Room correction - my case with measurement

[jb90]

Hello!
I made measurements of my spekers in my room with measurement mic and pink noise. The freq response looks like that below. How can I improve sound of my speakers with eq or maybe with placement of the speakers? In graphs like this where is "0db" middle of the road? Should I use particular target response curve to eq speakers? I'm really curious and wanna learn about how to eq any speakers/monitors/PA systems to make them sound great in any place. Every recommendations and tips would be appreciated Thanks!

 

[sejarzo]

Download and use Room EQ Wizard. There are many online tutorials and YouTube videos that will guide you, as REW is so powerful that it seems too complex at first--but you can do a lot without understanding or using all the capability.

As far as your "0 dB" reference, one typically uses a meter or that function in REW or another app to set both channels at 75, 80, or 85 dB (depending on your typical listening volume.) You would then use that specific level as the reference for developing correction filters. REW can help you with that.

 

[jb90]

Download and use Room EQ Wizard. There are many online tutorials and YouTube videos that will guide you, as REW is so powerful that it seems too complex at first--but you can do a lot without understanding or using all the capability.

As far as your "0 dB" reference, one typically uses a meter or that function in REW or another app to set both channels at 75, 80, or 85 dB (depending on your typical listening volume.) You would then use that specific level as the reference for developing correction filters. REW can help you with that.

Thanks. I would try it later for sure. I must borrow mic from my friend to make measurements anyway. But any quick tips how to correct somewhat my freq response without REW just looking at the graph? I listen not so loud. About 60db is the average. 70db at most (I measured it too).

 

[sejarzo]

Thanks. I would try it later for sure. I must borrow mic from my friend to make measurements anyway. But any quick tips how to correct somewhat my freq response without REW just looking at the graph? I listen not so loud. About 60db is the average. Maybe 70db at most (I measured it too).

You haven't posted anything about your room or system. Please don't take this the wrong way--but if you would take the time to learn a bit more about room acoustics on your own first, you would realize how it's not possible to recommend anything with the info you have provided.

 

[staticV3]

I made measurements of my spekers in my room with measurement mic and pink noise. The freq response looks like that below.

Depending on what mode your RTA is set to, the measured frequency response of your speakers will change:

For the Moving Microphone Method with Pink noise, you need to set your RTA to 1/x octave, otherwise the frequency response measurement will be wrong.
I assume you don't know which mode your RTA was set to in the graph above?

 

[jb90]

Depending on what mode your RTA is set to, the measured frequency response of your speakers will change:
View attachment 271210

For the Moving Microphone Method with Pink noise, you need to set your RTA to 1/x octave, otherwise the frequency response measurement will be wrong.
I assume you don't know which mode your RTA was set to in the graph above?

The only available options in this software are:

 

[jb90]

Ok I just applied some eq on what I measured with 7 band eq (lows and highs I use SF). Here are measurements before and after eq (just 1db boost or cut here and there not more!).

 

[No. 5]

The only available options in this software are:
View attachment 271215

Try changing the FFT size to 8192 bins, that should give you better detail at lower frequency. If you are looking for more temporal smoothing, you can change the "desired transform interval" up to 100ms or so.

As with any tool, keep it's limitations in mind. Firstly, with regard to the Spectroid app specifically, the level delineations are at 5dB intervals and it covers a range of 125dB, so changes of 1-2dB are going to be mighty hard to see. Secondly, in regard to in room measurements, they take a number of perceptually distinct acoustic elements and blend them all together into a kind of single average. So you need to determine if any issues that you see are real issues, or are just artifacts of that averaging. And lastly, repeatability is key when it comes to using measurements (and listening test for that matter) to improve any system. If you cannot get consistent measurements, you cannot be sure that what you want to address is being addressed, and you cannot be sure that what you want to address is worth addressing. So for example, can you be absolutely certain that the differences between your before and after screen caps is because of the EQ? Your before has a broad lack of energy centered around 200Hz that catches my attention as something that I would want to fix, but it's not there in the after pic, and there is way more than a 1dB difference in that area, so it doesn't seem like that change was caused by application of EQ.

All that being said, that lack of energy around 200Hz and a slight excess of upper bass are the biggest things I would address based on that measurement... but without knowing if that measurement is repeatable, I really wouldn't hold to that.