Home Theater Forum and Systems banner

1 - 6 of 6 Posts

·
Registered
Joined
·
13 Posts
Discussion Starter #1 (Edited)
Hello,

from my research, the answer seems to be no, but maybe I'm lucky and I've missed something.

I'm playing around with REW's "Match Response to Target" feature and I'm amazed of what it can do and how close the predicted curves are to the target response, unless - of course - there's not enough filters involved.

Unfortunately the amount of biquads on the minidsp 96k setting is limited to 5, which is sometimes not enough.
However I've just found out the minidsp 4x10 can process up to 13 biquads per output channel, if you incorporate the advanced crossover section. For example for a 24dB LR filtered woofer, this leaves you with 11 instead of 5 biquads for equalisation. Way better.

Now is the any way to make use of this, other than iteratively equalize, measure - equalize, measure...
Or is there a way to limit the "generic" EQ setting from 20 to 11 filters? One would have to do the conversion to minidsp's biquad format manually, but still better than only 5 filters.

Many thanks in advance
 

·
Registered
Joined
·
13 Posts
Discussion Starter #2
After a little more thinking, I think I could just make to seperate matches using the "match range" feature and then combine them to get 10 biquads? :rubeyes: Close enough. :R
 

·
Super Moderator
Joined
·
9,253 Posts

For example for a 24dB LR filtered woofer, this leaves you with 11 instead of 5 biquads for equalisation. Way better.
Not necessarily. Over-equalization typically does more harm than good, audibly-speaking. Typically no more than 4-5 filters are needed to address a subwoofer’s response, often less than that. If you haven’t already, I suggest reviewing my Minimal EQ article, which you can find in my signature.

The same is true for full-range EQ as well (not sure exactly which you’re doing, subs or mains). Lots of good information about that in thread linked below.


Spridle’s Experiment


Regards,
Wayne
 

·
Registered
Joined
·
13 Posts
Discussion Starter #4
Hi Wayne!
Thank you for your reply.

I can imagine that over-equalizing the on-axis response might not always lead to good results, but I want to find it out the hard way ;) Learning is part of the process.

The speaker system I'm currently working on is a 3-way with 15" aluminium cone woofer and 3"+1" soft domes.
Because of the limited range of the domes, the 15" will have to work up to 400Hz. Besides trying some other crossover designs/strategies I'd like to try one with the drivers equalized flat way beyond their crossover frequencies to achieve kind of perfect LR4 slopes.

The woofer response in its passband is already very good without EQ, I'd say not more than 3 biquads are needed here, but it starts to behave badly above 800Hz with high breakup resonances, so 5 biquads aren't enough in this case.


Anyway, splitting the matching process into to 2 seperate parts via frequency range seems to work well and solves my problem. Manually importing the coeff's into minidps's advanced x-over works well, too. I'm happy :D

kind regards,
thomas
 

·
Premium Member
Joined
·
6,711 Posts
You could also use the nanoAVR filter setting in beta22 to get 10 filters. REW will only make use of filters that are set to Auto and enabled (leftmost check box is ticked).
 

·
Registered
Joined
·
13 Posts
Discussion Starter #6
Great! Just checked my REW version and it's beta21. Time to update.

Thank you!
 
1 - 6 of 6 Posts
Top