Correcting for soundstage?

Discussion in 'Sound Science' started by oblongdarts, Nov 15, 2017.
First
 
Back
1 2
Page 3 of 3
  1. sander99
    Ah, I did not think about that yet. Using two head-trackers may or may not work, if the slightest difference occurs in the detected head-positions for the A and B part then the result would get messed up I think. Better would be if the signal from one head-tracker could be split, but I have no idea what kind of signals are going through those cables or if it would be possible to split at all. And the A8 has a totally different head-tracker, wireless I think(?), I have no clue what is possible for that either.
     
  2. Erik Garci
    I think Smyth would have to update the firmware to add a new mode that allows one head-top to control both users at the same time (without changing the hardware). I have not found a work-around.
     
  3. Erik Garci
    I tested this, and it works, providing up to 24 dB of crosstalk reduction.

    The global delay must be set to the minimum value of 1.0f in order for both users to be perfectly in sync (otherwise, user A will be delayed relative to user B, because global delay affects user A but not user B).

    In addition, the master volume for user B should be set 9 dB higher than the master volume for user A, to match the levels of the analog RCA outputs.
     
    Last edited: Dec 14, 2017
    jgazal likes this.
  4. sander99
    Beautiful!
    But could you somehow test/measure that it is in sync? Or you base your judgement on theoretical specifications combined with a listening test?
    I am just asking because indeed ITD is about very tiny time differences and hence very critical I suspect.

    By the way, it is still not guaranteed to work in sync using the A16 of course. First of all because the A8 handles 2 users in a very different way compared to the A16 as far as I understand.

    And how did it sound? Did you compare with your cross-talk free PRIR?
     
  5. Erik Garci
    I tested sync by switching between the right output of user A and user B, while keeping the left output of user A (without reducing crosstalk during this test). They sounded the same to me. When I increase the global delay, even just slightly to 1.1f, the sound changes drastically. I might do a more precise test later.
    I compared them, and the effect is very similar. The dual-user method seems to sound more natural for far-left and far-right sounds. The other PRIR was created using different speakers in a different room, and the speakers were spaced 60 degrees apart, and these differences affect my comparison to some extent.
     
    jgazal likes this.
  6. sander99
    One more question: did you also try with head tracking with 2 head-top units? (I don't know if you have 2 head-top units?)
     
  7. Erik Garci
    I only have one head-top. I plan to buy the wired one when it comes out.
     
First
 
Back
1 2
Page 3 of 3

Share This Page