Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bruno putzeys]

I'm talking sampled filters of course. This conversation is becoming very cryptic

[Vladislavs Korehovs]

Bruno Putzeys wrote on Fri, 01 June 2007 02:49

I'm talking sampled filters of course. This conversation is becoming very cryptic

[Quince]

Is this decramping usable for other filters like crossovers?  Or, assuming there's already oversampling in the system, then wouldn't it be best to do a crossover after oversampling filter, just before the DACs?

Also, Bruno mentioned numerical optimization from a given response.  I'm wondering what the limitations are though, surely one can come up a set of magnitude and phase response curves that are not achievable in a digital equalizer?

[bruno putzeys]

Usually crossover/speaker correction filters are designed/optimised directly in the z-domain, obviating the need to worry about transform methods from s to z domains and their various foibles. Otherwise the direct z transform works well enough for speakers.

Depends what you mean by achievable. If a FIR filter is allowed to become infinitely long and you're allowed to add an infinite amount of delay, any response is achievable. Otherwise, any response can be merely approximated.