One correction, before you invert the A-to-B gain matrix, you have to account for the -3dB gain on W in B-format FuMa.
On Thu, Mar 8, 2018 at 8:35 PM, Aaron Heller <ajhel...@gmail.com> wrote: > A friend bought one, so I've been doing some reverse engineering. > > In the 96kHz stereo file, the samples are interleaved > > L: 1313131313 > R: 2424242424 <(242)%20424-2424> > > litepc2aformat reformats this to a 48kHz, 4-channel file > > 1: 11111 > 2: 22222 > 3: 33333 > 4: 44444 > > I also measured the impulse responses of Twirling720 Studio.app when > converting from 4-channel A-format to 4-channel B-format FuMa. It is just a > 4x4 matrix, no coincidence correction or any other filters. (view this in > a fixed-width font) > > 1 2 3 4 > FRU BRD FLD BLU > W 0.0354 0.0354 0.0354 0.0354 > X 0.0107 -0.1220 0.1220 -0.0107 > Y -0.1220 -0.0107 0.0107 0.1220 > Z 0.0866 -0.0866 -0.0866 0.0866 > > > If you invert that matrix and convert the velocity components to polar, > they assume the mics are pointed at, > > az > -85.0000 > -175.0000 > 5.0000 > 95.0000 > > el > 35.2644 > -35.2644 > -35.2644 > 35.2644 > > so the whole mic array is rotated 40 degrees clockwise. The elevations are > correct for a tetrahedral array. > > Also the euclidean length of the X, Y, Z components equals W*sqrt(2), so > the gains are for perfect cardioid capsules. > > Looking at some sample recordings that were made well into the diffuse > field of a fairly reverberant hall, W looks about 3 dB too low, so I'm > guessing the capsules are more directional that cardioids, something > between a cardioid and supercardoid (which would make W 4.6 dB too low). > > Without coincidence correction filters, the diffuse field response will be > wrong. > > Aaron Heller (hel...@ai.sri.com) > Menlo Park, CA US > > > > > On Thu, Mar 8, 2018 at 4:27 PM, Marc Lavallée <m...@hacklava.net> wrote: > > > > Hi Steven, and thanks for the useful info. > > > > The fact that the Twirling720 presents itself as a stereo 96Khz > > device and the possibility that the 4 channels are "encoded" in 2 > > channels is very interesting, because it could potentially be used with > > any computer. I still don't use mine because the Android app is asking > > for too many permissions on my phone. I'm waiting for the official SDK, > > hoping I can program simple and safe custom apps. I also plan to build > > a mount because holding it only from its USB plug is very risky; a > > modified phone shell could be a good start. > > > > Le Thu, 8 Mar 2018 18:40:20 +0000 > > Steven Boardman <boardroomout...@gmail.com> a écrit: > > > > > I have one too. Been using it successfully for a while with my android > > > phone. > > > It does seem to present itself to other audio apps as 2 channel 96khz > > > device. > > > When using the apk it records either stereo, or 4 channel A and B > > > format at 48khz. I think theres some sort if matrixing going on. > > > I haven't done any vertical tests, but the horizontal works well for > > > the price. > > > It is also pretty easy to rotate the capsule spindle, so not sure how > > > accurate the positioning is. Mine is also not quite perpendicular! > > > The manual is useless. > > > They are quick to fix bugs, and implement suggestions thoigh. (the > > > A-Fornat one was mine.) > > > Its way better than a h2n in my opinion, and really easy to carry, as > > > i always have my phone anyway. > > > Because if this i use it a lot, as i carry it at all times. > > > I just have to make a mount for use with my Samsung gear 360. > > > > > > Best > > > > > > Steve > > > > > > On 8 Mar 2018 16:05, "John Leonard Main" <j...@johnleonard.uk> wrote: > > > > > > > Mine (pre-ordered for some small amount) arrived a couple of days > > > > ago and I’ve got it hooked up to my MacBook via a suitable USB > > > > adapter and an old Apple keyboard extension cable. At first, I > > > > couldn't get a sensible signal out of it, but then discovered that > > > > it needs to be connected via USB3, or it won’t work. Then I took a > > > > look at the capsule orientation, which, although it is indeed a > > > > tetrahedral array, seems to be skewed by 45º off centre, but as the > > > > output is encoded in some way into two channels, this may not be a > > > > problem. By using their 720 Studio app, I can get a sort-of > > > > surround signal out of it, although it appears to have no vertical > > > > information. The skimpy on-line manual is pretty useless for Mac > > > > users, so I wonder in anyone else has had better or more consistent > > > > results? > > > > > > > > Bruce - I could send it to you for chamber analysis, if you’re > > > > interested. > > > > > > > > All the best, > > > > > > > > John > > > > _______________________________________________ > > > > Sursound mailing list > > > > Sursound@music.vt.edu > > > > https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe > > > > here, edit account or options, view archives and so on. > > > > > > > -------------- next part -------------- > > > An HTML attachment was scrubbed... > > > URL: > > > <https://mail.music.vt.edu/mailman/private/sursound/ > attachments/20180308/5e465316/attachment.html> > > > _______________________________________________ Sursound mailing list > > > Sursound@music.vt.edu > > > https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe > > > here, edit account or options, view archives and so on. > > > > _______________________________________________ > > Sursound mailing list > > Sursound@music.vt.edu > > https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, > edit account or options, view archives and so on. > -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20180308/28467883/attachment.html> _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit account or options, view archives and so on.
