Topping LA90 Discrete Amplifier Review

[stefanovesa]

Isnt more headroom preferable than inaudible sinad chase?

If I have understood correctly, a source like a DAC or a CD player doesn't need any headroom in the next input device, unlike e.g a pre phono where the internal voltage headroom and the next chain link headroom, must accomodate the highly variable output of a huge number of different cartridges.

 

[terryforsythe]

If I have understood correctly, a source like a DAC or a CD player doesn't need any headroom in the next input device

Theoretically that should be true. But, in reality, sometimes engineers designing equipment overlook something. The miniDSP HTx, for example, EDIT 2: has had a 0dBFS multitone issue, but it goes went away with a lower output voltage below -10dB. Side note: That is why I run my amplfiers in high gain - that way I am sure to keep my HTx output voltage below the range where the multitone issue is was present.

MiniDSP Flex HTx Measurement (FW1.1.17 Update)

This is a set of measurements of the Minidsp Flex HTx. It was sent to me by a friend. I have no affiliation to Minidsp. This set of measurements should not be directly compared to Amirm's test as I'm using Comos ADCiso here, but it should provide an general idea of performance of this gear. I'm...

www.audiosciencereview.com

EDIT: There have been firmware updates since then, but I don't know whether any of the updates have corrected the issue.

EDIT 2: I just went back through the thread and indeed it was fixed with an update. I guess I can go back to low gain on my amplifiers if I ever choose to do so.

 

[bigpilot]

Is this a class A or class AB design? And what chip is the IC version using?

 

[terryforsythe]

Is this a class A or class AB design?

Given its thermal output, I think it is class AB. If it were class A I would expect it to run much hotter than it does given the size of its enclosure.

And what chip is the IC version using?

Topping claims it uses "discrete NFCA power amplifier circuitry", i.e., it does not use a power amp IC. Hence the name of the unit.

The original LA90 used a power amp IC, though, if my understanding is correct.

 

[dcolak]

I connected Topping LA90 Discrete to my JBL Studio 4410's and it sounded perfect.

As I have two LA90's, I put them into bridge mode, each driving a single speaker.

I got less bass! It was no longer "blooming" it was missing in action.

Here is what ChatGPT had to say on it, does it make sense?

Why ASR didn’t mention it​

1) They don’t evaluate “bass feel”​

Audio Science Review focuses on:

• SINAD
• THD+N
• frequency response
• power into resistive loads

2) Their load is NOT your speaker​

ASR tests with:

• resistors (4Ω / 8Ω dummy loads)

Your JBL Studio 4410 is:

• reactive load
• impedance varies with frequency
• produces back EMF

None of that appears in ASR graphs.

3) Bridging is measured, not stressed​

They might show:

• more power in bridge mode

But they don’t test:

• sustained low-frequency current demand
• woofer control
• damping under reactive load

So the failure mode you hear = completely invisible in their tests

4) LA90 design philosophy​

The Topping LA90 Discrete is built for:

• ultra-low distortion
• not maximum current delivery

In stereo → fine
In bridge → current limits show up first in bass

Again: not captured in SINAD charts.

5) Measurement blind spot (important)​

ASR assumes:

if frequency response is flat → bass is correct

Reality:

bass quality depends on control, not just amplitude

---

The core issue​

ASR answers:

“Is the signal accurate into a resistor?”

You are hearing:

“Can this amp control a real woofer under load?”

Those are different problems.

---

One-liner​

ASR didn’t miss it — they simply never measured the thing that fails.

---

Your result is actually expected (engineering-wise)​

• Bridge → impedance per amp ↓
• Current demand ↑
• LA90 → limited current capability

Bass is the first thing to degrade.

 

[Panelhead]

For chuckles, try a Topping PA5 II. They are under 200.00. I found the bass control was incredible. This driving is a four ohm, 15 inch woofer.
Tried it with subwoofer and without. The subwoofer helps fill in the bottom of the speaker.

 

[testp]

would LA90 Discrete be fine, with speaker that has:

min impedance 3.2,
min EPDR 1.6.

+ i would use rca-to-xlr.

i'm worried about load, and power as well (if rca-to-xlr halves the power, but don't know how this amplifier is designed, so maybe that is fine)

i saw Amir testing AsciLab A6 which has same impedance + similar curve with LA90 non-D, but can't see EPDR from review.


i was thinking maybe finding one used, ~869€ new looks little ridiculous for the amount of power at least to me.

EDIT - nrvmd the above question, i got it sorted!

 

[dcolak]

would LA90 Discrete be fine, with speaker that has:

min impedance 3.2,
min EPDR 1.6.

+ i would use rca-to-xlr.

i'm worried about load, and power as well (if rca-to-xlr halves the power, but don't know how this amplifier is designed, so maybe that is fine)

i saw Amir testing AsciLab A6 which has same impedance + similar curve with LA90 non-D, but can't see EPDR from review.


i was thinking maybe finding one used, ~869€ new looks little ridiculous for the amount of power at least to me.

I'd suggest buying a real amp with real driving power. The "100w" of LA90 DISCRETE (bridged x2) are nothing like real 100w of a 20kg amp.

Even 30w typical amps drive with more authority real speakers.

Ebay is full of real, vintage amps, see Yamaha a-2000 you can get for cheap.

 

[Jcrow]

View attachment 519806

I connected Topping LA90 Discrete to my JBL Studio 4410's and it sounded perfect.

As I have two LA90's, I put them into bridge mode, each driving a single speaker.

I got less bass! It was no longer "blooming" it was missing in action.

Here is what ChatGPT had to say on it, does it make sense?

Why ASR didn’t mention it​

1) They don’t evaluate “bass feel”​

Audio Science Review focuses on:

• SINAD
• THD+N
• frequency response
• power into resistive loads

2) Their load is NOT your speaker​

ASR tests with:

• resistors (4Ω / 8Ω dummy loads)

Your JBL Studio 4410 is:

• reactive load
• impedance varies with frequency
• produces back EMF

None of that appears in ASR graphs.

3) Bridging is measured, not stressed​

They might show:

• more power in bridge mode

But they don’t test:

• sustained low-frequency current demand
• woofer control
• damping under reactive load

So the failure mode you hear = completely invisible in their tests

4) LA90 design philosophy​

The Topping LA90 Discrete is built for:

• ultra-low distortion
• not maximum current delivery

In stereo → fine
In bridge → current limits show up first in bass

Again: not captured in SINAD charts.

5) Measurement blind spot (important)​

ASR assumes:



Reality:

---

The core issue​

ASR answers:



You are hearing:



Those are different problems.

---

One-liner​

---

Your result is actually expected (engineering-wise)​

• Bridge → impedance per amp ↓
• Current demand ↑
• LA90 → limited current capability

Bass is the first thing to degrade.

I put this contents of this post in chat gpt to check it. It disagreed with most of it and mentioned that a change in gain structure in bridge mode and potentially mismatched levels being the main issue.
Don’t trust AI too much. I’d actually be keen to hear what someone with knowledge has to say on the issue.