REW: In-room response matches anechoic response?

[AudioLimey]

Hello

I've taken my first REW measurements...6 of them, using a UMIK-1, with calibration file. The averaged response, with Physcoacoustic smoothing applied is below. I measured:

• Wharfedale Diamond 9.1 speakers
• Desktop system (inc. speaker stands)
• Nearfield listening ~50 cm from baffles

Below the in-room response is a graph of the anechoic response, source: https://www.stereophile.com/content/wharfedale-diamond-91-loudspeaker-measurements

I'd be very grateful if you could help with these questions please...

1) To my untrained eye, the general shape of the graphs look pretty similar. Because of that, would I be correct to say that the room is not negatively affecting the frequency response?

2) For the anechoic response graph, the majority of the response is between 6 dB and 0 dB i.e. a 6 dB range. For the in-room graph, the majority of the response is within a 13 dB range. Why is the in-room range twice the anechoic range?

3) Would it be beneficial to to EQ the following unevenness?

i) 100 to 150 Hz:

ii) 300 to 400 Hz:

Thanks again

 

[DVDdoug]

A 13dB bump is certainly audible. A 6dB bump MIGHT be OK, especially at lower listening levels and it might help to make-up for the drop-off in deep bass.

It looks like you might want to boost the highs a bit too. Measurements are worthwhile but don't totally ignore your ears, especially if it sounds worse after EQ.

You have to be careful boosting the low bass because you can end-up driving your amplifier or the speaker into distortion and with a ported speaker the woofer movement is unrestrained.

Because of that, would I be correct to say that the room is not negatively affecting the frequency response?

The room always makes a difference, especially at low frequencies. And it depends on speaker & listener location in the room. With nearfield monitoring the room reflections are less significant but it doesn't make as much difference with the low frequency standing waves (longer wavelengths).

 

[AudioLimey]

Hello @DVDdoug . Thank you for your reply.

Are you saying it's the room/speaker positioning that is causing the 7db difference compared to the anechoic response?

Do you think it's worth trying to EQ the unevenness from 300 to 400 Hz?...

300 to 400 Hz:

Thanks

 

[Elkerton]

It is likely that your speakers' positions so close to the corners are boosting the bass even more than the anechoic measurements from 50 to 100 Hz. While it's a lot of work and may not work either aesthetically, nor practically, have you considered placing your TV in one of the corners, and the speakers flanking it. You may find that the one speaker in the middle of the wall (I don't know about the other) will help cancel some of that boost. It may be worthwhile just moving the speakers at first to give your ears some idea.
Happy playing!

 

[AudioLimey]

Hi @Elkerton . I actually like the bass response from the speakers. It's not boomy at all, and actually doesn't sound out of proportion to the rest midrange. Do you have any thoughts on the unevenness from 300 to 400 Hz? Thank you.

 

[Elkerton]

Hmmm, your response suggests you listen at modest volume levels since the bass sounds good to you. We hear differently at different volume levels (check out the Fletcher Munson curves though there are more recent, more accurate ones), needing a boost in the bass at lower volume levels. The wee suck-out you have in the 3 to 4 hundred Hz range could be caused by a lot of things: floor bounce, reflection off the side walls or ceiling. It is in the transition zone (the Schroeder frequency) between how the room affects the response in the bass, which can and should be equalized usually, and what shouldn't be touched by equalization in the higher frequencies. Again, moving the speakers around may improve the dip, but don't hold your breath.

 

[AudioLimey]

Thanks @Elkerton . Yes, I'm only about 50cm from the baffles, so don't need to crank-up the volume. Speakers are on a desk, so I guess the 3-400 unevenness is caused by the desk, despite the speaker stands? I'm a long way from side walls and rear wall.

can and should be equalized

OK, I'll have a look at doing that. Thanks

 

[Elkerton]

My apologies- I didn’t expand your picture to notice that you have a desktop system. You should be able to find suitable parametric equalization online though I don’t know any to recommend. One thing: bring peaks down; trying to boost troughs has little success so scientists like Floyd Toole discovered.

 

[RayDunzl]

3) Would it be beneficial to to EQ the following unevenness?

i) 100 to 150 Hz:

If it sounds "boomy" in the low frequencies, I'd knock them down with EQ.

If you like how it sounds, leave it alone.



I prefer measurable flat across the spectrum, and listen at levels where the old Fletcher Munson curve is rather flat - 80dB spl or so average, 100dB peaks

.

Without EQ mine have a big hump in the lows, too.


MartinLogan request - 12" woofer and 15x48 inch electrostatic panel. with and without room EQ. Psychoacoustic smoothing.
The dip at 48Hz is a room cancellation, but i don't notice it. It disappears to a large degree when the bass is stereo, and changes the phase of the bass between the two speakers.

I noticed the boom, though.

The bass hump is not really the speakers, as a pair of little JBL LSR 308 show a similar room hump and cancellation:

Why flat?

The peak room measurements (red) playing music match the musical peaks (also red) when the source is measured.

Ignore black trace.

Peaks of room and two channels of source overlaid. Nothing sticks out, and not much falls short.

Room or measurement gear has noise down low, so the room level mostly appears higher than the source up to 40hz or so.

That's my story, and I'm sticking with it.

 

[ozzy9832001]

OP, the drop off in the high frequencies is probably because the mic isn't perfectly centered between the 2 speakers. Even an inch will cause a drop off to some degree. It may not even be the speakers response.

Aside from the bump in the bass, which given the overall SPL level, probably does sound good, the rest of the response looks fine. If you want to nit pick, I'd say you'd probably want to EQ down from 100-200 hz to be inline with the rest of the mids, since that area can cause some very unpleasent sounding songs.

 

[AudioLimey]

Hi @ozzy9832001 . Thanks for that - it's helpful.