• 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). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Topping LA90 Review (Integrated Amplifier)

Rate this amplifier:

  • 1. Poor (headless panther)

    Votes: 27 4.3%
  • 2. Not terrible (postman panther)

    Votes: 42 6.7%
  • 3. Fine (happy panther)

    Votes: 155 24.6%
  • 4. Great (golfing panther)

    Votes: 405 64.4%

  • Total voters
    629

restorer-john

Master Contributor
Joined
Mar 1, 2018
Messages
9,128
Likes
25,782
Location
Gold Coast, Queensland, Australia
Clearly there is some form of that effect in many modern low-distortion amplifiers (including the Topping LA90), as you can see that as the output decreases from the maximum there is kind of a "sawtooth" of the noise/distortion floor until it finally reaches some minimum and then it just slopes up as the output level goes down.

The 'sawtooth' as you describe it, is the Audio Precision switching gain/attentuation ranges to keep the signal for the onboard data aquisition A/Ds in their relative sweet spots for the greatest accuracy/lowest noise.

At extreme low input levels, there is a heap of gain applied to the signals, using very low noise amplifiers, in order to keep the signals applied to the A/Ds in their linear ranges. The rise before the vertical drop is the AP, not the DUT. As the applied input levels increase, the gain/attentuation is switched out, and at high levels, way above the A/D's input ranges, the signal is attentuated in ever increasing multiples to keep the A/Ds right in their full scale ranges. Less gain equals less intrinsic noise.

The AP's automatic range selection matrix would be quite impressive to look at all the combinations of nested gain stages and attenuation stages.
 

usa_satriani

Member
Joined
Jan 20, 2022
Messages
12
Likes
6
Location
Aix-en-Provence, France
Hello,

I have a active and efficient setup.
15" BMS sealed and corrected, 6.5" waveguided phl 1140, 1" faital hf108r sth100 horn.
Digital filters with RME DAC.
I have been using lab gruppen amps for years and I love them.

We are making a similar system for a friend, we got a lab gruppen for the lows.

I don't want to start a debate but has someone used this amp on a compression driver ? How does it react and play ?
The little brother in class d, same question and what would you pick between high end class D and high end class ab for the compression drivers (1.5-20khz).

Regards Greg
 

Talisman

Active Member
Forum Donor
Joined
Mar 27, 2022
Messages
121
Likes
237
ASR's point is measurement.
I would say that the point of ASR is conscious spending, in fact in ASR users are invited to demonstrate whether or not they hear a blind difference between two pieces (dacs, amplifiers, cables)
Can you hear a blinded difference between an LA90 and a PA5? And between a Marantz PM6007 and a LA90?
If you have a dac with sinad 115, do you think having one with sinad 125 will improve your system?
Psychoacoustics works for both shiny, heavy audiophile instruments and the wonders shown in the graphs, but the point of ASR is
CAN YOU EAR A DIFFERENCE THAT JUSTIFIES THE SPEND?
The answer can only be given by you
 
Last edited:

manisandher

Senior Member
Joined
Nov 6, 2016
Messages
403
Likes
315
Location
Royal Leamington Spa, UK
... what would you pick between high end class D and high end class ab for the compression drivers (1.5-20khz).

Hi Greg, I don't really have a view on this, but wanted to let you know that I'm very happy with my Neurochrome amps (class AB, with clever error correction) driving my high horns (compression drivers, 1-20kHz). The horns are 107dB/[email protected], and with your ear right next to the mouth of the horn, all you hear is the faintest of hiss. And I can't find any fault in the sound (that could be due to the amps, at least).

But I've never compared the Neurochromes directly to any class D amps - never had the inclination, as I'm perfectly happy with the sound I currently have.

Good luck with your search.

Mani.
 

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83
So looking at the tests Amir did beyond the rated specs, ~120W into 4ohms bridged before clipping is pretty good. Topping advertises an 8ohm minimum load in bridged mode. 1) Are they being conservative since so many speakers rated at 8ohms nowadays actually dip to 4ohms or even lower or 2) is it dangerous for the LA90 to run it bridged into a "real" (as opposed to nominal) 4 ohm load?

5792916.jpg


Topping rates power here at 1% THD, but that point occurs during clipping. I think Amir's estimations are more useful -- the max power before clipping.

(This adds an appreciation for soft clipping circuitry, such as that found in NAD amps.)
 
Last edited:

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83
While WolfX-700 did not drive the LA90 up to 1% or 10% THD+N, his THD+N vs output power sweeps show that the Amp would most likely meet its spec:
https://www.l7audiolab.com/f/topping-la90/

To clarify, Amir's measurements more or less match Topping's advertised specs. 1% THD just happens to occur during clipping, and Amir rated max power BEFORE clipping.
 
Last edited:

Billy Budapest

Major Contributor
Forum Donor
Joined
Oct 11, 2019
Messages
1,251
Likes
1,770
To clarify, Amir's measurements more or less match Topping's advertised specs. 1% THD just happens to occur during clipping, and Amir rated max power BEFORE clipping.
It’s an informal convention for many manufacturers, reviewers, etc., to *define* clipping as 1% THD and then use the power measurement at 1% THD as the rated power at clipping, even if that specific 1% THD threshold is more or less arbitrary and does not really represent anything in particular. The real definition of clipping is when the power supply cannot supply enough juice to the output circuitry for an amp to fully reproduce the sine wave of the music signal and the top gets squared off (“clipped.”)

Now, realistically there probably isn’t much difference between the two.
 
Last edited:

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,151
Likes
4,718
Location
Berlin, Germany
The 1% distortion point for power spec was historically established as this amount of clipping you'll start to recognize on an oscilloscope screen shot.
Personally, the whole fixation on "precision" clipping level measurement is moot as in practice an amp sounds louder and cleaner when it has clean soft clipping starting early (say, raising to 1% at 70% max voltage which relates to 50% power) than the same amp with hard clipping starting late and showing a bad recovery behavior ("rail sticking").
A distortion point at 1% or 0.1% or whatever (like Amir's "onset of visible change of the plot") only makes sense for hard clipping amps.
It would be helpful if signal waveforms would be posted at various distortion levels like 1% and 10%
 

KSTR

Major Contributor
Joined
Sep 6, 2018
Messages
2,151
Likes
4,718
Location
Berlin, Germany
Since 36wpc into 8ohm is apparently all that anyone would ever need it makes me wonder why all those relatively small active studio monitors have hundreds of watts.
Universal studio monitors are used to, well, monitoring signals during any stage of music production and thus need way more headroom than speakers aimed at listening to the final product. You don't want to clip the internal amps on transients.
Not all products labelled studio monitors are in this universal class, however.
 

MaxBuck

Major Contributor
Forum Donor
Joined
May 22, 2021
Messages
1,028
Likes
1,310
Location
SoCal, Baby!
The 1% distortion point for power spec was historically established as this amount of clipping you'll start to recognize on an oscilloscope screen shot.
Personally, the whole fixation on "precision" clipping level measurement is moot as in practice an amp sounds louder and cleaner when it has clean soft clipping starting early (say, raising to 1% at 70% max voltage which relates to 50% power) than the same amp with hard clipping starting late and showing a bad recovery behavior ("rail sticking").
A distortion point at 1% or 0.1% or whatever (like Amir's "onset of visible change of the plot") only makes sense for hard clipping amps.
It would be helpful if signal waveforms would be posted at various distortion levels like 1% and 10%
You've put your finger on one reason amplifiers may not all sound alike, at least when they're driven near their limits. And as for measurements and how they describe these behaviors, as you suggest it's not a matter of measurements being unable to describe them, it's the fact that the measurements that could describe them are rarely performed.
 

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83

Based on this circuit layout, is this amp going to face any difficulties under vertical bi-amping, where one of the channels is going to be under much higher load than the other? Will the total max power effectively double compared to conventional amping?

To my completely untrained eyes, I perceive a kind of symmetry that would indicate that each channel's circuitry is largely independent of the other's.
 

nUUK25

Member
Joined
Jun 18, 2022
Messages
8
Likes
3
I'm thinking of buying the Topping LA90 - I already have the DAC/preamp Topping D90SE and it is great in dual-mode.
I want to replace 4 DIY Icepower 125ASX2 bridged/Ghentaudio case, based power amps. for something better.
I will preferably buy for a mono/bridged setup (2 or 4 identical amps.) - but I see the warning from Topping about
8 ohm speakers when LA90 are bridged.
My electrostatic/dynamic speakers are Martin Logan Electromotion ESL X (Bi-amped, now) and they are 6 ohm nominal. M.L. also says 1,6 ohm at 20 KHz and
they recommend 4,6 or 8 Ohm amps. HIFinews.com has done some measurements/Tech Lab Report, on ESL X: Impedance Min. and Max.
(20 Hz-20 KHz): 2.87 ohm at 30 Hz and 20.5 ohm at 2,5 KHz.
Will it be safe to bridge (mono) one LA90 amp. for each speaker ? (Using jumper clips between speaker terminals - on the same speaker)
Will it be safe to bridge (mono) two LA90 amps for each speaker ? (Removing jumper clips between speaker terminals - on the same speaker)
and using a XLR signal splitter from the D90SE, as I need 2 signals, one for each power amp and thus one for each speaker terminal
(which I already do with the Icepower setup right now - when I added two more IcePower amps in the setup - the sound improvement was huge).
I hope some of you 'shining' people can figure out if it is too risky and if there is any real difference bewteen '2' or '4' in terms of risk.
I mostly listen to 'unplugged' baroque; Bach, Vivaldi, Corelli and that kind !! :)
 

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83
I'm thinking of buying the Topping LA90 - I already have the DAC/preamp Topping D90SE and it is great in dual-mode.
I want to replace 4 DIY Icepower 125ASX2 bridged/Ghentaudio case, based power amps. for something better.
I will preferably buy for a mono/bridged setup (2 or 4 identical amps.) - but I see the warning from Topping about
8 ohm speakers when LA90 are bridged.
My electrostatic/dynamic speakers are Martin Logan Electromotion ESL X (Bi-amped, now) and they are 6 ohm nominal. M.L. also says 1,6 ohm at 20 KHz and
they recommend 4,6 or 8 Ohm amps. HIFinews.com has done some measurements/Tech Lab Report, on ESL X: Impedance Min. and Max.
(20 Hz-20 KHz): 2.87 ohm at 30 Hz and 20.5 ohm at 2,5 KHz.
Will it be safe to bridge (mono) one LA90 amp. for each speaker ? (Using jumper clips between speaker terminals - on the same speaker)
Will it be safe to bridge (mono) two LA90 amps for each speaker ? (Removing jumper clips between speaker terminals - on the same speaker)
and using a XLR signal splitter from the D90SE, as I need 2 signals, one for each power amp and thus one for each speaker terminal
(which I already do with the Icepower setup right now - when I added two more IcePower amps in the setup - the sound improvement was huge).
I hope some of you 'shining' people can figure out if it is too risky and if there is any real difference bewteen '2' or '4' in terms of risk.
I mostly listen to 'unplugged' baroque; Bach, Vivaldi, Corelli and that kind !! :)

If your room is small and your speakers aren't inefficient, you'll be fine. Your speakers seem like they could be pretty power-hungry, which *might* not be a good fit for the LA90. Can you post the impedance curve?
 

nUUK25

Member
Joined
Jun 18, 2022
Messages
8
Likes
3
I think Martin Logan Electromotion ESL X, is fairly efficient: M.L states: 91 db / 2.783 volts / meter
HIFInews.com at their Lab report states:
Sensitivity (SPL/1m/2.83V – 1kHz/Mean/IEC)88.8dB / 89.6dB / 86.3dB
Impedance modulus: minimum & maximum (20Hz–20kHz)2.87ohm @ 30Hz, 20.5ohm @ 2.5kHz
Impedance phase: minimum & maximum (20Hz–20kHz)–64° @ 6.45kHz, +55° @ 1.23kHz
Pair matching/Resp. error (200Hz–20kHz)1.7dB/ ±4.8dB/±5.9dB
LF/HF extension (–6dB ref 200Hz/10kHz)38Hz / 31.1kHz/26.7kHz
THD 100Hz/1kHz/10kHz (for 90dB SPL/1m)0.6% / 0.4% / 3.5%

The room is 7,5 x 3,7 meters (low celing: 2,15 m). Effectively I use 3,7 m x 3,7 meters but of cause the sound 'moves' on to the rest of the room, sideways !! The distance between speakers are 2 meters and I'm 2,5 meters away from them. I don't have an impedance curve. I can try and ask M.L. for one !
 
Last edited:

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83
I think Martin Logan Electromotion ESL X, is fairly efficient: M.L states: 91 db / 2.783 volts / meter
HIFInews.com at their Lab report states:
Sensitivity (SPL/1m/2.83V – 1kHz/Mean/IEC)88.8dB / 89.6dB / 86.3dB
Impedance modulus: minimum & maximum (20Hz–20kHz)2.87ohm @ 30Hz, 20.5ohm @ 2.5kHz
Impedance phase: minimum & maximum (20Hz–20kHz)–64° @ 6.45kHz, +55° @ 1.23kHz
Pair matching/Resp. error (200Hz–20kHz)1.7dB/ ±4.8dB/±5.9dB
LF/HF extension (–6dB ref 200Hz/10kHz)38Hz / 31.1kHz/26.7kHz
THD 100Hz/1kHz/10kHz (for 90dB SPL/1m)0.6% / 0.4% / 3.5%

The room is 7,5 x 3,7 meters (low celing: 2,15 m). Effectively I use 3,7 m x 3,7 meters but of cause the sound 'moves' on to the rest of the room, sideways !! The distance between speakers are 2 meters and I'm 2,5 meters away from them. I don't have an impedance curve. I can try and ask M.L. for one !

2.5m is between near and mid-field. I think it's strange that Martin Logan recommends multiple impedance classes for amplifiers. Either an 8-ohm targeted amplifier can drive them or not. Saying 4, 6, or 8 just muddies the waters.

Well, the LA90 can drive a 4ohm load with some headroom to spare. So, odds are that the LA90 is going to have sufficient power for your needs as long as you don't blast your music to ear-bleeding volumes!
 

nUUK25

Member
Joined
Jun 18, 2022
Messages
8
Likes
3
Right ! -
2.5m is between near and mid-field. I think it's strange that Martin Logan recommends multiple impedance classes for amplifiers. Either an 8-ohm targeted amplifier can drive them or not. Saying 4, 6, or 8 just muddies the waters.

Well, the LA90 can drive a 4ohm load with some headroom to spare. So, odds are that the LA90 is going to have sufficient power for your needs as long as you don't blast your music to ear-bleeding volumes!
Thanks kenshone,
Right,..... 'unplugged' baroque composers are best at realistic sound levels anyway and does not improve on higher levels. The same goes mostly !! ...for Jazz (Return to Forever and some Sting). I do listen to Jennifer Warnes, Leonard Cohen, Janis Ian ...stuf like that, but also on modest levels (I kan converse 'normal' to another person 1 meter away without problem). Not being an expert on this: I recond, the load on my present 4 Icepowers are mostly between 5-25 watt pr. amp - thats all !!
Using the Topping D90SE as preamp and DAC will you recommend gain set at 10 db for the 4 bridged LA90 ? - or should it be 20 db ?
 

kenshone

Active Member
Joined
May 15, 2022
Messages
195
Likes
83
Right ! -

Thanks kenshone,
Right,..... 'unplugged' baroque composers are best at realistic sound levels anyway and does not improve on higher levels. The same goes mostly !! ...for Jazz (Return to Forever and some Sting). I do listen to Jennifer Warnes, Leonard Cohen, Janis Ian ...stuf like that, but also on modest levels (I kan converse 'normal' to another person 1 meter away without problem). Not being an expert on this: I recond, the load on my present 4 Icepowers are mostly between 5-25 watt pr. amp - thats all !!
Using the Topping D90SE as preamp and DAC will you recommend gain set at 10 db for the 4 bridged LA90 ? - or should it be 20 db ?

D90SE output voltage:
6010549.jpg


LA90 input sensitivity:
5792924.jpg


If you want to be able to run the LA90 at max volume from your D90SE, it looks like you'll need to set the gain to high. Why? Max output is ~5Vrms, which is below the low gain sensitivity of ~6.9Vrms. At high gain, you still get record-setting performance, but you'll be able to drive the LA90 to max volume at less than 50% of the D90SE's output capability. Be careful about overdriving, or you'll induce clipping...I'm sure your ears will stop you from overdriving before your hands do though!

I'm not sure what you mean when you say "4 bridged," but I would really try stereo mode before bridging. Odds are quite good that stereo mode will give you more than enough volume.
 

nUUK25

Member
Joined
Jun 18, 2022
Messages
8
Likes
3
Ocay, LA90 gain in mono mode is 15,5 db or 25,2 db ....or as you suggest just try with one amp in stereo as a start and see how it goes at 1/4 of the expence - maybe 4 amps are overkill. Stereo and one amp will give me an idear about quality level at lower and modest audio levels - can always expand later if I feel it is necessary.
 
Top Bottom