WiiM Amp Pro Streaming Stereo Amplifier Review

[Emiyanez]

The info you guys shared here is a bit over my head, so I’d really appreciate a simpler explanation...

I don't think there would be any relevant difference between the WiiM Amp Pro and the M1. The Win Amp Pro (and I guess the M1) has a flat frequency response (unlike the previous version) and the power is IMHO enough for your needs. The slight difference in power between the WiiM and the M1 will not have much impact in your system.

If you're not happy with the sound in the bigger room, I would rather look for new speakers. That being said, you could also apply room correction or PEQ using the WiiM App or add a sub later if needed, which could help tweaking the sound to your liking. I guess you can do similarly with the M1 so choose the one you like more.

 

[Emiyanez]

Maybe. My speakers definitely cannot reproduce the low frequencies...

You could use the sub crossover menu in the WiiM App to get rid of the frequencies that your speakers cannot reproduce. Just activate the sub and select the crossover point slightly lower than the minimal frequency your speakers can accurately reproduce. This will simulate a HPF and a aliviate the load to the Amp.

 

[shuppatsu]

You could use the sub crossover menu in the WiiM App to get rid of the frequencies that your speakers cannot reproduce. Just activate the sub and select the crossover point slightly lower than the minimal frequency your speakers can accurately reproduce. This will simulate a HPF and a aliviate the load to the Amp.

Yes, I do this. I turned it and the EQ off for the purpose of this test, to be an apples to apples comparison. Though the gain is the same so it probably has little to no effect on the SPLs my speakers put out. It certainly doesn’t sound louder or quieter when I toggle the crossover on and off.

 

[shuppatsu]

If you're not happy with the sound in the bigger room, I would rather look for new speakers. That being said, you could also apply room correction or PEQ using the Win App or add a sub later if needed, which could help tweaking the sound to your liking. I guess you can do similarly with the M1 so choose the one you like more.

I’m pretty sure the M1 just has bass and treble tone controls, no PEQ or RC.

 

[Old_School_Brad]

Maybe. My speakers definitely cannot reproduce the low frequencies. I've been saying that over and over again. I don't know if it's clipping. From the (non-pro) review: "Unless you use analog input, the gain settings is designed such that you never clip. This means that the "peak" power is the same as what I have measured above. Spec is 120 watts at 4 ohm and we are achieving that so no issue there. Same with 60 watt specified for 8 ohm." Now, maybe it can't clip on his test, but it can clip if fed a low frequency signal? I don't know enough about audio to know whether that's likely.


That's also what I was suggesting when I said, "Maybe your system goes down really low. Mine doesn't, no subwoofers." And showed you the FR graph showing sharp reduction in bass below 60Hz. But we are drawing different inferences from this.

What I'm saying is that all things being equal (it's not, but close enough), the fact that the Harbeth goes down lower means that for the same wattage, my CPeak would be lower than the Harbeths. That means the Harbeths would be pumping out the bass that my system can't handle, plus it would outputting the mids and highs which makes it unbearable for me to listen for more than a second. In reality, my speakers are 2dB more sensitive, so it's a little more complicated than that, but then again 90W and 700W is like a 9dB difference. My point is my setup could be half as loud (9W?) and there's no way I'd be sticking my ears right next to the speaker for any length of time. Yet they are doing this casually in the Youtube video!

It really sounds like I'm against you and Doodski but I'm not. I'm trying my best to understand this stuff, and apart from a brief dalliance 25 years ago I've only gotten into audiophilia in the last year or so. So I'm just trying to expand my mental model by pushing against it.

I think I see where you’re coming from, but I believe your assessment may be missing a few key aspects.

When measuring that track with dBA at the same average level I’m using, your amplifier may not be able to produce the necessary power to cover the low-frequency range fully. This would allow you to play it louder in terms of dBA average, leading to a perception that the video’s results might be inaccurate. However, the system in the video (and mine) can handle the low-frequency transients, which yours may struggle with. I'll try to explain it in a couple of different ways.

The reason you might not realize this difference is due to a few factors: first, the dBA vs. dBC measurements, and second, the fact that distortion in low frequencies is often inaudible, especially with the bass-heavy content of this track. If you took an SPL measurement at your listening position using the NIOSH app on an iPhone (which is accurate with iPhones), you'd likely notice that while it doesn't take much power to reach uncomfortable volume levels on your system, you wouldn’t achieve as high a dBC level due to limited power and capability to reproduce those transient low frequencies.

However, the amplifier in question doesn’t appear to clip or distort -at least based on the test from page one. Instead, it may instead be causing a perceived compression effect, often seen when speakers are driven at high power levels. In this case, the low-frequency content is reproduced only up to the amplifier’s power limit, while the higher frequencies, to which our ears are more sensitive, continue, relatively unaffected to a degree, to be reproduced with whatever power the amplifier provides at that moment.
Analogy warning; this effect resembles a high-pass filter or high-shelf EQ, where the low-frequency content reaches its maximum output relative to the rest of the frequency spectrum. In essence, the bass doesn’t scale proportionally with the rest of the audio.

By the way, I don’t see your posts as oppositional -just as healthy curiosity and skepticism.

 

[shuppatsu]

I think I see where you’re coming from, but I believe your assessment may be missing a few key aspects.

When measuring that track with dBA at the same average level I’m using, your amplifier may not be able to produce the necessary power to cover the low-frequency range fully. This would allow you to play it louder in terms of dBA average, leading to a perception that the video’s results might be inaccurate. However, the system in the video (and mine) can handle the low-frequency transients, which yours may struggle with. I'll try to explain it in a couple of different ways.

The reason you might not realize this difference is due to a few factors: first, the dBA vs. dBC measurements, and second, the fact that distortion in low frequencies is often inaudible, especially with the bass-heavy content of this track. If you took an SPL measurement at your listening position using the NIOSH app on an iPhone (which is accurate with iPhones), you'd likely notice that while it doesn't take much power to reach uncomfortable volume levels on your system, you wouldn’t achieve as high a dBC level due to limited power and capability to reproduce those transient low frequencies.

However, the amplifier in question doesn’t appear to clip or distort -at least based on the test from page one. Instead, it may instead be causing a perceived compression effect, often seen when speakers are driven at high power levels. In this case, the low-frequency content is reproduced only up to the amplifier’s power limit, while the higher frequencies, to which our ears are more sensitive, continue, relatively unaffected to a degree, to be reproduced with whatever power the amplifier provides at that moment.
Analogy warning; this effect resembles a high-pass filter or high-shelf EQ, where the low-frequency content reaches its maximum output relative to the rest of the frequency spectrum. In essence, the bass doesn’t scale proportionally with the rest of the audio.

By the way, I don’t see your posts as oppositional -just as healthy curiosity and skepticism.

This is very interesting and a good explanation. Something I really don't know anything about! Can you point me to any resources where this kind of compression is discussed? ChatGPT tells me quite a bit about it, but ChatGPT is extremely unreliable for audio stuff since it trains on all the wacky audiophile stuff out there. My searches have all been in vain, in part because compression is used in so many ways in the audio world.

ChatGPT thinks this "perceived compression effect" is akin to, but not exactly the same as "soft clipping," where the peaks get rounded off gently instead of getting cut off and generating square waves. I'm trying to figure out what the difference is from "soft clipping," and also what the mechanism is whereby the amp delivers less power to the low frequencies than the higher ones.

 

[Doodski]

ChatGPT thinks this "perceived compression effect" is akin to, but not exactly the same as "soft clipping," where the peaks get rounded off gently instead of getting cut off and generating square waves. I'm trying to figure out what the difference is from "soft clipping,"

You just explained soft clipping yourself. You nail'd it.

what the mechanism is whereby the amp delivers less power to the low frequencies than the higher ones.

The amp is not delivering less low frequency. It is the low frequency capability of the speakers to reproduce the extremely low stuff and the crossovers in the speakers doing all the work.

 

[shuppatsu]

You just explained soft clipping yourself. You nail'd it.

The amp is not delivering less low frequency. It is the low frequency capability of the speakers to reproduce the extremely low stuff and the crossovers in the speakers doing all the work.

So ChatGPT says it is like soft clipping, but that it is not soft clipping. It also says that soft clipping is more for tubes, and maybe for amps that have a soft clipping circuit. So I'm still confused how that applies, knowing of course that ChatGPT is often wrong about these sort of things. Can this amplifier clip? Is Amir wrong? Or is he supposing a more normal situation? Can the gain structure be such that the amp doesn't normally clip even at 0dBFS, but can if that 0dBFS is at a really low frequency?

I still don't understand the second paragraph. I assume that if the amp delivers the low frequency and the speaker can't handle it, that manifests either in distortion or heat. But what I think @Old_School_Brad is saying is that somehow the speaker/amp pairing interacts in such a way that the LF waveforms are not given as much power as the HF waves. Which confuses me.

 

[Doodski]

So ChatGPT says it is like soft clipping, but that it is not soft clipping. It also says that soft clipping is more for tubes, and maybe for amps that have a soft clipping circuit. So I'm still confused how that applies, knowing of course that ChatGPT is often wrong about these sort of things. Can this amplifier clip? Is Amir wrong? Or is he supposing a more normal situation? Can the gain structure be such that the amp doesn't normally clip even at 0dBFS, but can if that 0dBFS is at a really low frequency?

Let's start with the basics from the beginning and also I have always worked in Watts, Volts and Amps so you will see me using those terms. I simply never had a strong need for decibels metering in the kind of mechatronics service work that I specialized in. Plus to understand clipping one must approach the operation of the audio amp circuitry and that is in Volts and Amps. So here we go. Hopefully one topic at a time.

Soft clipping.
It can occur with tube amps and amplifiers that have special soft clipping circuitry like a NAD does for example. Soft clipping rounds out the waveform @ clipping so that it is not a hard clipping comprised of DC voltage and current that will fry a speaker. All audio amps can and do clip.

Hard clipping.
Occurs with most amplifiers and that means the voltage used for the operation of the amplifier that makes the speaker drivers move in and out has been reach to the maximum available voltage and current. Clipping can occur on low frequency audio signals and high frequency too.

Example for clipping points of the amps.
-The amps in the video we don't know the actual clipping point because they never showed us but we saw ~750 Watts peak reached and so we will use 750 Watts for rough calculations so that we may compare the Wiim to the video's mono block amps.

-The Wiim hard clips at 59Wrms@8R and so that means there is a ~ +/- 30.7 Volts DC power supply in the device for the speakers and the Wiim will hard clip @ ~ +/- 30.7 Volts whether it is a 8 Ohms speaker or a 4 Ohms speaker. The reason the Wiim clips at this voltage whether the speaker is a 4 Ohms or a 8 Ohms speaker is due to the type of power supply used in the Wiim. So the current at 8R at peak output will be maximum @ ~ +/- 3.8A

-The big mono block power amps in the video test are seen to peak at 750Wrms. So the voltage output there will be ~ +/- 109.5 V peak @ ~ +/- 13.7A peak for those Harbeth speakers used in the video.

-Obviously the big mono block amps voltage and current output exceeds what the Wiim can put out and by a large margin.

-The actual dB level reached by the Wiim is substantial and will satisfy many users @ ~30.7 Volts peak but it is not as dynamic in transients as the mono blocks @ ~109.5 Volts peak which go louder but in actuality the total dB increase with the mono block amps is not massive due to the way the dB scale works and with the way our ears work. The mono blocks at that power output are for people who are very serious about their sound and or have very difficult speaker impedances and phase angles.

I still don't understand the second paragraph. I assume that if the amp delivers the low frequency and the speaker can't handle it, that manifests either in distortion or heat. But what I think @Old_School_Brad is saying is that somehow the speaker/amp pairing interacts in such a way that the LF waveforms are not given as much power as the HF waves. Which confuses me.

If the amp can deliver non-clipped voltage and current that exceeds the speakers range of operation then what usually happens first is the woofer(s) hits the stops and that makes a horrible snap/cracking sound and the operator must turn down the power to the speakers to prevent damage to the voice coil(s) and the voice coil former(s). If the signal to the speakers does not have low frequency energy that the crossovers would divert to the woofers then the tweeters and or midrange drivers convert that higher frequency energy to sound waves and the sound output will usually sound smeared, may crackle a bit and ultimately the voice coil(s) will be overheated and carbonized etc if the power rating is exceeded.

I don't think Brad meant that low frequency energy contains less energy than higher frequency energy in the case of audio amplification and speaker operation. I think what he meant is that if the speakers cannot reproduce the low frequency energy due to limitations in the speaker then the operator will not hear that low frequency energy even though the amplifier is sending the low frequency to the speakers. The amplifier will still hard clip and then DC is sent to the speakers and will cause damage. The point at which hard clipping occurs is the difference between 30.7 Volts @ 3.8 Amps for the Wiim and a theoretical 109.5 Volts @ 13.7 Amps for the mono block amplifiers.

 

[shuppatsu]

Let's start with the basics from the beginning and also I have always worked in Watts, Volts and Amps so you will see me using those terms. I simply never had a strong need for decibels metering in the kind of mechatronics service work that I specialized in. Plus to understand clipping one must approach the operation of the audio amp circuitry and that is in Volts and Amps. So here we go. Hopefully one topic at a time.

Soft clipping.
It can occur with tube amps and amplifiers that have special soft clipping circuitry like a NAD does for example. Soft clipping rounds out the waveform @ clipping so that it is not a hard clipping comprised of DC voltage and current that will fry a speaker. All audio amps can and do clip.
View attachment 401566

Hard clipping.
Occurs with most amplifiers and that means the voltage used for the operation of the amplifier that makes the speaker drivers move in and out has been reach to the maximum available voltage and current. Clipping can occur on low frequency audio signals and high frequency too.

Example for clipping points of the amps.
-The amps in the video we don't know the actual clipping point because they never showed us but we saw ~750 Watts peak reached and so we will use 750 Watts for rough calculations so that we may compare the Wiim to the video's mono block amps.

-The Wiim hard clips at 59Wrms@8R and so that means there is a ~ +/- 30.7 Volts DC power supply in the device for the speakers and the Wiim will hard clip @ ~ +/- 30.7 Volts whether it is a 8 Ohms speaker or a 4 Ohms speaker. The reason the Wiim clips at this voltage whether the speaker is a 4 Ohms or a 8 Ohms speaker is due to the type of power supply used in the Wiim. So the current at 8R at peak output will be maximum @ ~ +/- 3.8A

-The big mono block power amps in the video test are seen to peak at 750Wrms. So the voltage output there will be ~ +/- 109.5 V peak @ ~ +/- 13.7A peak for those Harbeth speakers used in the video.

-Obviously the big mono block amps voltage and current output exceeds what the Wiim can put out and by a large margin.

-The actual dB level reached by the Wiim is substantial and will satisfy many users @ ~30.7 Volts peak but it is not as dynamic in transients as the mono blocks @ ~109.5 Volts peak which go louder but in actuality the total dB increase with the mono block amps is not massive due to the way the dB scale works and with the way our ears work. The mono blocks at that power output are for people who are very serious about their sound and or have very difficult speaker impedances and phase angles.


If the amp can deliver non-clipped voltage and current that exceeds the speakers range of operation then what usually happens first is the woofer(s) hits the stops and that makes a horrible snap/cracking sound and the operator must turn down the power to the speakers to prevent damage to the voice coil(s) and the voice coil former(s). If the signal to the speakers does not have low frequency energy that the crossovers would divert to the woofers then the tweeters and or midrange drivers convert that higher frequency energy to sound waves and the sound output will usually sound smeared, may crackle a bit and ultimately the voice coil(s) will be overheated and carbonized etc if the power rating is exceeded.

I don't think Brad meant that low frequency energy contains less energy than higher frequency energy in the case of audio amplification and speaker operation. I think what he meant is that if the speakers cannot reproduce the low frequency energy due to limitations in the speaker then the operator will not hear that low frequency energy even though the amplifier is sending the low frequency to the speakers. The amplifier will still hard clip and then DC is sent to the speakers and will cause damage. The point at which hard clipping occurs is the difference between 30.7 Volts @ 3.8 Amps for the Wiim and a theoretical 109.5 Volts @ 13.7 Amps for the mono block amplifiers.

First of all, thanks for going into such depth and engaging in this back and forth with someone who knows comparatively much less.

I agree with most of what you have written. What I have trouble reconciling is the clipping behavior. I have the non-pro amp, which has significantly higher distortion at 20Hz than the Pro, but it’s still -60dB down and there’s no “elbow” all the way up to 0dBFS:

To me this means that the amp will not clip no matter how much low frequency energy there is in the signal, due to insufficient gain.

Is that inference wrong, and if so, why?

 

[Doodski]

To me this means that the amp will not clip no matter how much low frequency energy there is in the signal, due to insufficient gain.

I don't know any specifics about the gain and clipping behavior in this model in question. In amplifiers the clipping can be and is initiated by low frequency energy and high frequency energy too.

 

[shuppatsu]

I don't know any specifics about the gain and clipping behavior in this model in question. In amplifiers the clipping can be and is initiated by low frequency energy and high frequency energy too.

Does the graph I included not give sufficient information? To me it does, but again I’m not confident in my interpretation. The graph basically shows no clipping at any audible frequency (20KHz not tested). Would a multitone induce clipping where a pure tone sweep would not?

 

[Doodski]

Does the graph I included not give sufficient information? To me it does, but again I’m not confident in my interpretation. The graph basically shows no clipping at any audible frequency (20KHz not tested). Would a multitone induce clipping where a pure tone sweep would not?

The graph shows a elbow or knee (Onset of clipping.) at the 200 Hz to 5 kHz test frequencies although the 20 Hz does not show a elbow/knee in this test. I can't answer why this specific Wiim has this behavior other than think it is not full bandwidth. That makes me think there is a deficiency/limitation in the design. This is unusual, not standard behavior and is specific to this model. Perhaps if the amp was pushed a bit more then there would be more obvious knees/elbows and the 20 Hz waveform would have clipped. I don't know the amp in question (Can you link me to the test that you are referring too?) so I can't comment further. If you think the amp won't clip you are mistaken because it is commencing clipping in that graph although it is not a severely shaped/sloped knee/elbow. I would not rely on this test to crank up your Wiim with low frequency output as if it can't clip nor damage your speakers. As per multitone if enough signal strength is provided then it will clip. Why the strong interest in this distortion and clipping behavior?

 

[shuppatsu]

The graph shows a elbow or knee (Onset of clipping.) at the 200 Hz to 5 kHz test frequencies although the 20 Hz does not show a elbow/knee in this test. I can't answer why this specific Wiim has this behavior other than think it is not full bandwidth. That makes me think there is a deficiency/limitation in the design. This is unusual, not standard behavior and is specific to this model. Perhaps if the amp was pushed a bit more then there would be more obvious knees/elbows and the 20 Hz waveform would have clipped. I don't know the amp in question (Can you link me to the test that you are referring too?) so I can't comment further. If you think the amp won't clip you are mistaken because it is commencing clipping in that graph although it is not a severely shaped/sloped knee/elbow. I would not rely on this test to crank up your Wiim with low frequency output as if it can't clip nor damage your speakers. As per multitone if enough signal strength is provided then it will clip. Why the strong interest in this distortion and clipping behavior?

WiiM Amp Streaming Amplifier Review

This is a review and detailed measurements of the WiiM Amp streaming amplifier. It was sent to me by the company and costs US $299. I was pleasantly surprised by the good bit of weight this small box has. It will help it not slide with the weight of the speaker wires and such. Back panel shows...

www.audiosciencereview.com

The strong interest is in reconciling my mental model of how this stuff works with the possibility of comfortably placing your ear right next to 85dB speakers @750W.

 

[Doodski]

with the possibility of comfortably placing your ear right next to 85dB speakers @750W.

That does not jive. LoL...

As per the review link you provided Amirm stated,
"Unless you use analog input, the gain settings is designed such that you never clip. This means that the "peak" power is the same as what I have measured above. Spec is 120 watts at 4 ohm and we are achieving that so no issue there. Same with 60 watt specified for 8 ohm."

I have no idea how or what they did to get it this way with the digital input but with analogue input clipping is a reality.

 

[shuppatsu]

That does not jive. LoL...

As per the review link you provided Amirm stated,
"Unless you use analog input, the gain settings is designed such that you never clip. This means that the "peak" power is the same as what I have measured above. Spec is 120 watts at 4 ohm and we are achieving that so no issue there. Same with 60 watt specified for 8 ohm."

I have no idea how or what they did to get it this way with the digital input but with analogue input clipping is a reality.

But you keep telling me that I’m clipping! I told you that I’m using the non-pro Wiim amp. I’m not using analog input.

And if I’m not clipping, we’re back at square one. Painfully loud with my Wiim amp, how can they stand being so close to the speaker with the monos?

 

[Inertiaman]

I don't know any specifics about the gain and clipping behavior in this model in question. In amplifiers the clipping can be and is initiated by low frequency energy and high frequency energy too.

My understanding was the Wiim Amp has a very conservative gain control arrangement, such that the amp will essentially never clip. The upside of this for Wiim is they surely mitigate chances for amp failure; the upside for owners is they never experience clipping. The downside is that the max output power is notably lower than what the TPA3255 can do if "unconstrained."
Any/all inputs are either natively digital or digitized on "entry", so even if you wanted to contrive a situation for clipping, you effectively can't feed it an intentionally high input signal like you could with a conventional analog-in/analog-out amp.
Perhaps I've misunderstood something along the way, but as a Wiim Amp owner, this behavior is consistent with my anecdotal experience.

 

[Doodski]

My understanding was the Wiim Amp has a very conservative gain control arrangement, such that the amp will essentially never clip. The upside of this for Wiim is they surely mitigate chances for amp failure; the upside for owners is they never experience clipping. The downside is that the max output power is notably lower than what the TPA3255 can do.
Any/all inputs are either natively digital or digitized on "entry", so even if you wanted to contrive a situation for clipping, you effectively can't feed it an intentionally high input signal like you could with a conventional analog-in/analog-out amp.

Upon being linked to the review and reading it I was not aware that the digital input is regulated to limit the amplifier power output. I now know. The analogue input is not regulated and apparently will clip if pushed.

 

[Doodski]

how can they stand being so close to the speaker with the monos?

I already went into detail about that.

But you keep telling me that I’m clipping! I told you that I’m using the non-pro Wiim amp. I’m not using analog input.

Yes, and now that you linked me to the review on the amp you refer to I now know the digital input is regulated. There's no way I could have known this. Back to the graph the onset of clipping is shown with the slight knee in the graph. It is not severe and is regulated to prevent hard clipping as we all now know. Why did you not advise me that the model has anti clipping on the digital inputs? The review clearly stated that and you where looking at it all along. It seems you are deceiving me by not giving me full disclosure.

 

[Inertiaman]

Upon being linked to the review and reading it I was not aware that the digital input is regulated to limit the amplifier power output. I now know. The analogue input is not regulated and apparently will clip if pushed.

I realize when Amir says "unless you use analog inputs" that one might assume you can get clipping w/ the analog input. But Amir never actually did a power sweep with the analog input. No clipping has been demonstrated, even with analog input. I don't think Erin showed clipping in his review, either, but I may need to go back and re-read.
My perception -- and I could be wrong -- is that Wiim is actively regulating the analog input when it is digitized. The analog input doesn't go directly to the amplifier circuitry; it is first digitized so it can be treated like all the other inputs (equalizer filtering, transmission over the network, etc). It certainly wouldn't surprise me if Wiim "normalized" the input level during this digitization; in fact I would expect that. Once digital, that input is subject to the same gain control as the natively digital sources.