Topping LA90 Review (Integrated Amplifier)

[jmillar]

I won't be home until tomorrow. I can then try to test it. For now what is the application for 2 ohm load? I have not tested one speaker that goes that low.

Quite a few do, at some frequencies. (sometimes with amp clipping when this happens) Am I misled?

 

[Billy Budapest]

I know I am sounding like a broken record, but because the LA90’s volume control and source selector are redundant when used with a preamp or certain DACs with volume and switching, I think the following would make a perfect follow-up product:

1) delete volume pot and source selector (although keep gain switch on back)
2) include two bridged amplifiers within single chassis
3) include power supply (or supplies) within chassis

 

[AnthonyH]

I'm not enthusiastic about this amplifier. Over the years I've tried several different amplifiers but I always found that my Mackie M1400 sounded the best when driving the front speakers, which are Magnepan MG1.6 planar speakers. I'd assumed it was something subtle, like a weird impedance curve, but it turned out to be obvious. Even in a mid-sized room (20 feet x 13 x 8) these speakers need loads of power. The best estimate I have of how much power is based on playing a CD-quality track at 85 dB SPL and looking for peak transient voltage on the speaker terminal with an oscilloscope. I measured 70 V pk-to-pk. The amplifier, rated 500 W RMS into 4 ohms, can deliver 125 V pk-to-pk so I'm running out of headroom. I know that this is extreme case - the reviews of the Magnepans have commented on their low efficiency - but a more efficient speaker in a different room could still run into the same problem. Anyhow, the Topping amplifier does not suit my situation at all.
The next question for me is: should I spend the money for a high-output, high-quality Class D amplifier. I'll deal with that in a different thread.
Should you consider replacing your amplifier with a low-power, low-distortion amplifier? I would advise measuring how much power your current speakers actually need before deciding. Borrow an oscilloscope and GET DATA. Or even buy one - a Pico 2204A is only $165 or you could buy a used Tektronix.

 

[lashto]

I have a lot of respect for your knowledge and experience of a number of class D amplifiers - but that is an "audiophile folklore" type of statement. NFB improves the numbers because it reduces distortion, and output impedance. Both of these improve fidelity, and hence sound quality.

From the point of view of fidelity, improved numbers are improved sound quality.

according to Putzeys that is no "folklore". Fig 14 should be pretty clear about the negative effects of NFB (at so called moderate levels):

negative feedback .. creates distortion components that weren’t there before

 

[manisandher]

Volume control on the LA90 is fine (not so handy when using it as a mono block in which case an extra pre-amp or digital volume control (in or before a DAC) would be handy.

I hear clear degradation using the LA90's vol control, compared to using my old Pass X1 pre.

(And no, I haven't done any bloody double blind tests.)

Mani.

 

[manisandher]

Has anyone built a linear power supply for LA90? If yes, please share your experience. Thanks

I've tried it with a linear lab supply outputting 64V. Sounds fine. I haven't compared to the stock SMPS.

Mani.

 

[Hckd]

I hear clear degradation using the LA90's vol control, compared to using my old Pass X1 pre.

(And no, I haven't done any bloody double blind tests.)

Mani.

What do you mean by “clear degradation”? (Serious question)

 

[antcollinet]

according to Putzeys that is no "folklore". Fig 14 should be pretty clear about the negative effects of NFB (at so called moderate levels):

Implementation of "moderate" NFB is where the myth has come from that NFB is bad.

In other words badly implemented negative feedback (just like any other badly implemented deisign feature) is bad, not that negative feedbak per se is bad. It is why Putzeys designs use much more than "moderate" amounts - as do any well measureing apms using negative feedback. You should note the article you've referenced is sub-titled:

why there is no such thing as too much feedback

You should also note the post I replied to was referring to negaitve feedback that generates "excellent" numerical performance, yet still being bad for audio quality (that is the myth right there). Moderate feedback that causes incresing harmonc distortion by definition also generates bad numerical performance.

 

[manisandher]

What do you mean by “clear degradation”? (Serious question)

I could describe what I hear, but what's the point? I'm not going to back it up with a DBT.

But it's easy to assess the quality of the LA90's volume control, if you have the means to attenuate digitally (preferably at 64 or 32 bits) and set the LA90 to bypass mode. If you can't hear any difference, you're good to go.

Mani.

 

[Hckd]

I could describe what I hear, but what's the point? I'm not going to back it up with a DBT.

But it's easy to assess the quality of the LA90's volume control, if you have the means to attenuate digitally (preferably at 64 or 32 bits) and set the LA90 to bypass mode. If you can't hear any difference, you're good to go.

Mani.

The point is me trying to understand what you ear, I think it is a completely valid question…

When a forum like this starts getting full of statements of A, B or C without any debate about either, it just loses its usefulness.

Thanks anyway.

 

[manisandher]

... I think it is a completely valid question…

I agree.

When a forum like this starts getting full of statements of A, B or C without any debate about either, it just loses its usefulness.

Again, I agree.

I was involved in a blind ABX a few years ago. Scored 9/10 (1% probability of guessing). But because I wasn't 'supposed' to hear a difference, it was dismissed as a fluke. I don't share the results of my DBTs now (but still do them). I'm pretty confident that what I hear is 'real', especially as it seems to corrrelate well with some null testing that I've been doing with DeltaWave.

If people on an audio forum don't talk about how things sound to them, and what might be causing things to sound that way, then it becomes a bit of a flatland place where reductionism reigns supreme. What a shame.

Mani.

 

[mdsimon2]

I could describe what I hear, but what's the point? I'm not going to back it up with a DBT.

But it's easy to assess the quality of the LA90's volume control, if you have the means to attenuate digitally (preferably at 64 or 32 bits) and set the LA90 to bypass mode. If you can't hear any difference, you're good to go.

Mani.

I think it useful to explain what you think you are hearing even if it isn't backed up by anything more than sighted listening.

Channel imbalance? Increased noise? More distortion? Frequency response seems different?

Michael

 

[Billy Budapest]

I agree.


Again, I agree.

I was involved in a blind ABX a few years ago. Scored 9/10 (1% probability of guessing). But because I wasn't 'supposed' to hear a difference, it was dismissed as a fluke. I don't share the results of my DBTs now (but still do them). I'm pretty confident that what I hear is 'real', especially as it seems to corrrelate well with some null testing that I've been doing with DeltaWave.

If people on an audio forum don't talk about how things sound to them, and what might be causing things to sound that way, then it becomes a bit of a flatland place where reductionism reigns supreme. What a shame.

Mani.

You hit the nail on the head, though—a person describing how things sound “to them” is introducing a variable in the equation that may not be reproducible by others. Namely, the differences in sound might not be attributable to the equipment, but to the listener.

 

[dimedrol]

I was involved in a blind ABX a few years ago. Scored 9/10 (1% probability of guessing).

That's some interesting math right there

 

[manisandher]

That's some interesting math right there

The probability of scoring 9/10 in an ABX by guessing alone is 0.977%. Call it "interesting" if you want...

Mani.

 

[antcollinet]

That's some interesting math right there

Here is an explanation of the stats for you. And you are right - it is interesting. It is also correct:

Statistics of ABX Testing

Statistics of ABX Testing By Amir Majidimehr ABX is by far the most talked about type of listening test on the Internet to demonstrate audible difference between two devices, or two audio files. As the name implies, the test involves two inputs, "A" and "B" and the listener is asked to vote...

www.audiosciencereview.com

 

[Hckd]

If people on an audio forum don't talk about how things sound to them, and what might be causing things to sound that way, then it becomes a bit of a flatland place where reductionism reigns supreme. What a shame.

Yep, and that’s why I think you should elaborate on what degradation you ear, it’s a completely valid point as long as we don’t dismiss that what you ear can be linked to something else and not the equipment itself…

 

[manisandher]

OK, OK. But don't say I didn't voice my concerns at the outset.

Describing what you hear in language is always difficult. But here goes... The LA90's vol pot seems to 'dullen'/'flatten' the sound. It becomes 'matt' as opposed to 'gloss'.

Once I've figured out how to capture the LA90's output with an ADC, I strongly suspect that I will find that the vol pot affects the channel separation, but that it does NOT affect the frequency response at all.

In any event, my LA90's vol bypass remains ON at all times (unless I'm physically switching inputs).

Mani.

 

[juliangst]

Topping should have made it just a power amp without a pot imho.
It’s meant to go with their Premium A90/D90/Pre90 products which all have volume control anyways.

Also heard about an upcoming PA5 2023 version. I wonder if it’ll just be a fixed PA5 or something more

 

[Deleted member 35357]

Topping should have made it just a power amp without a pot imho.
It’s meant to go with their Premium A90/D90/Pre90 products which all have volume control anyways.

Also heard about an upcoming PA5 2023 version. I wonder if it’ll just be a fixed PA5 or something more

It's about reaching as many possible customers as they can with a single product.
If all the features are in the one product they have a higher chance of not missing out from a purchase because they are missing a critical feature like a volume control being present.
Many others buyers without any other Topping products might get interested in this product because of it's performance and you never know what other gear they are using and might need the pot in these.