In Reply to: Digital equalizer experiment over posted by mls-stl on July 2, 2006 at 09:54:59:
Practical limitations of EQHi
EQ seems like a great idea, the problem is the people that sell equalization like to think it is more powerful than it is.
In reality, EQ is a double edge sword, in fact it has an edge hidden on the grip too haha.EQ can perfectly “fix’ anything which is a minimum phase problem.
That is to say, changes in response that also have a linked phase change can be fixed as the EQ correction has the proper phase as well.EQ can’t fix things which are non-minimum phase AND trying to fix those things CAUSES big phase problems. This is what / why most people think of as why EQ is bad, the phase problems cause by improper use.
Also, an issue is what to measure with. Understand, 1/3 rd octave resolution RTA using pink noise is THE minimum resolution one should consider and at that, it does not reveal the transfer function of the speaker.
Most importantly, keep in mind what you can’t fix and shouldn’t try to fix.
Lets say you see an LF or HF roll off slope, this can be fixed.
Lets say you see a broad mound in the response, this can be fixed.
Lets say you see a narrow peak in the response, this you can usually fix.
Lets say you see a deep dip (the depth of which is likely NOT visible with an RTA), this is from cancellation (a reflected signal) and is NOT a minimum phase problem and CANNOT be fixed and it is BAD to tryA speakers response in room at the listening point is likely to be full of minimum AND non minimum phase problems.
Directivity is the only way around that but its not spouse friendly.
Hope that helps.Tom Danley
This post is made possible by the generous support of people like you and our sponsors:
Follow Ups
- Practical limitations of EQ - tomservo 08:24:55 07/05/06 (3)
- Crossover networks are equalizers too and by comparison they are awful - Soundmind 06:24:48 07/07/06 (2)
- 'Awful' passive crossovers - thoriated 10:38:48 07/07/06 (1)
- Re: 'Awful' passive crossovers - Soundmind 11:05:28 07/07/06 (0)