Share your in-room measurements?

[YSC]

One Question here for experts, I saw Dirac used to do EQ to target for full spectrum while someone here referenced to researches to EQ only below 200hz regardless of the actual listening position measurement, I wonders which one would sound more "correct" as for studio/ live condition?

also I am using the speakers really near field in desktop config (~0.8m distance from speakers) with a wardrobe door directly behind my chair, which I suppose will give some reflection interference during measurement even for high frequencies (?) any suggestion to what I should do to the EQ tactic? only using Equalization APO so channel independant EQ isn't possible

 

[Eetu]

One Question here for experts, I saw Dirac used to do EQ to target for full spectrum while someone here referenced to researches to EQ only below 200hz regardless of the actual listening position measurement, I wonders which one would sound more "correct" as for studio/ live condition?

also I am using the speakers really near field in desktop config (~0.8m distance from speakers) with a wardrobe door directly behind my chair, which I suppose will give some reflection interference during measurement even for high frequencies (?) any suggestion to what I should do to the EQ tactic? only using Equalization APO so channel independant EQ isn't possible

EQ'ing individual channels is most certainly possible with Equalizer APO. Just select channel L instead of ALL, add Parametric EQ, then below that add another channel R, add PEQ. Then you can see their effect in the graph below flicking between L/R. (Even more channels is possible if you have a multichannel DAC but you likely need something like Voicemeeter to output to.)

My take is that you don't want to EQ whatever the wardrobe door is doing. I would probably EQ the direct sound separately and what I mean by that is that you could take the speakers to a more open space (middle of the room) then measure the speakers ~0.6m away from tweeter and use a gate in REW to filter out first reflections. You could use the gating technique in your current speaker/listening position as well but if the wardrobe door is very close (you have to add the gate before that reflection) you won't get much resolution to work with.. See part 3 of this guide for a walkthrough.

Once you have created EQ filters for both speakers individually, do measurements (sweeps/MMM) again in your normal listening position and then EQ only below ~400Hz. Now you can combine those two sets of EQ settings (room+'speaker' correction), done!

 

[YSC]

EQ'ing individual channels is most certainly possible with Equalizer APO. Just select channel L instead of ALL, add Parametric EQ, then below that add another channel R, add PEQ. Then you can see their effect in the graph below flicking between L/R. (Even more channels is possible if you have a multichannel DAC but you likely need something like Voicemeeter to output to.)

My take is that you don't want to EQ whatever the wardrobe door is doing. I would probably EQ the direct sound separately and what I mean by that is that you could take the speakers to a more open space (middle of the room) then measure the speakers ~0.6m away from tweeter and use a gate in REW to filter out first reflections. You could use the gating technique in your current speaker/listening position as well but if the wardrobe door is very close (you have to add the gate before that reflection) you won't get much resolution to work with.. See part 3 of this guide for a walkthrough.

Once you have created EQ filters for both speakers individually, do measurements (sweeps/MMM) again in your normal listening position and then EQ only below ~400Hz. Now you can combine those two sets of EQ settings (room+'speaker' correction), done!

IC, I think the left channel is affected by the really close stuffs like files put right aside the speaker, so really difficult to do, but with Genelec I am pretty sure it's direct sound should be flat, just wondering if the file right next to the speaker tweeter protruding a bit creating the broad dip at 1-6khz, if so maybe I should leave anything above 400hz untouched?

 

[Eetu]

IC, I think the left channel is affected by the really close stuffs like files put right aside the speaker, so really difficult to do, but with Genelec I am pretty sure it's direct sound should be flat, just wondering if the file right next to the speaker tweeter protruding a bit creating the broad dip at 1-6khz, if so maybe I should leave anything above 400hz untouched?

Yeah certainly possible that the objects next to the speaker can have such an effect but I'm hesitant if EQ'ing it would be an improvement audibly since you are basically making the direct, non-reflected, sound less flat. If it's possible to physically clear the objects away from the speakers that would of course be best.

 

[thewas]

MMM measurements of my original LS50 desktop setup with a bit of EQ (mainly subtractive)

and my LS50 Meta + subwoofer classic couch listening stereo triangle with EQ only below 500 Hz and only subtractive

 

[YSC]

Yeah certainly possible that the objects next to the speaker can have such an effect but I'm hesitant if EQ'ing it would be an improvement audibly since you are basically making the direct, non-reflected, sound less flat. If it's possible to physically clear the objects away from the speakers that would of course be best.

right, maybe I just try EQ lower end, I kind of revisited the correction, if I will do EQ on the channel with the harman curve full range it mostly means lowering the peak around 10khz+, so basically a treble roll off, which I think shouldn't be a really big issue, will try both do the channel correction vs nothing and see if the stereo image change. coz if not then likely the combined reflection of the stuffs beside is what I hear IRL, but if direct sound dominated then I bet it's better to not EQ it as currently the image is very centered without any EQ

 

[bjmsam]

Audyssey OFF vs. both curves toggling Dynamic EQ

Update: To help mitigate influence of aforementioned comb filtering, I re-calibrated Audyssey using five measurements taken laterally across width of seat at MLP, then used REW to measure those same five positions for each of five Audyssey permutations, and averaged each set of five to produce the two graphs below (I also did the same with RTA MMM, but these sweep graphs differentiate more effectively).

Audyssey FLAT with Dynamic EQ at 0 dB Reference Level sounds best to me now, through I have yet to audition or measure the other DEQ Reference Level settings. For now, averaging the five FLAT DEQ OFF measurements with the five FLAT DEQ 0 measurements results in this attractive approximation (+/- 10 dB from 20 Hz - 20 kHz):

According to what Acoustic Research published in 1978, the Teledyne AR-9 measures -3 dB at 28 Hz with the curve for "a typical listening room" depicted below (I am driving them with an Adcom GFA-585).

That said, frequency response is not the problem. Regardless of settings, comb filtering remains annoyingly pronounced when moving my head laterally in the MLP (typically, my right ear is at position 2 while my left ear is at position 3; position 1 is dead center, 4 is far right, 5 is far left). I believe it is due to primary reflections off of the "acoustic" ceiling tiles so will try removing some from the grid temporarily to simulate absorption while investigating options for diffusion. Thoughts?

 

[YSC]

right, maybe I just try EQ lower end, I kind of revisited the correction, if I will do EQ on the channel with the harman curve full range it mostly means lowering the peak around 10khz+, so basically a treble roll off, which I think shouldn't be a really big issue, will try both do the channel correction vs nothing and see if the stereo image change. coz if not then likely the combined reflection of the stuffs beside is what I hear IRL, but if direct sound dominated then I bet it's better to not EQ it as currently the image is very centered without any EQ

strange that after I applies the EQ and use REW to remeasure it nothing shown changed, the driver in REW is JAVA as default

 

[Deleted member 43441]

My room is 16‘ x 26‘ x 7‘ high. I know the ceiling is not optimal. I just started from scratch in another house a little over a year ago, and I’m so used to the sound… and I’m afraid room correction is going to ruin what I have. I run three subwoofers in my two channel set up, and my home theater, two 0° phase and two 180° phase so they don’t cancel each other out. I have run Audyssey XT32 from my Marantz pre-pro in my other residence and I was not especially happy about the results for two channel. So I decided to do a little EQ adjusting within the preamp and I like this a lot better.

Also in my room I could not get the bass to sound good with my limited placement capabilities. Doorways, lack of space on the front wall, lack of space on the side walls. So I went with a nearfield set up which has incredible base in the MLP and all across the seated positions.

I also have different preamps that I have on an XLR switcher. My preamp is the now 9 year old Marantz AV8801, and it’s a keeper, I’ve seen some measurements and it seems to measure well, and actually one of the last ones were Marantz actually posted specs.

So my question is… Are my fears of running my room correction unfounded?

Will room correction always sound better for two channel?

I need to ask this because the last time I ran it it was so opposite that I couldn’t get used to it, or never gave it a chance.

Edit;

I wanted to add that I have base traps in the four corners of the room, and some wall treatments and my room is not exactly what you would call lively, and it’s also not dead sounding.

 

[Flageborg]

Will room correction always sound better for two channel?

Yes, but a picture of your room makes it easier for suggestions...

 

[Deleted member 43441]

Yes, but a picture of your room makes it easier for suggestions...

My room is a little bit of a nightmare, including a large fireplace. One of these days I will post a picture, but basically it’s 16’x26’ with 7 foot high ceilings. I know the ceilings are not high enough but it is what it is. Right now I’m happy with my sound, and as soon as I get done treating the walls some more i’m going to run XT32 and see how it does. Although I was happy with its results for home theatre in a different home, I was never happy with the results for two channel audio. So now I have it disabled, and I just use the EQ in my Marantz to shape the sound how I want. With my size room I realize that at the MLP there’s a lot of midbass, and the bass traps do help. All this being said I’m not to the point where I can run any room correction yet, and although my electronics are in place as are my speakers, my wall treatments are not done.

 

[DWPress]

Here's a pic of my current filter implementation. MMM method, unsmoothed = +/- 8dB from 30 to 20K. Could be a little better.

 

[Deleted member 43441]

This is a very helpful thread, and I definitely will be reading.

 

[bjmsam]

I have yet to audition or measure the other DEQ Reference Level settings.

I just measured (as above, averaging the same five positions for each Audyssey permutation).

That seems to support my auditioning last night (Hans Zimmer Live in Prague), when ultimately I preferred the 10 dB reference level offset (bass) and the FLAT curve (treble) represented by the purple trace below.

Measurements were made with the AVR volume at -25.0 dB and the REW level check reporting -25.7 dBFS / 66 dB SPL.

Now if I can just eliminate the comb filtering...

 

[bjmsam]

That said, frequency response is not the problem. Regardless of settings, comb filtering remains annoyingly pronounced when moving my head laterally in the MLP (typically, my right ear is at position 2 while my left ear is at position 3; position 1 is dead center, 4 is far right, 5 is far left). I believe it is due to primary reflections off of the "acoustic" ceiling tiles so will try removing some from the grid temporarily to simulate absorption while investigating options for diffusion. Thoughts?

After reading the Comb Filter Effects thread and reviewing the Impulse Responses page in the REW documentation and being inspired by the string method, this morning I measured L and R separately from the MLP, plotted the results in the Impulse window, and used CTRL - Right Mouse Button to estimate the primary reflection distance to confirm that the ceiling indeed is at fault. I'm finding few options for drop ceiling diffusion so most likely will end up adding more absorption to a room that already is pretty dead...

Physical measurements:
50” from ceiling to between AR9 tweeter and upper-midrange dome
50” from ceiling to tip of Audyssey microphone
180” from driver to mic

206" - 180" = 26" = 2' 2"

I'm chalking up the 1" discrepancy to tape measure sag!

 

[ernestcarl]

After reading the Comb Filter Effects thread and reviewing the Impulse Responses page in the REW documentation and being inspired by the string method, this morning I measured L and R separately from the MLP, plotted the results in the Impulse window, and used CTRL - Right Mouse Button to estimate the primary reflection distance to confirm that the ceiling indeed is at fault. I'm finding few options for drop ceiling diffusion so most likely will end up adding more absorption to a room that already is pretty dead...

View attachment 185649

View attachment 185650

Physical measurements:
50” from ceiling to between AR9 tweeter and upper-midrange dome
50” from ceiling to tip of Audyssey microphone
180” from driver to mic

View attachment 185651
206" - 180" = 26" = 2' 2"

I'm chalking up the 1" discrepancy to tape measure sag!

If you have any spare acoustic panels very absorbent blanket around, you might want to measure while temporarily propping it to cover the suspected problem boundary(ies) first. For a ceiling on the other hand... eh, that is admittedly harder to do.

 

[rdenney]

I think I would look for diffusion rather than absorption in that case. With your drop ceiling, try making up tiles with inverted pyramids on them to deflect the sound to the side instead of reflecting it directly to the LP. That has been one strategy for me to kill echo without killing reverberation. But it requires a space used only for sound, because giant clos d'Paris on the ceiling has a low spousal approval probability (SAP).

(I know this because I pitched the idea to the Redhead for one wall of our living room. She thought I was joking, and such was the response that I thought it politic to pretend I was joking.)

Rick "the picture indicates such a space, however" Denney

 

[TurtlePaul]

Here is where I am, have manually placed six filters so far. Sounds much much better than uncorrected. Primarily am fixing a 72 hz bass mode and a 140 hz suckout which has dual peaks caused by the back wall and the TV console/coffee table. The highest filter is 335 hz.

Equipment is:
Philharmonic Affordable Accuracy (latest version)
Yamaha Receiver HTR-5540 (from 2002)
Hsu STF-2 (budget 10” sub from 2003)
The receiver is doing bass mgmt at 90 hz, with everything else done in EAPO.

 

[YSC]

figured out previously the EQ APO was wrongly configured, updated post EQ to harman target curve:

Green: Left, Purple: Right

6 measurement MMM method averaged

seems not bad to my ears, in an un treated room with stuffs littering around this is as good as I can go

 

[cottonballs]

Would it be useful to have a forum thread where we can gather people's in-room frequency response measurements and discuss them? Those who did something to improve their sound based on measurements might also describe what they did and show the results. I think this could be quite interesting. What do you think?

I guess it will be really stupid of me if I share my tiny room measurments here in this forum. But here is what I noticed. I have two 20 inches wooden speakers that were standing facing the longer side of the room. And the sound was harsh and quite loud. Yesterday I moved them up to the shelf, facing the shorter side of the room. So the sound improved greatly. I wonder if this makes sense to the point of your discussion. But I am planning to lay these speakers on the back facing the ceiling and see what happens.