On Tue, Jun 11, 2013 at 12:44:01PM +0100, Paul Power wrote: > Hi Jorn, > Why do you say that you would not want to use the Eigenmike for musical > recordings?
(pre scriptum: please do not copy the entire digest when replying...) Joern will probably answer your question and provide some good arguments for his opinion. But I do share his view on this. Of all the musical recordings made using the Eigenmike I've heard so far there were none that I really liked. It's all too easy to get carried away by the initial quite spectacular surround effect, but once that wears off, in all cases I want to reach for some EQ. And using EQ usually improves things, but never to the point that the result is really OK. When you use a recording technique such as Blumlein, XY, MS, or ORTF everything depends on how well the mics preserve their polar pattern in function of frequency - or their frequency response in function of direction - for the simple reason that almost all sources will be off-axis. That's an entirely different situation compared to multi-mic techniques, where such things matter much less, and it's why people tend to use high quality expensive mics in such a role. When using the Eigenmic (or a first order tetrahedral), you're in the same situation. You will use either virtual directional mics or Ambisonic components which are polar patterns as well. Simple fact is that with the current state of the art these synthesised patterns are not really what they should be - they are only correct over a rather limited frequency range which gets narrower as either Ambisonic order or directivity goes up. Which in the end means you can either get the direct sound right, or the diffuse (reverb) part, but not the two at the same time. Which is why EQ fails to correct things. In some applications that is not a problem, for example when you need some virtual spot mics to follow one or a few actors on a film set or stage. But when the result depends on capturing sound from all directions in a balanced way - as it does for music recordings which include e.g. the concert hall acoustics as well, this is a real problem. There are currently two SW apps for processing the Eigenmic signals into either a set of directional 'spot' mics or into third order Ambisonics: the original SW coming with the Eigenmic, and the one developed by Farina's team here in Parma. While I'm not personally involved in the latter, I'm following that project closely. But so far neither of them have things as right as they could be. Part of the problem may be that even while the Eigenmic capsules are of very high quality, the remaining random differences in sensitivity and frequency response are still too large to allow accurate higher order processing without individual calibration. And until someone finds a better method than renting an anechoic room for half a day, such calibration is an expensive exercise. Earlier today I've been listening to some organ recordings made using the Eigenmic and processed into 3rd order Ambisonics. This is one of the applications that work reasonably well - frequency response errors on organ sounds are much more forgiving than on e.g. a string quartet, or voices. But I found the difference between a 1st order decode and a 3rd order one rather marginal while it is normally quite easy to hear that difference. Which probably means that higher order components are not as accurate as they should be. 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
