Some people claim that there are large differences in how well AVPs decode audio, etc. Then again, those claims are based only on listening tests. It would be nice to be able to measure more aspects of AVP decoding and room correction. But I do understand that comprehensive AVP testing is complicated and can be very time-consuming.
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Hyperion APR-16 AV Processor Review
- Thread starter amirm
- Start date
I don’t know what that is.
So what would the 10th channel be?
Never mind—just downloaded it. It will take me a little while to read all 156 pages.
Using the tables in the appendix, I would categorize it between level 2 and level 3. It has the required number of non-subwoofer speakers for level 3 (13) but most of its attributes fall within the level 2 parameters. That’s eyeballing/estimating as I have not conducted measurements where indicated in the appendix.
Hi Amir, I now remember seeing something similar with the Motu Ultralite MK5. The EQ filter was on by default and generating an overflow at low frequencies too and at 192kHz SR, even with all EQ settings flat. Maybe there’s something similar here?
welwynnick
Major Contributor
- Joined
- Dec 26, 2023
- Messages
- 1,334
- Likes
- 1,715
Thinking out loud for a moment (bear with me) the low frequency noise issue sounds slightly familiar. Maybe it's some kind of bass management problem?
The Arcam AVR850 had a high noise floor and SINAD measured 81dB, when the front speakers were set to large.
When the speakers were set to small, the noise floor dropped and the SINAD increased to 92dB.
The Anthem MRX1140 was much the same - the noise floor dropped and SINAD increased from 91dB to 103dB when bass management was enabled.
What's peculiar about the APR-16 is that the noise floor is either high or low, depending on whether a low frequency tone is present. Amir tested 20, 50 & 100Hz - all of them bass frequencies - and the noise level rose around 20dB in each case. It was the same increase for all of them. More than that, this wideband chart shows the noise only increases up to around 22kHz.
That's the sort of "noise frequency response" that you might get if the 20Hz to 22kHz band went through a A>D>A conversion, and followed by a reconstruction filter, such as when DSP crossovers and equalisation were being applied .
This doesn't quite add up, like in the Arcam & Anthem cases, but maybe there's an automatic LFE detector and switch that's cutting in an additional DSP process?
EDIT: Here's an example from Stereophile of a Jurgen Reis (MBL) FFT of a DCS reconstruction filter. Ignore the green and blue lines.
The red line is the filter characteristic on wideband white noise. There's a roll-off just above 20kHz that looked somewhat similar to the Hyperion behaviour, above.
Last edited:
It went through the DAC indeed, with a cut around 24kHz from the digital filter as Amir reported with white noise.
So, like you showed, it seems there’s processing before conversion which could explain the issue, and it could come from a DSP that manages bass/speaker size/PEQ (even if all set to 0/flat).
Last edited:
This one is much slower than that of the Hyperion, because it is fully active at 30kHz only.
The Noctua fans are a surprise! This suggests they may very well want to fix the firmware ASAP.
welwynnick
Major Contributor
- Joined
- Dec 26, 2023
- Messages
- 1,334
- Likes
- 1,715
Sure, it was just an example of low pass filtered white noise.
The shape reminded me of the reconstruction filter tests in Stereophile and SoundStage. That was the clue.
Here's an example of a steep filter:
Same test, same player, different filter.....
Sokel
Grand Contributor
- Joined
- Sep 8, 2021
- Messages
- 11,865
- Likes
- 13,241
I think that the one with the fans is a different model, I see some speaker binding posts at the bottom.
Sokel
Grand Contributor
- Joined
- Sep 8, 2021
- Messages
- 11,865
- Likes
- 13,241
I'll drop this one here:
The pattern looks very familiar and can explain a lot, is the one I mentioned earlier.
Maybe the whole thing is about recourses and where to use them, don't know much about the horsepower of this piece of gear but at its price it must pack (ought to) some serious horsepower so to add the needed precision.
Nice that they used Noctua fans throughout the design!
I had Auro in my Marantz and I have it in my Maestro x9s, I don't use it - had some loudness issues in the Marantz when using the processing so I didn't try it again. I've yet to try it in the X9s.
Last edited:
OP
- Thread Starter
- #97
As far as I know, Dolby requires bit exact decoding before they sign off on a company's implementation. So there can't be differences unless there are bugs.
I could try testing it but need the bitstream version of the test signal available to play somehow. So it is not hard but I don't have a clear path to it right now.
AudioKC
Active Member
- Joined
- Jul 28, 2021
- Messages
- 209
- Likes
- 261
Why multi-tone is SOTA performance, but this one is measured badly, and next tests is -22 db in SINAD?
I don’t get it
Multi-tone is perfect, therefore it will be perfect in playback of real media, no?
p.s. I didn’t read all questions/responses, so advanced apologies if it’s already answered elsewhere.
OP
- Thread Starter
- #99
I don't know but multitone levels are lower because otherwise, the signal would clip.
beeface
Addicted to Fun and Learning
- Joined
- Dec 5, 2017
- Messages
- 583
- Likes
- 1,375
Not denying that it's expensive, but it seems like an industry-wide phenomenon where there's a steep price increase when the channel count on a processor increases.
- On paper, this unit seems most similar to the Marantz AV10, and is priced similarly. The AV30 is a far better value unit for everyone who doesn't need more than 11 channels.
- The Anthem AVM90 does seem like better value ($6375 when Amir reviewed it 2.5 years ago). Uses ARC instead of Dirac - whether this matters, I don't know.
- The miniDSP is much cheaper at $3.5k - 16 channels at this price makes it a huge outlier. I suspect some people won't get past the HDMI specs: no 8K for their non-existent 8K media and source, only 3 inputs instead of 7, no ALLM for their gamer son, the suggestion of adding extra boxes like the HDFury Vrroom as a workaround. But it's undeniably the best value solution for certain people with a particular setup. I think a lot of people perceive it (unfairly or not) as a niche, audio-focused, DIYer product. I'm interested to see if integrators will install it for budget-conscious clients.
- Conversely, Trinnov/JBL is way more expensive!
Last edited:
