This is a review and detailed measurements of the Onkyo PR-RZ5100 Audio/Video Processor (AVP). It is on kind loan from a member. The PR-RZ5100 was announced (I think) around 2017 and seems to be discontinued. It cost US $2,300.
The outside enclosure certainly gives a "high-end" feel to the PR-RZ5100 with its extra tall front face and large volume knob:
Unlike many modern AV processors, boot time is very fast.
The back panel shows the usual connector including XLR Outputs:
The unit is reasonably heavy although peaking inside, the front half is mostly empty other than a transformer there.
For my testing, I focused on HDMI input and Pure Audio mode for all tests. Pure Audio shuts off the front display as is the case with some other units, making a bit annoying to operate.
The on-screen display was decent quality and easy to navigate. Oddly I could not find a factory reset option so set everything to neutral manually. I only tested the Left and Right channels. I believe a lower quality 8-channel AKM dac is used for the rest.
AV Processor HDMI Performance
As usual we rely on our dashboard of feeding the unit a 1 kHz tone and see what else it produces besides that one tone:
First, it was great to see the output nicely going to 4 volts and beyond. This is the Achilles heel of many AVR and even high-end AV Processors. Then again, SINAD of 82 dB is nothing to write home about:
As you see, a whole bunch of AVRs outperform it. The problem is not distortion as the highest spike there is at or below -100 dB. It is extra noise that piles on top of that and drags the combined score down. You will see this as a common theme throughout the measurements.
Seeing if better performance could be extracted at different output voltage, I ran a sweep to quantify that:
This test is tricky as it seems these AV products look at the volume control position and engage or disengage different gain settings/stages. Here, if I set the volume to 84.5, it naturally maxes out at 4 volts and shows the best performance at that output level. Reason is simple: noise dominates so the more output voltage you have, the better THD+N/SINAD becomes. To see if this continues I had to dial the volume control higher. The moment I did that (regardless of it was at Max or lower), performance drops across the board with increased noise (red line). I suggest getting an amplifier that produces its maximum power at 4 volts or lower so you don't given up even more noise performance.
Signal to noise ratio shows the high noise level and hence poor performance:
High noise floor kills intermodulation distortion+noise performance to well below that of a cheap phone dongle for most of the output level:
The high noise level kills linearity as well in addition to some other issues in there:
32-tone signal tells us what we already know: high noise floor and relative high distortion level:
Jitter performance looks good but this is partially due to high noise floor which is likely hiding some sins:
Fortunately the spikes are either very close to our main tone which means they will be perceptually masked, or too low/below noise level to be heard.
I ran into a very strange problem in the above test. Even though the output was set to 4 volts, as soon as I fed it the J-test signal it shot up to 8 volt and slightly clipping! Went back to a sine wave without changing anything and the output became 4 volts again. J-Test appears to be a 12 kHz tone but it is not. It is actually a square wave at 12 kHz but because the DAC output gets filtered, it becomes a sine wave as the rest of its harmonics are out of band. Seems like the AVR is examining the digital samples and deciding to switch gain for some odd reason.
Wide-band THD+N versus frequency again emphasize the high level of noise:
The dashed blue line is a $99 DAC board.
Finally, filter response shows the typical "DAC chip" somewhat lazy response but not as bad as some others (Marantz: I am looking at you):
Conclusions
Someone told a designer to slap higher output on this unit and don't worry about anything else. While I am happy to see the higher output level, clearly the rest of the unit is not well-engineered. There are flaws everywhere with the most egregious being high noise level likely due to improper "gain staging." The jump in level with j-test jitter signal shows a clear bug in the implementation. Overall this is one of the most broken AV Processors out there with respect to audio. Shocking then to see a review of it with this indication: https://www.soundandvision.com/content/onkyo-pr-rz5100-surround-processor-review
There is some utterly useless "bench test" at the end which explains why they did not see the large set of problems with this processor.
It seems that not only the companies are not paying attention to proper audio fidelity, neither is the press to keep them honest.
It goes without saying that I can NOT recommend the Onkyo PR-RZ5100. Spend your money elsewhere.
-----------
As always, questions, comments, recommendations, etc. are welcome.
Hey, it is not easy to come up with lame jokes every day of the week. So today, I am going to be straight with you to ask you to donate what you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
The outside enclosure certainly gives a "high-end" feel to the PR-RZ5100 with its extra tall front face and large volume knob:
Unlike many modern AV processors, boot time is very fast.
The back panel shows the usual connector including XLR Outputs:
The unit is reasonably heavy although peaking inside, the front half is mostly empty other than a transformer there.
For my testing, I focused on HDMI input and Pure Audio mode for all tests. Pure Audio shuts off the front display as is the case with some other units, making a bit annoying to operate.
The on-screen display was decent quality and easy to navigate. Oddly I could not find a factory reset option so set everything to neutral manually. I only tested the Left and Right channels. I believe a lower quality 8-channel AKM dac is used for the rest.
AV Processor HDMI Performance
As usual we rely on our dashboard of feeding the unit a 1 kHz tone and see what else it produces besides that one tone:
First, it was great to see the output nicely going to 4 volts and beyond. This is the Achilles heel of many AVR and even high-end AV Processors. Then again, SINAD of 82 dB is nothing to write home about:
As you see, a whole bunch of AVRs outperform it. The problem is not distortion as the highest spike there is at or below -100 dB. It is extra noise that piles on top of that and drags the combined score down. You will see this as a common theme throughout the measurements.
Seeing if better performance could be extracted at different output voltage, I ran a sweep to quantify that:
This test is tricky as it seems these AV products look at the volume control position and engage or disengage different gain settings/stages. Here, if I set the volume to 84.5, it naturally maxes out at 4 volts and shows the best performance at that output level. Reason is simple: noise dominates so the more output voltage you have, the better THD+N/SINAD becomes. To see if this continues I had to dial the volume control higher. The moment I did that (regardless of it was at Max or lower), performance drops across the board with increased noise (red line). I suggest getting an amplifier that produces its maximum power at 4 volts or lower so you don't given up even more noise performance.
Signal to noise ratio shows the high noise level and hence poor performance:
High noise floor kills intermodulation distortion+noise performance to well below that of a cheap phone dongle for most of the output level:
The high noise level kills linearity as well in addition to some other issues in there:
32-tone signal tells us what we already know: high noise floor and relative high distortion level:
Jitter performance looks good but this is partially due to high noise floor which is likely hiding some sins:
Fortunately the spikes are either very close to our main tone which means they will be perceptually masked, or too low/below noise level to be heard.
I ran into a very strange problem in the above test. Even though the output was set to 4 volts, as soon as I fed it the J-test signal it shot up to 8 volt and slightly clipping! Went back to a sine wave without changing anything and the output became 4 volts again. J-Test appears to be a 12 kHz tone but it is not. It is actually a square wave at 12 kHz but because the DAC output gets filtered, it becomes a sine wave as the rest of its harmonics are out of band. Seems like the AVR is examining the digital samples and deciding to switch gain for some odd reason.
Wide-band THD+N versus frequency again emphasize the high level of noise:
The dashed blue line is a $99 DAC board.
Finally, filter response shows the typical "DAC chip" somewhat lazy response but not as bad as some others (Marantz: I am looking at you):
Conclusions
Someone told a designer to slap higher output on this unit and don't worry about anything else. While I am happy to see the higher output level, clearly the rest of the unit is not well-engineered. There are flaws everywhere with the most egregious being high noise level likely due to improper "gain staging." The jump in level with j-test jitter signal shows a clear bug in the implementation. Overall this is one of the most broken AV Processors out there with respect to audio. Shocking then to see a review of it with this indication: https://www.soundandvision.com/content/onkyo-pr-rz5100-surround-processor-review
There is some utterly useless "bench test" at the end which explains why they did not see the large set of problems with this processor.
It seems that not only the companies are not paying attention to proper audio fidelity, neither is the press to keep them honest.
It goes without saying that I can NOT recommend the Onkyo PR-RZ5100. Spend your money elsewhere.
-----------
As always, questions, comments, recommendations, etc. are welcome.
Hey, it is not easy to come up with lame jokes every day of the week. So today, I am going to be straight with you to ask you to donate what you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/