This is a review and detailed measurements of the Paradigm PW Amp streaming capable power (speaker) amplifier. It is on kind loan from a member. The PW Amp normally costs US $499 but is being sold currently for just US $225 from company's website. Needless to say this has generated a lot of interest in the unit.
Due to use of class-d amplification and likely switching power supply, the PW Amp is quite compact and light:
Please pardon the plastic over the controls. When someone send me a unit this way, I like to give them the privilege of removing it.
Beside the obvious controls, the one with arrow changes the input from streaming to back analog RCA and back. In use, I thought that when you streamed to it, it would take priority and play that input.
Here is the back panel:
Not wanting to mess with configuring Wifi, I did my testing using Ethernet port.
The speaker connectors are push in type, not banana. Fortunately the hole accommodates a banana plug and that is how I connected my dummy load to it.
The key differentiator for Paradigm PW Amp is inclusion of a measurement mic and Anthem (sister company) room correction (ARC). I have not had time to test it but in past experience, ARC works well and above other mass market solutions in consumer Audio/Video Receivers.
Streaming apps require obligator app on your portable device to control them and the PW Amp is no exception. So I downloaded a DTS Play-Fi app which proceeded to take any confidence I ever had in DTS away. While the App immediately recognized there was a Play-Fi device on the network, it gave me a four digit ID for it rather than its name. Really? Someone was too lazy to put in the protocol a way for the manufacturer to identify its devices?
Next thing after that was the DTS App telling the unit was out of date and wanted to update it. I said go ahead and it started with step 1 out of 2 only to hang there for minutes. I let it sit there until it gave up telling me it failed and whether I want it to try again. I said yes and this time it downloaded something and pushed it to the device as again, step 1. It then said it was rebooting the unit. I sat there for good 10 minutes with the Amp rebooting but the app hung there.
I restarted the app and this time it did not tell me anything about an update. So I told it to browse my music and found my test file to play. It did this a few times but then gave me the hour glass and proceeded to hang completely. At this point I gave up on using it. I mean really, how can the app/system be so untested as to not be able to play a dozen times a file locally on my phone? I shudder to think what it would do if I tried to play networked content.
There is a rule among software engineers that says to never trust a hardware company with software. DTS exemplifies this. Who thought they should trust this company with building a streaming solution? These things seem simple but even large companies have trouble pulling it off properly. DTS needs to close shop given this dismal experience. I also can see why Paradigm is likely trying to clear stock on this unit. Support costs must be through the roof!
Power Amplifier Audio Measurements
I started my testing using the RCA input as that allows me to run the full suite of tests:
Yuck. We not only have high distortion but also high noise floor to go with it, resulting in less than acceptable SINAD (signal over distortion and noise) of just 71 dB.
Thinking the input is digitized so digital streaming using the app may do better, I fired up the Play-Fi with the same signal:
Sadly the performance is the same. Don't mind the shape of the FFT. It is a bug in APx555 analyzer in that it can capture data continuously from external sources reliably. Likewise ignore the Gain measurements as the analyzer is not aware of the input signal level.
Seeing how the performance was the same, I did the rest of my testing with RCA input. Here is the signal to noise ratio:
We have confirmation of the high noise floor contributing to below average SINAD.
Most important test is power versus THD+N:
There is no escaping of the high noise and distortion. The competing products have far lower distortion and hence the reason their graphs keep declining until we get to or near clipping. Not so with Paradigm PW Amp. Distortion starts to creep in as low as a couple of watts and by max power we are at whopping 0.36% distortion. The Amazon Link Amp in contrast remains clean up to its clipping point with similar power rating.
Frequency response measurements shows a very sharp cut-off above 20 kHz:
Even at 20 kHz we are down 1.35 dB. Even with sample rate of 44.1 kHz, there should have been flatter response. Ideally it would use 48 kHz for sampling.
THD+N versus Power and Frequency confirms what we already know:
I had to rescale to graph to 10% THD to see the response in red. Protection trigger was a bit scary in that there is no indication that it has occured other than the unit muting. I thought I had damaged the unit until I power cycled it and it came back to life.
Removing the AES-17 (40 kHz) filter shows this as far as ultrasonic noise:
Some of you have asked for time domain/scope display and I have shown that in the inset. Naturally you see the thickening of the sine wave with high frequency riding on it.
Someone asked for measurements at lower frequencies and I happened to run that by accident and was surprised by its output (in streaming mode):
What's with all that garbage above our tone? I know, it is noise leaking from its streaming subsystem into the output of the amplifier.
Intermodulation distortion shows what we already know:
IMD using two high frequency tones (19 and 20 kHz) makes it hard to see our signal from all the distortion!
Intermodulation products rise up to just -50 dB. Admittedly this is at full power and such tones are not going to occur in your music.
Conclusions
As many of you, I was very hopeful going into this review. Room EQ and streaming functionality with decent amount of power would be a killer product. Alas, this was not meant to be. The streaming functionality is flat out broken. Use as a power amp alone shows very much sub-par performance. It is clear no attention was put in designing a clean amplifier. When a company can farm out a design to get a clean amp and Paradigm can't, you know there is trouble.
No doubt room EQ will make a bigger difference than any of these measurements do in the sound of the amplifier. So there is good value there. Then again if you can perform that function upstream of the amp, then there is no value here.
It is with a heavy heart that I say I can NOT recommend the Paradigm PW Amp.
----------------------
Questions, comments, critique, etc. are welcome.
I like to suggest to the owner that we take out the Paradigm PW Amp and bury it in my backyard to save him the aggravation of streaming functionality. The owner would want to be reimbursed for the cost of the unit so please donating sufficient funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
Due to use of class-d amplification and likely switching power supply, the PW Amp is quite compact and light:
Please pardon the plastic over the controls. When someone send me a unit this way, I like to give them the privilege of removing it.
Beside the obvious controls, the one with arrow changes the input from streaming to back analog RCA and back. In use, I thought that when you streamed to it, it would take priority and play that input.
Here is the back panel:
The speaker connectors are push in type, not banana. Fortunately the hole accommodates a banana plug and that is how I connected my dummy load to it.
The key differentiator for Paradigm PW Amp is inclusion of a measurement mic and Anthem (sister company) room correction (ARC). I have not had time to test it but in past experience, ARC works well and above other mass market solutions in consumer Audio/Video Receivers.
Streaming apps require obligator app on your portable device to control them and the PW Amp is no exception. So I downloaded a DTS Play-Fi app which proceeded to take any confidence I ever had in DTS away. While the App immediately recognized there was a Play-Fi device on the network, it gave me a four digit ID for it rather than its name. Really? Someone was too lazy to put in the protocol a way for the manufacturer to identify its devices?
Next thing after that was the DTS App telling the unit was out of date and wanted to update it. I said go ahead and it started with step 1 out of 2 only to hang there for minutes. I let it sit there until it gave up telling me it failed and whether I want it to try again. I said yes and this time it downloaded something and pushed it to the device as again, step 1. It then said it was rebooting the unit. I sat there for good 10 minutes with the Amp rebooting but the app hung there.
I restarted the app and this time it did not tell me anything about an update. So I told it to browse my music and found my test file to play. It did this a few times but then gave me the hour glass and proceeded to hang completely. At this point I gave up on using it. I mean really, how can the app/system be so untested as to not be able to play a dozen times a file locally on my phone? I shudder to think what it would do if I tried to play networked content.
There is a rule among software engineers that says to never trust a hardware company with software. DTS exemplifies this. Who thought they should trust this company with building a streaming solution? These things seem simple but even large companies have trouble pulling it off properly. DTS needs to close shop given this dismal experience. I also can see why Paradigm is likely trying to clear stock on this unit. Support costs must be through the roof!
Power Amplifier Audio Measurements
I started my testing using the RCA input as that allows me to run the full suite of tests:
Yuck. We not only have high distortion but also high noise floor to go with it, resulting in less than acceptable SINAD (signal over distortion and noise) of just 71 dB.
Thinking the input is digitized so digital streaming using the app may do better, I fired up the Play-Fi with the same signal:
Sadly the performance is the same. Don't mind the shape of the FFT. It is a bug in APx555 analyzer in that it can capture data continuously from external sources reliably. Likewise ignore the Gain measurements as the analyzer is not aware of the input signal level.
Seeing how the performance was the same, I did the rest of my testing with RCA input. Here is the signal to noise ratio:
We have confirmation of the high noise floor contributing to below average SINAD.
Most important test is power versus THD+N:
There is no escaping of the high noise and distortion. The competing products have far lower distortion and hence the reason their graphs keep declining until we get to or near clipping. Not so with Paradigm PW Amp. Distortion starts to creep in as low as a couple of watts and by max power we are at whopping 0.36% distortion. The Amazon Link Amp in contrast remains clean up to its clipping point with similar power rating.
Frequency response measurements shows a very sharp cut-off above 20 kHz:
Even at 20 kHz we are down 1.35 dB. Even with sample rate of 44.1 kHz, there should have been flatter response. Ideally it would use 48 kHz for sampling.
THD+N versus Power and Frequency confirms what we already know:
I had to rescale to graph to 10% THD to see the response in red. Protection trigger was a bit scary in that there is no indication that it has occured other than the unit muting. I thought I had damaged the unit until I power cycled it and it came back to life.
Removing the AES-17 (40 kHz) filter shows this as far as ultrasonic noise:
Some of you have asked for time domain/scope display and I have shown that in the inset. Naturally you see the thickening of the sine wave with high frequency riding on it.
Someone asked for measurements at lower frequencies and I happened to run that by accident and was surprised by its output (in streaming mode):
What's with all that garbage above our tone? I know, it is noise leaking from its streaming subsystem into the output of the amplifier.
Intermodulation distortion shows what we already know:
IMD using two high frequency tones (19 and 20 kHz) makes it hard to see our signal from all the distortion!
Intermodulation products rise up to just -50 dB. Admittedly this is at full power and such tones are not going to occur in your music.
Conclusions
As many of you, I was very hopeful going into this review. Room EQ and streaming functionality with decent amount of power would be a killer product. Alas, this was not meant to be. The streaming functionality is flat out broken. Use as a power amp alone shows very much sub-par performance. It is clear no attention was put in designing a clean amplifier. When a company can farm out a design to get a clean amp and Paradigm can't, you know there is trouble.
No doubt room EQ will make a bigger difference than any of these measurements do in the sound of the amplifier. So there is good value there. Then again if you can perform that function upstream of the amp, then there is no value here.
It is with a heavy heart that I say I can NOT recommend the Paradigm PW Amp.
----------------------
Questions, comments, critique, etc. are welcome.
I like to suggest to the owner that we take out the Paradigm PW Amp and bury it in my backyard to save him the aggravation of streaming functionality. The owner would want to be reimbursed for the cost of the unit so please donating sufficient funds using:
Patreon: https://www.patreon.com/audiosciencereview), or
upgrading your membership here though Paypal (https://audiosciencereview.com/foru...eview-and-measurements.2164/page-3#post-59054).
