[CanMad]

Looks like they're averaging 5 a week.

one per working day?

 

[teacher1000]

I have no idea why I've not had to wait in line, but I have notification that my (2U) unit will be arriving tomorrow
Kickstarter #46

 

[audiohobbit]

Did you get an email to confirm your address first?

 

[teacher1000]

Did you get an email to confirm your address first?

I got an email the same day they announced they were starting shipping, then nothing until shipping notification from the courier

 

[Fox1977]

I got an email the same day they announced they were starting shipping, then nothing until shipping notification from the courier

Someone got lucky !

 

[Richter Di]

I have no idea why I've not had to wait in line, but I have notification that my (2U) unit will be arriving tomorrow
Kickstarter #46

Congratulations. Somehow this is giving hope.

 

[CanMad]

Bit random. I got contacted later. Mine arrives today hopefully. Working from home waiting expectantly.

 

[CanMad]

It's arrived. Probably have time to play with it on Sunday lol

 

[bpinnell]

Longtime A8 owner patiently waiting for my A16 shipment notification. Apologies if this was mentioned earlier in this thread, but is the processing done by the A16 to create the PRIR still 48kHz? I know it can accept higher bit rates similar to the A8, but the A8 then converted that input to 48kHz when processing. I know the A16 has additional spatial capabilities, but assuming one is just making a two-channel PRIR, would the A16 create a higher quality PRIR than the A8 in the same conditions? Thanks!

 

[phoenixdogfan]

Longtime A8 owner patiently waiting for my A16 shipment notification. Apologies if this was mentioned earlier in this thread, but is the processing done by the A16 to create the PRIR still 48kHz? I know it can accept higher bit rates similar to the A8, but the A8 then converted that input to 48kHz when processing. I know the A16 has additional spatial capabilities, but assuming one is just making a two-channel PRIR, would the A16 create a higher quality PRIR than the A8 in the same conditions? Thanks!

I would imagine even if it could, the SINAD ratio of the binaural earphone would be the limiting factor on PRIR quality. Most of what I am reading leads me to believe that the binaural mics SINAD is l.t. 80db.

 

[Dixter]

one question for those with an A8 and now with your new A16... how do the two compare sound wise... just wondering if there is going to be differences in the quality of the sound...

 

[Sanctuary]

Probably just going to end up getting something like this I guess:

This looks almost the same as the Denon I was previously looking at, yet this has 7.1 in as well.

 

[Got the Shakes]

Probably just going to end up getting something like this I guess:

This looks almost the same as the Denon I was previously looking at, yet this has 7.1 in as well.

I was originally looking at buying something like that before I decided to take my measurements a speaker at a time. If you’re literally looking to buy something just for the purposes for analog inputs and wouldn’t be using it for actual home theater use, you can find something used on eBay for like $60 shipped.

 

[sander99]

Longtime A8 owner patiently waiting for my A16 shipment notification. Apologies if this was mentioned earlier in this thread, but is the processing done by the A16 to create the PRIR still 48kHz? I know it can accept higher bit rates similar to the A8, but the A8 then converted that input to 48kHz when processing. I know the A16 has additional spatial capabilities, but assuming one is just making a two-channel PRIR, would the A16 create a higher quality PRIR than the A8 in the same conditions? Thanks!

With the current firmware version the A16 does the processing in 48 kHz, 32 bits. Originally Smyth had plans for various "operating modes" including some with processing at 96 kHz (1 user, 16 channels) and 192 kHz (1 user, max 4 channels). Will they ever implement those? We don't know. (Now I have my own ideas about the relevance of higher sampling frequencies, but I won't start a discussion about that here, better go to the sound science forum for that.)
Processing in 32 bits is a good thing in any case, to keep the influence of accumulated rounding errors in the many many calculation steps that the A16 performs far below audible levels. (The end result is rounded back to 24 bits before output.)
Anyway, I think real improvements can be made elsewhere: although I don't think the noise of the microphones is a real problem (see below), I do think that the size and placement of the microphones are the first "bottleneck", and next maybe the quality/noise of the microphones. In VDT Magazine 2-2015 Mike Smyth writes something that I already was thinking myself: "Ideally the measurement would occur at the ear drum of the listener rather than at the entrance to the ear canal, but this is not practical as a DIY technique." And of course ideally, you would want to measure at the ear drum without microphones and cables sticking in the ear canals that would influence the measurement, so how to do that is a good question. Maybe similar to those spy laser eavesdropping things that can "see" vibrations on a far away glass window but now detecting vibrations of the eardrum itself? Or make a 1:1 scale physical model of your head (and torso), from materials that share the same acoustic properties as your own body, with high quality microphones build inside with the membranes at the position of the ear drums, the rest of the microphones and the cables inside the head. Difficult, expensive but not impossible. But my main point: you can improve the processing all you want, the bottleneck, or weakest link remains here at the measurement I think.

I would imagine even if it could, the SINAD ratio of the binaural earphone would be the limiting factor on PRIR quality. Most of what I am reading leads me to believe that the binaural mics SINAD is l.t. 80db.

I want to ask @Maestroso again to give his opinion about the following:

With the noise in the impulse response I think it is something like this (maybe one of the DSP experts here, like @Maestroso can correct me if I am wrong, or confirm if I am right):
The measurement is done with the sweep tones playing at a fairly loud and constant level. Let's say as an example that indeed in the resulting impulse responses of the PRIR there is noise 72 db under the full level. (By the way, I don't know if deconvolving the sweep response with the sweep to create the impulse response - I hope I describe how it works correctly - would change the noise level). Later during the speaker emulation the input signal for each speaker is convolved for each ear with the corresponding impulse response. If a certain virtual speaker is reproducing a soft sound, that soft sound is convolved with a full level impulse response, and the noise in the resulting soft sound caused by the noise in the impulse response is now 72 db under that soft level that is being reproduced, and hence inaudible. The noise is scaled down (and up) with the actual soundlevel coming from the virtual speaker so to speak, always 72 db under the actual sound level, always masked by the actual sound that is reproduced by the speaker. Only when direct sound stops and the reverberation is fading out the level difference with the noise decreases. And this probably is one of the other things that is solved by the reduced reverberation length. In other words: a 72 db s/n ratio in the impulse responses does not imply that the s/n ratio of the speaker emulation is limited to 72 db [Edit: not in the normal meaning that if your actual sound level is for example at -30 db the noise - floor - would be only 42 db below that.]. Again: this is just how I imagine it, experts please comment.

 

[audiohobbit]

They are part of Atmos specs:
https://www.dolby.com/us/en/guide/9.1.6-mounted-overhead-speakers-setup-guide.pdf

Be careful, there are two versions of 9.1.6 setups: with top speakers and height speakers!
See p. 48 in https://www.dolby.com/us/en/technol...tmos-home-theater-installation-guidelines.pdf for the version with height speakers (front, rear) which you linked. On the following p. 49 is the "normal" version with top speakers.
Front height speakers should have an elevation angle according to the white paper of 20 to 30 degrees, top front speakers should be at 45 degrees.
For x.x.4 setups there are only top speakers mentioned at 45 (front) and 135(back) degrees elevation.
If you look up the tables in the manual in appendices A and B you'll notice that the Realiser Atmos setups always refer to top speakers.
In appendix B the elevation angles for top front speakers are listed as 60 degrees opposed to 45 degrees in Dolby white paper. Also the back surrounds according to Dolby should be at an azimuth of 135 to 150 degrees. Dolby calls them "left and right rear surround speakers" whereas Smyth don't name them rear surrounds but just "left and right back" with azimuth angles of 120 degrees!
There are also back surround speakers mentioned (at an odd angle of 164 degrees) but they're not used in Atmos setups.

Smyth writes that these are only default angles and can be changed to match a phyiscal loudspeaker setup during PRIR measurement, but what does that mean??
Does the Atmos decoder remap the layout to those angles you entered?? (The hardware should be capable I think since it's the same MDS APM89L board that is used in the Storm Audio AVPs)
If it does not remap, which angles does it presume?? Those from the Smyth tables or those from Dolby?

I think this is a very important question and it's not answered.