It seems we're not going to reach a unanimous opition about an IsZero flag. (I would prefer not to settle on a simple majority opinion or super-majority, even if that's the case right now.) Vincent's point, which warrants at least a benchmark, is that the relative cost of zerofilling is negligible for any plugin of even modest complexity. Recursive DSP has "infinite" decay, so you can't say "zero-in, zero-out" without loss of quality and precision. I happen to *believe* -- I can't *know* without a benchmark -- that for some classes of DSP the cache penalty for reading samples out of a zeroed buffer is significant. Vincent *believes* otherwise -- unless he has benchmarks he is willing to share that prove otherwise. But right now we're not actually doing the design, or doing benchmarks. So let's defer the decision on whether or not to add this flag to a time where benchmarks are appropriate. -----Original Message----- From: Vincent Burel [mailto:vincent.burel@xxxxxxxxxx] Sent: Tuesday, June 03, 2003 10:41 AM To: gmpi@xxxxxxxxxxxxx Subject: [gmpi] Re: Topic 7: Audio packaging ----- Original Message ----- From: "Paul Davis" <paul@xxxxxxxxxxxxxxxxxxxxx> To: <gmpi@xxxxxxxxxxxxx> Sent: Tuesday, June 03, 2003 4:21 PM Subject: [gmpi] Re: Topic 7: Audio packaging > >zero fill code is may be the fastest operation we can do , so i remind me > > zero filling is fast in terms of instruction cycles, but it has cache > effects that you are ignoring. i missed you and remind me when you would participate. it' s always a pleasure when you open your mouth. this always bring a big amount of humour, intelligence and information. Thanks for all , Paul ! Vincent Burel ---------------------------------------------------------------------- Generalized Music Plugin Interface (GMPI) public discussion list Participation in this list is contingent upon your abiding by the following rules: Please stay on topic. You are responsible for your own words. Please respect your fellow subscribers. Please do not redistribute anyone else's words without their permission. Archive: //www.freelists.org/archives/gmpi Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe