This post is part of my bass integration guide series on Hifi Zine >
Many believe that the ideal system has a flat response. In a real room this is neither possible nor desirable. A speaker that has a flat response (+/- 3 dB) on axis in an anechoic chamber will not be flat in any listening room, whether it be a typical room or a professionally treated one. It would be more typical to achieve deviations from flat around +/- 15 dB.
It is possible with DSP to achieve a flat response in one listening position, however, the result is generally considered undesirable and unbalanced. Here I will focus on frequencies below the Shroeder frequency, typically 200 Hz, where the response is room-dominated. In this region we can safely apply EQ and treat the direct response of the speaker and the room response together. Above this point, the direct sound from the speaker and room effects should not be lumped together - they require separate treatment.
In the diagram above you can see an illustration of a bass target curve. This is more fully described in my bass integration guide.
Choices are made in the studio regarding the levels of bass and they vary between artists and albums. Unfortunately this means that with any given setting, one album may sound boomy while another anemic in the bass. This is part of the reason why you need to set a curve that tends to apply to the material you listen to. There is also the aspect of personal preference. This requires some trial and error until you find a curve that you like best. Before applying a curve, however, it's important to first achieve certain objective standards as outlined in my bass integration guide.
There are some skeptics who don't like the idea of applying a target curve, claiming that a flat response in the listening position is the ultimate goal. Many of these skeptics have never actually tested this idea. My suggestion is to test this out for yourself before forming an opinion. I also suggest that you follow the process outlined in my bass integration guide. This target curve can only be applied effectively in a system that has been properly optimised as outlined in the guide. Otherwise your result will be skewed by untamed modal resonances.
Paul, have you updated this at all? Inquiring minds want to know.
ReplyDeleteThanks for the reminder about this page!
ReplyDeleteI'm liking the info on your site and would like to hear the conclusions for your " house curve"
ReplyDeleteThanks for the great info. I am tuning my room and your info is helpful
Thanks for the ping. I've updated this post.
ReplyDeleteI’m wondering, isn't better idea is to implement dynamic loudness contour filters instead of room target curves? Human ear perception of lower frequencies at low levels is very weak. In my opinion no target curve can compensate it unless you listen music on certain loudness levels. But to hear sound linearly SPL should be at least 80 db. At this levels human can listen to music for 1h per day maximum. At louder volumes there is possibility to cause hearing damage and permanent hearing loss.
ReplyDeleteFair comment. This is a feature built in to many DSP systems but it doesn't eliminate the need for a target curve. It simply means the chosen target won't vary as much with SPL level.
Delete