ProSoundWeb Community
Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => LAB: The Classic Live Audio Board => Topic started by: Michael John on July 18, 2019, 06:26:56 PM
-
I've seen many questions on ProSoundWeb forums about FIR filtering - some dating back years. I've put together an article describing some DSP, FIR and IIR basics, and applications to loudspeakers.
https://eclipseaudio.com/fir-filter-guide/
Let me know if you find this helpful, and/or if there's anything else you'd like to see included.
BR,
Michael
-
I've seen many questions on ProSoundWeb forums about FIR filtering - some dating back years. I've put together an article describing some DSP, FIR and IIR basics, and applications to loudspeakers.
https://eclipseaudio.com/fir-filter-guide/
Let me know if you find this helpful, and/or if there's anything else you'd like to see included.
BR,
Michael
Love your work Michael, thats great :)
FWIW the one thing that I would like to see emphasized and explained in simple terms is why it takes time to implement an FIR filter / crossover. So many people are expecting the time or latency to get better as computing power increases.
-
Love your work Michael, thats great :)
FWIW the one thing that I would like to see emphasized and explained in simple terms is why it takes time to implement an FIR filter / crossover. So many people are expecting the time or latency to get better as computing power increases.
Thanks very much Peter!
We show some examples of how it takes a fairly long FIR to match some IIR's, but we could beef this up somewhat. And, yes, as computing power increases, we'd expect to see filters get longer and LF reach get lower and lower. We're thinking of another article explaining long FIR using segmented FFT techniques, like used in the old Lake Huron. :-)
-
Thanks very much Peter!
We show some examples of how it takes a fairly long FIR to match some IIR's, but we could beef this up somewhat. And, yes, as computing power increases, we'd expect to see filters get longer and LF reach get lower and lower. We're thinking of another article explaining long FIR using segmented FFT techniques, like used in the old Lake Huron. :-)
... also FWIW in live sound applications the (FIR) processing time for FOH applications is limited to some where between 10 and 25ms depending on the size of the stage i.e. the delay time from the FOH speakers to the loud back line instruments. This usually dictates an IIR crossover for the subs.
For monitor systems it has to be much less, as low 2ms in some situations.
-
Hi Michael,
I really enjoyed the article. It was an excellent read!
I'm thankful for the explanations of the filter implementations in DSP. I really like fundamental explanations. For instance, I wasn't able to drive a manual very well until my dad explained how the mechanism physically worked. Brought me back to discrete signals class :)
I know being all things to all people muddies the water but I got hung up on the impulse response and magnitude response for a sec. Maybe a quick blurb introducing it the way you introduced everything else in the article.
Going beyond the scope for future articles, going over filter ringing, 'horn correction' briefly mentioned, how to perform averages to make speaker curves and what speakers are equalizable (cocentric/coexit) and then finally steep filters for bi-amping in a future article(s)?
Also the new LIR filters would be cool to understand.
Overall thanks!