[GU1DO]

There are people in this thread that have experienced the demo, or own a Realiser A8, and will vouch for how well the Realiser works. Of course in theory you can never be sure of someone's objectivity but if you read in this thread I hope you get the feeling you can trust at least some of the people here. If you google for reviews of the A8 you will find a few and what is said in those will be a good indication for what the A16 is capable of.
I have expierenced a demo myself in Munich Germany last May, and I assure you I have no affiliation with Smyth, and nothing to gain by lying or exaggerating about this, this is what I experienced:
It is really hard to tell the difference between the real speakers and the simulated speakers. Initially when I first heared the sound over headphones I had to ask if I was really hearing the headphones and not the speakers. Like the first guy in the video of Munich 2016. In fact I could tell the difference at some moment, but only because there was a slight level mismatch (the simulation over headphones was a tiny bit louder or softer - I don't remember which of the two, and I think there was a ever so slight difference in sound colour but maybe I imagined that - possibly as a result of the small volume difference. I think this was caused by the fact that the headphone during the HPEQ measurement was a little bit out of the optimal position on my head, and hence in a different position than during the rest of the demo. This year they used a Stax instead of the Sennheisser they used in 2016 in Munich, and this Stax tended to sit a little bit to low on my head. Luckily this doesn't affect the PRIR measurement that was made, the HPEQ I have to do again at home anyway with my own headphones.) But the feeling of hearing speakers was 100% convincing. It is a strange feeling: it is as if the headphones on your head have nothing to do with the sound you hear, it feels like the sound is coming from outside, from the speakers. I have never experienced something like this before. (I have listened to samples of DTS Headphone X for example: there you sometimes think a sound comes a little more from the front of your head, or a little bit more from the back, nothing more, it stil is practically in your head or a little bit around, and could never for a millisecond be mistaken for sound coming from speakers or other sources far outside of your head.) After the measurement they first one by one let each of the speakers play alone, and let you compare the sound of that speaker with the sound over the headphones. Over the headphones the sound just seemed to stil be coming out of that speaker, and seemed to have nothing to do with the headphones on my head.
Like Fox1977 I also witnessed other people getting the demo so I witnessed that indeed the speakers were not playing when they put on the headphones. Another guy waiting to get the demo was present in the room during my demo, I thought I had noticed a Dutch accent so I asked him in Dutch (I'm Dutch also): "They are not fooling me are they? The speakers are really quiet when I put the headphones on?" and he confirmed that.

i am drooling over your demo description
i hope it meet my expectations specially after a very detailed post like yours,
thanks for sharing your experience.

 

[sahmen]

There are people in this thread that have experienced the demo, or own a Realiser A8, and will vouch for how well the Realiser works. Of course in theory you can never be sure of someone's objectivity but if you read in this thread I hope you get the feeling you can trust at least some of the people here. If you google for reviews of the A8 you will find a few and what is said in those will be a good indication for what the A16 is capable of.
I have expierenced a demo myself in Munich Germany last May, and I assure you I have no affiliation with Smyth, and nothing to gain by lying or exaggerating about this, this is what I experienced:
It is really hard to tell the difference between the real speakers and the simulated speakers. Initially when I first heared the sound over headphones I had to ask if I was really hearing the headphones and not the speakers. Like the first guy in the video of Munich 2016. In fact I could tell the difference at some moment, but only because there was a slight level mismatch (the simulation over headphones was a tiny bit louder or softer - I don't remember which of the two, and I think there was a ever so slight difference in sound colour but maybe I imagined that - possibly as a result of the small volume difference. I think this was caused by the fact that the headphone during the HPEQ measurement was a little bit out of the optimal position on my head, and hence in a different position than during the rest of the demo. This year they used a Stax instead of the Sennheisser they used in 2016 in Munich, and this Stax tended to sit a little bit to low on my head. Luckily this doesn't affect the PRIR measurement that was made, the HPEQ I have to do again at home anyway with my own headphones.) But the feeling of hearing speakers was 100% convincing. It is a strange feeling: it is as if the headphones on your head have nothing to do with the sound you hear, it feels like the sound is coming from outside, from the speakers. I have never experienced something like this before. (I have listened to samples of DTS Headphone X for example: there you sometimes think a sound comes a little more from the front of your head, or a little bit more from the back, nothing more, it still is practically in your head or a little bit around, and could never for a millisecond be mistaken for sound coming from speakers or other sources far outside of your head.) After the measurement they first one by one let each of the speakers play alone, and let you compare the sound of that speaker with the sound over the headphones. Over the headphones the sound just seemed to stil be coming out of that speaker, and seemed to have nothing to do with the headphones on my head.
Like Fox1977 I also witnessed other people getting the demo so I witnessed that indeed the speakers were not playing when they put on the headphones. Another guy waiting to get the demo was present in the room during my demo, I thought I had noticed a Dutch accent so I asked him in Dutch (I'm Dutch also): "They are not fooling me are they? The speakers are really quiet when I put the headphones on?" and he confirmed that.

Thanks for the detailed response.

People on this forum, just like me, are not sent by the Smyth brothers to advertise about the A16. I ordered the A16 on Kickstarter and yet i am what you call an "independent enthusiast". When i ordered through Kickstarter on summer 2016, i hadn't tried it yet but i knew that anyway, even if i was eventually disappointed by the device, i could easily resell it once it's released without losing a cent since the price on Kickstarter was very low compared to retail price. My enthusiasm came two months from the demo i had in Paris.
In a hotel room, an Atmos 7.4.1 setup was installed. They made a capture for me (it's very easy, you just put the microphones correctly in your ears and launch the program ; it takes less than 5 minutes). After the headphones was calibrated too, they adjusted the volume of the headphones to match the actual speakers, and then they launched the A/B comparison. I am enthusiast because i simply couldn't tell if i was listening to the speakers or the headphones. When i was taking headphones off, a slight click was letting me know that the speakers were switched back on, but when i was putting the headphones back on, there was no click at all and i could not tell that i was listening to headphones. Yet i could still exactly pinpoint each speaker.
When i was done, my nephew got also the demo, and since i was still in the room, i could then witness that the speakers were really turning off.
Many of the other enthusiasts here got a similar experience, and that's precisely why we are very excited about getting our own A16.
One more precision : you say that you prefer your music in 2.1 mode. It's not a problem, if you feed a 7.4.1 PRIR with stereo music, only the 2 virtual front speakers will play it (i think there will be also a feature that will allow to play it on all the speakers to get surrounded by the music)

Thanks for the helpful input. By the way, I was not implying any insinuations or aspersions against the Smyth Brothers or their business practices by asking for the impressions of "independent enthusiasts." I actually recognize and respect their rights to promote their product by posting favorable and supportive impressions and reviews of it on their site. Almost every business does that, and the Smyth brothers would be selling themselves short if they didn't do the same.... However, because most businesses with a hot new product do take the same approach, I find "independent," third party reviews and testimonies to be invaluable and vital, as a way of validating or serving as evidence to corroborate what the manufacturers themselves or their own handpicked reviewers claim about their product. I would think that most buyers (especially early adopters) take this same approach, and that most, if not all, manufacturers bringing a new product (such as the A16) to the market for the first time would even anticipate this, so I did not expect my insistence on "independent enthusiast" testimonies to ruffle any feathers.

yeah I was thinking the same thing, most of the testimonies on the kickstarter page are members we know who aren't affiliated in any way with Smyth. and I've seen them posting stuff very much in phase with the quotes in this very forum right after coming back from whatever show they were at. I think they just really got impressed by the demo.

Thanks.

Okay guys, I'm sold on the unit. I just pulled the trigger (paid the "pre-sale deposit", and it is all your fault <<==That is what I shall tell my wife's divorce lawyer if he should comes calling after today

Seriously though, I am grateful and am looking forward to the experience... I shall opt for the 2U "stealth" version, which I can slip into my rack very easily, and under the radar, without raising any red flags with the spouse. The other version (the prototype) looks too much like a medieval helmet or weapon, and hence too attention-grabbing to suit my "covert ops" approach .

By the way, what is supposed to happen after the "pre-sale deposit" is paid? I have a paypal receipt, but no response from Smyth Realiser... I need further instructions from them, and also I have to let them know that I am opting for the 2U version.

Thanks again for you help, guys

 

[sander99]

By the way, what is supposed to happen after the "pre-sale deposit" is paid? I have a paypal receipt, but no response from Smyth Realiser... I need further instructions from them, and also I have to let them know that I am opting for the 2U version.

When I payed my deposit, I afterwards sent a mail to sales@smyth-research.com (and added cc james@smyth-research.com to be on the save side) in which I explained that I had payed the deposit and asked to tell me what I stil owed them. I also asked to include a second head-tracker (that was $65 extra excluding taxes for me, I am not sure the price will be the same now). I would advice everybody to take a second head-tracker, the device can do the simulation for two people each with their own PRIR, HPEQ and headtracking, but standard comes with only one head-tracker. I think it would be silly to limit one of the two possible users to non-head-tracked operation only to save 65 dollars after spending so much money on the device.

 

[phoenixdogfan]

Okay, I have been reading much about the A16 lately, and am tempted to pull the trigger, taking advantage of the pre-order sale price before it expires on September 30th. I am only hesitating, because the operation and management of the unit (especially the PRIR/BRIR calibration processes) appear to be too complex sometimes, and I also do not know how much additional costs the related online operations will add up to...

So far the appeal for me of the unit lies in the access that it will provide to immersive object based audio formats (ATMOS, DTS-X, Auro-3d) via headphones... I would ideally like to have these formats operating in live "speaker-based" implementations, but that is not going to be possible for a while because of where I currently live, so the Smyth realiser solution seems the only viable option I have for now, if I want to go the 3d/object based audio route.

Now while I consider my options, could someone point me to a place where I can read actual impressions of independent enthusiasts who have actually demoed the A16 (I want "independent" impressions, as in those that are not listed by Smyth Realiser themselves either on their own website or on the Kickstarter page.

Could someone give me a layman's rundown on how difficult or expensive getting and implementing PRIR and BRIR calibrations might turn out to be for someone who cannot visit an actual professional studio, such as AIX, in order to get calibration profiles done? I ask because I am pretty much going to have to do everything by myself using my own speakers and measurement tools, because I am not going to have access to any pro studios and I need a realistic assessment of how manageable such tasks are going to be for me, as a layman..

Also, can this unit be exploited for speaker audio in a situation where one only has a pre-pro or avr that cannot handle Atmos/DTS-X/Auro 3d object based formats? (I do have the Oppo 203 which can handle blu ray discs with ATMOS/ DTS-X codecs, in case that helps)

Finally, how useful is this unit going to be for enhancing the quality of music that one listens to? This may not be terribly important, as I basically prefer my music in 2.1/2-channel mode, in which regard I consider myself to be decently well covered (for both headphone/speaker listening). I am asking only because I have not seen or read much literature about what possible enhancements the Realiser can bring to music listening, and I am willing to learn about whatever is out there on this front.

Any helpful thoughts would be highly appreciated.

So far the appeal for me of the unit lies in the access that it will provide to immersive object based audio formats (ATMOS, DTS-X, Auro-3d) via headphones... I would ideally like to have these formats operating in live "speaker-based" implementations, but that is not going to be possible for a while because of where I currently live, so the Smyth realiser solution seems the only viable option I have for now, if I want to go the 3d/object based audio route.

Think you might want to read the discussion from a few pages back. Since the A16 has 16 analog line outs. it should not only be able to function as a process/emulator in these advanced formats for headphones, but also for speakers as well. True, you would not employ your PRIRs or BRIRs when using with your own speakers, but it should be able to input a 16 channel digital feed and have the unit's ATMOS decoder do its stuff and output a 16 channel ATMOS feed for your amps and speakers if that's what suits your fancy.

 

[sahmen]

So far the appeal for me of the unit lies in the access that it will provide to immersive object based audio formats (ATMOS, DTS-X, Auro-3d) via headphones... I would ideally like to have these formats operating in live "speaker-based" implementations, but that is not going to be possible for a while because of where I currently live, so the Smyth realiser solution seems the only viable option I have for now, if I want to go the 3d/object based audio route.

Think you might want to read the discussion from a few pages back. Since the A16 has 16 analog line outs. it should not only be able to function as a process/emulator in these advanced formats for headphones, but also for speakers as well. True, you would not employ your PRIRs or BRIRs when using with your own speakers, but it should be able to input a 16 channel digital feed and have the unit's ATMOS decoder do its stuff and output a 16 channel ATMOS feed for your amps and speakers if that's what suits your fancy.

Nice. I agree that I have a lot of reading to catch up on.

 

[sahmen]

When I payed my deposit, I afterwards sent a mail to sales@smyth-research.com (and added cc james@smyth-research.com to be on the save side) in which I explained that I had payed the deposit and asked to tell me what I stil owed them. I also asked to include a second head-tracker (that was $65 extra excluding taxes for me, I am not sure the price will be the same now). I would advice everybody to take a second head-tracker, the device can do the simulation for two people each with their own PRIR, HPEQ and headtracking, but standard comes with only one head-tracker. I think it would be silly to limit one of the two possible users to non-head-tracked operation only to save 65 dollars after spending so much money on the device.

Thanks for these tips. The second head-tracker suggestion is really helpful. I have already sent one e-mail to the sales@smyth.. address.. I shall send them another one asking for the extra head-tracker to be included in the remaining balance, and I shall add james@smyth... to the list of addressees.

 

[raband]

Yep - Ordered and paid for the 2nd tracker when I got the balance due email.

 

[Richter Di]

Does anyone of you have some insights how the workflow of the production of the A16 is organized. Somehow I got the impression that all parts are shipped to the Smyth brothers and they assemble them themselfs.

 

[Fox1977]

Does anyone of you have some insights how the workflow of the production of the A16 is organized. Somehow I got the impression that all parts are shipped to the Smyth brothers and they assemble them themselfs.

That's also my understanding. In their communications, they seem to imply that they will assemble all the units themselves, test them and set them up (with the PRIR made during the exhibitions for example) before shipping them.

 

[sahmen]

Here's noob question : Does the A16 have any in-built amps (like receivers do), or do they rely on external amps to power the 16 virtual speaker channels the simulate? (I am a bit embarrassed to admit that I can't find any information about that in the literature I have read. I know it is supposed to have an on-board DAC, but I do not understand how it is going to manage or handle its amplification needs). I do have normal two channel (even balanced) head amps that I use with my headphones, and I am assuming that they should be okay to handle any signals that are fed them by the A16... I guess the confusion arises when I compare the way 3d multichannel/object based audio is handled by Pre pros and AVRS in which case, each speaker channel has its own individual channel of amplification, and the A16 which can simulate the action of 16 separate speakers---in which case, I am not getting how my regular two channel headamps are going to support all the 16 virtual speaker channels that the A16 can allegedly simulate...

Can someone kindly explain how that works (facepalm from asking what could be a potentially embarrassing question)

 

[VandyMan]

The A-16 has two built-in headphone amps. You can also use the analog output to feed your headphone amp. Or, you can use the digital output to feed a DAC and headphone amp. The output is two channels, not 16. Only two channels of amplification are required.

 

[sander99]

Here's noob question : Does the A16 have any in-built amps (like receivers do), or do they rely on external amps to power the 16 virtual speaker channels the simulate? (I am a bit embarrassed to admit that I can't find any information about that in the literature I have read. I know it is supposed to have an on-board DAC, but I do not understand how it is going to manage or handle its amplification needs). I do have normal two channel (even balanced) head amps that I use with my headphones, and I am assuming that they should be okay to handle any signals that are fed them by the A16... I guess the confusion arises when I compare the way 3d multichannel/object based audio is handled by Pre pros and AVRS in which case, each speaker channel has its own individual channel of amplification, and the A16 which can simulate the action of 16 separate speakers---in which case, I am not getting how my regular two channel headamps are going to support all the 16 virtual speaker channels that the A16 can allegedly simulate...

Can someone kindly explain how that works (facepalm from asking what could be a potentially embarrassing question)

For use with the headphone VandyMan is correct.

For use with real speakers you will have to need external (power-) amplification. There are analog outputs for all 16 channels (in the form of 8 stereo 3.5mm mini jack connectors).
[Edit: NOT TRUE:] Also up to 8 (or 7.1) channels can be assigned to be output as multichannel pcm over the HDMI out.

I am going to use 2 7.1 receivers, [Edit: NOT POSSIBLE:] my latest with HDMI input supporting 7.1 channel pcm, and my previous reveiver that I never sold (lucky coincidence) that has no HDMI but 7.1 analog inputs. And maybe get an old stereo amp to have all 15 or 16 channels.

Instead it is also possible to connect power amplifiers (or active speakers!).

Btw: note that the Realiser will not be able to do for example full 9.1.6 Dolby atmos, but as far as they know now at most 7.1.6 or 9.1.2 (earlier Smyth even thought it would be 7.1.4 maximum). However with the illusoic upmixer it can be further upmixed to 9.1.6 (but if that would give a usefull result or would be somehow unnatural I don't know).


I added some more:
Sahmen, I was a little bit to quick with my last post I did not read the question good enough.
I now see that you seem not to have understood the whole underlying principle of the device.
You have only 2 ears, yet you can hear where sound is coming from (within certain limitations of course). The brain uses all kind of information to determine that, and I could mention them all here but actually you don't even need to know exactly what it is. All that needs to be done to recreate the same experience as listening to 16 speakers in a room is making sure that the exact same signals that enter your ears when you listen to real speakers will enter your ears when listening to NORMAL STEREO headphones. That is what the realiser does. It measures the total net result that enters your ears as a result of the testsweeps being played over the speakers (per speaker per ear, so in fact 2x16 "transfer functions"), including all colourations from bending around your head and into your ears, including all reflexions and reverberations and their colourations from bending etc. etc.
With the result of this measurement (the PRIR) the Realiser can later apply the same changes to any source signal (music or whatever in any of the supported formats from mono to multichannel) you feed to it, creating the sensation of hearing the same source signal being played over the speakers.

Digital Signal Processing creates a stereo signal intended for the headphone based on up to 16 input signals and 32 channel to ear transferfunctions, a abstract description of what it calculates (without taking the headtracking into account) could be:

LeftOutput = TransferfunctionSpeaker1toLeftEar(InputSignalChannel1)
+ TransferfunctionSpeaker2toLeftEar(InputSignalChannel2)
...
+ TransferfunctionSpeaker16toLeftEar(InputSignalChannel16)

RightOutput = TransferfunctionSpeaker1toRightEar(InputSignalChannel1)
+ TransferfunctionSpeaker2toRightEar(InputSignalChannel2)
...
+ TransferfunctionSpeaker16toRightEar(InputSignalChannel16)

[Edit: October 11, 2017 have received information from Smyth: unfortunately the HDMI board that will be used in the Realiser A16 supports only 2 channels of LPCM over the HDMI out, see my post on October 10, 2017. The previous board did support 8 channels but not HDR. The next board (in future versions) does support 8 channels and HDR. That's why I am making some corrections to statements that are based on the original Smyth plans that are no longer possible.]

 

[sahmen]

The A-16 has two built-in headphone amps. You can also use the analog output to feed your headphone amp. Or, you can use the digital output to feed a DAC and headphone amp. The output is two channels, not 16. Only two channels of amplification are required.

I realize the my question was probably not framed clearly enough. This is what I meant: because I am going to use the A16 with headphones, I know that only two channels of amplication will be required. So my question was more about the 16 channels of **simulated** speakers one is supposed to hear through one's headphones thanks to those two channels of amplification... If you like, how does the A16 enable those 2 amplified channels on your ears to create the illusion of up to 16 virtual separate "amplified" channels that resemble those you would hear were you actually listening to 16 actual speakers in a 3d immersive system, without headphones...

I hope that has made my question clearer...

For use with the headphone VandyMan is correct.

For use with real speakers you will have to need external (power-) amplification. There are analog outputs for all 16 channels (in the form of 8 stereo 3.5mm mini jack connectors). Also up to 8 (or 7.1) channels can be assigned to be output as multichannel pcm over the HDMI out.

I am going to use 2 7.1 receivers, my latest with HDMI input supporting 7.1 channel pcm, and my previous reveiver that I never sold (lucky coincidence) that has no HDMI but 7.1 analog inputs. And maybe get an old stereo amp to have all 15 or 16 channels.

Instead it is also possible to connect power amplifiers (or active speakers!).

Btw: note that the Realiser will not be able to do for example full 9.1.6 Dolby atmos, but as far as they know now at most 7.1.6 or 9.1.2 (earlier Smyth even thought it would be 7.1.4 maximum). However with the illusoic upmixer it can be further upmixed to 9.1.6 (but if that would give a usefull result or would be somehow unnatural I don't know).


I added some more:
Sahmen, I was a little bit to quick with my last post I did not read the question good enough.
I now see that you seem not to have understood the whole underlying principle of the device.
You have only 2 ears, yet you can hear where sound is coming from (within certain limitation of course). The brain uses all kind of information to determine that, and I could mention them all here but actually you don't even need to know exactly what it is. All that needs to be done to recreate the same experience as listening to 16 speakers in a room is making sure that the exact same signals that enter your ear when you listen to real speakers will enter your ear when listening to NORMAL STEREO headphones. That is what the realiser does. It measures the total net result that enters your ears as a result of the testsweeps being played over the speakers (per speaker per ear, so in fact 2x16 "transfer functions"), including all colourations from bending around your head and into your ears, including all reflexions and reverberations and their colourations from bending etc. etc.
With the result of this measurement (the PRIR) the Realiser can later apply the same changes to any source signal (music or whatever in any of the supported formats from mono to multichannel) you feed to it, creating the sensation of hearing the same source signal being played over the speakers.

Digital Signal Processing creates a stereo signal intended for the headphone based on up to 16 input signals and 32 channel to ear transferfunctions, a abstract description of what it calculates (without taking the headtracking into account) could be:

LeftOutput = TranferfunctionSpeaker1toLeftEar(InputSignalChannel1)
+ TranferfunctionSpeaker2toLeftEar(InputSignalChannel2)
...
+ TranferfunctionSpeaker16toLeftEar(InputSignalChannel16)

RightOutput = TranferfunctionSpeaker1toRightEar(InputSignalChannel1)
+ TranferfunctionSpeaker2toRightEar(InputSignalChannel2)
...
+ TranferfunctionSpeaker16toRightEar(InputSignalChannel16)

Sander99 : you're correct that I really don't/didn' understand the underlying principle governing the way the A16 works, but the second part of your answer has been very helpful . I am understanding that the A16 uses the PRIR measurements to create some magic that virtually replicates the sound of the appropriately positioned live speakers whose sounds have been measured in the room or setting, and, then when the headphones are positioned on my head, it tricks my brain into feeling as if I am actually listening to those live speakers rather than headphones... If that is the basic mode of operation, then I get it in that basic form. A lot of the intricate technicalities involved in that "magic" of calibration and virtualization still escape me, but the basic clarification should do.

Thanks a lot.

 

[VandyMan]

I realize the my question was probably not framed clearly enough. This is what I meant: because I am going to use the A16 with headphones, I know that only two channels of amplication will be required. So my question was more about the 16 channels of **simulated** speakers one is supposed to hear through one's headphones thanks to those two channels of amplification... If you like, how does the A16 enable those 2 amplified channels on your ears to create the illusion of up to 16 virtual separate "amplified" channels that resemble those you would hear were you actually listening to 16 actual speakers in a 3d immersive system, without headphones...

I hope that has made my question clearer...

Here is a non-technical way to think about it. You have two ears, yet you hear sounds all around you. So, even though hear in two channel (an over-simplification), your brain is able to perceive where sounds come from. This is called Sound Localization. Wikipedia describes our brain's ability to localize sound pretty well: "The auditory system uses several cues for sound source localization, including time- and level-differences (or intensity-difference) between both ears, spectral information, timing analysis, correlation analysis, and pattern matching." What the A16 is doing is converting multi-channel to two channel in such a way as to preserve/enhance this information that your brain uses.

[Edit: Just noticed that sander99 said pretty much the same thing above, but I'll leave this incase it is helpful.]

 

[castleofargh]

I want to try too. having a lifetime experience of being a noob, maybe I'll somehow make some sense. ^_^

when a sound comes from a given direction, let's say 30°left, it reaches the left ear first and then the other ear with a given delay(depending on the size of the listener's head and distance between the ears). some frequencies being blocked by the listener's face, a given attenuation and signature will enter the right ear. that change will be specific to the position of the sound source and the head of the listener. in real life we can tell the general direction of a sound source thanks mainly to delays and signature changes/attenuation between the 2 ears(ILD and ITD cues, Loudness and Time).
we use some improved triangulation between the source and both ears. and what our brain analyzes anytime something makes a noise, is a stereo signal(one per ear) for what really was one sound source/one channel.
when you calibrate the A16, you will look at a specific positions and the A16 will analyze the sound reaching your two ears for that sound source and that position. then you'll calibrate using the headphone, and the A16 will use all that information to simulate and send into the headphone a stereo signal the ears should normally receive for that speaker at that given position. it's simplified a good deal but that's really the main idea.

now no matter how many speakers you have in your system, you will make a calibration for each one of them. so when you play some multichannel music on your headphone, the channel that was normally sent to the left speaker will be altered by the A16 based on the left speaker calibration. the channel usually going into the right speaker will be altered based on the right speaker calibration, etc. each channel will be treated independently and turned into an independent stereo signal. only when getting out of the A16 will they all end up sent into the same headphone.
in case the idea of mixing many sounds down to a single stereo signal isn't intuitive, I'll simply remind that we only have 2 ears and yet we do enjoy a band playing. and we can close our eyes and tell the general direction of each instrument and recognize said instruments. there is no issue with having several sounds mixed down to stereo because that's what our ears do in the end anyway. (and the signal doesn't mind because of wave magic. which might deserve a specific topic about sound waves in general. )

and of course what I explained here is only about position cues, in reality the signature of each speaker, the reverb from the room and even more are also captured and replicated to some level by the A16. which should make for a pretty good result as long as the headphone used isn't total low- fidelity garbage.