Topping LA90 Review (Integrated Amplifier)

[ObjectiveSubjectivist]

I'm more concerned about power handling mono into 4-3ohms. As many speakers even if they are 8ohms nominal, have deeps as low as 4-3ohms.

ps. @amirm please consider measuring it in mono into some lower loads. Also I'm curious about measurements into headphones loads. As I might be considering buying one of these for Susvaras.

 

[abdo123]

No problem: just get one pair of those close to 100dB efficient speakers they had at the concert and hook up the LA90

It’s all about making choices, and these days it’s out of fashion to have large efficient speakers in your room, so high-powered amps have become the norm.

As simple as it is to call out lower powered amps, I can just as well call out low efficiency speakers. It’s neither here nor there…

Powering a 95 dB efficient speaker with the LA90, it would be the same as powering a 85dB efficient speaker with a 10x more powerful amp.

Jokes aside, i’m very happy with the current status quo.

Compression drivers and line arrays have terrible high frequency linearity. It’s really annoying to be honest and i almost forgot about that because of Covid. I was hearing things at my shoulder because of how phasy it was and i even heard sound coming from the back of the venue.

The current compromise (exchange linearity for sensitivity) is way better imo.

The venue i went to was decent and you would expect them to have some Tom Danleys or something but why bother when the majority of your clientale is just under 18 or above 18 with boom boxes and stuff?

 

[abdo123]

Rock concerts are poor examples of what a good listening room is. I was at one recently as well and measured 100+ dB on my phone. I, like the performers, were wearing musicians earplugs. I, like the performers, would like to have good hearing when I'm older, thank you very much. Let's not use a place, which will eventually ruin your hearing, as a reasonable point of reference. BTW you weren't hearing an iota of the actual live instruments there; rather you were hearing whatever came out of the excessively loud PA speakers. But why stop there? Gunshots are >140 dB, do you really want that level of realism in say movie/theater applications?

You’re not getting my point.

A state of the art amplifer should appeal to all customers, that’s why it’s expensive.

and just because you don’t need more power doesn’t mean other people don’t or that their opinions are invalid.

 

[H-713]

I'm impressed by the numbers, but the limited power sort of limits this to desktop use or on the top end of a bi-amped system. Remember, as soon as the amplifier clips, SINAD is at the bottom of the lake. Clipping on short transients doesn't usually sound like what most people assume a clipping amplifier sounds like, so most people will just hear it as "degrade sound quality", but the cause won't be immediately obvious.

What topology is being used in this thing? The suspiciously high SINAD and relatively low output power makes me wonder if it's a huge parallel array of monolithic op-amps, or some variation of that. Maybe monolithic op-amps directly driving the output devices?

Also, can we get standard 20mm spacing for the output connectors so they're compatible with MDP connectors? Please?

 

[voodooless]

Jokes aside, i’m very happy with the current status quo.

I’m not saying either option is bad. It’s just that people make other choices. From the responses here it’s clear what is currently in fashion.

Compression drivers and line arrays have terrible high frequency linearity.

They don’t have to be. It’s a design choice. For a venue it’s SPL over linearly. For HiFi other choices can be made. It’s not easy to do, but nor is creating a good coaxial.

A state of the art amplifer should appeal to all customers, that’s why it’s expensive.

It’s (relatively) expensive because it’s advances certain objective performance metrics by a good deal. Expensive stuff almost never appeals to all people.

and just because you don’t need more power doesn’t mean other people don’t or that their opinions are invalid.

A coin has two sides

 

[Bruce Morgen]

...And with 90 dB sensitivity, some roomgain and two loudspeakers the acoustic output will be higher than 100 dB. Most loudspeakers cant reproduce 100 dB without very high levels of distortion.

I think its nice to be able to buy the worlds best amplifier ( highest SINAD ) for 1000 dollars.

But:
To be a realist - two Aiyima a07 and a two-way active loudspeaker with a dsp crossover such as t racks 4x4 will probably play the music much better overall compared to an average passive bookshelf loudspeaker with this topping amplifier. Its the passive crossovers in most loudspeakers thats to badly engineered to give a really good result, even if the amplifier is the worlds best.

A pair of excellent active three-way studio monitors -- which could comprise an entire two-channel system with the addition of any decent DAC -- can be had for the same MSRP. Anybody want to take a stab on how many $$$ worth of passive speakers in addition this (truly outstanding!) amp it would take to equal (e.g.) a pair of "2nd Wave" Kali IN-8s in audible performance?

 

[pma]

What topology is being used in this thing? The suspiciously high SINAD

Low noise is due to low gain (too low). Regarding THD part of the THD+N number, it is achievable in linear designs, if you make a sophisticated feedback network.
To me, this amplifier, similarly as PA5, is designed for numbers and top SINAD chart position, as a marketing and sales goal. It is not designed for rationality and long term reliability.

 

[abdo123]

They don’t have to be. It’s a design choice. For a venue it’s SPL over linearly.

I don’t want to start this discussion cause it’s heavily off-topic but for high-frequency linearity to 10KHz for a line array you need to have the drivers be 2.5cm apart.

No popular commercial design does this for several reasons. And if they do they’re not really popular.

It’s both physics and demand. No line array can defeat this with design.

 

[voodooless]

I don’t want to start this discussion cause it’s heavily off-topic but for high-frequency linearity to 10KHz for a line array you need to have the drivers be 2.5cm apart.

Well, skip the line array then . You gave a second option, I’ll take that one

 

[H-713]

Low noise is due to low gain (too low). Regarding THD part of the THD+N number, it is achievable in linear designs, if you make a sophisticated feedback network.
To me, this amplifier, similarly as PA5, is designed for numbers and top SINAD chart position, as a marketing and sales goal. It is not designed for rationality and long term reliability.

Its general lack of usefulness ($800 35W amplifier with a power brick...) does make me think it might have been made for the sole purpose of having the top place on SINAD charts.

 

[Talisman]

Leaving aside a reliability to be verified (in the light of the pa5 thread). It seems to me that all the expendable is being sacrificed on the altar of a sinad that ultimately interests us relatively, beyond a certain level. External power supply to avoid any interference, no display, no remote control, low power .... And I agree with PMA in this case, this is a race for marketing and approval of a slice of the audio market that looks at these numbers as an indicator absolute. Topping is clearly not a charity company, it goes where it believes the money will go, and the money is evidently going on pumping values well beyond the level where they can make a real difference to the user. But if we let ourselves be enchanted by the sinad like a snake by its charmer, how are we different from an audiophile who lets himself be enchanted by exotic materials and strange shapes? 800 euros, can you feel the difference with a 350 pa5? do you have better features? Is it just the psychological satisfaction of having "the state of the art"?

 

[chych7]

You’re not getting my point.

A state of the art amplifer should appeal to all customers, that’s why it’s expensive.

and just because you don’t need more power doesn’t mean other people don’t or that their opinions are invalid.

I suppose the claim of state of the art is not clear. I never said such a thing on this Topping amp. My original point was that 50W is more than plenty for typical home audio applications, which is something a lot here are getting stuck on.

 

[JeremyFife]

Personal view: looks like an outstanding amp that would be high up my shopping list if I was buying. I listen to stereo in a smallish room.

Would I have gone for the PA5 anyway? Perhaps, because it's cheaper, but I'd have been very tempted.
I'm not likely to upgrade to this as that will coincide with a move to a larger room, new speakers and a (possibly misguided) desire for more power.

However, when I do come to upgrade I have high hopes that this fine piece of engineering will have stimulated the competition and my choices will be even better than they are now.

I don't need SOTA, just clean power that doesn't colour the sound.
These are good times for hifi

 

[ObjectiveSubjectivist]

I suppose the claim of state of the art is not clear. I never said such a thing on this Topping amp. My original point was that 50W is more than plenty for typical home audio applications, which is something a lot here are getting stuck on.

Can we say that LA90 is STATE OF THE chART (?)

 

[JSmith]

think it might have been made for the sole purpose of having the top place on SINAD charts

Regardless of whether that is the case or not, this should not detract from the engineering achievement here, as it is no easy feat to top the SINAD charts for an amp. Keep in mind this may be directed at those parts of the market that have unfounded "issues" with class d based amps being ab. I am also getting the impression that both the PA5 and the LA90 are the beginning path for Topping for performance amps and would expect the next iterations to have more power on tap, so in that respect they have laid the groundwork well here... get the numbers sorted, then improve on the power output.


JSmith

 

[BoredErica]

Part of me feels like as a forum people should figure out minimum SINAD required for nobody on the planet to ever hear a difference with. I'm talking about an expensive sound isolated room, ear touching the tweeter on the most efficient speaker ever. And past that point I stop caring about noise. Many people will stop long before that though, so how about same setup but at 30db room at 2 feet away.

I don't feel like the amp's gain is too low. Even at 20db gain noise is lower than ahb2. But ahb2 is a tried and true product. We're reaching skyscraper levels of SINAD so at this point I do think there is validity in critiquing other things, like heatsink solution. I'm not going to go that hard on pa5, but for a $800 class AB amp seeking to be SOTA? I would care. This is coming from a person who otherwise loves the PA5 if not for the fact that there are some serious QC issues with units failing and them saying they are struggling to find the issue. Not a Topping hater by any means.

3. Just look at the back of that thing. WTF??? What kind of connectors are that? Is that a pro product? No, no pro amp I know of only makes 38 Watts. Is it a consumer product? No, for that it would have to have some RCA inputs. No pro and not consumer. Who is this for then, Amir so he can gush about how well the thing measures???

Why does this thing annoy me so much? Because I know Topping can make a great amplifier. For some reason they just seem to keep messing it up.

Dear Topping. Please make a high performance power amp with single ended RCA inputs and binding posts for a pair of speakers. That is all I ask, no unusual connectors, no volume control, no this, no that. Just give me more than 200 Watts into 4 Ohms and make the price sane. I know you can do it.

I think balanced audio is just better. Last thing I want is to set up a system and hear stupid ground loops and odd noises. There are many topics on ASR asking about how to fix those kinds of problems. I am never going back to single ended.

 

[amirm]

Its general lack of usefulness ($800 35W amplifier with a power brick...) does make me think it might have been made for the sole purpose of having the top place on SINAD charts.

Where do you get 35 watts? I have never measured an 8 ohm speaker. Most are around 4 ohm or less. There with 1% thd you have nearly 90 watts per channel.

 

[KSTR]

People seem to get the power measurements wrong and also they get the implications wrong.

For all practical purposes, this amp is actually 50W/8R, 100W/4R and 200W/8R-bridged (with distortion below 0.3%). Amir's decision to read the "max power" at the eyballed point the distortion start to rise signficiantly in relative terms is extremely unforgiving.

In general, Watts do not scale in a linear fashion. Going from 50W to 100W is not twice as loud. Technically (sound pressure level), it's only 1.41x as loud, and perceptually it's even less (rule of thumb: twice perceived loudness requires about 10x the power -- or 3.2x the voltage).

A 50W amp with good and fast soft-clipping can perceptually go much louder without perceived distortion than any hard-clipping 50W one and will still sound clean with the top of occasional transients being clipped. Other than it is hard-clipping, we don't know the overload behavior of this thing and it might or might not have signifcant clipping artifacts ("rail-sticking" for much longer than the actual overload duration).

In the quest for "best numbers" designers tend to avoid soft-clipping like the plague as it competely "spoils" the THD+N vs power graph at the top end. In the same quest Topping is sort of cheating with their low input sensitivity. With a THX-spec'd gain of 23dB (for balanced sources) all of the noise advantage would be gone, but that's academical anyway.

Of much higher practical interest are other things than that dreaded SINAD numbers.
- effect of PSU mains leakage current with unbalanced input cables
- clipping behavior
- thermal behavior with low impedance loads
- ruggedness and reliability, fully working speaker protection etc

 

[amirm]

. It is not designed for rationality and long term reliability

Don't make up stuff. You have no idea of its reliability factor. And what on earth is rationality?

 

[amirm]

With a THX-spec'd gain of 23dB (for balanced sources) all of the noise advantage would be gone, but that's academical anyway.

Performance is superb at nearly 20 dB gain. No way it gets smashed by going to 23 dB.