Hyperion APR-16 AV Processor Review

[JD_Spoon]

Haha, the AVM90 is great, but if you believe in even half of what people say about ART, especially Mr. Robinson's review on the Denon AVR-X3900H (not about the Denon hardware, but about the ART performance part only), you might change you reconsider.

I usually only watch his reviews if I get bored and want to watch something entertaining, but his comments on ART seems consistent with most, probably all ASR ART user.

My own experience working with the two has been that while they have some different strengths and focus areas, ARC Genesis and Dirac Live full range are broadly equivalent, but Dirac Live ART has created a huge capability gap in what Anthem offers in their room correction platform, and thus far they have not been able to close the gap in matching what ART can do for optimizing multichannel audio.

It frustrates me to no end that there isn't a device available yet that supports ART in a purely stereo, 2.n configuration. You're stuck either buying an overkill multi-channel approach, paying the Trinnov premium for the Amethyst and its myriad capabilities, or accepting that you'll only get Dirac Live full range in the limited devices that support that.

 

[dogmamann]

$8000 is a lot of money than a Denon X 3x00 receiver.

 

[Sokel]

For reference AudioControl have used Noctua fans in all their units where they need them since the Bijou product line. Their hardware engineers are massive nerds so fully appreciate choosing the best option. There are no fans in the APR-16. It's a fairly sparce device in terms of volume usage due to the form factor needed for all the back panel connections needed. The DSP board has a TI Sitara chip and there's a pair of headers on the mainboard next to it that says INSTALL STREAMING MODULE HERE, so some future hardware expansion prepared.

Dante wise I believe it's all done in software rather than a potentially limiting hardware chip and will be 16 channel output with a later firmware update. I suspect this will coincide with the DPR-16 becoming available as obviously it's somewhat core to that product and they'll have the same DSP/ARM/processing internals.

The nerdy, god level about fans (depending the task) would naturally be papst .
Noctua is probably more recognizable amongst PC enthusiasts though.

 

[peng]

My own experience working with the two has been that while they have some different strengths and focus areas, ARC Genesis and Dirac Live full range are broadly equivalent,

Is that based on subjective listening experiences? On the objective side, based on numerous REW sweeps, ARC Genesis did not do well versus Dirac Live, or even Audyssey, though it does work well enough if I spent hours tweaking. I like the visual look of the interface, and the mic kit, other than that, I wish they offer a paid option to use Dirac Live.

It frustrates me to no end that there isn't a device available yet that supports ART in a purely stereo, 2.n configuration. You're stuck either buying an overkill multi-channel approach, paying the Trinnov premium for the Amethyst and its myriad capabilities, or accepting that you'll only get Dirac Live full range in the limited devices that support that.

I will likely eventually upgrade my AVM70 to an ART capable AVP, and relocate the AVM70 to my stereo (2ch) system that currently runs DLBC, and will upgrade to ART if and when that (PC standalone version) becomes available, apparently some time this year. It that's what you mean by ART in a "purely stereo, 2.n config", that's what I will be doing as soon as the PC standalone version ART becomes available. The "device" would be the PC/laptop/miniPC etc.).

 

[welwynnick]

My own experience working with the two has been that while they have some different strengths and focus areas, ARC Genesis and Dirac Live full range are broadly equivalent, but Dirac Live ART has created a huge capability gap in what Anthem offers in their room correction platform, and thus far they have not been able to close the gap in matching what ART can do for optimizing multichannel audio.

I thought that ARC was essentially an amplitude/frequency correction tool, while Dirac extended that to time domain correction as well.

Nothing against Anthem here, I've used ARC to great effect in my system, but I thought Dirac took things further? Just trying to keep learning.

 

[Oddball]

Your mileage with ART 2.n (say 2) system will greatly vary based on the room response. If you have just L sub support L speaker, and R sub supporting the R speaker, and subs are closely co-located with respective mains, you are at risk that dips in the response will not be fixed. Also, there might be impact on signal cancelation ability due to limited support. While people report good results with 2.2 systems, it did not work really well in my system. Stereo is still best when I have 4 subs (2 front 2 back) supporting the LR as a group and then have at least support from front wides that are close, but still 1m away from mains.

@welwynnick - ART is billed primarily as reducing decay. Bass control is already doing all the advanced Dirac functions and there is e.g. no great difference in frequency response graphs between the two. It is a bit of a black box what ART actually does, but many graphs show that decay is under better control. Another significant difference is that ART is not using traditional crossover concept as Bass Control but rather virtual channels that consist of the main channel and other supporting channels that you assign and set the parameters for that support.

 

[JD_Spoon]

Is that based on subjective listening experiences? On the objective side, based on numerous REW sweeps, ARC Genesis did not do well versus Dirac Live, or even Audyssey, though it does work well enough if I spent hours tweaking. I like the visual look of the interface, and the mic kit, other than that, I wish they offer a paid option to use Dirac Live.

I intentionally left a lot of room for interpretation in my use of "broadly." However, I was speaking more in terms of their functional capabilities, not their raw output, because my functional experience is that both require varying degrees of tweaking to get them dialed in to preferences. I freely admit I got sold a lot of hype on the ARC Genesis capabilities, and IIRC at the time I bought them, the only real options for a Dirac Live room correction-enabled AVP were the NAD M17 v2i, and maybe an Arcam AV40. ART wasn't a thing at all, even when it was announced it was a Storm-exclusive option.

 

[welwynnick]

@welwynnick - ART is billed primarily as reducing decay. Bass control is already doing all the advanced Dirac functions and there is e.g. no great difference in frequency response graphs between the two. It is a bit of a black box what ART actually does, but many graphs show that decay is under better control. Another significant difference is that ART is not using traditional crossover concept as Bass Control but rather virtual channels that consist of the main channel and other supporting channels that you assign and set the parameters for that support.

I think JD Spoon was talking about DLRC, and that was what my reply was about. Dirac specifically include Impulse Response Correction in their description of the key features of DLRC. I think ART is another thing completely, and in the absence of anything more than marketing hyperbole and (very convincing) pretty charts, I've been trying to figure it out for a while. This was my guess about how it's done:

Generically, I suppose you take the output of each channel, invert it, and apply a filter function and a time delay. I think it becomes a more complicated problem to solve that simply applying a correction profile. You could see it either as a frequency domain, time domain, or a geometric problem.

In the frequency domain, you look at the peaks and nulls generated by one channel, send an inverted bass signal to a correction sub, and juggle the time delay and frequency response to minimise the overall response variation as best you can. Then repeat with the other correction subs. I guess the correction signals might have to be multiple superimposed correction signals - one with a time delay of say 20ms, one with a delay of 30ms, ... etc. It sounds iterative and difficult.

In the time domain you could generate some test pulses or steps, and see when the responses come back from the room. The use those as the correction signal time delays.

It might be possible to imagine the LF room modes problem geometrically. Draw a plan of the room with all the speakers and walls. Consider the walls to be mirrors, and draw the mirror images of each LF-capable speaker on the plan. Draw and measure the direct and indirect path lengths from each speaker, and each image, to the listening position. Calculate the path differences, then calculate the associated time delays. That's kind of similar to above, except you measure distances rather than time delays. One problem is that it's a 3D problem, not 2D, so you really need to consider the floor and ceiling.

Maybe it's not that difficult, and you don't have to generate a correction signal filter function, and you just pass everything through 20Hz to 200Hz and apply a global time delay or delays to all of it. That might be OK if you assume that subs are omni-directional and the walls reflect all bass frequencies equally. A DBA sounds a lot simpler, and wave-forming sounds positively one-dimensional in comparison.

Using high power correction subs to remove the waves generated by the primary subs seems wasteful and resource-heavy. But part of me thinks it makes sense to cancel the cause of the room effects at source, rather than boosting the primary signal quite a lot in a vain attempt to cover up any nulls.

 

[Oddball]

I think JD Spoon was talking about DLRC, and that was what my reply was about. Dirac specifically include Impulse Response Correction in their description of the key features of DLRC. I think ART is another thing completely, and in the absence of anything more than marketing hyperbole and (very convincing) pretty charts, I've been trying to figure it out for a while. This was my guess about how it's done:

There are lots of speculations about ART and how it really works. I have no idea or theory as there is no white paper that would cover that. If what ART is doing would be so simple, perhaps others would have it by now. ART has been out for like 3 or more years - albeit on Storm only until Oct 2025.

ART is definitively not a marketing hyperbole. Take a look at the Dirac filter charts. The virtual channels they create were never done before - plus you have the ability to influence all that and tailor to your liking. If ART was no good wonder why Mini, Hyperion and Tonewinner would go for it or why Onkyo is flirting with it for their next release.

While there are no confirmations, I am pretty sure that there is canceling in the bass area. Otherwise would be difficult to defeat the room. This is the similar principle as Trinnov waveforming, but without the strict placement required by Trinnov. I did not experience any wastefulness - can play as loud as I did with Audy or Bass Control.

 

[welwynnick]

There are lots of speculations about ART and how it really works. I have no idea or theory as there is no white paper that would cover that. If what ART is doing would be so simple, perhaps others would have it by now. ART has been out for like 3 or more years - albeit on Storm only until Oct 2025.

There is a white paper that explains how MIMO is used in ART and Unison (for cars, which pre-dated ART by several years) , but it hardly explains what's going on. the closest it gets is :

Now, picture placing another speaker at the origin of the echo and playing back the sound at precisely the right moment but in reverse phase, adjusting for how the sound travels, e ectively cancelling out the echo. This concept is close to Active Noise Cancellation (ANC), where unwanted sound is neutralized by introducing its opposite.

And there's no mention of cross terms at all.

ART is definitively not a marketing hyperbole. Take a look at the Dirac filter charts. The virtual channels they create were never done before - plus you have the ability to influence all that and tailor to your liking. If ART was no good wonder why Mini, Hyperion and Tonewinner would go for it or why Onkyo is flirting with it for their next release.

I wasn't suggesting that ART was just marketing hyperbole, only that there's very little explanation about what it does and how it does it. The post above was my attempt to try and reason what the functionality is. People often talk about cross terms, being the filter that is applied between a pair of bass speakers, but I think it's more complicated than that, as the filter will need a time domain component to account the propagation delay, so I think it's a 3D filter rather a 2D filter (and there are a lot of them).

While there are no confirmations, I am pretty sure that there is canceling in the bass area. Otherwise would be difficult to defeat the room. This is the similar principle as Trinnov waveforming, but without the strict placement required by Trinnov. I did not experience any wastefulness - can play as loud as I did with Audy or Bass Control.

I'm wrestling with the bit about cancelling. I think the point of ART, that other schemes can't do so effectively, is avoid cancelation without having to resort to boosting signals at an in-room null point. And it helps the peaks as well, of course. That's one of the ways I look at it.

 

[Oddball]

There is a white paper that explains how MIMO is used in ART and Unison (for cars, which pre-dated ART by several years) , but it hardly explains what's going on. the closest it gets is :

And there's no mention of cross terms at all.

I wasn't suggesting that ART was just marketing hyperbole, only that there's very little explanation about what it does and how it does it. The post above was my attempt to try and reason what the functionality is. People often talk about cross terms, being the filter that is applied between a pair of bass speakers, but I think it's more complicated than that, as the filter will need a time domain component to account the propagation delay, so I think it's a 3D filter rather a 2D filter (and there are a lot of them).

I'm wrestling with the bit about cancelling. I think the point of ART, that other schemes can't do so effectively, is avoid cancelation without having to resort to boosting signals at an in-room null point. And it helps the peaks as well, of course. That's one of the ways I look at it.

Not sure what else to say - try ART as rationalizing about it is probably too complicated even for the greatest minds. So the speculations you offer are of anecdotal use.

Cross-terms is a limitation to how far you can go in cross-supporting your speakers. D&M have limited that to a certain number under 100 for their various gear, for unknown and undisclosed reason. HTP-1 does not have this limitation, but also not 19 discrete channels. Storm's limitation is 500 which is quite far out but they do note that in highest channel count systems it could run out. In general, the more supporting speakers you have the spread will get tighter aka more accurate reproduction. But then there is the opposite side of the coin where support could be better off following some rules other than everything supports everything. There is no common wisdom in both approaches. But certainly would not be the first time that pretty graphs are not preferred when you engage your hearing organs. BTW, there will likely be no bad measuring ART graph, so decision becomes a bit difficult.

 

[Billy Budapest]

The nerdy, god level about fans (depending the task) would naturally be papst .
Noctua is probably more recognizable amongst PC enthusiasts though.

I haven’t heard of Papst fans, but Noctua are known to be very quiet and are readily available in the supply chain.

 

[amirm]

I received special access to beta version of the firmware which fixed the issues I found. Review is updated with IMD and THD+N results. Here is the latter:

The update process was somewhat glitchy so I hope they fix that by the time it is released.

 

[Billy Budapest]

I like what I see. Is it just a coincidence, or do I spot a bit of the ESS IMD hump in the new measurements? (I don’t even know if the unit employs an ESS D/A converter).

 

[rhale64]

Good to see they fixed the issue. Thank you for the review of this. I am now glad I bought it.

 

[Sokel]

I haven’t heard of Papst fans,

Industrial fans used in very hostile environments, etc.
Highly regarded here in Europe.

 

[rhale64]

I like what I see. Is it just a coincidence, or do I spot a bit of the ESS IMD hump in the new measurements? (I don’t even know if the unit employs an ESS D/A converter).

It has the new ESS Sabre 9039 Dacs.

 

[W2APS]

Good to see they could roll out a fix so swiftly. I'll keep an eye out for this on the next update of our demo unit.

 

[thcdru2k]

I think it's fair to say people should change their votes to golfing panther. This seems like an incredible value/$ when compared to Stormaudio/Trinnov.

 

[grobalt]

I haven’t heard of Papst fans, but Noctua are known to be very quiet and are readily available in the supply chain.

Sorry, as a German i have to add a bit of education ​

ebm-papst: The Industrial Heavyweight (Est. 1963)​

Papst is not just an old company; it is an foundational pillar of modern ventilation technology.

• The Origins: The company we know today as ebm-papst was born from the merger of three pioneering German companies, most notably Papst Motoren (founded by legendary inventor Hermann Papst in the Black Forest in the 1940s) and ebm (Elektrobau Mulfingen, founded in 1963).
• The Innovations: Hermann Papst invented the external rotor motor, which revolutionized how compact cooling fans operate. In 1975, ebm developed the very first electronically commutated (EC) motor, boosting fan efficiency from a terrible 15% (old shaded-pole motors) to a massive 65%.
  
  
• The Scale: ebm-papst is a massive global enterprise with billions in annual revenue, tens of thousands of employees, and manufacturing plants worldwide. They design ventilation for automotive giants, medical equipment, massive server farms, data centers, and heavy industrial HVAC systems.
  

When you buy a Papst fan, you are buying into over 60 to 80 years of hard industrial engineering heritage.

Noctua: The Premium Boutique (Est. 2005)​

Compared to Papst, Noctua is absolutely a "new player," but they completely flipped the consumer market on its head.

• The Origins: Noctua was founded in Austria in 2005 as a strategic joint venture between the Austrian distributor Rascom and the Taiwanese manufacturing specialist Kolink International.
  
  
• The Focus: While Papst cools everything from high-speed trains to commercial refrigerators, Noctua was created with a laser-sharp focus: low-noise premium PC cooling.
  
  
• The Engineering: Despite being younger, Noctua operates like a high-end research lab. They pioneered ultra-tight blade tip clearances (just 0.5 mm on their NF-A12x25) using proprietary materials like Sterrox liquid-crystal polymer (LCP) to prevent blades from stretching over time.
  
  
• The Cult Status: Noctua didn't care about industrial volume; they cared about acoustic optimization. By focusing entirely on premium computer fans, heatsinks, and keeping things whisper-quiet, they built a massive cult following and have won over 7,000 media awards.
  

The Verdict: How Do They Compare?​

If you look at the raw facts, they operate in completely different leagues, but their paths cross at the very top:

Feature	ebm-papst	Noctua
Primary Domain	Industrial HVAC, Automotive, Server Infrastructure	High-end Consumer PCs, B2B Quiet Cooling
Engineering DNA	Ultimate reliability, raw power, massive static pressure, industrial efficiency	Psychoacoustics, specialized fan-blade geometry, long-term consumer support
Lifespan	Rated for extreme industrial environments (often 100k+ hours under harsh conditions)	Legendary reliability in clean/home environments (backed by a 6-year warranty)