Speaker placement and Room Correction Advice

[kharan]

View attachment 524549

Would you consider this?

When I used to live in a small apartment, I would have my dining table pushed against the wall. It was fine, I lived alone. But when friends came over, I pushed the listening sofa closer to the speakers and pulled the dining table out. You might have to do the same.

Thank you for the suggestion, speakers on the kitchen island is certainly A choice.

I guess the expensive solution is to close off the kitchen with a new wall, but I don't have that option in practice.

Is there anything you can glean from the measurements? I think my room is still a little too lively and it can get messy on some tracks.

As the C8C are more or less omni in the subbass it might be interesting to simulate the room and move speaker and listener around a bit with https://www.vesalaasanen.com/tools/room-mode-calculator
There is a thread here, too.

Room simulator for non-shoebox rooms

Couldn't find a room mode simulator that handled anything other than shoebox rooms, so I made one. Been using it for my own projects, but decided to throw it online. https://www.vesalaasanen.com/tools/room-mode-calculator Rectangular rooms do the usual analytical mode calculation. The more...

www.audiosciencereview.com

Didn't know about this website, thanks.

Tried to get as close as reasonably possible to my room. The results don't quite match, I assume that partially because I have a door leading to the corridor that is open most of the day, and rather thin and undercut for airflow.

Room Mode Calculator

Calculate room modes, standing waves, and resonant frequencies for any room shape. Free online acoustics tool with pressure mapping and source optimization.

www.vesalaasanen.com

Dimensions are more or less correct, do you have any takeaway from this?

Placing a loudspeaker in front of a window is a compromise to be sure, but one that lots of people accept.

That's fair. I considered the option but this window is the main light source in the room and sunlight enters the living room from that it. Moving some things around, I did end up with 2.2 m between the acoustical centres of my speaker and 2.75 m from the acoustical centres to my listening position.

Initially I thought the same, but a closer look and looking at it by decade, below 100 Hz is highest, 100-1000 Hz is a bit lower, and 1000-10000 Hz is lower again. Pretty good! If @kharan smoothed his response curves to half octave or even full octave, it might look quite sweet.

Also, looking at the EQ above the bass, there is already a couple of dB of lift through the midrange, and the treble is left relatively untouched. I wouldn't want to increase that for fear of changing the direct / anechoic sound quality, which I believe is excellent for those speakers. Remembering that direct sound FR is the primary component of perceived sound quality. For speakers that are naturally excellent in that regard, it is highly desirable to minimise its EQ above the bass.

@kharan if you are finding it a bit bright consistently, with excellent recordings, then (with such excellent speakers) I would suggest that you 'soften the room' a bit with absorption, such as fatter couches, heavier curtains (even where there is no window!), and dense carpeting/rugs.

Yes, I'm thinking of more things to dampen the room a bit, the question is what. Another thing is that I switched to the Harman Curve with is more bass heavy that the AcoustiX curve, and it is more pleasing for the moment, but I'm also losing some of the head room.

In general it's not a good idea to boost dips in the frequency response. It might conceivably be done in a very few cases, like when a speaker is located near the null of a room mode. But your speakers are located in corners, so they will excite the modes maximally. The dip is likely due to the listening position being located in a null.

How does the L+R measurement look like? Use Var smoothing to see full resolution and how narrow the dip at 50 Hz really is? Is it audible? It could possibly the null of the lowest lying mode in the width direction. If so, try to move the MLP a little to the side, and see if it makes a difference. It may sacrifice exact stereo symmetry, but if you gain a smoother bass it might be an acceptable tradeoff.

Here you go, left channel first, Raw, with AcoustiX curve, with Harman curve, with L+R Harman, and then Harman left at a higher SPL 90 dB with pink noise. With Var Smoothing.

Then right channel with the same.

I'm not sure what's happening with the treble L+R measurement exhibits a dip, is that typical, some kind of destructive interference?

 

[staticV3]

I'm not sure what's happening with the treble L+R measurement exhibits a dip, is that typical, some kind of destructive interference?

Correlated L+R measurements can result in comb filtering in the treble with treble roll-off as the result.

More here:

Need help with in-room measurements

Can anyone please tell me how these measurements look? This was done from a stationary mic from listening position (Beyerdynamic MM-1 with 0 degree cal file.) I have a custom setup with two different size subwoofers next to each speaker and am using crossovers to make it work. I played with the...

www.audiosciencereview.com

Use REW Beta version with "Uncorrelated" option enabled in the Signal generator to fix it.

OTOH, music is usually correlated at bass frequencies, so measuring L+R Uncorrelated can give false bass response.

 

[Paul7777x]

I don't think the 50Hz null is a huge deal as it is quite narrow. It's due to some asymmetry at the seating position which I think your drawing clarifies. I do think the resulting curve is a bit flat, I'd aim for some downward tilt from bass to treble. Post EQ looks quite nice to be honest but it's possibly a bit bright.

I agree with the above.

Personally I’d buy a MiniDSP flex or DDRC24 and use REW, either with or without Dirac.

 

[kharan]

Correlated L+R measurements can result in comb filtering in the treble with treble roll-off as the result.

More here:

Need help with in-room measurements

Can anyone please tell me how these measurements look? This was done from a stationary mic from listening position (Beyerdynamic MM-1 with 0 degree cal file.) I have a custom setup with two different size subwoofers next to each speaker and am using crossovers to make it work. I played with the...

www.audiosciencereview.com

Use REW Beta version with "Uncorrelated" option enabled in the Signal generator to fix it.

OTOH, music is usually correlated at bass frequencies, so measuring L+R Uncorrelated can give false bass response.

Yes that's what I was suspecting.

 

[olieb]

Dimensions are more or less correct, do you have any takeaway from this?

I do not know, what kind of difference between simulation and measurement you see, but you can try to simulate a door with an approach like this. Maybe you can get closer? Probably it's hit or miss.
EDIT: modified link
Copy and paste of the link works for me.
[COLOR=rgb(41, 105, 176)]https://www.vesalaasanen.com/tools/room-mode-calculator#poly,2.50,-4.50,5.50,-5.75,4.25,-5.75,2.00,-11.37,6.54,-13.77,2.72,-12.37,-0.59,-5.75,1.00,-5.75,-0.25,-2.00,-0.25,-2.00,1.75,-1.25,1.75,-1.25,7.25,-4.50,7.25|s,-4.00,7.00,0.00,0.0,0.0,1,1|s,-1.81,7.06,0.00,0.0,0.0,1,1|l,-2.90,5.20,0.96|t23|a0.11[/COLOR]

I placed the listener (significantly) closer which seems to help (in the simulation).
More absorption will help too, of course, but below 100Hz that is probably not realistic.
There is a "problem" around 70Hz coming from a longitudinal mode. Difficult to get rid of.

 

[Mr. Widget]

Is there anything you can glean from the measurements? I think my room is still a little too lively and it can get messy on some tracks.

I recently moved and my new living room/listening room is also quite lively. Before I dive into measurements, EQ, and other gyrations, I am working with my wife to find as many aesthetically pleasing elements of decor that I can use to deaden the room and bring in diffusion. I haven't even powered up the system yet... you can get a pretty good feeling about a room and have a pretty good idea if a room is going to be a stinker or will play nicely by carefully listening to the room.

Once you have a decent baseline, you can dial in the best location for the speakers, listening position, and EQ settings.

As we have been adding design appropriate items to the room it has been getting better and better. Should be able to power up the speakers and grab the mic stand by the weekend.

 

[janbth]

Here you go, left channel first, Raw, with AcoustiX curve, with Harman curve, with L+R Harman, and then Harman left at a higher SPL 90 dB with pink noise. With Var Smoothing.

Then right channel with the same.

I'm not sure what's happening with the treble L+R measurement exhibits a dip, is that typical, some kind of destructive interference?
View attachment 524821
View attachment 524820

L+R looks pretty good as far as I can see - disregarding the above mentioned issues with treble. How does it sound? If you don’t have any audible problems in the bass region, I wouldn’t touch this!

 

[Keith_W]

I'm not sure what's happening with the treble L+R measurement exhibits a dip, is that typical, some kind of destructive interference?

It's going to be a situation similar to this.

I have said "post your MDAT" so many times on ASR now that I don't bother any more. People don't know where to look to get their answers, and without the MDAT, we can't show you where to look.

 

[kharan]

I do not know, what kind of difference between simulation and measurement you see, but you can try to simulate a door with an approach like this. Maybe you can get closer?

Room Mode Calculator

Calculate room modes, standing waves, and resonant frequencies for any room shape. Free online acoustics tool with pressure mapping and source optimization.

www.vesalaasanen.com

I placed the listener (significantly) closer which seems to help (in the simulation).
More absorption will help too, of course, but below 100Hz that is probably not realistic.
There is a "problem" around 70Hz coming from a longitudinal mode. Difficult to get rid of.

I'm not sure what I'm supposed to see with your link, it links to a 4*5 m generic room. Also, I don't see a way to add a door to the calculator but it's possible I missed it.

L+R looks pretty good as far as I can see - disregarding the above mentioned issues with treble. How does it sound? If you don’t have any audible problems in the bass region, I wouldn’t touch this!

TBH, since there was no issue with the separate left and right measurements, I suspected some destructive interference from the start with correlated treble, which almost never happens with real musical content.

It's going to be a situation similar to this.

I have said "post your MDAT" so many times on ASR now that I don't bother any more. People don't know where to look to get their answers, and without the MDAT, we can't show you where to look.

There is an MDAT in the very first post. I made sure to include it from the start. There treble thing with L+R sweep is more of an idle curiosity, but as I mentioned I suspected a measurement artifact rather than something reflective of the behaviour of the speakers or the room since there is nothing showing in the separate left and right measurements.

 

[Fslag]

 

[olieb]

I'm not sure what I'm supposed to see with your link, it links to a 4*5 m generic room. Also, I don't see a way to add a door to the calculator but it's possible I missed it.

I guess it is me who missed it. I thought I could post a link with room information in it. I should have checked, my bad.
How did you do it? I would like to update my link.

In the meantime here is what I was talking about as screenshots.
Left is the pressure map at 71Hz and right is the FR for this geometry/positions

 

[marin.weigel]

I suspected some destructive interference from the start with correlated treble, which almost never happens with real musical content.

Haven't REW installed at my work PC to check into it, but if you look at the impulse response of the measurement with both channels driven, most likely there'll be a slight time delay between left and right speaker's first arrival at the single small microphone diaphragm, causing just that dip in the high frequencies - and the larger the delay, the lower in frequency the resulting comb filter effect will kick in and become visible as such in the graph with low enough averaging.
But ...
... as one's listening with two spaced ears and an auditory cortex, trained to deal with comb filter effects as an additional cue to decipher sound sources' positions in 3D-space, the effect may audibly affect only the perception of soundstaging, but not tonality.

 

[kharan]

I guess it is me who missed it. I thought I could post a link with room information in it. I should have checked, my bad.
How did you do it? I would like to update my link.

In the meantime here is what I was talking about as screenshots.
Left is the pressure map at 71Hz and right is the FR for this geometry/positions
View attachment 525047 View attachment 525046

I just copied the link in the address bar and checked that it worked in another browser tab before posting it here.

For example, this should work.

Room Mode Calculator

Calculate room modes, standing waves, and resonant frequencies for any room shape. Free online acoustics tool with pressure mapping and source optimization.

www.vesalaasanen.com

View attachment 525004

I'll have to try later, with a lounge chair and see if I can move the couch elsewhere rather than backwards or forwards

Haven't REW installed at my work PC to check into it, but if you look at the impulse response of the measurement with both channels driven, most likely there'll be a slight time delay between left and right speaker's first arrival at the single small microphone diaphragm, causing just that dip in the high frequencies - and the larger the delay, the lower in frequency the resulting comb filter effect will kick in and become visible as such in the graph with low enough averaging.
But ...
... as one's listening with two spaced ears and an auditory cortex, trained to deal with comb filter effects as an additional cue to decipher sound sources' positions in 3D-space, the effect may audibly affect only the perception of soundstaging, but not tonality.

Yes, that dip disappeared as soon as I moved the microphone by 2 cm laterally.