Hello,
are there easy to use programs out there that can create FIR convolution files (for camillaDSP) by simply typing in parametric EQ`s ?
the only real solution i know is using rephrase + rew, but could i just enter the parametric eq settings from rew correction to get a FIR convolution filter instead of IIR?
best regards
are there easy to use programs out there that can create FIR convolution files (for camillaDSP) by simply typing in parametric EQ`s ?
the only real solution i know is using rephrase + rew, but could i just enter the parametric eq settings from rew correction to get a FIR convolution filter instead of IIR?
best regards
Agree - rePhase works great.
However, speaker and room EQ should be done with IIR filters, as both behave as minimum-phase.
However, speaker and room EQ should be done with IIR filters, as both behave as minimum-phase.
IIR is certainly normally computationally simpler but I can't see a reason why it should have less delay.
Fir filters do a far better job at room related issues and speaker related issues
Iirs work fine because they behave minimum phase and are usually stable
Firs can be both…… and speakers in a room do not behave purely minimum phase and with an fir you can restore invert-ability
@abraxalito
Iirs don’t have latency
For reproduction sound who cares about a little latency from an fir
A full room correction is about 60-100ms of latency
That’s a fraction on a single video frame….. there no sync issues and anyone that says that has no clue about a proper fir, and no you don’t need to upsample and run trillion tap firs to get good sq from an fir…..
A 48k or 96k fir works fantastic and very low latency.
Iirs don’t have latency
For reproduction sound who cares about a little latency from an fir
A full room correction is about 60-100ms of latency
That’s a fraction on a single video frame….. there no sync issues and anyone that says that has no clue about a proper fir, and no you don’t need to upsample and run trillion tap firs to get good sq from an fir…..
A 48k or 96k fir works fantastic and very low latency.
Sure, but no reason for a min-phase FIR to have latency either. Latency would render it non-minimum phase.
FIR filter is the type of filter realization. It is not linear phase or minimum phase. You can use any characteristics filter with FIR method. That is the beauty of it. Digital filtering in other words.All analog filter physics applies, but you also have the posibility of linear phase which is imposible in analog way because you have to travel “back in time” 😀 before real impulse started
i currently use fedora with easyeffects (i switched from raspberry pi to desktop pc because i noticed fedora set up right actually sounds superior..)
with easyeffects you can choose on the parametric eq if you wanna use IIR/FIR/FFT/SPM
actually SPM sounds the best... FIR is also better than IIR and from what i read FFT is just a compute improved FIR
for me easyeffects is the best sounding EQ i tried so far
also IIR vs FIR... if you wanna have phase changes for crossovers IIR is definitely the way to go, BUT FOR EVERYTHING ELSE you want linear (unaltered) phase which FIR/FFT/SPM can do unless your multiway system asks for it
what definitely is on my plan: trying to linearize the phase with rephase but from way i read this is a linear/delay compromise
with my current 2way studio monitors i dont really need phase linearization, specially if the EQ isnt messing with the phase!!
with easyeffects you can choose on the parametric eq if you wanna use IIR/FIR/FFT/SPM
actually SPM sounds the best... FIR is also better than IIR and from what i read FFT is just a compute improved FIR
for me easyeffects is the best sounding EQ i tried so far
also IIR vs FIR... if you wanna have phase changes for crossovers IIR is definitely the way to go, BUT FOR EVERYTHING ELSE you want linear (unaltered) phase which FIR/FFT/SPM can do unless your multiway system asks for it
what definitely is on my plan: trying to linearize the phase with rephase but from way i read this is a linear/delay compromise
with my current 2way studio monitors i dont really need phase linearization, specially if the EQ isnt messing with the phase!!
In my experience they don't, especially at very low frequencies where you will need very long FIR filters. There is no need for FIR either, since rooms and speakers usually are minimum-phase.
I use FIR to create crossover filters, and all my EQ is IIR. The result is flat phase.
Yes I agree for LF….
I think I was speaking mainly about phase issues that can’t be dealt with in an iir. Reflections vs coherence at listening position….
I’ve Eq many drivers to be flat with no crossover filters applied and still had some pretty ugly phase issues where an APF simply made matters worse…. Then trying to get the other channels to mimic any residual phase…. An fir is far more easy and frankly does a better job.
But yes there are limitations…. And LF usually behaves itself, so an iir works fine 99% of the time…. In the 250-1k range an fir can be a godsend
@ppp000
Yes it’s noticeable 1/10/sec
Most tvs have a video sync and 1/10th sec I haven’t noticed any issues…. I have 64ms on my home system and movies play fine… I don’t sit there and try to read lips during a movie…. You’ll drive yourself crazy doing that
It’s fine …. All the sounds come out at the proper time unless I try to notice it ….. just watch the movie and don’t try to notice and you won’t notice
Yes it’s noticeable 1/10/sec
Most tvs have a video sync and 1/10th sec I haven’t noticed any issues…. I have 64ms on my home system and movies play fine… I don’t sit there and try to read lips during a movie…. You’ll drive yourself crazy doing that
It’s fine …. All the sounds come out at the proper time unless I try to notice it ….. just watch the movie and don’t try to notice and you won’t notice
This is a puzzling statement. I don't know why you'd have "ugly phase issues" (I'm not sure what that means actually) in the driver passband if correcting the amplitude response to be flat.
And of course adding an all-pass filter would make matters worse (all other things being equal.) Why would you think it wouldn't? The purpose of all-pass filter is to do exactly that. 🙂
Dave.
Dave ,
Okay I should try state that I am measuring in a highly reflective environment.. and yes things get ugly
There’s a but tho …. But , regardless, an fir seems to be a lot easier to deal with such issues , especially when summing more then one speaker and when one speaker has a different reflection character then the other
There’s not much that can be done otherwise… at least we’re on agreement that APF isn’t usually the answer, I suppose it would be OK if they were exactly the same on both sides but that rarely happens anywhere above 100 Hz and below 1k
Okay I should try state that I am measuring in a highly reflective environment.. and yes things get ugly
There’s a but tho …. But , regardless, an fir seems to be a lot easier to deal with such issues , especially when summing more then one speaker and when one speaker has a different reflection character then the other
There’s not much that can be done otherwise… at least we’re on agreement that APF isn’t usually the answer, I suppose it would be OK if they were exactly the same on both sides but that rarely happens anywhere above 100 Hz and below 1k
Please be more specific. Lots of people can be confused when you talk about FIR vs IIR. IIR can only be minimum phase while FIR can be both. Talk about min phase vs linear phase so it is clear what it is all about.