortion levels that are a complete joke. Consider the n-core specifies its (headlined) maximum output at 1% THD. No amplifier, aspiring to High Fidelity pretensions since the 1950s would have dared printed such a hideous number. That is the realm of the bad old days of ghetto blasters and car stereos.

Harold Leak specified point one (0.1%) as the point where HiFi started in 1958 with his tube gear. 1958! It's only at half power numbers where these Class D modules approach THD levels that have been common in Class AB amplifiers for many decades.

As an example let's take the "NC500 OEM Definitive performance class D amp" with "its measured and sonic performance (that) actually raises the bar for audio amplifiers of any description."

The headline numbers are:

1545513601559.png


Let's look at their own measured AP THD vs Po:

1545515110424.png

I have several amplifiers and power amplifiers where the ratings were XXX watts per channel from 20-20KHz, at any power from 250mW to rated power output and THD of less than 0.003% or similar.

Based on this AP graph, the amplifier module would be a [email protected] (not a [email protected]) module at similar distortion levels to a typical high performance 35 year old power amplifier. Proper HiFi manufacturers didn't specify their headline numbers halfway up the skyward exponential trajectory of their THD curve!

Let's look at FR.

I have plenty of amplifiers with small signal frequency responses out to 200KHz+, many over 300KHz and some up to 600KHz. They are ruler flat within the audible bandwidth, above and below.

The NC500 by their own FR plot is half a dB down at 20KHz at a measly 1 watt. Ask yourself, is that really 'raising the bar'? Clearly, the bar is drooping rapidly, and in the audible bandwidth too.

1545515522084.png


Anyway, I want to see some scope shots of square waves on these things driven into capacitive and reactive loads.
I want to see expanded FR plots that go from DC to 50Khz with tight deviations.
I want to see overload recovery plots.
I want to see proper FTC style, soak testing then full power testing as per the guidelines, not short term peak power, pretend numbers.

:)
 

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#8
It would be very difficult to narrow down to even a top 50 or 100. There are simply so many quality, integrated amplifiers out there made in the last 40+ years.

The OP appears to have a noise issue which apparently is 'solved' by going balanced. Using an input topology to solve a problem instead of curing the issue is a band aid approach IMO.

:)
How would you suggest I fix this issue? It is a ground loop caused by my GPU when the GPU starts to go towards full load. I tried some iFi USB thing that helped but also introduced its own problems.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,234
Likes
1,369
#9
It would be very difficult to narrow down to even a top 50 or 100. There are simply so many quality, integrated amplifiers out there made in the last 40+ years.

The OP appears to have a noise issue which apparently is 'solved' by going balanced. Using an input topology to solve a problem instead of curing the issue is a band aid approach IMO.



These hyped class D amplifier designers are very clever by extolling what they see as virtues and conveniently ignoring, dismissing, or failing to specify parameters that have for decades defined the very essence of high fidelity amplification.

The focus seems to be on biggish power numbers, but power at rated distortion levels that are a complete joke. Consider the n-core specifies its (headlined) maximum output at 1% THD. No amplifier, aspiring to High Fidelity pretensions since the 1950s would have dared printed such a hideous number. That is the realm of the bad old days of ghetto blasters and car stereos.

Harold Leak specified point one (0.1%) as the point where HiFi started in 1958 with his tube gear. 1958! It's only at half power numbers where these Class D modules approach THD levels that have been common in Class AB amplifiers for many decades.

As an example let's take the "NC500 OEM Definitive performance class D amp" with "its measured and sonic performance (that) actually raises the bar for audio amplifiers of any description."

The headline numbers are:

View attachment 19245

Let's look at their own measured AP THD vs Po:

View attachment 19246
I have several amplifiers and power amplifiers where the ratings were XXX watts per channel from 20-20KHz, at any power from 250mW to rated power output and THD of less than 0.003% or similar.

Based on this AP graph, the amplifier module would be a [email protected] (not a [email protected]) module at similar distortion levels to a typical high performance 35 year old power amplifier. Proper HiFi manufacturers didn't specify their headline numbers halfway up the skyward exponential trajectory of their THD curve!

Let's look at FR.

I have plenty of amplifiers with small signal frequency responses out to 200KHz+, many over 300KHz and some up to 600KHz. They are ruler flat within the audible bandwidth, above and below.

The NC500 by their own FR plot is half a dB down at 20KHz at a measly 1 watt. Ask yourself, is that really 'raising the bar'? Clearly, the bar is drooping rapidly, and in the audible bandwidth too.

View attachment 19247

Anyway, I want to see some scope shots of square waves on these things driven into capacitive and reactive loads.
I want to see expanded FR plots that go from DC to 50Khz with tight deviations.
I want to see overload recovery plots.
I want to see proper FTC style, soak testing then full power testing as per the guidelines, not short term peak power, pretend numbers.

:)

This is why I’m looking forward to @amirm’s review of March Audio’s Hypex based units (reich I presume will happen at some point). It would be nice if a fuller suite of measurements than usual were run on one of these units (hint hint).

Re: distortion testing, could you also explain John how the power testing used in these Hypex AP graphs differs from full power testing used in the guidelines?
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
5,930
Likes
1,876
Location
Riverview, FL
#10
Re: distortion testing, could you also explain John how the power testing used in these Hypex AP graphs differs from full power testing used in the guidelines?
One thing, the Class-D type amps typically require the use of a low-pass filter before "measurement".

"The main limitation is measuring class-D amps as their high frequency noise can cause problems. I plan to build a passive filter for it (AP charges some $1,500 for theirs!). Until then, measuring them accurately may be a problem." - https://www.audiosciencereview.com/...-500-watt-minimum-4ohm.3939/page-2#post-96613
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,234
Likes
1,369
#11
One thing, the Class-D type amps typically require the use of a low-pass filter before "measurement".

"The main limitation is measuring class-D amps as their high frequency noise can cause problems. I plan to build a passive filter for it (AP charges some $1,500 for theirs!). Until then, measuring them accurately may be a problem." - https://www.audiosciencereview.com/...-500-watt-minimum-4ohm.3939/page-2#post-96613
That's true, but if I understood @restorer-john correctly, what he's saying is that the Power vs Distortion measurements used by Hypex show short term peak power rather than continuous power.

It's this statement that I'd like to understand better...
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#12
...It is a ground loop caused by my GPU when the GPU starts to go towards full load...
The GPU clearly is pulling a ton of current when heavily loaded and modulating the power supply rails and/or the 0v common to different potentials all over the MB. It could be a nightmare to track down or perhaps impossible. The power supply capability, the cable dressing, the placement all can have effect.

I guess I am wondering why a heavy GPU load even occurs during listening to music in the first place. If the machine is extreme gaming while listening to music, maybe.
 

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#13
The GPU clearly is pulling a ton of current when heavily loaded and modulating the power supply rails and/or the 0v common to different potentials all over the MB. It could be a nightmare to track down or perhaps impossible. The power supply capability, the cable dressing, the placement all can have effect.

I guess I am wondering why a heavy GPU load even occurs during listening to music in the first place. If the machine is extreme gaming while listening to music, maybe.
It is an issue when I am trying to play games and listen to the game! Balanced seems to be a decent solution?

Also the noise starts to leak in pretty quickly, right when it starts to render the actual game.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#14
That's true, but if I understood @restorer-john correctly, what he's saying is that the Power vs Distortion measurements used by Hypex show short term peak power rather than continuous power.
It's their specification, not mine. Peak power has been used by shysters in the industry for ever. So much so, the FTC brought in the amplifier rule in 1974 to stamp out the practice.

Now, with their collective eyes off the ball, we see silly shelf systems with 2000W ratings that bear no resemblance to reality.

Let's see some real amplifier testing, you know, like they used to do in the old days. :)

Amir will need a bunch of non-inductive high power dumy loads and obviously build the LPF that kills all the extraneous crap that comes out of Class D amplifiers. Someone has reverse engineered the overpriced AP one and I have a schematic someplace. Maybe I should build one for him and one for me.
 
Last edited:

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#15
It is an issue when I am trying to play games and listen to the game! Balanced seems to be a decent solution?
It's difficult to argue with something that works for you and your particular situation. Go for it with balanced. Alan (March Audio) will have something for you soon enough I would say.

:)
 

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#16
I'm just hoping for any solution (if it involves me not replacing my speaker amplifier with one with balanced inputs, it can save me a lot of money)! If you have any recommendations, I would love to hear them. I guess I will wait for March Audio as well. Thanks for the insight.
 

andreasmaaan

Major Contributor
Patreon Donor
Joined
Jun 19, 2018
Messages
2,234
Likes
1,369
#17
It's their specification, not mine. Peak power has been used by shysters in the industry for ever. So much so, the FTC brought in the amplifier rule in 1974 to stamp out the practice.

Now, with their collective eyes off the ball, we see silly shelf systems with 2000W ratings that bear no resemblance to reality.

Let's see some real amplifier testing, you know, like they used to do in the old days. :)

Amir will need a bunch of non-inductive high power dumy loads and obviously build the LPF that kills all the extraneous crap that comes out of Class D amplifiers. Someone has reverse engineered the overpriced AP one and I have a schematic someplace. Maybe I should build one for him and one for me.
Ok I think I'd misunderstood your previous post..

When I look at the graph you posted in your link ("6.2 Large Signals Test"), I see a test of continuous power output, not "peak power" blah blah. The maximum continuous power could be stated as 400W if you take 1% as the yardstick, or 300W if we set it at 0.1% ( and I agree that the latter is a more useful yardstick). But it's 100% the max. continuous power that the graph is giving us, right?

Just wanted to clarify as your word on amplifier measurements is gospel to me and the way I'd understood your previous post seemed at odds with what I thought I knew.
 

restorer-john

Major Contributor
Joined
Mar 1, 2018
Messages
1,333
Likes
1,649
Location
Gold Coast, Queensland, Australia
#18
Just wanted to clarify as your word on amplifier measurements is gospel to me and the way I'd understood your previous post seemed at odds with what I thought I knew.
Don't take me as gospel. :)

Until we see proper tests (1hour preconditioning and a full 5 minute (min) continuous), we really don't know what their 'peak' numbers really are. 'Peak power' in terms of the testing magazines/reviewers and guys like me do is usually the equivalent of the EIA/J toneburst tests which are viewed on an CRO/DSO looking for waveform mild clipping on the 'bursts' and doing a peak power calculation from that. Depending on the power supply and design, you can have results similar or identical to the continuous rating or considerably larger numbers that look great on paper.

Consider NAD/Proton/Hitachi et al who 'gamed' and practically invented the 'high dynamic headroom' push in the 70s/80s. It gave spectacular transient numbers in lightweight packages at a low cost. But continuous high power amplifiers they were not.

Much of the hype of this Class D gear is the 'efficiency' and the consequent smaller packages and smaller heatsink capacities. Basically, I want to see these things put on the bench and pushed hard- for hours. I want to see what they really can do long-term, not just flashy big numbers on a page.
 
Last edited:

Sylafari

Active Member
Joined
Jan 20, 2018
Messages
106
Likes
49
#19
I'm considering NAD C268 (Class D) or the much older Parasound A23 (Class AB). Was wondering if also using a passive preamplifier with balanced inputs and outputs with some XLR to RCA adapter could work out?
 
Joined
Apr 28, 2018
Messages
59
Likes
86
Location
Columbia, SC
#20
Get a bigger power supply for your PC? I had chirping from my graphics card, but this went away when I went to a Platinum Seasonic 800W power supply. Of course this is a guess as I don't know any details about your setup, but google GPU chirp/noise and power supply. Maybe this is what you are hearing, not sure... My noise was coming from the PC obviously, sounds like your noise is from your speakers.

How would you suggest I fix this issue? It is a ground loop caused by my GPU when the GPU starts to go towards full load. I tried some iFi USB thing that helped but also introduced its own problems.
 

Similar threads

Top Bottom