• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Topping LA90 Discrete Amplifier Review

Rate this stereo amplifier

  • 1. Poor (headless panther)

    Votes: 14 3.3%
  • 2. Not terrible (postman panther)

    Votes: 21 5.0%
  • 3. Fine (happy panther)

    Votes: 65 15.4%
  • 4. Great (golfing panther)

    Votes: 321 76.2%

  • Total voters
    421
This is an interesting amplifier (size is about nothing).
And something tells me that this is a very well-known and proven design (it just has to be).
It may very well be something like a CFA (ultra-low distortion, the third harmonic is slightly larger than the second and then a decline) but in a more modern implementation (mostly it's the switching power supply about).
The Ovation was and is still formally better at a full declared power, 70 ppm at 20 kHz at 80 W RMS output:

But in general, it's interesting, yes.
An autopsy will show :)
 
… how does it quadruples the power instead of doubling it in bridged configuration?
Isn’t it what a bridged configuration is supposed to deliver—given impedance and assuming no other limitation: 2x voltage, 4x power?
 
So we have, in real terms, a three input power amplifier with a 'massive' 41wpc@8R channel amplifier stage that hits the wall so hard it shuts down. Wow. Impressive. Not.

40wpc per channel was pretty much the minimum, the absolute bare minimum entry into high fidelity in the early 1970s! 40wpc amplifiers were less than $150(US) and they were a genuine, under-rated 40wpc. And that was with speakers boasting 90-95dBSPL@1W/M sensitivity. Most two ways these days are 80-87dB.

This is kid's basement stuff.
I hear what you are saying and agree to a large extent.
As a possible use case though, given their very low distortion at very low power I wonder whether they might not be an excellent pairing with super high-efficiency horns?
 
Thanks, @amirm !

I salute some aspects, the stellar THD+N VS Frequency in particular (unseen performance with any amp so far?). As the previous model, tho, the power output is unremarkable for 900€.

2 of these or hypex ncx500?
nearly same price.
Hypex NCx500 will be much more stable/capable for low impedances loads, not to mention more powerful if that is needed.

For $1,600 (bridged), is it going to be a real AHB1 killer this time?
You mean bridged ? AHB2s are capable of 500W/4Ohms. If not, I don't understand the comparison of two integrated amps bridged VS one power amp.;) When it comes to quality/durability and support, both offerings would share nothing close to comparable. Some have to understand that these things come at a certain cost and purchase choices can't be made on SINAD Vs SINAD only. I mean, that's a great accomplishment from Topping, no contest... How long will that last is my first wondering right now. And at 900€ (MSRP in France), "budget" is not an excuse anymore.
 
Last edited:
Thanks, @amirm !

I salute some aspects, the stellar THD+N VS Frequency in particular (unseen performance with any amp so far?). As the previous model, tho, the power output is unremarkable for 900€.


Hypex NCx500 will be much more stable/capable for low impedances loads, not to mention more powerful if that is needed.

You mean bridged ? AHB2s are capable of 500W/4Ohms. If not, I don't understand the comparison of two integrated amps bridged VS one power amp.;) When it comes to quality/durability and support, both offerings would share nothing close to comparable. Some have to understand that these things come at a certain cost and purchase choices can't be made on SINAD Vs SINAD only. I mean, that's a great accomplishment from Topping, no contest... How long will that last is my first wondering right now. And at 900€ (MSRP in France), "budget" is not an excuse anymore.

I thought they would be comparable based on price. $1,5980 for a paired of the LA90 vs $3,299 for a single AHB2. So if I need a power amp rated 100 W/200W 8/4 ohm, either one can do the job nicely, and with >100 dB SINAD at up to 100 W/200 W 8/4 ohm output, yes/no/maybe?
 
But the trick is that the price for such an amplifier is too high.
No preamp, no motorized low distortion high quality analog volume control, no remote control.
These microscopic distortions do not particularly excite anyone and have not been of interest for a long time.
Everything is the same, 0.003% or 0.0003%, there is no difference in any of possible senses.

But it's still interesting what's inside (maybe they owe someone for something, he he).
 
But the trick is that the price for such an amplifier is too high.
No preamp, no motorized low distortion high quality analog volume control, no remote control.
These microscopic distortions do not particularly excite anyone and have not been of interest for a long time.
Everything is the same, 0.003% or 0.0003%, there is no difference in any of possible senses.

But it's still interesting what's inside (maybe they owe someone for something, he he).

That could be true, and we can add other factors such as aesthetic (subjective), build quality, and more importantly, reliability though that is not going to be easy to determine... If I had $3,300 to spare, I would likely still go with the Benchmark as it looks better to me and likely will last longer.
 
This is a review and detailed measurements of the Topping LA90 Discrete balanced stereo amplifier. It was sent to me by the company and costs US $799.
View attachment 278528
The case is the same as the IC-based Topping LA90 which is fine with me given its high quality look and feel. The power switch doubles as input selector once the unit is on. Volume control can be bypassed using switch in the back:
View attachment 278530

As before, power is provided by a hefty, 64 volt switching power supply rated at 4 amps (likely provides higher current in short bursts). The amplifier as you see is all balanced with three inputs. You have a selection of low and high gains (although the latter should really be called medium gain). It can be bridged and nicely tolerates even 4 ohm loads in that mode. Trigger is provided to enable one power switch to turn on a series of devices.

Note that this is a traditional amplifier and not class D. There is no remote so if you need that, you have to use a DAC/pre-amp with that capability.

Topping LA90 Discrete Measurements
There is a lot to measure here given the fact that we have dual gains and bridge mode. Let's work our way through it starting with low gain in stereo mode:
View attachment 278532

Incredibly, the LA90 Discrete matches the performance of its IC based unit, landing on its perch on top of all other amplifiers ever tested:
View attachment 278533

Switching to high gain costs you a few dBs but you are still ahead the competition:
View attachment 278534

So feel free to use this mode. Noise performance continues to be superb:
View attachment 278535
View attachment 278536

Frequency response is flat of course and load independent:
View attachment 278537

Crosstalk is better than average amplifier but is a step down from the regular LA90:
View attachment 278538

Multitone showcases the superbly low distortion levels:
View attachment 278539

Let's see how we are doing in power department starting with 4 ohms:

View attachment 278540

We wind up good bit more power at 70 watts (vs 56 watts for normal LA90) while giving up nothing in spectacular noise and distortion department. Allowing for more distortion gets us more power:
View attachment 278542

8 Ohm performance follows the same path:
View attachment 278543

Tests so far have been at 1 kHz (other than multitone) so let's vary that and see what happens:
View attachment 278544
In a word, wow! The amplifier simply doesn't care what frequency you throw out it! This is a quite a step ahead of normal LA90 which itself was state of the art.

Power on is silent but off may create a faint noise depending on your speaker sensitivity:
View attachment 278548

Warm up showed one channel struggling a bit but later I found the power connector not fully tight so maybe that was the cause:
View attachment 278549

Regardless, it behaved correctly after a minute.

Topping LA90 Discrete Bridged Measurements
As with normal LA90, bridge mode still provides superb performance:
View attachment 278550
View attachment 278551

The name of the game is power though so let's see what we get into 4 ohm:
View attachment 278552

Notice how we get more power but distortion is nowhere to be found until we clip.

View attachment 278553

Wow! This makes a great case for using a pair of these amps. You get so much power out of each one, rivalling much larger amplifiers.

Reactive Load Testing
The LA90 has a very sharp clipping point. This means that it goes from vanishingly low distortion to full clipping. Combine this with aggressive protection circuit which shuts the unit off and you have the make up of a system that can't be measured in batch mode as my Powercube setup uses. If I set the voltage for 8 ohm load, then it causes the amp to shut down when driven at 4 ohm. Good news is that the amplifier doesn't care about reactive loads. Here are the results for 8 ohm for example:
Code:
Maximum Output
4/11/2023
12:13 AM
Maximum Output Level Results
Selected Load    Ch1 (Vrms)    Ch2 (Vrms)   
8 Ohm; 60 deg Capacitive    21.2    21.3   
8 Ohm; 30 deg Capacitive    21.1    21.2   
8 Ohm; 0 deg Resistive    21.0    21.0   
8 Ohm; 30 deg Inductive    20.2    20.3   
8 Ohm; 60 deg Inductive    20.9    21.0

I could not get a similar output at 4 ohm due to limitation of the instrumentation as stated above. But if I drove the parameters manually, I could get it to tolerate reactive loads just as well.

Selecting 2 ohm caused the protection circuit to kick in. I didn't have time to mess with that.

I spent nearly 2 hours trying to get this test done with the amplifier being pushed to clipping and shut down probably 50 to 100 times. Even though the case got pretty warm, the amp continued to function without complaining.

Conclusions
Traditionally in high-end audio when a company says they are using discrete designs, I get prepared for bad performance. Topping once again proves that this is no barrier for their designers. The discreate LA90 not only matches the exemplary performance of the IC based LA90, but betters it in power handling, producing more power (a needed feature). Pushing noise and distortion down in our top tier performance is very hard but Topping has no problem maintaining its significant lead over even best of the best amplifiers. Their skills and execution in this regard leaves me speechless.

Needless to say, it is my absolute pleasure to add Topping LA90 Discrete to my recommended list. Today is a happy day! :)

P.S. Power amplifiers are least reliable audio products (due to high currents and voltages involved). New designs sometimes need to prove themselves in this regard. So if you are risk averse, please don't be the first to purchase this amplifier.

-----------
As always, questions, comments, recommendations, etc. are welcome.

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
wow... exceptional!
 
It does provide 279 watts at 4 Ohm bridged according the measurement so should be something around 100 - 120(?) at 8 ohm.
Bridging an integrated amplifier is stupid. I know it will work, but I'd rather buy a pure mono block instead.
 
Putting aside my disappointment that my original LA90 is now discontinued after a very short lifespan, I am wondering if there is anything in the specs that suggest the discrete version will sound better than the original?

My original LA90, with reasonably efficient 8ohm speakers, goes plenty loud enough so a few extra watts is no big deal...but for those who understand the specs better than I, do any of the other numbers mean anything in terms of what comes out of my speakers?
 
It already is when you bypass pre-sektion.
Cheers!
sure BUT (IMHO):

(1) it has a big arsed volume control that at least aesthetically is a turn off, esp when running as 2 x monoblocks.
(2) removing the features I mentioned would would reduce cost a little
(3) at least for me, it doesnt sit right using an integrated amp as a power amp

I suppose, given the form factor, you could hide them out of sight.

Peter
 
Putting aside my disappointment that my original LA90 is now discontinued after a very short lifespan, I am wondering if there is anything in the specs that suggest the discrete version will sound better than the original?

My original LA90, with reasonably efficient 8ohm speakers, goes plenty loud enough so a few extra watts is no big deal...but for those who understand the specs better than I, do any of the other numbers mean anything in terms of what comes out of my speakers?
Nah you're good. Both will sound the same
 
For $1,600 (bridged), is it going to be a real AHB1 killer this time?
you probably meant AHB2 - and it is likely no. Benchmark is known for their brand, build quality, support and service. not to mention reliability and resale value. yes some will be tempted to buy 2 of these instead of one AHB2, but most people will stick with AHB2 if they are looking for something that clean.
 
I was just thinking about the power amir got in 4Ω bridged. The power supply is 64V 4.5A that's 288W yet we got 287W at the knee. 100% efficiency?? Somthings off here class AB is not known for being efficient particularly at low impeadance (2ohm internal). 50%-ish would be around what I'd expect. Even allowing for extra some extra headroom in the power supply.
Edit: it must be inverting one rail in bridged mode. That gives exactly 288W @50% efficiency with +64/-64 @4.5 amps.
 
Last edited:
I haven't seen any discrete speaker amplifiers been reviewed on ASR. Is this a more difficult implementation than those for headphone amps? Cheers
 
Back
Top Bottom