On 18 Jan 2011, at 17:00, [email protected] wrote: > > > ------------------------------ > > Message: 7 > Date: Mon, 17 Jan 2011 22:34:52 +0100 > From: [email protected] > Subject: Re: [Sursound] Available UHJ encoders? > To: Surround Sound discussion group <[email protected]> > Message-ID: <20110117213452.GC4090@zita2> > Content-Type: text/plain; charset=us-ascii > > On Mon, Jan 17, 2011 at 08:46:04AM +0000, Geoffrey Barton wrote: > >> UHJ needs all the help it can get, and the analogue design Dave referred to >> is 6 pole/zeroes in each of the 0 and 90 degree sections, designed for a >> phase ripple of <1degree 20~20k. Too much ripple gives blurring and >> mis-location in the images. IMHO 4 degrees is way too much. > > I very much doubt it is. If such small errors matter that would imply > that the UHJ coefficients themselves are very critical or that the > encoding/decoding matrices are ill-conditioned, but they are not.
How do you justify that assertion? Phase shifts changing with frequency also translate into amplitude errors when the phase shifter chains are summed. You need to maintain the asymptotic behaviour outside the band too. These things are audible on critical material and, especially in an encoder, should therefore be as accurate as possible. Not having to use analogue now makes this easily possible with a good design, and at reasonable mips cost. Why accept less? Geoffrey -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20110118/485e1d25/attachment.html> _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
