• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is this a good way to digitally correct a room?

kokoko3k

Member
Joined
Nov 29, 2021
Messages
8
Likes
4
Hi there,
Last weekend I played with my speakers and my pc to correct the room.
Sound was very booming with several peaks at, 110hz, 220hz and others (I'm speaking of peaks/deeps in the order of 12dbs!I don't know if this is common), i use no sub, but just a couple of Edifier R1900TII, supposed to have a good flat response.
By now i decided to correct just the basses until 600hz, because i used a good microphone (Samson SAQ7 Cardioid Dynamic Microphone), but not a measurement one (waiting for one to be shipped)
What i did was to place the mic in the listening position and keep playing a continuos white noise with audacity and analyzing the responce with REW set to smoothing=1/6 and a large window.
Initially i tried to use rew itself to generate the equalizer settings, but the results was very poor to my ears.
So i decided to do things manually, being on linux, by using:
https://lsp-plug.in/?page=manuals&section=para_equalizer_x16_stereo , and keep changing the equalizer until i saw the rew analysis as flat as possible.
The good thing of software equalizer i used is that it supports several Bell filter modes (RLC,BWC,LRX,APO,BT,MT,DR) that REW doesn't support and that i found absolutely necessary to correct the signal.
I've to say that my edifier r1900tii never sounded so well, and, to my surprise, even changing the listening position alot, produced a better sound than before.
Now I can hear some resonance in the 1k..2k range never heard before, but maybe it is just because the sound was so bad before that now i can focus on less imoprtant "details", i didn't touched that range because i plan to do that work when the berhinger ecm8000 arrives.
I didn't expected to be able to do such improvements with a mic that is targeted to human voice, nor, as said, i expected that i can freely walk in the room and have good sounding music.

Is because i didn't go the convolution way that the listening position is wider?
Are there hidden problems i not yet faced with the approach i used?

Thanks!
 
Last edited:

Jim Matthews

Major Contributor
Forum Donor
Joined
Mar 25, 2021
Messages
1,051
Likes
1,286
Location
Taxachusetts
Search this sub forum.

Amir posted the basics, as a series of steps.

 
OP
K

kokoko3k

Member
Joined
Nov 29, 2021
Messages
8
Likes
4
Hi Jim, thanks for answering, i already know about the standard way of using REW, but I was asking about something more (see the relation between the listening position and the convolution, the way REW is able to create equalization parameters and the fact that i used a white noise versus the standard way of using sweeps, because i did the work manually and in real time).
I apologize if those are infos already mentioned in the thread you pointed, but still i'm unable to get the answers by myself.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,204
Likes
16,986
Location
Riverview FL
What i did was to place the mic in the listening position and keep playing a continuos white noise with audacity and analyzing the responce with REW set to smoothing=1/6 and a large window.

Pink Noise might be more appropriate for ear-balling the preferred EQ using the RTA window.

Use the RTA window and averaging with pink, or Spectrum if using white noise, for an indicated flat response.

Pink noise will slope down with Spectrum display, White noise will slope upwards with RTA averaging display, confusing your attempts to flatten the sound.
 
Last edited:
OP
K

kokoko3k

Member
Joined
Nov 29, 2021
Messages
8
Likes
4
Pink Noise might be more appropriate for ear-balling the preferred EQ using the RTA window.

Use the RTA window and averaging with pink, or Spectrum if using white noise, for an indicated flat response.

Pink noise will slope down with Spectrum display, White noise will slope upwards with RTA averaging display, confusing your attempts to flatten the sound.
I suppose there would be not much differences when dealing with peaks or dips, right?
But it is definitely good to know, thanks.
 

Jim Matthews

Major Contributor
Forum Donor
Joined
Mar 25, 2021
Messages
1,051
Likes
1,286
Location
Taxachusetts
Hi Jim, thanks for answering, i already know about the standard way of using REW, but I was asking about something more (see the relation between the listening position and the convolution, the way REW is able to create equalization parameters and the fact that i used a white noise versus the standard way of using sweeps, because i did the work manually and in real time).
I apologize if those are infos already mentioned in the thread you pointed, but still i'm unable to get the answers by myself.
You already know more than I do.

As a freshly minted user, it's difficult to gauge experience levels.

As Ray is pointing out there's a proscribed method for getting best results from tools like REW, and that follows an established playbook - without hacks or shortcuts.

One thing I will say is that the fundamentals of room acoustics (size limitations, wall construction, diffusion from furniture and floor coverings) cannot yet be overridden entirely by equalization.

Most of Amir's favorable speaker reviews have limited "tweaks" to the response - often producing (audibly) acceptable results that don't fall directly along ideal response curves.
 

Willem

Major Contributor
Joined
Jan 8, 2019
Messages
3,659
Likes
5,277
And do realize that yo may be dealing with two diferent kinds of equalization tasks: equalizing the speakers, i.e. flattening their response irrespective of the room, and equalizing to compensate for the room.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,204
Likes
16,986
Location
Riverview FL
I suppose there would be not much differences when dealing with peaks or dips, right?
But it is definitely good to know, thanks.

If you account for the differences in slope, yes, I suppose.

It's your choice as to how you use your tools for your purposes.



The_Colors_of_Noise.png
 
OP
K

kokoko3k

Member
Joined
Nov 29, 2021
Messages
8
Likes
4
And do realize that yo may be dealing with two diferent kinds of equalization tasks: equalizing the speakers, i.e. flattening their response irrespective of the room, and equalizing to compensate for the room.

Won't equalizing the resulting signal to be as flat as possible address both speakers and room in a single shot or maybe there is the need to do it separately?
I mean... I don't plan to use the resulting equalizer settings in the same room with different speakers ofc
 
Top Bottom