Jump to content

Article: Digital Signal Processing - The Ultimate Guide To High End Immersive Audio

Recommended Posts

8 hours ago, Keith_W said:

Yes Mitch I know that Acourate is not for everybody. I know someone who is putting together a 5.2 system and asked me to help him correct his system with DSP. He wanted to use Acourate. I strongly advised against it, my recommendation was Audiolense + send you an email.


I go through that time alignment procedure for my 8 channel system (2x active 3 way speakers + 2 subwoofers) and it takes me hours to do it. So I was pretty impressed when a friend who used Audiolense came over with his laptop. 30 minutes was spent explaining my system architecture to him, downloading drivers, performing channel checks, and all the usual futzing around before a single sweep can be done. But when he actually got to it, the whole thing was done within 15 minutes, from crossover generation to usable filters and verification measurement (had to be done in REW, because Audiolense can't do verification measurements?). I was amazed when I saw the step response, it was textbook perfect. Not that I can't get the same result, but despite my proficiency, it takes me a long time. 


Many years ago, DSP correction via Acourate/Audiolense was much less well known and it was difficult to get help. This was even before your book came out. I posted a question asking about it in another forum, and a very kind member rang me from the USA to talk me through it. I have never forgotten his help and I am grateful for as long as I remain in this hobby, which will stay with me as long as I have intact hearing! I hope to do everything I can to help people see the benefits of DSP, in the same way that he helped me.


I am not partial to one software solution over another, I think there are different advantages and disadvantages to different software packages that might suit some folk more than others. Acourate has a very "Teutonic" approach which is both good and bad depending if you are the kind of person who enjoys doing your own car maintenance.


When I first bought your book I thought that I am way out of my depth here, but after 8 years and multiple re-readings, I have gone on to look up all the references you cited and come to my own conclusions. For example, you recommend measuring without the sofa. I asked myself why I should do that, when the listening sofa is always at that position? So I performed the experiment - measure with and without sofa, correction with/without sofa, and verification measurement with/without sofa. There is a noticeable difference measurably and audibly. Conclusion: measuring and correcting without sofa and then performing the verification measurement with the sofa in situ messes up the correction, but it actually sounds better. 


Another example: Toole says that the Harman target is not a target, it is the result of putting a speaker that measures flat under anechoic conditions in a room, this will naturally roll off the higher frequencies. In your book you suggested choosing a target to preference. So I tried Toole's suggestion, I knew that limitations of measurement means that any freqs < 425Hz (transition zone in my room calculated from 4x Schroder) would be meaningless, but that is OK because I was planning to use a different bass correction strategy anyway. I corrected the nearfield response to flat, then applied correction < Schroder. To my surprise, verification measurements showed a rising treble response at MLP instead of a falling one! Anyway, after a lot of investigation it turns out that the directivity of the horns was causing them to behave differently to the more omni woofer. So what does that say about Toole? I think it may not apply to horns, although I am not brave enough to say that to him ;) My system, I do what I want, and I use my own target curve, as per your recommendation in your book. 


Anyway, to other readers of AS: DSP is a really worthwhile pursuit. I think that in 2024, every system should have DSP. I would go further than Chris and say that anybody who refuses to consider it is stuck in outdated thinking and misplaced priorities. 

What an enjoyable read @Keith_W. Its about the journey and destination. 

When I started going down the DSP road I assumed there was a single objective destination that math and physics would lead me to. I quickly learned, with tons of information from Mitch, that the objective hands off to the subjective once the fundamentals are taken care of. Target curves and final adjustments are all about preferences. Once I grasped that, I had much more fun working with Mitch to get the best sound possible. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
8 minutes ago, SQFIRST said:

Thank you for a very informative article. It is very interesting to see immersive playback unfold and you are spearheading that front to a large extent. However, don't forget about the antiquated 2-channel setup entirely :) and we can use some network DSP there as well. Had to smile that you did draw the immersive audio line by subtracting the use of a screen.

Thanks for the kind words. 


I will never forget about two channel. I listened to a great two channel album this morning on my main system and I listen to two channel at my desk all the time. Immersive is just another option for people. 


Network DSP for all, would be great. We need to get HiFi manufacturers to embrace Ravenna. It puts everything on the network and opens up a world of possibilities. If it's good enough to capture a live orchestra, where there are no second chances, it's good enough for HiFi. 


No screens allowed in my room :~)

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
5 minutes ago, Keith_W said:

BTW, aren't you the guy who blew up his speakers when trying out the Merging Anubis? ;) Not easy to set up, are they.

Ha! yes, I blew a pair of Wilson Audio TuneTots up with an Anubis. I've come a long way. Now I can configure it with my eyes closed and I know the Anubis and HAPI Mk2 inside and out. Right now I have the Anubis setup to receive 7.1.4 from my MacBook Pro, and 4.0, 5.1, and 7.1.4 from an Aurender ACS10. I just switch "inputs" on the Anubis screen at my listening position. It's so flexible. I love it. 


This complexity is only there for people like us who want to know everything about the components and how to configure them. Normal people (we aren't normal), could have it configured by an expert, then never touch it. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
2 hours ago, Confused said:

I have a Denon AVR, although this is more used as a processor as I have external amps for all but the height speakers.


This comes with Audyssey as the default "built in" software. Optional software is also available, Dirac Live (limited Bandwidth), Dirac Live (Full Bandwidth) and Dirac Live Bass Control.


I am assuming that Dirac Live Bass Control is of no use to me, I do not use a subwoofer.


So I am thinking of getting the licence for for the Full bandwidth Dirac Live Room Correction. The licence is $349.


I presume that the Full Bandwidth Dirac option is superior to Audyssey?

I have no direct experience with Audyssey, but I know many people who believe your assumption is correct. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
19 hours ago, maxijazz said:

One of hybrid approaches could be use of Genelec SAM speakers with GLM software.

It allows for speaker-level bass management with optional 9301B interface (~$1k).
Linear phase from 100Hz up or low latency (~3ms); your choice.

Not as good spec as computer based DSP, but pretty good (and expensive).

You only need Merging Horus or Hapi (depending on number channels) without and DAC cards. DACs are built in to speakers.

If that is not enough, you can optionally use computer based DSP before sending stream to Merging.




Hi @maxijazz Thanks for bringing up this option. It raises some questions I've been thinking about over the weekend. 


If using Genelec speakers with built-in DACs and DSP, where does the volume control take place and how does the user control it? For example, if I'm streaming from JRiver to a HAPI to Genelec speakers. Without an Anubis, the user would have to be physically turning the knob of the HAPI for volume control. 


Also, if volume control is done prior to DSP inside the speakers, is this a big no-no? In my head this is an issue, but I honesty don't know if it plays out this way. 


I've been thinking of putting a DSP box between my Anubis and HAPI, but my hesitation is that the Anubis does volume control and this will reduce resolution for the DSP filters. 


@DigiPete do you have experience with Genelec DSP?

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
1 hour ago, Confused said:

So I am only thinking of Dirac for the 7.4 ATMOS side of things. Am I right that Audiolense is for PC only? For my surround system the source will be Blu-ray discs, Apple TV and similar, nothing streamed from a PC. So this takes me back to Dirac being my best option I think, for multi-channel at least.


One of the really cool benefits of using a protocol like Ravenna / AES67 is that you can route audio from non-computer devices to a computer for DSP. For example, I could use a Blu-ray player or Apple TV connected to an Arvus H-1D (HDMI in / AES67 out) and send that audio stream to Hang Loose Convolver for room correction, then out to my audio system. One can also just rip the Blu-ray and stream from Apple Music to get the same content if desired. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
  • 2 weeks later...
6 minutes ago, asibbald said:

I'd add one more device to the list of options - the Smythe Research Realiser A16.

Smythe focus on its unique ability to provide REAL immersive sound through headphones.

I can attest to its breathtaking realism through headphones once properly set up, although the setup requirement of a proper multiway HRTF is a considerable obstacle unless one already has a full surround or Atmos system.


In addition though, it is a full 24 channel Atmos/DTS:X etc decoder - and as well as the usual HDMI inputs has 16 inputs and 16 outputs. Depending on the version you get, these 16 in / 16 out are available in unbalanced (3.5mm TRS), Balanced XLR or Digital AES/EBU.


I have no connection to Smythe other than having purchased an A16 from them.


The hardware itself is excellent - and the headphone virtualisation spectacular, once you have your HRTF.


Having sung the praises of the Realiser A16, I have to say though that Smythe are one of the most frustrating companies I have ever dealt with. Emails and phone calls often go unanswered. They have no marketing to speak of - despite an excellent and unique product.


One upside though is that when they DO answer your call or email, they are extremely knowledgeable and helpful.


I get the feeling they are small team that are a) Overworked and b) more interested in the technology R&D than anything else.


But despite these negatives, I would still buy an A16 if I didn't have one. It's that good - and the headphone virtualisation is unique.



Hi @asibbald, I kind of forgot about Smyth Research. After attempting to contact the company numerous times and hearing nothing in return, I gave up on it. 


I'd love to use the A16 with the Dante card to input encoded Atmos and output decoded Atmos as PCM over the network. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
  • 2 months later...
4 hours ago, catastrofe said:

From what I've seen in the marketplace (MiniDSP, DSPeaker, etc.), most room correction DSP is done at 24/48 or 24/96 (my Lyngdorf TDAI-1120 resamples everything to 24/96). How does one rationalize that resampling against the desire for high quality DSP (say via Roon or HQPlayer). If the DSP occurs before a stand-alone DAC, the DAC is receiving a "bastardized" signal from the DSP unit. If the DSP occurs after the DAC, the DSP is changing the result of the DAC's work.


I'm really struggling with this as I'm trying to design a system that incorporates room correction (I'm currently using convolution filters in Roon to good effect), bass management (I run a 2.2 stereo system and my subs don't have a HP pass-through), and an external DAC so I can use Roon/HQPlayer...I purchased a lifetime Roon license when Roon first launched.


I know I can load FIR filters in HQP, so I could move that function from Roon to HQP. I also believe there's a way to handle bass management in Roon using Procedural EQ, but I have no idea how to accomplish that and there isn't any type of guide available. Roon provides for speaker distance entries, but only for the mains, not the subs. There are a couple of analog crossover products available from Behringer and DBX, but I've read that they're noisy.


A MIniDSP or DSPeaker solution would be easy single box solutions, but they wouldn't be my first choice for audio quality. A solution using HQP and/or Roon would be preferable (I think).





I recommend state of the art DSP for room correction. I can run 12 channels of DXD (24/352.8) at native rate with room correction, no problem. 


Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
23 minutes ago, catastrofe said:



Thanks Kal and Chris.


Kal, you make a good point regarding the value of room correction superseding the perceived disadvantage of downsampling. What hardware supports Dirac Live at 24/192? 


Chris, similar question...how are you running DXD with DSP at its native rate? HQPlayer?


Thanks gents!

I’ve done with HQPlayer, but lately with Hang Loose Convolver inside JRiver. It works perfect. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
21 minutes ago, Magnificat said:

i am new here and do not know why DLBC and DLART are not mentioned often. Also puzzling is the widespread use of complex setups with PCs, multi channel DACs, DIY room correction rather than out of the box solutions like AV receivers and processors, media players e.g. Zidoo, disk players e.g. Magnetar and room correction from Dirac and Anthem.

As you can see in the article, the solutions you mention aren’t near state of the art. They are very limited. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment
27 minutes ago, Magnificat said:

Interesting that you do not rate Storm Audio.


As I mentioned in my initial post, I have a humble setup, with standard components.  It does seem that I am out of step with many people here who are chasing state of the art bespoke Atmos setups.  

I wouldn’t say out of step. We all make choices for what works best in our own situations. 

Founder of Audiophile Style | My Audio Systems AudiophileStyleStickerWhite2.0.png AudiophileStyleStickerWhite7.1.4.png

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Create New...