Re: [Sursound] BBC Radio Three Surround Streaming Trial (15. to 31. March)
At 23:33 18-03-14, Paul Hodges wrote: --On 18 March 2014 23:18 +0100 David Pickett d...@fugato.com wrote: BBC server went down several times this evening. What I don't understand is why things like that go down so easily; it may be hard to get a computer system running, but if it's done even reasonably right, once you succeed it should just keep running until a real change is made. That agrees with my experience, too. One of the concomitant frustrations of these experiments is not atually knowing whether the loss of signal is due to the originating server or something along the way between there and here! It stopped working altogether towards the end of the concert. Continually pressing CTL-F5 and then attmepting to reconnect is not fun! I suspect that most of the problems last night were at the originating end, though there were cases when there were beats missing as the stream caught up, which seemed more likely to be delays in the Internet. David ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] BBC Radio Three Surround Streaming Trial (15. to 31. March)
On 19 Mar 2014, at 07:33, David Pickett wrote: I suspect that most of the problems last night were at the originating end, though there were cases when there were beats missing as the stream caught up, which seemed more likely to be delays in the Internet. from the BBC blog: 21. Rupert Brun, 18TH MARCH 2014 - 22:20 I am sorry we lost the stream before the end of the concert this evening, this was due to a problem with the internet connection to the server at the Southbank. What would the bitrate be ? I'm also curious about the delay. Has anyone been able to compare the streamed audio to fast radio ? Kees de Visser ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Periphonic Irregular HO Ambisonics Decoder
Thanks Fons! I will start from this! Ciao Davide -- next part -- An HTML attachment was scrubbed... URL: https://mail.music.vt.edu/mailman/private/sursound/attachments/20140319/b12dba8b/attachment.html ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sony's VR announcement
I'm watching these developments with interest due to my activities in virtual worlds... Indeed, as we're talking about VR headsets here, where head-tracking is vital for the visual operation, I wouldn't worry about that being available for audio - it's already there and I don't see how you could do a VR device without it. Just listening what happens to the audio stage on headphones in a current virtual environment such as Second Life suggests that the addition of head tracking would make existing models work fine. Whether or not VR content producers will bother to develop more detailed spatial sound models is another question, of course. [Actually, I must say that at this point I am wondering how VR headsets will get around the issue of eye convergence without focus shift, which is likely to be a much bigger problem than anything in the audio arena :) ] -_R On 19/03/2014 16:11, Stefan Schreiber wrote: (Head-tracking probably included, as this is a VR device. My source says that the correct position of noises will be calculated and reproduced via headphones. ) Sony's head-tracker is said to have a temporal accuracy of 1000Hz, which should be enough for any video and audio purposes. ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] BBC Radio Three Surround Streaming Trial (15. to 31. March)
At 17:47 19-03-14, Aaron Heller wrote: I downloaded the MPD file on the FAQ page with wget http://rdmedia.bbc.co.uk/dash/ondemand/channel_test/1/5.mpd It asked me what you wanted to do with it automatically! Just selected download. If I'm reading it correctly, the channel announcements are 320 kbits/sec, 48k sample rate. 48k is what the Head of Technology for BBC Radio told me too. I changed to that for last night. A bit scary that, as Paul said, Microsoft employ SRC automatically if you get it wrong... I heard from one of the engineers responsible that they have been working on the server today and hope to have a more robust connection tonight. The engineers update and monitor their Twitter account during the concert, and appreciate comments: #BBCR3surround David Aaron On Wed, Mar 19, 2014 at 1:01 AM, Kees de Visser k...@galaxyclassics.comwrote: On 19 Mar 2014, at 07:33, David Pickett wrote: I suspect that most of the problems last night were at the originating end, though there were cases when there were beats missing as the stream caught up, which seemed more likely to be delays in the Internet. from the BBC blog: 21. Rupert Brun, 18TH MARCH 2014 - 22:20 I am sorry we lost the stream before the end of the concert this evening, this was due to a problem with the internet connection to the server at the Southbank. What would the bitrate be ? I'm also curious about the delay. Has anyone been able to compare the streamed audio to fast radio ? Kees de Visser ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound -- next part -- An HTML attachment was scrubbed... URL: https://mail.music.vt.edu/mailman/private/sursound/attachments/20140319/cac673b8/attachment.html ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sursound Digest, Vol 68, Issue 10
in front, for a decent approximation of ITU 5.1 and 7.1, if necessary. * the satellites in a lower ring-of-eight, an upper ring-of-eight, another ring of six, one zenith speaker. then you have two spares, and they will come in handy some day. the bass management will be tricky. first of all, each speaker needs to be perfectly delay-compensated to the listening spot. then i'd try to create different layers of decoding: * separate first-order decode for the subs, low-passed at 60, 24dB/oct * fourth-order decode for everything else * horizontal speakers high-passed at 120/24 * satellites high-passed at 120/24 * a separate horizontal-only decode (of the same full-sphere input signal) for the range from 60 to 120 hz, again at 24dB/oct this lets you drive all speakers to the best of their abilities, and puts the missing bass frequencies in the correct direction. $DEITY help you if anything is not perfectly phase-aligned, though. disclaimer: i've toyed with such hacked-up multiband setups, but none of them ever went to production (or had to), so there may be pitfalls i've overlooked. First order decode for the four subs in the corners was what I was thinking. Didn't think about going to fourth order on everything else though, as I didn't think the increase in channel count was worth the little improvement. I also want to leave some processing power for mixing plugs (I use a lot) :) well, i started from the number of speakers you said you had available. Agreed on the full range horizontal ring. I was more thinking of a dodecahedron for the satellites, either only 20 on the vertices, or get 5 more, and would it be possible to use the edges? you mean you want to create entirely separate horizontal and full-sphere systems? Is it better to use platonic solids, or doesn't it matter? with the recent advances in optimizing for irregular layouts by zotter et al and heller et al, there is no longer a compelling reason to go for platonic solids, except that they are kind of pretty :) layouts based on a horizontal ring have the big advantage of better horizontal-only performance, without much degradation in the 3d case. Thanks again, and needless to say I will be asking a few more questions as I progress. The build won't start for another month, and when it's finished I would love for all you ambisonic heads to have a listen. can't wait to. where is your studio located? -- J?rn Nettingsmeier Lortzingstr. 11, 45128 Essen, Tel. +49 177 7937487 Meister f?r Veranstaltungstechnik (B?hne/Studio) Tonmeister VDT http://stackingdwarves.net -- next part -- An HTML attachment was scrubbed... URL: https://mail.music.vt.edu/mailman/private/sursound/attachments/20140319/3c24a4c7/attachment.html ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Construction of purpose built ambisonic studio (Fons Adriaensen)
Thanks Fons This looks like great space and probably a good goal to aim for. I think the room will have a roughly rectangle footprint now , with the vertical sides tapering out towards the back. The vertical front wall will be slightly angled from the furthest away centre position to the sides.The ceiling will take advantage of the all ready sloped roof at the front, and then be just off parallel to the floor rising all the way back. The back vertical wall will be completely parallel to a fictional wall at the front, (I.e if the front wall wasn't angled at the centre). Haven't worked out listening position yet, as will need to do some calculations, and tests when built, to see which is best for sound and available space. It's always a compromise, but bass traps will be built! I am wondering whether it would be better to get a few more speakers, and go to fourth order, and if so how many more, and would the improvement be that noticeable for a treated space this size? Also although the rotation would provide 5.1/7.1 compatibility for the smaller satellites, it doesn't include the larger 10 of horizontal full range speakers. These are really important to me as they are a speaker that I am very used to mixing on, and need to be included as it will mean less time getting acquainted to. In fact I think the whole system needs to be built around these as they will be the dominant force in my mixing. Everything will be done in HOA but decoded realtime for A/B' ing. Would it be better to replace the head height horizontal ring of satellites with 8 full range ones or add these in between, with a rotation on the same decoder? I would probably always upscale lower orders to 3rd or forth, would this be a problem, or would it be better to have a dedicated sub set for each order? In which case not sure on the best sub set. This is where it will get really complicated! Cheers, Steve Message: 4 Date: Tue, 11 Mar 2014 00:20:20 + From: Fons Adriaensen f...@linuxaudio.org To: sursound@music.vt.edu Subject: Re: [Sursound] Construction of purpose built ambisonic studio. (Aaron Heller) Message-ID: 20140311002020.ga5...@linuxaudio.org Content-Type: text/plain; charset=us-ascii On Mon, Mar 10, 2014 at 09:50:43PM +, Steve Boardman wrote: Stanford's CCRMA room does look (and undoubtably sound) good, but the space below is maybe a bit over board for what I want to achieve, in the space I have. The actual area of the build space is probably around 180 square foot within a bigger space of 700 square foot on two floors. As an example of what can be done without digging holes in the ground have a look a this: http://www.rossinispace.org/. This is at the conservatory of Pesaro, Italy, and the best sounding and most accurate higher order Ambisonics studio I know of. Size should be comparable to your 180 sq.ft. Shape is approximately a square, but with no parallel walls. The space has a very low RT60 down to LF (bass traps are planned but not yet operational), the idea being that in AMB mixes most of the space should be provided by the signal and not by the room (which makes sense, creating virtual spaces is one reason to use full surround). The control desk, shown against the wall in the panaromic picture, can be moved to the center. The speaker system consist of * a ring of six at elevation -33 degrees (ideally this should be -45 degrees, but this requires an elevated listening position), * a ring of eight at ear height, * a ring of six at +45 degrees * a speaker at the zenith. * one subwoofer for a total of 21+1 speakers. This is an excellent setup for third order, in the sense that the decoder matrix is very well-conditioned (it doesn't rely on signals that would cancel acoustically). If you have four subs there's no reason for not using them (put them in the corners, with a dedicated decoder). One thing that could be improved is that the current ring of eight is oriented such that there is no front speaker. The alternative, rotating it 22.5 degrees, would provide a layout that is more compatible with formats such as 5.1 or 7.1. One point not yet mentioned in the replies so far is that for lower order (and in particular first) you should use less speakers. Also for this the rotated ring of eight would be better - the subset used for first order at the moment does not have L-R symmetry. Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) -- next part -- An HTML attachment was scrubbed... URL: https://mail.music.vt.edu/mailman/private/sursound/attachments/20140319/987d6c10/attachment.html ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sursound Digest, Vol 68, Issue 17
Hi, Surely the best approach is to feed the noise signal (post decoder) into each speaker channel in turn and adjust the amplification on each channel until the level measured at the centre listening point is the same for each speaker. The panning approach won't work, as all speakers would be excited at various different levels. It would be useful after the above calibration to see if the sound of the noise of the noise was consistent everywhere everywhere it was panned. Ciao, Dave From: Iain Mott m...@reverberant.com Date: 18 March 2014 21:38:29 GMT To: Surround Sound discussion group sursound@music.vt.edu Subject: Re: [Sursound] calibrating ambisonic speakers using the k- system? Reply-To: m...@reverberant.com, Surround Sound discussion group sursound@music.vt.edu Em Tue, 2014-03-18 às 19:52 +, Fons Adriaensen escreveu: On Tue, Mar 18, 2014 at 03:32:55PM -0300, Iain Mott wrote: Thanks a lot Fons. When I pan pink noise with at W at -20dBFS RMS, the individual X and Y channels peak at about 3dB higher. Is that why you said to meter at 86dB and not 83? No. In a stereo system, with the levels as 0dB on the K-20 meter, each speaker produces 83 dB SPL. Assuming the signals are mostly decorrelated, the total level will be 86 dB. So the 'reference SPL' is 86 dB. OK - I see what you intended. For my current purposes, I'd like to reproduce as best as possible, ambiental b-format recordings over an array of speakers - and preferably try to match SPL measurements taken at each recording location. Do you think the formula above would be correct to match levels in this way? ie. if I make a recording at a site where the SPL is 70dB, during playback I meter this material (the W channel) at -13dB RMS on a k-20 meter, and in the case of a 14 channel system, calibrate each speaker channel at 71.5dB SPL (x = 83 - 10log14). Your only chance to get this right is to calibrate *via the decoder*. If you follow the procedure I explained, then 0 dB on the K-20 meter for W will corresponds to 86 dB SPL, no matter how the sound is distributed over the speakers. That's assuming you don't pan two or more strongly correlated signals to different directions (if you do that the result is no longer really Ambisonic). I now understand that W in the metering has a direct relationship to the total audio output of the array - no matter what the configuration - but sorry, I'm still in doubt as to how to go about adjusting the speaker output levels. I initially assumed that during the panning of the signal and the output adjustment, the speaker that is most in focus (at the peak level) would be soloed - but this wouldn't work because it wouldn't factor in the additional output from the other channels. Are you suggesting that all channels should be left open and the system tuned in a number of passes? Dare I say it: might the -10 log (N) level be a good starting point for each channel? Thanks -- next part -- An HTML attachment was scrubbed... URL: https://mail.music.vt.edu/mailman/private/sursound/attachments/20140319/7d24ab83/attachment.html ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sursound Digest, Vol 68, Issue 17
On Wed, Mar 19, 2014 at 07:50:43PM +, Dave Hunt wrote: Surely the best approach is to feed the noise signal (post decoder) into each speaker channel in turn and adjust the amplification on each channel until the level measured at the centre listening point is the same for each speaker. That would be a prerequisite for the method I explained. But it still leaves you with an uncalibrated system, as the decoder gain (no matter how you define it) isn't included. The panning approach won't work, as all speakers would be excited at various different levels. It would be useful after the above calibration to see if the sound of the noise of the noise was consistent everywhere everywhere it was panned. On the contrary, it's the only one that will give the correct result. *** Calibration means to have a defined relation between *** the level of the W channel and the measured SPL. This can be done only with the decoder in the path. Another approach would be to sent W only (at reference level) to the decoder, and then measure each individual speaker (by soloing it, ambdec provides the function) and adjusting for reference SPL - 10 * log(number_of speakers). This would be less accurate as it doesn't allow for the partial correlation between speaker signals (which will depend on frequency if you use dual band decoding). Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sursound Digest, Vol 68, Issue 17
On Wed, Mar 19, 2014 at 07:50:43PM +, Dave Hunt wrote: The panning approach won't work, as all speakers would be excited at various different levels. Anything to substantiate that claim ? Practice ? Theory ? FYI, I have used this method a number of times, with excellent results. I've also done the maths that show it do be correct. Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] Sursound Digest, Vol 68, Issue 17
Em Wed, 2014-03-19 às 20:21 +, Fons Adriaensen escreveu: Another approach would be to sent W only (at reference level) to the decoder, and then measure each individual speaker (by soloing it, ambdec provides the function) and adjusting for reference SPL - 10 * log(number_of speakers). This would be less accurate as it doesn't allow for the partial correlation between speaker signals (which will depend on frequency if you use dual band decoding). Great - yes, I was hoping this would work. The SPL meter arrived today by post - but i've still only two channels of amplification. Will test as soon as the equipment is organised. Thanks a lot for your help. Iain ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
Re: [Sursound] BBC Radio Three Surround Streaming Trial (15. to 31. March)
Tonight's effort a bit of a dog's breakfast with glitches, drop-outs and confusion over the recorded link pieces in the interval. However, when it was working, it was pretty damned impressive. They seem to have tweaked the surrounds a bit, so no need to up the gain any more. Together with some really nice playing (and bar someone's personal alarm going off a couple of minutes into the Tod und Verklärung, necessitating a restart) it was rather an encouraging evening: once the internet connection is sorted out, this would appear to be a viable approach to broadcasting at least some form of surround home systems, just using a browser and a sound-card. I have bits of it recorded, if anyone wants a listen. Mind you, when I say bits, I mean bits. Several rather large gaps in transmission means that it's by no means a complete performance. Off list to me if you want, and I'll prepare a file. It'll be a 48/24 wavex of the uninterrupted bit of the Strauss. Cheers, John On 16 Mar 2014, at 17:28, David Pickett d...@fugato.com wrote: I just stumbled on this: http://rdmedia.bbc.co.uk/radio3/faq.html I have an appropriate soundcard; but am not sure I want to install Google Chrome... (Bummer that it doesnt work on Firefox!) has anyone David ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound ___ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
