KVR Audio

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 5:45 pm

Just out of curiosity, cpu usage aside. I'm wondering, since there is enough info on it now. Can you now build a zero-delay filter? And not just add someone else's code ie. build your own.

Or is zero-delay for the math savvy few? Poll runs for for 7 days. Please vote, the results would be interisting to know.

Regards
Andrew

antto · Post by **antto** » Wed May 23, 2012 8:16 pm

Can you now build a zero-delay filter?

well, of course!.. not

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 8:20 pm

antto wrote:
Can you now build a zero-delay filter?
well, of course!.. not

Yes I know you can't, I can't either - I am just wondering what the general consensus is. I know of only about 7 people around here, that can definetly do it - wondering if they are the majority or minority.

Thanks for voting!

Andrew

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 8:24 pm

Hmmm...88reads -> only 3 votes. Voting is anonymous btw.

Richard_Synapse · Post by **Richard_Synapse** » Wed May 23, 2012 8:33 pm

I've built a few of my own, so I voted yes. But it's certainly tough. Not only is there many places where you can screw up, the model you choose as a starting point may be flawed to begin with.

Linear models are usually easier than the nonlinear ones, so that's what I'd recommend to try first.

Richard

tony tony chopper · Post by **tony tony chopper** » Wed May 23, 2012 8:40 pm

proof!

tony tony chopper · Post by **tony tony chopper** » Wed May 23, 2012 8:40 pm

I can build a negative-delay filter, & thus go back in time to right before I posted this message

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 8:44 pm

Richard_Synapse wrote:I've built a few of my own, so I voted yes. But it's certainly tough. Not only is there many places where you can screw up, the model you choose as a starting point may be flawed to begin with.

Linear models are usually easier than the nonlinear ones, so that's what I'd recommend to try first.

Richard

Thanks for the insight Richard. And the vote. 50/50 so-far.

kmonkey · Post by **kmonkey** » Wed May 23, 2012 8:44 pm

tony tony chopper wrote:I can build a negative-delay filter,

Don't forget about proper delay compensation then..

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 8:49 pm

Also forgot to say, that I don't mean any disrespect to Z1202 and the exceptional paper he wrote, it will probably be considered a "classic must-read" in years to come. And also gets a +1 for adding some actual code snippets. Isn't it strange that papers on DSP contain no actual DSP implementation?

mystran · Post by **mystran** » Wed May 23, 2012 9:14 pm

I seriously find them much easier than any other form of filter, except maybe direct forms. If you asked "how many can design their own traditional filter structures" I bet you'd actually get an even lower number.

JJBiener · Post by **JJBiener** » Wed May 23, 2012 9:33 pm

Are you kidding? Maybe at one point, years ago. I have been out of coding so long, I can barely put together a drupal or joomla website.

cheppner · Post by **cheppner** » Wed May 23, 2012 9:42 pm

well, yes.

To be honest though, I have never written a bi-quad. I guess it's a bit about what you grew up with!

Big Tick · Post by **Big Tick** » Wed May 23, 2012 10:15 pm

Not sure if I should vote yes or no... I can definitely build one with the iterative solver method. But something like the mystran optimization, or take an analog schematics and convert this into a digital zdf implementation... I guess I could if I worked really hard and dug back into my EE courses, from way back... but to be honest, I would rather contract mystran to do it for me

Ichad.c · Post by **Ichad.c** » Wed May 23, 2012 11:20 pm

Big Tick wrote:Not sure if I should vote yes or no... I can definitely build one with the iterative solver method.

Then that should be a yes - the method doesn't matter, just the can or can't.

KVR Audio

Poll - can you build a zero-delay filter?

Can you build a zero-delay filter?