This is a review and detailed measurements of Yamaha RX-A1080 Home Theater Audio/Video Receiver (AVR). I purchased this solely for the purpose of testing through my company (Madrona Digital is a dealer for Yamaha among many other brands). The RX-A1080 is part of the upper tier of Yamaha AVRs consisting of RX-A1080, RX-A2080 and RX-A3080. It costs US $1,300 and is price protected so you won't find it discounted for less online.
The front panel of the RX-A1080 resembles other AVRs such as Pioneer:
I watched a review of the Yamaha online and the person praised the feel of the volume control. Good thing I don't eat cereal or I would have thrown up all of it after hear that! The control is plastic, and stiff and in no way makes you feel good playing with it. Fortunately you won't be using it much and instead, resort to remote control so practically doesn't matter other than longing for great volume control of 1970s and 1980s audio gear.
The front panel opens up and gives a bunch of direct access buttons, again much like Pioneer. And as with Pioneer, we have a "Pure Direct" button.
The back panel is what you expect to see in an AVR:
Hopefully the money spent on analog component video input and all of those composite inputs (in yellow) did not take away from expense elsewhere to produce great performance. I imagine they put these on the back just for show as to make it look like they do a lot. I mean who needs 7 analog inputs plus Phono? Speaking of phono, I will test that in a future installment.
Anyway, you are here to see measurements rather than read my rants so let's get into that.
AVR DAC Audio Measurements
I tried to save myself work and not test this but you all wouldn't let me.
After all, the most common application is HDMI In, speakers out. But let's see how digital input and pre-amp output performs. I started with testing HDMI and took a lot of messing with menus to get the darn thing to output 2 channels and no folding of other channels into Left and Right. Disabling processing, etc. Here is that output with volume level set to 0 dB:
As noted, enabling Pure Direct made no difference with this setup. As is typical of these AVRs, nominal output falls short of 2 volts we like to see out of stand-alone DACs. To get there, I turned up the volume a couple of dBs:
Distortion increased but so did the output level so SINAD (signal over noise and distortion) did not change.
The unit though was at the verge of shutting down despite me leaving speaker terminals disconnected. If I just dialed up the volume 0.5 to 1 dB, it would shut down. So for the rest of the measurements I went with 0 dB volume level.
With a SINAD of 85 dB, you are not going to brag to your neighbors about RX-A1080 in this regard:
Forth and bottom quarter performance. Within the much smaller subset of AVRs, we get somewhat better news:
Mostly because we have much worse measuring AVR DAC subsystem measurements.
I had request to measure with S/PDIF so I choose to use Toslink to avoid any chance of ground loops and got this:
What??? Noise floor goes way down but harmonic distortion -- especially third harmonic -- shoots up resulting in much lower SINAD of 78 dB.
Let me spell this out: there is only one DAC which converts digital to analog. Whether you feed it with audio samples extracted from HDMI or Toslink, the final output should be very similar, sans noise and jitter which could be source specific. What we are seeing is radically different analog distortion profile. I don't understand how this is possible.
To rule out any changes, I immediately retested HDMI and got the same better performance as I have shown above.
Maybe there is some processing going on here but why would it be specific to Toslink and how would it add non-linearities this way and nothing else? Notice how the levels are almost identical to HDMI input.
I plan to investigate this more. If you have some ideas, let me know.
This I think is the first AVR we are measuring with ESS DAC chip. I think it uses an obscure ES9007S. The ES9007 came out back in 2007 I think and has far better performance. Here is the performance of the ES9008 from back then (can't find ES9007):
The SINAD is 114 dB which is far, far higher than our 86 dB number we got in our first dashboard view let alone what the Toslink did. The third harmonic distortion is highest at just below 120 dB.
Back to HDMI input, this is the dynamic range we have:
No better than what a 16 bit audio input could produce. Forget about any high-res audio aspirations with regards to 24-bits and such.
Jitter and noise performance was reasonable actually:
The higher noise floor than state-of-the-art stand-alone DACs hides some sins here but still, it is clean and definitely inaudible tones.
Intermodulation distortion relative to level was revealing:
We see the classic ESS IMD Hump in all of its glory in this graph, showing this problem has existed for more than a decade before we discovered it!
Despite having the RX-A1080 at 0 dB volume, it still shut down before reaching max level. That is why distortion shot up through the roof.
Linearity was decent:
Multitone at 192 kHz sampling was a head scratcher as well:
Can't figure out why the response drops between 50 and 300 Hz. Tested a single tone at a time but did not see it there. So there is some kind of inverted distortion product from the rest of the tones that is interfering. As may be the case with the risen noise floor in low frequencies. For now, don't run with this.
AVR Amplifier Measurements
Per our recent discussions, I am standardizing the AVR dashboard with HDMI input and 5 watt output:
0 dB was fed the the AVR and volume adjusted to produce 5 watt. I also did it for 1 watt as noted above the SINAD in red.
Distortion products are actually not material here: what defines the SINAD/THD+N is the rise spike at 120 Hz or double the mains frequency. This is NOT ground look or mains leakage. Nothing changes it with respect grounding. Instead, 120 Hz is from the power supply rectifier which doubles the incoming frequency (by flipping the negative wave into positive). This is usually smoothed with capacitors and good grounding. One, the other, or both are not done here causing that large spike at -78 dB or so which sets SINAD to the same value. Without it, the SINAD could have been as good as 90 dB. We have to run with what we have and this is where the RX-A1080 ranks:
So slightly below average for all amps tested (in AVRs or otherwise). Among AVRs alone, it is actually decent ranking:
For the rest of the tests, I switched to using the analog input.
Frequency with and without pure mode shows the difference:
Even when the input is being digitized, the bandwidth is pretty good, stretching to 48 kHz or so indicating ADC sampling rate is 192 kHz or thereabouts.
Re-running our dashboard with analog input we get:
We lost 5 dB of SINAD. Switching on Pure Direct gets us that back:
In other words, allowing analog inputs to be digitized to allow bass management, DSP and Room EQ, costs you 5 dB of noise and distortion. This is on top of lackluster value to begin with.
There is just no excuse here for not having a transparent ADC given how low the performance bar of the amplifier is. CD's 16 bits demands 96 dB of signal to noise ratio. If we had that in the ADC, there would be no impact on these measurements with or without ADC.
Notice all the junk in the spectrum of the FFT. We put in 1 kHz, and we get all that noise and garbage. Why or why?
Here is our signal to noise ratio:
At 5 watts, we only have 86 dB or so blended SNR. Watch out for noise with sensitive horn speakers and such. Max power improves fair bit, exceeding CD's spec a bit.
By now you may be asking why the decapitated pink panther is not adorning this AVR. This is why. Power versus distortion at noise at 4 ohm:
Seems like Yamaha does not participating in "AVR mediocracy cartel" when it comes to Amplifier performance. We pure-direct mode enabled in Blue, we see much lower noise floor than its competitors. You can see what happens when we allow ADC to get in the way though (in red) with significantly higher noise floor. But even then, it is still better than other AVRs.
Same thing repeats with 8 ohm:
Give that amp designer a carrot!
Peak power as expected is much higher:
Power versus level and frequency shows almost no frequency dependence which is unusual:
It could be that the noise/power supply spikes are so high that mask the distortions.
Conclusions
Almost any hope we had of the Yamaha RX-A1080 being a stand-out among other AVRs is dashed. We are still in the red when it comes to DAC performance. We still have strangeness in audio pipeline. We still have ADCs that are not transparent to analog inputs.
The one highlight was better noise/distortion at levels below max power. Here, there is a rather wide gap against other AVRs recently tested. Nothing remotely coming close to state-of-the-art amplifiers but at least better than the competition.
As I mentioned, as time permits, I will perform more tests on this.
------------
As always, questions, comments, recommendations, etc. are welcome.
Halloween is just around the corner and the panthers are demanding cool outfits for that. So please be generous with your donations using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
The front panel of the RX-A1080 resembles other AVRs such as Pioneer:
I watched a review of the Yamaha online and the person praised the feel of the volume control. Good thing I don't eat cereal or I would have thrown up all of it after hear that! The control is plastic, and stiff and in no way makes you feel good playing with it. Fortunately you won't be using it much and instead, resort to remote control so practically doesn't matter other than longing for great volume control of 1970s and 1980s audio gear.
The front panel opens up and gives a bunch of direct access buttons, again much like Pioneer. And as with Pioneer, we have a "Pure Direct" button.
The back panel is what you expect to see in an AVR:
Hopefully the money spent on analog component video input and all of those composite inputs (in yellow) did not take away from expense elsewhere to produce great performance. I imagine they put these on the back just for show as to make it look like they do a lot. I mean who needs 7 analog inputs plus Phono? Speaking of phono, I will test that in a future installment.
Anyway, you are here to see measurements rather than read my rants so let's get into that.
AVR DAC Audio Measurements
I tried to save myself work and not test this but you all wouldn't let me.
As noted, enabling Pure Direct made no difference with this setup. As is typical of these AVRs, nominal output falls short of 2 volts we like to see out of stand-alone DACs. To get there, I turned up the volume a couple of dBs:
Distortion increased but so did the output level so SINAD (signal over noise and distortion) did not change.
The unit though was at the verge of shutting down despite me leaving speaker terminals disconnected. If I just dialed up the volume 0.5 to 1 dB, it would shut down. So for the rest of the measurements I went with 0 dB volume level.
With a SINAD of 85 dB, you are not going to brag to your neighbors about RX-A1080 in this regard:
Forth and bottom quarter performance. Within the much smaller subset of AVRs, we get somewhat better news:
Mostly because we have much worse measuring AVR DAC subsystem measurements.
I had request to measure with S/PDIF so I choose to use Toslink to avoid any chance of ground loops and got this:
What??? Noise floor goes way down but harmonic distortion -- especially third harmonic -- shoots up resulting in much lower SINAD of 78 dB.
Let me spell this out: there is only one DAC which converts digital to analog. Whether you feed it with audio samples extracted from HDMI or Toslink, the final output should be very similar, sans noise and jitter which could be source specific. What we are seeing is radically different analog distortion profile. I don't understand how this is possible.
To rule out any changes, I immediately retested HDMI and got the same better performance as I have shown above.
Maybe there is some processing going on here but why would it be specific to Toslink and how would it add non-linearities this way and nothing else? Notice how the levels are almost identical to HDMI input.
I plan to investigate this more. If you have some ideas, let me know.
This I think is the first AVR we are measuring with ESS DAC chip. I think it uses an obscure ES9007S. The ES9007 came out back in 2007 I think and has far better performance. Here is the performance of the ES9008 from back then (can't find ES9007):
The SINAD is 114 dB which is far, far higher than our 86 dB number we got in our first dashboard view let alone what the Toslink did. The third harmonic distortion is highest at just below 120 dB.
Back to HDMI input, this is the dynamic range we have:
No better than what a 16 bit audio input could produce. Forget about any high-res audio aspirations with regards to 24-bits and such.
Jitter and noise performance was reasonable actually:
The higher noise floor than state-of-the-art stand-alone DACs hides some sins here but still, it is clean and definitely inaudible tones.
Intermodulation distortion relative to level was revealing:
We see the classic ESS IMD Hump in all of its glory in this graph, showing this problem has existed for more than a decade before we discovered it!
Despite having the RX-A1080 at 0 dB volume, it still shut down before reaching max level. That is why distortion shot up through the roof.
Linearity was decent:
Multitone at 192 kHz sampling was a head scratcher as well:
Can't figure out why the response drops between 50 and 300 Hz. Tested a single tone at a time but did not see it there. So there is some kind of inverted distortion product from the rest of the tones that is interfering. As may be the case with the risen noise floor in low frequencies. For now, don't run with this.
AVR Amplifier Measurements
Per our recent discussions, I am standardizing the AVR dashboard with HDMI input and 5 watt output:
0 dB was fed the the AVR and volume adjusted to produce 5 watt. I also did it for 1 watt as noted above the SINAD in red.
Distortion products are actually not material here: what defines the SINAD/THD+N is the rise spike at 120 Hz or double the mains frequency. This is NOT ground look or mains leakage. Nothing changes it with respect grounding. Instead, 120 Hz is from the power supply rectifier which doubles the incoming frequency (by flipping the negative wave into positive). This is usually smoothed with capacitors and good grounding. One, the other, or both are not done here causing that large spike at -78 dB or so which sets SINAD to the same value. Without it, the SINAD could have been as good as 90 dB. We have to run with what we have and this is where the RX-A1080 ranks:
So slightly below average for all amps tested (in AVRs or otherwise). Among AVRs alone, it is actually decent ranking:
For the rest of the tests, I switched to using the analog input.
Frequency with and without pure mode shows the difference:
Even when the input is being digitized, the bandwidth is pretty good, stretching to 48 kHz or so indicating ADC sampling rate is 192 kHz or thereabouts.
Re-running our dashboard with analog input we get:
We lost 5 dB of SINAD. Switching on Pure Direct gets us that back:
In other words, allowing analog inputs to be digitized to allow bass management, DSP and Room EQ, costs you 5 dB of noise and distortion. This is on top of lackluster value to begin with.
There is just no excuse here for not having a transparent ADC given how low the performance bar of the amplifier is. CD's 16 bits demands 96 dB of signal to noise ratio. If we had that in the ADC, there would be no impact on these measurements with or without ADC.
Notice all the junk in the spectrum of the FFT. We put in 1 kHz, and we get all that noise and garbage. Why or why?
Here is our signal to noise ratio:
At 5 watts, we only have 86 dB or so blended SNR. Watch out for noise with sensitive horn speakers and such. Max power improves fair bit, exceeding CD's spec a bit.
By now you may be asking why the decapitated pink panther is not adorning this AVR. This is why. Power versus distortion at noise at 4 ohm:
Seems like Yamaha does not participating in "AVR mediocracy cartel" when it comes to Amplifier performance. We pure-direct mode enabled in Blue, we see much lower noise floor than its competitors. You can see what happens when we allow ADC to get in the way though (in red) with significantly higher noise floor. But even then, it is still better than other AVRs.
Same thing repeats with 8 ohm:
Give that amp designer a carrot!
Peak power as expected is much higher:
Power versus level and frequency shows almost no frequency dependence which is unusual:
It could be that the noise/power supply spikes are so high that mask the distortions.
Conclusions
Almost any hope we had of the Yamaha RX-A1080 being a stand-out among other AVRs is dashed. We are still in the red when it comes to DAC performance. We still have strangeness in audio pipeline. We still have ADCs that are not transparent to analog inputs.
The one highlight was better noise/distortion at levels below max power. Here, there is a rather wide gap against other AVRs recently tested. Nothing remotely coming close to state-of-the-art amplifiers but at least better than the competition.
As I mentioned, as time permits, I will perform more tests on this.
------------
As always, questions, comments, recommendations, etc. are welcome.
Halloween is just around the corner and the panthers are demanding cool outfits for that. So please be generous with your donations using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited: