What improvements can i make to my process? 2 channel audio w/ UMIK-1

[pimogo]

Hello All ! New to the community and looking for help to fine tune the process below which I've undertaken after a bit of research both in the forums and on YouTube. I plugged settings I've seen others use but I'm not entirely certain if these settings are best for my specific case. I should also state that my objective is primarily to address any room issues. I have some bass traps behind each of the speakers, but nothing along the walls to address early reflection, etc.. Hoping i can do some EQ magic for now.

My equipment:

1. Yamaha A-S1200
2. Cambridge CXN v2
3. Forte IVs
4. Umik-1
5. REW Software
6. Moving Mic Method

Using a Umik-1 with calibrated 90 degree file, I began my measurements by pointing away from the speakers from about the listening spot using the moving mic method I learned from watching this video (link removed), I took several measurements.

I started w/ taking measurements of each speaker first (L and R) and then comparing those measurements to the composite measurement of L&R together, seen below.

REW Preferences:

Generator Settings

Using those settings, I generated the following set of measurements....

Measurements with 1/24 Smoothing

1. Left only - red
2. Right only - blue
3. Both #1 - orange
4. Both #2 - green

Average

1. I averaged all 4 measurements above to create this Average

Now is the part I probably need the most help with! Creating a set of filters from these measurement so I can plug into ROON. I took a stab as follows, but please correct me if I'm wrong. Watching several videos on the topic helped but also confused to an extent as everyone tends to use their own preferred settings.

EQ Settings-
This part is still a bit of a mystery to me. I used a harman curve which i added via preferences.

Settings i changed were:
1. Generic
2. Full range Speakers - forte iv's??
3. I hit the 'Calculate target level from response' to get the curve to overlap the house curve. Yes , I'm using a harman house curve for this example.
4. Match range 38 to 20,000 (my fortes only go down to 38 so i figured i would cap it there)

With those settings in tow, I hit the 'match response to target'.


That yielded the following:

Questions:

1. Should i be using Full Range Speakers?
2. What should I try next to create better picture of my room?
3. The PEQ it generated yielded several high Q filters, should i drop high Q filters?
4. Several filters in the 100-200 hz range, should they be consolidated?
5. Should i have a high pass filter at 38hz? The lowest range my speaker can go, theoretically.
6. Also why isn't REW adusting for the dip between the 200-800 hz region?

I tried using these in ROON, but i felt some of the energy was sucked out when ABing before and after results.

Any help or guidance would be awesome!

Thanks!!!.

 

[JStewart]

In terms of figuring out REW, you've done a great job thus far. The measurements you've obtained track very well with the @Erin 's measurements found here:

Klipsch Forte IV Review

Klipsch Forte IV Review

www.erinsaudiocorner.com

The measurements show a directivity mismatch in the 1kHz to 2kHz area centered at 1.5kHz which very strongly suggests Eq of that range will make it sound worse.

Erin has explanations of the measurements on his site if you're not familiar and want a better understanding. There is also Dr. Floyd Toole's book for a deep dive

https://www.amazon.com/Sound-Reproduction-Psychoacoustics-Loudspeakers-Engineering-ebook/dp/B074CHY128/ref=sr_1_1?crid=3LUV7NPZAS70F&keywords=Sound+Reproduction%3A+The+Acoustics+and+Psychoacoustics+of+Loudspeakers+and+Rooms&qid=1658583683&rnid=2941120011&s=digital-text&sprefix=sound+reproduction+the+acoustics+and+psychoacoustics+of+loudspeakers+and+rooms%2Caps%2C127&sr=1-1

My personal preference to improve the response aspect of the sound would be to add a subwoofer(s) to deal with the issues below 65Hz.

And now back to our regularly scheduled programming:

Should i be using Full Range Speakers?

Yes.

The PEQ it generated yielded several high Q filters, should i drop high Q filters?

Yes because you are using Roon and it does not support a Q greater than 10 IIRC.

Several filters in the 100-200 hz range, should they be consolidated?

1. Try removing the high Q filter option first and see what you get.
2. I'm not so sure lopping off the low end peaks will make much of or a positive difference in the sound. You'll have to try it and see what think.

Should i have a high pass filter at 38hz? The lowest range my speaker can go, theoretically.

Looking at the distortion charts on Erin's site I would say not needed. (Insert add subwoofer here.)

Also why isn't REW adusting for the dip between the 200-800 hz region?

It's in the settings for max individual boost and max overall boost.
You can add a PEQ manually with REW also, if you want a boost for this region. Like this:

I tried using these in ROON, but i felt some of the energy was sucked out when ABing before and after results.

Not surprising given the anechoic response and the directivity characteristics. I would shy away from any PEQ for the higher frequencies. If you want want to adjust the speaker balance then perhaps a tone control or a shelving filter will work.

Keep in mind that EQ results are also a matter of taste/preference. Matching a target response (which is not the same thing as correcting for room modes) may or may not sound better to you. So, good sound for you is the goal as opposed to an arbitrary target.

Good luck! Have fun!

 

[pimogo]

SUPER HELPFUL!

Didn't understand all of it, but ordering that book. If i can lean on you one more time, i have a couple follow up questions i was unclear about after reading (and rereading) your post:

The measurements show a directivity mismatch in the 1kHz to 2kHz area centered at 1.5kHz which very strongly suggests Eq of that range will make it sound worse.

Not familiar w/ the concept so my question will likely be nonsensical, but how were you able to determine a directivity mismatch from the graph? I would like to be able to spot that myself. Is EQing such a a mismatch a known problem? Is there a course of action i can take to remove such mismatch?

1. Try removing the high Q filter option first and see what you get.

What option is that?

My personal preference to improve the response aspect of the sound would be to add a subwoofer(s) to deal with the issues below 65Hz.

What would a subwoofer do here other than add lower end, which to be fair is not entirely nothing. I'm not opposed to the idea, but had hoped the Forte's would provide enough bass by themselves. If i integrate a sub, i then need to add a cross-over and more importantly know how best to use it.

Matching a target response (which is not the same thing as correcting for room modes) may or may not sound better to you. So, good sound for you is the goal as opposed to an arbitrary target.

I like this and am more interested in room correction, just wanted to carry out the house curve example for experimentation. I will take the room correction route. If i'm not mistaken i simply use the 'room curve' settings below, right? . Should i stick w/ the default settings or should they be adjusted?

Many thanks and hope you're having a great weekend.

 

[JStewart]

but how were you able to determine a directivity mismatch from the graph?

You can see in the top circled area that as the measurements move off access the response changes. This means the reflections from the walls are of a different character than the direct sound. The reflections combine with the direct sound and change it. When you measure your in-room response you are actually measuring a combination of the direct and reflected sound together. The bottom two lines of the graph are two directivity indicies. If the off axis character of the sound is the same as the direct sound then these will be smooth lines.
You might go to Erin's website and look up the BMR, Ki, or Dutch and Dutch speakers for a couple of examples of speakers with excellent directivity to compare the measurement graphs to the Forte..

Is EQing such a a mismatch a known problem?

Yes.

Is there a course of action i can take to remove such mismatch?

No. It's baked in to the speaker design. You can still like the sound of your speakers and you can use tone controls or shelving filters to try to tonally adjust them to personal preference. PEQ filters will have unexpected results.

What option is that?

Uncheck the box that says use narrow filters below 200Hz.

What would a subwoofer do here other than add lower end, which to be fair is not entirely nothing. I'm not opposed to the idea, but had hoped the Forte's would provide enough bass by themselves.

Looking at your graph there's a large dip between 45 and 65Hz. Erin's measurement shows the actual response going smoothly to about 45Hz, therefore the dip is an effect of the speaker location or listening location within the room. A subwoofer might be placed to mitigate the dip. There's a bit of music and harmonics in this area including kick drums.
Anyway, having a sub or not is your personal choice. Mine would be to have one as a solution for an even low end response to the bottom of the range for most music.

If i integrate a sub, i then need to add a cross-over and more importantly know how best to use it.

It's a bit of chore to integrate properly, but it's also really common these days so there's lots of folks around who will help.

I like this and am more interested in room correction, just wanted to carry out the house curve example for experimentation. I will take the room correction route. If i'm not mistaken i simply use the 'room curve' settings below, right? . Should i stick w/ the default settings or should they be adjusted?

I think I didn't convey the point I was trying to make. Here is what I said...
"Keep in mind that EQ results are also a matter of taste/preference. Matching a target response (which is not the same thing as correcting for room modes) may or may not sound better to you. So, good sound for you is the goal as opposed to an arbitrary target."

The point being that any curve is arbitrary and just because you can use the tools to match a target doesn't mean you'll like how it sounds.

Taking the Harman curve for example, it was never intended to be a target. It's actually the naturally occurring response of a loudspeaker with good directivity behavior in a typical listening room. So it's a result, not a target. EQing to it may work out and may not. Not saying not to try it, just trying to help you understand why you matched a target and it didn't sound good. Adjust it so it sounds good to you.
Hope that wasn't too wordy.

For a more comprehensive look at room curves and EQ you can read these comments from Dr. Toole:

Some comments from Floyd Toole about room curve targets, room EQ and more

Room curve targets Every so often it is good to review what we know about room curves, targets, etc. Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses. Those with...

www.audiosciencereview.com

If you'd like to learn what room modes and standing waves are, you could start here:

Room Modes 101

What are room modes and how do they impact sound quality? Some unavoidable facts: ALL rooms have naturally occurring resonance frequencies called room modes or standing waves. If you read around the web you’ll find some more esoteric terms like eigentones or eigenmodes. I quite like calling them...

www.acousticfrontiers.com

 

[pimogo]

Thank you! I've read both resources and have a better understanding of room modes and the sound imparted by a given room. How to correct that is still a bit fuzzy and at times overwhelming and discouraging.

I'm looking into the idea of adding a sub. I have a spare one from a leftover project but its nothing fancy. My concern with integrating a sub is partly due to me having an all analog device in the AS1200. The AS1200 does not contain any bass management but I can try the onboard crossover options to begin. Just not sure how to level match sub to speakers effectively.. Will need to research further.

As for room treatment, a topic covered in one of your linked resources... I do have a couple of 7" thick bass traps , one behind each of the speakers. Clealry not helping w the massive dips seen on my chart. or they are, but the two are not enough on their own. I had been debating , even prior to this conversation, adding an additional two 4" free standing 5 foot panels along both sides of the wall for early reflection and additional bass trapping. I wonder if that would help address some of the issues seen on the chart.. I'm not in a room I can completely makeover and so need to be strategic in my use of treatment.. Biggest bang... which is why I turned to Room EQ in the first place.. as a supplement to room treatment.

anyway I'm early on in this journey and I'm sure I'll arrive at a satisfactory spot in time, but right now I can't help but feel I should have stuck w/ headphones!

 

[JStewart]

anyway I'm early on in this journey and I'm sure I'll arrive at a satisfactory spot in time, but right now I can't help but feel I should have stuck w/ headphones!

LOL, another rabbit hole unto itself!

 

[ddude003]

As for room treatment, a topic covered in one of your linked resources... I do have a couple of 7" thick bass traps , one behind each of the speakers. Clealry not helping w the massive dips seen on my chart. or they are, but the two are not enough on their own. I had been debating , even prior to this conversation, adding an additional two 4" free standing 5 foot panels along both sides of the wall for early reflection and additional bass trapping. I wonder if that would help address some of the issues seen on the chart.. I'm not in a room I can completely makeover and so need to be strategic in my use of treatment.. Biggest bang... which is why I turned to Room EQ in the first place.. as a supplement to room treatment.

The best place for Bass traps is in the corners... See http://arqen.com/bass-traps-101/placement-guide/

 

[thothsong]

When you said you did MMM with the mic pointing away from the speakers, did you mean vertically using the 90 degree cal file, or something else? FWIW, if you have leeway, I'd try altering the distance from the speakers to the front wall, both closer and farther, to see what effect it has on the dips.

 

[thothsong]

Regarding traps behind the speakers, they may be helping with reflections, decay, and clarity measures, but you can't analyze those facets using MMM.

 

[pimogo]

When you said you did MMM with the mic pointing away from the speakers, did you mean vertically using the 90 degree cal file, or something else? FWIW, if you have leeway, I'd try altering the distance from the speakers to the front wall, both closer and farther, to see what effect it has on the dips.

Hi. Thanks for asking! I held the microphone parallel to the floor with the head (or grill part) facing the right wall. l then moved the mic in medium sized circles (about a foot in diameter) in and around the listening area for about 30 seconds or until the graph was no longer moving.

The bass traps were added behind the speakers on instruction from GIK Acoustics.. Ideally i should have more, including on the corners, but this is a multi-use room and sadly can not cover it up as much as i'd like . The Forte's sit about 5 " - 6" away from the bass traps in the narrowest spot and are angled towards the listening area. I've experimented w/ placement for many hours until settling on the spot they are now. I feel sound-stage is best now and the floors are marked with the new defaults. Occasionally i will experiment w/ placement when i feel bass response can be improved but end up going back to default. I hope to add freestanding 4" panels to address early reflections next, with the hopes of improve clarity and bass tightness further. I did notice an obvious improvement after adding the basstraps so encouraged to try adding another pair.

Images below for context.. The subwoofer you see below was added today on recommendation of @JStewart. I have not measured the room w/ them yet, still experimenting.

Thanks for any and all input/critiques, etc!

 

[thothsong]

I don't use MMM, but I would have thought it best to either hold the UMIK-1 horizontal pointing to the front and use the 0 degree cal file, or hold it vertical pointing at the ceiling and use the 90 degree cal file. Perhaps it doesn't really matter if you only look at low frequencies. Have you done normal frequency sweep measurements, to compare results? Sweeps, with and without the traps in place, would let you see if the traps are having any other benefits.

 

[ddude003]

Since these Forte speakers have 15" passive radiators in the back (for bass), I would be tempted to remove the bass traps/absorbers from behind speakers and position the speakers as Klipsch suggests, 2" to 10" from the wall...

I am also wondering if the traps/absorbers come down low enough on the wall to fully cover the passive radiators or maybe only half and half way? Don't know exactly what that might do to the bass if half and half...

Also makes me wonder about the accuracy of the spin-o-rama of a speaker that depends on support of a rear wall, like di-poles or rear firing speakers or radiators...

 

[thothsong]

Since these Forte speakers have 15" passive radiators in the back (for bass), I would be tempted to remove the bass traps/absorbers from behind speakers and position the speakers as Klipsch suggests, 2" to 10" from the wall...

I was about to suggest trying the same thing (having just looked at the manual for them).

 

[pimogo]

Since these Forte speakers have 15" passive radiators in the back (for bass), I would be tempted to remove the bass traps/absorbers from behind speakers and position the speakers as Klipsch suggests, 2" to 10" from the wall...

I am also wondering if the traps/absorbers come down low enough on the wall to fully cover the passive radiators or maybe only half and half way? Don't know exactly what that might do to the bass if half and half...

Also makes me wonder about the accuracy of the spin-o-rama of a speaker that depends on support of a rear wall, like di-poles or rear firing speakers or radiators...

Thats a fair point and was unsure about that myself. Let me follow up w/ GIK to see what they say.

I don't use MMM, but I would have thought it best to either hold the UMIK-1 horizontal pointing to the front and use the 0 degree cal file, or hold it vertical pointing at the ceiling and use the 90 degree cal file. Perhaps it doesn't really matter if you only look at low frequencies. Have you done normal frequency sweep measurements, to compare results? Sweeps, with and without the traps in place, would let you see if the traps are having any other benefits.

I'll rerun my sweeps in that configuration. SHould be a useful data point regardless. Thanks for suggesting it!

 

[pimogo]

Actually just looked at the back of the Fortes. The bottom of absorber meets the bottom of the radiator so the radiator port is completely covered by the absorber.