Room EQ Measurements - what next - Guidance needed

[Satanica]

18 minutes ago, JkSpinner said:

I thought, (through my research that Dirac is what I would look at after room treatment, like the final step.  And as far as I know, I would need a new amp, and I am not sure that is on the cards at this stage, but who knows what the future holds

 

When it comes to bass frequencies, they are known to be very hard to trap. And if even possible, require enormous devices. Digital room correction should be so much more targeted and effective.

 

Dirac can be configured to only correct a subset of frequencies. E.g. I configure it to only correct from about 200hz and down. 

Edited August 12, 2024 by Satanica

[JkSpinner]

3 hours ago, frednork said:

revisit the bass

I agree, maybe move the speaker closer to wall may improve this a little, 

I am thinking a step by step approach, and I would not have put bass at the top of the priorities prior to measuring, so your suggestion after measuring works with that

[JkSpinner]

4 hours ago, Keith_W said:

If you don't listen from that position

I agree with not worrying about it, however I feel if I can hear the issues when standing in the entrance, I can also hear the issues from the listening position, just a hunch.  But I do take your point

[JkSpinner]

3 hours ago, frednork said:

panel next to the couch

Hi Fred, it’s a privacy screen, we put it there to give us some privacy from people walking in the street, they can look straight into our lounge.  Nothing to do with sound.  Although, I would not mind doing something different now to help improve the sound.  I would like to replace the quad sliding door also, we do not use it, replace it with two one metre windows or something, would give me some more wall space to put staff against, and probably help my SQ

 

21 minutes ago, Satanica said:

you are thinking

 

22 minutes ago, Satanica said:

thinking about the idea.

Hi, I am thinking about many of your comments.  Thinking I may put up some absorption panels, 2 or 4 to bring the highs down a little, and move the speakers back the length of the speaker, to where they use to be 6 months ago, most of my mates use to comment on how bassy it was, so two things at once, then remeasure.

[Satanica]

31 minutes ago, JkSpinner said:

Hi, I am thinking about many of your comments.  Thinking I may put up some absorption panels, 2 or 4 to bring the highs down a little, and move the speakers back the length of the speaker, to where they use to be 6 months ago, most of my mates use to comment on how bassy it was, so two things at once, then remeasure.

I don't know enough about room treatments. But from what I've read from respected sources, you need to be careful as you can easily go backwards. The easiest example of this are treatments that effect the wrong frequencies that should be targeted and/or effect the right ones too much. I put curtains up in my previous house and it reduced the mids an highs too much, while doing nothing to bass frequencies and this caused and overall inbalance. Bottom line is, be scared to put in room treatments and thus make sure that if you put some in, you absolutely have a good idea what is going to happen. Consider a professional, once again, you want a good idea what is going to happen before it does.

[frednork]

You can think of the absorption and RT60 as mostly a separate thing to the bass. regardless of where you put the speakers your RT 60 will still benefit from some absorption. It may affect the bass a little but it will have a much bigger impact on everything else.  And yes if the entrance is audible than a small amount of absorption will help alot.

 

Moving speakers back and forth will affect the bass  as closer to the wall you will get more reinforcement. What comes with that is usually  greater variation in the evenness of bass so then big traps or dsp may be required.  With some broad absorption the ringing in the top end will reduce and it wont measure or sound as top heavy.  Of course dont forget aspects like soundstage width and depth.  

 

@Satanica is correct in that if you apply a lot of absorption which does not match the problem you are trying to solve then it can easily make it worse but if you are measuring you will see where you came from and where  you are at after treatment and can make some informed decisions on the effectiveness of your particular approach. 

[Keith_W]

1 hour ago, JkSpinner said:

Would moving the speaker back closer to the front wall, or closer to the side wall help, or  increasing bass on the tone controls?

 

Peaks and dips in your room will be due to one of three things: (1) room modes. This is related to room dimension, speaker position, and listener position; (2) Speaker Boundary Interference Response / SBIR. This is related to the distance of the speaker to the wall and listener position; and (3) crossover phase cancellation which is independent of the room and listener position. 

 

The outcome can be predicted theoretically if you have a rectangular room. If that were the case, I would simply direct you to REW's room simulator and look at the prediction. It is usually pretty close to reality. But you don't have a rectangular room ... so, who knows what will happen. 

 

Pushing speakers closer to the wall will shift the SBIR cancellation frequency upwards (good) but also cause forward arriving reflections to arrive earlier and with almost equal loudness to the direct signal (bad). This is one reason why subwoofers are so advantageous, you can place your speakers where they are optimal for imaging, and place your subs where they are optimal for bass. 

 

1 hour ago, JkSpinner said:

I will read this again, and have a further look at my charts.  My idea around the first reflection point is to Create them - which will prevent some of the sound from leaving the room,  might help with the leaky room, and also create some symmetry, as my speakers are different distance from each side wall.  Using diffusion may also help with the dead or flat room.  Just a guess here.

 

IMO creating a first reflection point is not such a bad idea. At present the reflections from the left and right side of your room is asymmetrical and you have a "bigger" room on the right side. 

[JkSpinner]

7 hours ago, Satanica said:

be scared

Hi Paul, I absolutely agree, I never really wanted to measure, I already thought I had a good idea of what was needed, I live in the room, I spend countless hours walking the room and listening.  The measuring has really only backed that up, although provided some valuable feedback on issues I did not really know about.

i plan on a step by step method, starting with 2 absorption panels on the back wall, and maybe 2 diffusion panels at first reflection point, + move me speakers 15cm closer to the wall in increments. Then remeasure.  Of, and try and understand some of these graphs, so I understand the results.

[JkSpinner]

7 hours ago, Keith_W said:

IMO creating a first reflection point is not such a bad idea. At present the reflections from the left and right side of your room is asymmetrical and you have a "bigger" room on the right side. 

I am stuck on this one Keith, I mentioned in another response, that I live and breath this room, I walk this rooms boundaries constantly trying to understand what is happening.  I may put the 600mm wide diffusion panels on wider 1.2m office divides  to to create a boundary, similar to a wall - who knows

[JkSpinner]

7 hours ago, Keith_W said:

Speaker Boundary Interference Response

I have been moving my speakers around of late, moved them even further out then the are now, on guidance from people on this forum, but have been slowly moving them back, mainly due to lack of bass, and hollowness of sound.  Where they are now is a good balance, although as the results have indicated, I to think it is lacking in some bass.  You mentioned SBIR.  After I set my 2 channel up, I then re set my 7 channel up, but when using Audesey it continued to tell me one of my front speakers is out of phase.  It’s not.  I have read this can be caused by SBIR

[JkSpinner]

7 hours ago, Keith_W said:

subwoofers are so advantageous

I will play around with speaker placement and minimal treatments, then if not satisfied, will look at subwoofers.  I have moved items around on the front wall, so probably have room now.  I HT subwoofer use to be beside the couch

[Satanica]

4 hours ago, JkSpinner said:

I never really wanted to measure, I already thought I had a good idea of what was needed, I live in the room, I spend countless hours walking the room and listening.  The measuring has really only backed that up, although provided some valuable feedback on issues I did not really know about.

i plan on a step by step method, starting with 2 absorption panels on the back wall, and maybe 2 diffusion panels at first reflection point, + move me speakers 15cm closer to the wall in increments. Then remeasure.  Of, and try and understand some of these graphs, so I understand the results.

 

What microphone, are you measuring with?

UMIK-1, UMIK-2 or something else?

Are you using a calibration file?

Are you pointing the microphone at the speakers or up in the air?

What holds the microphone in place?

 

Just my opinion, but I think for the time being you should ignore the graphs above about 300Hz.

As it has been mentioned, you need to interpret the measurements of different frequencies differently.

Below about 200Hz, there is good correlation between what is measured and what is heard and this is one of the reasons that equalisation is so effective in this region.

But, as we move up in frequency, the correlation is not anywhere as close, so it gets harder and harder to correlate the measurements to what is actually perceived by our brain.
Remembering this, what is measured by the microphone and what is perceived are not the same thing, especially above about 200Hz.

The microphone captures everything, and then everything is stored on the computer.

Ours ears capture everything (well sort of), but our brain doesn't perceive all of it and/or does so differently (psychoacoustics).

Really high frequencies (above about 10kHz) where a speaker typically beams the sound rather than spread it out evenly and where reflections are not as strong, the correlation between measurement and perception is going to be stronger (I think).
This leads me to another question, what are your speakers?
The reason I ask, is if one knows the dispersion characteristics of a speaker, this information should influence positioning in a particular room, somewhat.

 

Anyway, I think you should concentrate on bass first and if you do change your bass, your perception of mids and highs is going to change (psychoacoustics).

For example, if your bass becomes louder, then you are likely to not turn the volume up as much to get "that" bass and so now without directly changing them, you have actually changed the mids and highs because now there will be less of them.

So, from this new bassline you can then listen again and again taking note.

Just a reminder, when it comes to speaker position, be weary about sacrificing the quality of mids and highs for better bass.
Better bass can be achieved with equalisation.

Edited August 13, 2024 by Satanica

[JkSpinner]

17 minutes ago, Satanica said:

What microphone, are you measuring with?

UMIK-1, UMIK-2 or something else?

Are you using a calibration file?

Are you pointing the microphone at the speakers or up in the air?

What holds the microphone in place?

What microphone, are you measuring with?

UMIK-1,

Are you using a calibration file?

yes

Are you pointing the microphone at the speakers or up in the air?

up in the air

What holds the microphone in place?

Small tripod which sits on top of lounge, right about ear height.

Speakers - Paradigm Tribute Floor standers 30th Anniversary 

https://www.paradigm.com/en/floorstanding/tribute

I have already moved the speakers back a little, about 30cm, the front of the speaker is still about 97cm from the wall, which is about 1/7 the length of the room.  It is approx where they use to be before I started moving all my stuff around from room to room, trying to separate 2 channel from HT.

thanks Paul

[Keith_W]

Paul was very astute and asked incredibly relevant questions! I had forgotten that most people use those crappy microphone tripods that come with the UMIK-1. 

 

When you take measurements, you have to take precautions to make sure that your measurements are not contaminated by reflections. That is, unless it is your intention to measure those reflections. To put it simply, we window out reflections to look at a speaker's frequency response above the Schroder frequency, and we measure the combined result of room reflections + speaker output below the Schroder frequency. If you use that mini tripod, you do not have adequate clearance from reflective surfaces so the reflections arrive early and are impossible to window out. 

 

Let's say for example that your microphone is 10cm from a reflective surface. A 10cm wavelength is 3430Hz. This means that only measurements from 3430Hz up can be trusted, anything less than that is meaningless. If your mic is pointing upwards, it will capture even more reflections from anything that is behind it (e.g. your sofa). Point mics upwards if you are measuring surround speakers. But for measuring front speakers, point them forwards. 

 

Whether you need proper measurements depends on your intention. For those of us who DSP, a proper measurement is absolutely critical because we will be correcting it. It is also important if you are designing speakers or publishing reviews. But if all you are doing is taking a look, then you don't have to take as much care. You just have to be aware of the limitations and not read too much into it. 

[Keith_W]

6 hours ago, JkSpinner said:

I am stuck on this one Keith, I mentioned in another response, that I live and breath this room, I walk this rooms boundaries constantly trying to understand what is happening.  I may put the 600mm wide diffusion panels on wider 1.2m office divides  to to create a boundary, similar to a wall - who knows

 

Ears don't hear the same as microphones. Understanding the difference can be the topic of a book. 

 

Re: Audyssey saying your speakers are out of phase. A lot of Dirac and Audyssey users (in another forum) complain of inconsistent results or strange errors. The usual problem is that the measurements were not taken properly, so the software does not know what to do. 

[Satanica]

53 minutes ago, Keith_W said:

Paul was very astute and asked incredibly relevant questions! I had forgotten that most people use those crappy microphone tripods that come with the UMIK-1. 

 

When you take measurements, you have to take precautions to make sure that your measurements are not contaminated by reflections. That is, unless it is your intention to measure those reflections. To put it simply, we window out reflections to look at a speaker's frequency response above the Schroder frequency, and we measure the combined result of room reflections + speaker output below the Schroder frequency. If you use that mini tripod, you do not have adequate clearance from reflective surfaces so the reflections arrive early and are impossible to window out. 

 

Let's say for example that your microphone is 10cm from a reflective surface. A 10cm wavelength is 3430Hz. This means that only measurements from 3430Hz up can be trusted, anything less than that is meaningless. If your mic is pointing upwards, it will capture even more reflections from anything that is behind it (e.g. your sofa). Point mics upwards if you are measuring surround speakers. But for measuring front speakers, point them forwards. 

 

Whether you need proper measurements depends on your intention. For those of us who DSP, a proper measurement is absolutely critical because we will be correcting it. It is also important if you are designing speakers or publishing reviews. But if all you are doing is taking a look, then you don't have to take as much care. You just have to be aware of the limitations and not read too much into it. 

 

Hi Keith, I'm a bit on the fence as to whether to point upwards or towards the speakers.

Dirac recommend upwards using a UMIK-1 as to capture as much reflected sound from around the room and this is what I do.

For the purpose of speaker-room correction as opposed to pseudo-anechoic speaker correction, would you not agree that capturing as much of the room is a good idea?
JKSpinner seems only interested in speaker-room correction and not trying to obtain measurements that separate the speaker from the room.

Edited August 13, 2024 by Satanica

[Keith_W]

1 hour ago, Satanica said:

 

Hi Keith, I'm a bit on the fence as to whether to point upwards or towards the speakers.

Dirac recommend upwards using a UMIK-1 as to capture as much reflected sound from around the room and this is what I do.

For the purpose of speaker-room correction as opposed to pseudo-anechoic speaker correction, would you not agree that capturing as much of the room is a good idea?
JKSpinner seems only interested in speaker-room correction and not trying to obtain measurements that separate the speaker from the room.

 

Take a look at this article. In short: omnidirectional microphones are not perfectly omnidirectional, they show variation in frequency response depending on orientation, mostly in the high frequencies. The least variation and most accurate result is when it is pointed at whatever you are measuring. 

 

As I understand it, JKSpinner wants to understand the behaviour of his room. This is why I am not telling him to get the most accurate measurement possible, because that requires additional effort from him (move furniture, take speakers outside) additional expense (get a proper mic tripod, get a ladder, rent a forklift, etc), and it would mean I would be sitting here typing long instructions on how to do it  

 

This is why I agree with your suggestion of ignoring what is going on > 200Hz because his measurement technique means he won't get accurate results in the high freqs anyway. Totally different matter if he was planning to DSP his high freqs though! 

[Satanica]

34 minutes ago, Keith_W said:

 

Take a look at this article. In short: omnidirectional microphones are not perfectly omnidirectional, they show variation in frequency response depending on orientation, mostly in the high frequencies. The least variation and most accurate result is when it is pointed at whatever you are measuring. 

 

As I understand it, JKSpinner wants to understand the behaviour of his room. This is why I am not telling him to get the most accurate measurement possible, because that requires additional effort from him (move furniture, take speakers outside) additional expense (get a proper mic tripod, get a ladder, rent a forklift, etc), and it would mean I would be sitting here typing long instructions on how to do it  

 

This is why I agree with your suggestion of ignoring what is going on > 200Hz because his measurement technique means he won't get accurate results in the high freqs anyway. Totally different matter if he was planning to DSP his high freqs though! 

 

The UMIK-1 is quite an omnidirectional microphone, more so than the UMIK-2 according to Gene DellaSala of Audioholics and he actually prefers the UMIK-1 because of this.
The UMIK-1 and 2  come with correctional files for both 0 and 90 degrees.

I see no issue with doing a room measurement using 90 degrees, especially as Dirac Live gives recommendation to do so for both two and multi channel for up to full frequency correction.
I've done measurements using my UMIK-1 at 0 and 90 degrees and the differences from memory were quite minimal at all frequencies.

So now that I've thought about it a bit more, I guess I'm no longer on the fence.

Edited August 13, 2024 by Satanica

[Satanica]

Were the measurements performed without subwoofer(s) and without any EQ/Room Correction enganged? Do you currently have a device (AVR, Processor, miniDSP etc.) with EQ/Room Correction (Audyssey, Dirac Live etc) ?

[JkSpinner]

1 hour ago, Satanica said:

Were the measurements performed without subwoofer(s) and without any EQ/Room Correction enganged? Do you currently have a device (AVR, Processor, miniDSP etc.) with EQ/Room Correction (Audyssey, Dirac Live etc) ?

Test performed on 2 channel amp.  
no sub, no EQ or room correction.

i also have a Denom AVR with Audyssey only used for HT duties, along with the sub.

[JkSpinner]

Hi gents

before I take a second measurement, thought I would make sure my settings are correct, can someone guide me through this, or advice if any changes need to be made.  I am assuming that I do not need to calibrate the microphone as I have the calibration file.  I then set up volume etc as per GIK, increase the volume until the meter reads 76DB.

On the preference tab, I am guessing I set everything to either R or L.

is there anything else I need to be aware of or change

 

[Satanica]

It looks OK to me.

[JkSpinner]

A second set of measurements.

speakers moved back about 35cm, now about 1m off the front wall (front of speaker back to wall) moved the speakers out slightly, as I thought a wider soundstage was possible,  right speaker now 650mm from the short side wall.  
speakers are 2.7m apart, 3.1m to the Listening position.

curtains are closed on the left side

bifolds are closed on the rear left side

Edited August 14, 2024 by JkSpinner

[Keith_W]

1 hour ago, JkSpinner said:

 

OK before we begin, here are some pointers. 

 

1. You are on a Windows 11 laptop. Are you aware you can hit the "PrtSc" button to copy what is on your screen to the clipboard? You can click and drag to copy part of a window, or the whole window. 

 

2. Choose sensible zoom levels. By default REW zooms out too much and the graphs become hard to read. It's like taking a photo of your wife when she is standing 100m away. For e.g. this graph is far too zoomed out, so it looks flat. Press the "limits" button, and you can either choose "Fit to data" or manually choose your own limits - in this case I would choose -10dB for the upper limit, and -60dB for the lower limit. Another way to zoom is to Ctrl-right click and drag to draw a box. Click inside the box and it will zoom. 

 

3. What question are you asking? If you are comparing your left vs. right speaker, then show a graph with your left/right speaker on "All SPL" as you have done. But if you are asking what is the difference between "before" and "after", load the curve with "before", and load the graph with "after". Adjust the two curves so that they overlap, and then post that. Clearly label which is before and which is after. 

 

It would be better to post the REW .MDAT file. SNA won't let you do that, so I suggest you upload it to a Google drive or Dropbox account and post the link to the .MDAT.

 

If you want, you can PM me and we can have a chat on WhatsApp and I can guide you through the settings. 

Edited August 14, 2024 by Keith_W

[JkSpinner]

14 hours ago, Keith_W said:

If you want, you can PM me and we can have a chat on WhatsApp

Hi Keith

thank for your comments mate.

i am aware I can print screen,  I just use an IPAD for this forum and most things, and I am over sending emails with attachments, but you have point though, there is a better way I could do this.

i will play around with the graphs settings and zoom, and try and link a google drive link.

what am I asking - Good question.  I posted my initial results several days ago, and several responded that I am lacking bass.  I had recently moved my speakers closer to the listening position, and did notice the reduction in base, so after the initial test results and comments, I moved the speakers back closer to the front wall, and out a little closer to the side wall.  @Satanica suggested some would like to see these results, so I posted the results.  So not really comparing right to left, although the results for right to left differ, more comparing the results comparing this test to the initial measure at the start of this post, seeing if moving the speakers back towards the front wall made a measurable difference.

The left and right comparison, that is another question, and I think I need to create some sort of symmetry (such as creating first reflection points) before comparing this.

 

Edited August 14, 2024 by JkSpinner