[johnn29]

How about they just release a 5.1 a 7.1, a stereo, a 16 channel atmos, and 8 or 12 pairs of speakers which constitute the 16-24 channel lay out. If necessary someone could do 12 separate PRIRs and combine them into a room, or 8 pairs or 9, or 4 or 2 or whatever strikes their fancy. Use the naming system which is common to Atmos, DTS-X, and Auro 3D. There is one if you look in the manual, so you don't need three different naming schema to make virtual speakers which will work in your format of choice (it could even be done as 24 separate single channels). And this will greatly facilitate anyone with two speakers, and sturdy, height adjustable speaker stands in their effort to get a good PRIR. I could do 24 channels tomorrow with my LS-50s if I had the Realiser back. (It made it to Smyth yesterday, btw.)

It's not hard to code something that'll encode a sweep into a chosen channel layout! Impulcifer does it, and it's free. It's limit to only bed channels of course, because Windows has no channel output support of Atmos.

I believe what the hold up might be is that the Atmos encoder is costly. I've spoken to another acoustic consultant who would like to run sweeps for height channels - but he looked it up and the software licensing was cost prohibitive for his business. Or it could be Smyth are relying on revenue from measurements in pro-studios where they've reserved the sweep files for.

That's the major thing that really puts me off getting one. The most frustrating thing is the measurement process - a lot can go wrong. I want to be able to just sit my butt down in the sweet spot and record all at once.

 

[phoenixdogfan]

It's not hard to code something that'll encode a sweep into a chosen channel layout! Impulcifer does it, and it's free. It's limit to only bed channels of course, because Windows has no channel output support of Atmos.

I believe what the hold up might be is that the Atmos encoder is costly. I've spoken to another acoustic consultant who would like to run sweeps for height channels - but he looked it up and the software licensing was cost prohibitive for his business. Or it could be Smyth are relying on revenue from measurements in pro-studios where they've reserved the sweep files for.

That's the major thing that really puts me off getting one. The most frustrating thing is the measurement process - a lot can go wrong. I want to be able to just sit my butt down in the sweet spot and record all at once.

" could be Smyth are relying on revenue from measurements in pro-studios where they've reserved the sweep files for "

Are you saying they've already created these files, and won't release them to their customers b/c they want those customers to go to the studios like the one in Germany and pay 1200Eu a crack to get a PRIR with the sweep files they gave the studio and for which they are getting a kickback? Do you know that for a fact? How do you KNOW that?

And do the height channels have to go through Windows. Somehow people are using their windows Pc to reproduce atmos with Netflix streams, are you telling me that those streams don't do the height channels? Can;t the Windows mixer be entirely bypassed by using an Asio driver?

 

[Litlgi74]

It's not hard to code something that'll encode a sweep into a chosen channel layout! Impulcifer does it, and it's free. It's limit to only bed channels of course, because Windows has no channel output support of Atmos.

I believe what the hold up might be is that the Atmos encoder is costly. I've spoken to another acoustic consultant who would like to run sweeps for height channels - but he looked it up and the software licensing was cost prohibitive for his business. Or it could be Smyth are relying on revenue from measurements in pro-studios where they've reserved the sweep files for.

That's the major thing that really puts me off getting one. The most frustrating thing is the measurement process - a lot can go wrong. I want to be able to just sit my butt down in the sweet spot and record all at once.

I'm sorry... Am I missing something? I thought the sweeps were somewhat automated? At least that's what I've heard in some of the setup videos... But then again that was only for the LCR channels.

 

[Dr. Strangelove]

So much BS and nonsense in this thread. Out of all the conspiracy theories that have come up here over the years, the idea that Smyth is intentionally sabotaging their niche product's functionality is probably the one that makes my head hurt the most.

 

[phoenixdogfan]

So much BS and nonsense in this thread. Out of all the conspiracy theories that have come up here over the years, the idea that Smyth is intentionally sabotaging their niche product's functionality is probably the one that makes my head hurt the most.

I don't buy it either. If someone has the evidence, let them bring it forth. Otherwise, it's just b.s. But the Smyth's need to get the digital sweeps to the Realiser owners. This was one of the promised deliverables, and it's long overdue!

 

[NigelJ]

Back in the 11 March 2020 Kickstarter update Smyth Research stated (I have added red highlighting):

"6) Realiser Exchange​

We aim to launch a limited functionality Realiser Exchange website within the next 6-8 weeks provided our prelaunch testing concludes successfully. While the forum aspect of the Realiser Exchange should be running from launch, the PRIR store with shopping cart will not function initially except to allow members to post their own PRIRs for free share with other members. We also plan to host a PRIR merge function that will allow members to modify non-personalised PRIRs with data from their own measured files. Other functions such as building listening rooms and constructing Async measurement files are also in development. It is also our intention to distribute the PRIR files that were measured at Can Jam shows through the Exchange website as an incentive for those individuals to create an account."​

Obviously, as we have seen, their timeline estimates are extremely optimistic; however, this does indicate an intention to provide async measurement files to all owners. (The potential advantage of constructing async files rather than providing a limited set of pre-constructed files is that they can be matched to the specific set of speakers being used, rather than a limited number of default configurations.)

However I strongly agree with @phoenixdogfan that they are long overdue; additionally, timely (not certain if Smyth Research know the meaning of this word) access to the PRIR files measured at CamJam shows is important to those backers that had them done and do not have easy access to a suitable speaker setup.

 

[sander99]

I'm sorry... Am I missing something? I thought the sweeps were somewhat automated? At least that's what I've heard in some of the setup videos... But then again that was only for the LCR channels.

There are different methods to run the A16 measurements sweeps through the system that you want to measure. The "standard" way, that currently is the only one that is "officially" available to us, is that the A16 generates the sweeps itself and outputs them to the (up to) 16 analog outputs.
This poses practical problems for measuring a complete system in one go.
Because what we normally would need then is a system with analog inputs for all channels.
Some av receivers have multichannel analog inputs, but:
-most of them will bypass all dsp (room correction eq!) when using the analog inputs
-and virtually none have analog inputs for the height channels, only for the 7.1 part

However there are workarounds. And it is less of a problem if you don't need the room eq of the receiver.
Also it is not a problem at all if you use only one or two speakers (in a fixed position) to measure all the channels.

Then there is the so called asynchronous PRIR measurement method:
(That has been implemented a few months ago, but we don't yet have the special files that are needed for this method.)
In this method the A16 doesn't generate the sweeps, but a media file is played over the system that you want to measure.
The file contains the sweeps, plus some additional signals. The A16 can sync to the signals and do the measurement.
Such a file can only include the height channels if it is encoded as an atmos bitstream! And as I now understand from one of the previous posts that involves big licencing fees.)

(By the way: that would also be a problem for outputting all the sweeps via HDMI by the A16 itself. And 7.1 lpcm sweeps over HDMI out are also not possible due to technical limitations of the used HDMI board.)

However there is another possibility, I recently came across this post again:
[Edit: but still not including the height channels of course...]

Hey guys,
you don't have to wait for the asynchronous mode to be implemented (at least for up to 8 speakers)! Just record the measurement signal, save it as multichannel wav or flac and play it back on your streaming device or blu-ray player.
I've done this when I received my Realiser last month. If you have a legacy 5.1 system, you can use my recording:
https://www.dropbox.com/s/k42hj391gwq2w8c/SMYTH Realiser - 5.1 Longsweep 30deg.flac?dl=0
Just go to PRIR measurements, press SPK, play back the file and press the SPK-button when the playback hits the 3 second mark.

You must use the 12 sec non-overlapping routine with one look-angle at 30 deg, HT assist must be turned off. The LFE signal is -10dB, so you have to switch hp LFE +10dB on afterwards (Listening rooms / manage bass)

 

[sander99]

@SobiC: The link to that sweep file of yours doesn't seem to work anymore. Did you remove it?

By the way: How did you make this recording? Can you make recordings of 8 analog inputs also?

 

[stevoz]

Hi all. I'm a KS backer and received my unit a few months ago. Haven't used it yet because was waiting for the async functionality to come along.

I've recently changed my set up though and now my speakers are connected to a power amplifier taking input from the pre-outs on my receiver.

I'm guessing if I plug the Realiser in between the receiver and power amplifier using the analogue inputs/outputs, that I'll be able to run the sweeps direct from the Realiser?

If so, I was hoping those of you with knowledge of this sort of set up could perhaps help out with some questions I have.

Set up this way, with the Realiser in AV mode, will it essentially pass-thru the analogue signal from my receiver to the power amplifier untouched? Will it do this only when on or also in stand-by?

If pass-thru is possible, will my receiver still be able control the speaker volume or will it now need to be controlled by the Realiser? Or perhaps a combination of both?

Does it matter which numbered ins/outs I connect in the Realiser? Or can specific speakers be mapped individually to input/output connections?

If I'm measuring a PRIR from analogue sweeps, they will obviously not include any room correction as the speaker sweep is coming direct from the Realiser. However, is that strictly a bad thing in this sort of connectivity scenario?
My thinking is that the speaker measurement should be uncorrected so that when the Realiser is decoding content from the receiver via analogue input, any applied correction by the receiver at playback comes through as part of the surround signal. If room correction was used as part of the speaker measurement, and the corresponding playback was also room corrected by the receiver, then this would end up correcting twice, and may not sound that great.
Am I right in thinking that baking in room correction would only be important if the Realiser was the source of the decoding rather than taking a decoded analogue signal?

Thanks in advance for all your help!

 

[sander99]

@m8o: My previous posts hint at one last possible solution for your problem: instead of using the asynchronous measurement files just use a recording of the sweeps.
Like with the asyncronous measurement files you will need in-ear-microphones and a recording interface (or a digital pocket recorder with suitable mic inputs), and later we will have to find a way to input the signals into the mic inputs of the A16.
Even with the one 5.1 sweep file that we have now (I still have it on my computer) it can be done.
(For your 7.2.4 in 3 runs. The sweeps are all the same except the .1 sweep.)

If you would make the decision to buy in-ear-mics and a recording interface then I advice to do both the above and capture the system with Impulcifer (because then at least you can check the result directly).

But, I repeat here your request: if anyone (that lives on Long Island or the NY Tri-State area) with an A16 is willing to help m8o that would be soo much better!
(https://www.head-fi.org/threads/smyth-research-realiser-a16.807459/page-682#post-15869304)

 

[sander99]

I've recently changed my set up though and now my speakers are connected to a power amplifier taking input from the pre-outs on my receiver.

I'm guessing if I plug the Realiser in between the receiver and power amplifier using the analogue inputs/outputs, that I'll be able to run the sweeps direct from the Realiser?

Assuming you have a multichannel power amplifier (or several power amplifiers) for all your channels, then yes.

Set up this way, with the Realiser in AV mode, will it essentially pass-thru the analogue signal from my receiver to the power amplifier untouched? Will it do this only when on or also in stand-by?

Not in standby I fear. Only in AV mode and "line" input selected. You can also run your power amps using the A16 as decoder or preamp with other connected sources but then you won't have room correction.

If pass-thru is possible, will my receiver still be able control the speaker volume or will it now need to be controlled by the Realiser? Or perhaps a combination of both?

Both would be working. I would set the volume of your receiver at some high setting, but not max, something like -10 dB. Of course you should not have any speakers connected to the speaker outputs of your receiver.
(I have had a borrowed A16 in my house for a few weeks. I used a Yamaha RX-V771 receiver with the pre-outs connected to the analog inputs of the A16 to be able to decode DTS. At first I put it's volume at 0 dB but then one day the Yamaha switched off automatically at some loud part of a movie. Protection mechanism kicking in! - The movie was Blade Runner 2049, the scene where he lands on the protien farm.)

Does it matter which numbered ins/outs I connect in the Realiser? Or can specific speakers be mapped individually to input/output connections?

For PRIR measurements the outputs can be freely assigned to channels (determined by how you setup the PRIR Sound Rooms).
In AV Mode you can see in the listening rooms which speaker is mapped to what input/output number. What speaker names/labels can be used depends on the listening mode in the listening room, and whether it is an Atmos room, a DTS room or a PCM room. In PCM rooms - that you will have to use for analog line input signals - you can freely change the mapping of channels to input/output numbers I think, but I don't think you should want to do that.

If I'm measuring a PRIR from analogue sweeps, they will obviously not include any room correction as the speaker sweep is coming direct from the Realiser. However, is that strictly a bad thing in this sort of connectivity scenario?
My thinking is that the speaker measurement should be uncorrected so that when the Realiser is decoding content from the receiver via analogue input, any applied correction by the receiver at playback comes through as part of the surround signal. If room correction was used as part of the speaker measurement, and the corresponding playback was also room corrected by the receiver, then this would end up correcting twice, and may not sound that great.
Am I right in thinking that baking in room correction would only be important if the Realiser was the source of the decoding rather than taking a decoded analogue signal?

Correct.
Except one thing:
(Also relevant for @audiohobbit, I wanted to bring this up earlier but never did.)
The A16 will remove the initial delays when creating a PRIR. Normally the AV receiver or processor will delay several speakers such that simultanious sounds in different channels reach the listeners ear simultaniously. By simply removing all initial delays the A16 achieves the same goal. But when your AV receiver still has different delays set while feeding the A16 then simultanious sounds in the input channels will not reach the listeners ears simultaniously! So when using the A16 with virtual speakers over headphones you would have to set all your AVR's delays equal (zero or minimal)! While when throughputting your AVR's signal to the power amp you would need to have the normal delay settings! I suspect something similar for levels. The A16 will normalise the volume of the different speakers in one PRIR with eachother (I think).

 

[stevoz]

Assuming you have a multichannel power amplifier (or several power amplifiers) for all your channels, then yes.

Not in standby I fear. Only in AV mode and "line" input selected. You can also run your power amps using the A16 as decoder or preamp with other connected sources but then you won't have room correction.

Both would be working. I would set the volume of your receiver at some high setting, but not max, something like -10 dB. Of course you should not have any speakers connected to the speaker outputs of your receiver.
(I have had a borrowed A16 in my house for a few weeks. I used a Yamaha RX-V771 receiver with the pre-outs connected to the analog inputs of the A16 to be able to decode DTS. At first I put it's volume at 0 dB but then one day the Yamaha switched off automatically at some loud part of a movie. Protection mechanism kicking in! - The movie was Blade Runner 2049, the scene where he lands on the protien farm.)

For PRIR measurements the outputs can be freely assigned to channels (determined by how you setup the PRIR Sound Rooms).
In AV Mode you can see in the listening rooms which speaker is mapped to what input/output number. What speaker names/labels can be used depends on the listening mode in the listening room, and whether it is an Atmos room, a DTS room or a PCM room. In PCM rooms - that you will have to use for analog line input signals - you can freely change the mapping of channels to input/output numbers I think, but I don't think you should want to do that.

Correct.
Except one thing:
(Also relevant for @audiohobbit, I wanted to bring this up earlier but never did.)
The A16 will remove the initial delays when creating a PRIR. Normally the AV receiver or processor will delay several speakers such that simultanious sounds in different channels reach the listeners ear simultaniously. By simply removing all initial delays the A16 achieves the same goal. But when your AV receiver still has different delays set while feeding the A16 then simultanious sounds in the input channels will not reach the listeners ears simultaniously! So when using the A16 with virtual speakers over headphones you would have to set all your AVR's delays equal (zero or minimal)! While when throughputting your AVR's signal to the power amp you would need to have the normal delay settings! I suspect something similar for levels. The A16 will normalise the volume of the different speakers in one PRIR with eachother (I think).

Thanks @sander99! Makes sense.

Re the analogue inputs/outputs, its too bad there's no bypass mode in which the A16 doesn't bother processing the signal at all.

In regards to the room correction, I never thought of the fact that the speaker delays are no longer necessary once listening through headphones since the "virtual" speakers are no longer at different distances!

I'm using a Yamaha RX-A830 receiver. I wonder if I could just temporarily hook up the A16 to my power amp purely to measure a PRIR, then connect the power amp back to the receiver for speaker listening and hook up the HDMI output of my receiver to the A16, cutting the A16 out of the speaker path. I think the receiver will output decoded multi channel PCM if it senses a multi channel device connected to its HDMI out, albeit probably without any room correction.

 

[sander99]

I think the receiver will output decoded multi channel PCM if it senses a multi channel device connected to its HDMI out

I expect not.

I'm using a Yamaha RX-A830 receiver.

I discovered a little bit unexpected behaviour from my Yamaha RX-V771 receiver, maybe more Yamaha's do that, that can be "misused" to quickly swith between different delay settings:
With my Yamaha you can choose to set the distances in feet or in meters. When auto calibrating he speaker setup he sets them both. I discovered that when I manually change the distances in meters, he keeps the old settings in feet! So when you change all the settings in meters to one constant value then afterwards you can switch between those constant values or the other set of values by only toggling the meters/feet setting.

 

[audiohobbit]

@SobiC: The link to that sweep file of yours doesn't seem to work anymore. Did you remove it?

By the way: How did you make this recording? Can you make recordings of 8 analog inputs also?

This may work somehow (in Sync mode! NOT in Async mode, because the Async mode waits for the preamble DTMF tones and the Phase Locking Tone), however the Sync Mode has its name because the Realiser expects that the recorded sweeps and the played back sweeps are in sync (phase), like all measurement gear I know that works with sweeps (like Room EQ Wizard etc.) they all expect a closed loop. So I don't know what repercussions this will have on sound quality/impulse response of the recorded PRIR.
You cannot correctly sync this manually. So don't expect the result to be correct!

With the Async signals there is one long sine tone to sync the player to the Realiser down to a fraction of samples per second, See https://smyth-research.com/wp-conte...-rev-1.80-Dec-12-2019-update-instructions.pdf p. 9

@m8o: My previous posts hint at one last possible solution for your problem: instead of using the asynchronous measurement files just use a recording of the sweeps.
Like with the asyncronous measurement files you will need in-ear-microphones and a recording interface (or a digital pocket recorder with suitable mic inputs), and later we will have to find a way to input the signals into the mic inputs of the A16.

I don't think this is feasible in any way. He really needs s.o. with a Realiser.

(Also relevant for @audiohobbit, I wanted to bring this up earlier but never did.)
The A16 will remove the initial delays when creating a PRIR.

Thanks but I know this already aftere extensively asking Stephen.
But you're right I guess, when driving the Realiser with an AVR over the analogue inputs you have to set the delays all to the same value I think. The same distance for all speakers would suffice, it don#t have to be the minimum setting, afaik the AVRs only calculate the difference between speaker distances, if all speakers are set to the same distance then there will be no delay.

If you let the AVR decode and send PCM via HDMI (which will only do 7.1 if it even works) then there will be no room correction, I'm pretty sure about that.
For the room correction you'll have to use the analogue outputs.

What probably can be done is to measure 2 speakers at a time with the sync signals over a stereo input. With analogue stereo input the signal should be AD-converted and room correction should be applied. We did this last year, but it's a bit complicated though.
If you want to do this, approach me/us again here.



Now to the Async signals:

Are you saying they've already created these files, and won't release them to their customers b/c they want those customers to go to the studios like the one in Germany and pay 1200Eu a crack to get a PRIR with the sweep files they gave the studio and for which they are getting a kickback?

I agree, this is BS, sorry.
I have a little bit insight, but not about the cash flows. Even if they get a little money from the studios for using the Asnyc Signals: How many studios do you know that offer PRIR recording to customers?
One? Two with Gilles, but he don't really have studio afaik.

I know for fact that a few hundred Async sequences exist for studios. From 2.0 to 9.1.6, with different combinations of look angles, with and without vertical looks, and each combination of speakers and look angles for all types of sweeps: 4s with and without overlap, 12s with and without overlap.
All these combinations result in several hundred different sequences!
As the files are at the moment you need a DAW (Digital Audio Work Station) to play them and it wouldn't work with normal playing equipment.

When I look at the picture on p. 5 here https://smyth-research.com/wp-conte...-rev-1.80-Dec-12-2019-update-instructions.pdf
then I guess they want to create video files for different combinations which is a bit of work I think.
I will ask Stephen if they can upload a few files maybe without video, at least 2.0, 5.1, 7.1 with -30, 0, 30* and 4s without overlap ( I don't really hear a difference between 4 and 12s with my wooden ears... With overlap I heard a difference, some type of crossfeed between the channels. As far as I'm concerned I'd rather user 4s without overlap than 12s with overlap.).
*) I'd recommend going to 60 deg azimuth, so -60, -30, 0, 30, 60, but that will already be too many looks for some people I think (Stephen recommends finer increments across the center, so one sequence of the existing signals is -60, -30, -15, 0, 15, 30, 60. With my wooden ears I can't really hear a great differenec with and without 15 deg, but going to 60 is convenient in some situations)

Everything above 8 channels will have to be encoded in Atmos and I think this is complicated.
But will not be necessary, since the Override Preamble ID function https://smyth-research.com/wp-content/uploads/2020/05/A16-firmware-1.90-26-May-2020.pdf (p. 10).
So with a 7.1 signal you could measure 16 speakers in 2 passes, however you have to re-plug the speakers to the terminals o your AVR and reapply DSP/EQ to the reassigned channels. Will still be a bit complicated.

 

[sander99]

This may work somehow (in Sync mode! NOT in Async mode, because the Async mode waits for the preamble DTMF tones and the Phase Locking Tone), however the Sync Mode has its name because the Realiser expects that the recorded sweeps and the played back sweeps are in sync (phase), like all measurement gear I know that works with sweeps (like Room EQ Wizard etc.) they all expect a closed loop. So I don't know what repercussions this will have on sound quality/impulse response of the recorded PRIR.
You cannot correctly sync this manually. So don't expect the result to be correct!

With the Async signals there is one long sine tone to sync the player to the Realiser down to a fraction of samples per second, See https://smyth-research.com/wp-conte...-rev-1.80-Dec-12-2019-update-instructions.pdf p. 9

I know about the sync signals, and maybe they offer some benefits but I don't see how it could be absolutely necessary. When connecting the analog outputs of the A16 to for example only a stereo input of an AVR, so that the AVRs DSP including room correction are in the path, then: the analog signals first pass an ad conversion (and the AVR and the ad converter will run on their own clock, not related to the A16's clock, and that doesn't matter), then there will be the dsp processing, delays, delays of sound traveling through the air, all adding up to some unknown latency between the analog signals coming out of the A16 and the microphone signals coming back in. Yet that is no problem! (And all initial delays will not end up in the PRIR). I think for it to work at all only a approximate timing is enough (like @SobiC apparently demonstrated, unless he made it up). The thing with the sync signals is extra synchronisation at another, much finer, scale that apparently has some advantage, but I can not see how it could be essential because then the normal PRIR measurement using the analog outputs couldn't work either! Oh, and starting the playback of earlier recorded sweep responses shouldn't be any more difficult to time than starting the recorded sweeps for a "real time" PRIR measurement the way SobiC did.

But okay, I am not 100% certain of this. That is one more reason to advice: if buying mics, then try this and also capture everything with Impulcifer. Then he is betting on 2 horses so to speak.

[Edit: The primary reason for Smyth to add the sync signals would have been to relieve the user of any concern about it, maybe they just thought: when we do it anyway, let's do it very precise.]