Wiim Roomfit questions, please help

[harkpabst]

Sounds a little better but I still feel unsatisfied ...

... because ... of what?

... and not sure if this is good.

These graphs alone, as mentioned before, tell almost nothing about if the RoomFit correction is good or not. This is just the correction applied, it says nothing about how it's been before and nothing about how it is now.

If anything, your subwoofer level is set too high. The broad (~30 Hz and ~90 Hz) and deep (-12 dB or even more) cut tells pretty clearly that the level is way too high before RoomFit and that this is probably the frequency range covered by the subwoofer.

This is just a guess, since you didn't show us your subwoofer settings yet and I can only assume that the crossover frequency is set to ~80 Hz.

Once again, make absolutely sure to use the Left (LFE) input and turn the LOW PASS CROSSOVER FREQUENCY knob to the BYPASS position, as recommended by @DanielT. Leave the PHASE knob at 0⁰, the WiiM Home app is the much better place to deal with absolute phase and delay/latency! This is what Subwoofer Calibration is for. Start with the VOLUME knob in the 12 o'clock position and the subwoofer level set to 0 dB in the WiiM Home app. If you find that there is too much or too little bass (with RoomFit disabled) adjust the volume, preferably in the WiiM Home app. Remember that the goal is a seamless integration with the speakers, so the sub should not stand out. If in doubt, a slightly higher level is better, as this gives RoomFit something "to play with" in non-boost mode. However, the current sub level is too high.

You can also tell this by comparing the "Flat" and the "Harman" target curves. Harman aims for an overall higher bass level than Flat, so in turn, the Flat correction tries to tame your subwoofer even more. Since each individual peak filter is limited to -12 dB, which is not enough in this case, RoomFit starts to pile upmmultiple filters at very close frequencies to be able to reach the required amount of negative gain. As a result, the 10 available filters are eaten up just by the bass region.

The dip in the ~4.2 kHz region with the Harman target curve would have been in the Flat correction as well, if not all available PK filters would have been already used up. It also tells us that you have set the upper frequency limit for RoomFit to something higher than 4 kHz, although you've been advised to lower the upper limit. In this case it makes just a tiny difference. But it's all easier if you stay focused and don't make changes we're not aware of.

RoomFit has undergone some massive improvements in recent months. Switching to an external tool without having exhausted all the possibilities it brings would be a bit of a hasty move.

Sticking to your phone's mic for now can be a good idea, but this also depends on the make and model of the phone. I phone mics are generally consistent and their properties are known to software vendors. Some models don't work well below 60 Hz. Android phones can be good enough for bass correction, too, but especially MediaTek SoCs are known to have almost unusable mic implementations. They will often measure much too low levels below 80 Hz and there's no way for the app to work around this.

But most importantly we need to see the frequency response before and after correction as displayed in the WiiM Home app before you save your RF profile!

 

[Ruediger]

All the needed info.

If the bass level from 80Hz crossed is set in a good relation and the measurement comes up with the same deep corrections for bass, consider as next step.

You might start a powerful room mode down there in the 30Hz region. What you then want to try is a room gain compensation. Either your sub had that as an option (I don't think so) or you simply do it as a WiiM RoomFit filter and type it in.

But go through the suggestions first and make sure the sub is not just at silly levels or it is turned up in the app.

 

[tkdj123]

All the needed info.

If the bass level from 80Hz crossed is set in a good relation and the measurement comes up with the same deep corrections for bass, consider as next step.

You might start a powerful room mode down there in the 30Hz region. What you then want to try is a room gain compensation. Either your sub had that as an option (I don't think so) or you simply do it as a WiiM RoomFit filter and type it in.

But go through the suggestions first and make sure the sub is not just at silly levels or it is turned up in the app.

Ive attatched my sub settings here, thanks for your help!

 

[Ruediger]

The hardware looks alright.

You could add screens of your sub pages. And from your measurement the PEQ list.

 

[Chrispy]

Ive attatched my sub settings here, thanks for your help!

Curious, what did you do to place the sub and set phase/level originally (i.e. before roomfit)?

 

[tkdj123]

Curious, what did you do to place the sub and set phase/level originally (i.e. before roomfit)?

It was the same, I never changed it.

 

[Chrispy]

It was the same, I never changed it.

I mean setting it up initially. Not sure what you might have changed. Best to set up sub as best as possible on its own before applying dsp.....for example, why is the phase set that way? Level?

 

[tkdj123]

... because ... of what?


These graphs alone, as mentioned before, tell almost nothing about if the RoomFit correction is good or not. This is just the correction applied, it says nothing about how it's been before and nothing about how it is now.

If anything, your subwoofer level is set too high. The broad (~30 Hz and ~90 Hz) and deep (-12 dB or even more) cut tells pretty clearly that the level is way too high before RoomFit and that this is probably the frequency range covered by the subwoofer.

This is just a guess, since you didn't show us your subwoofer settings yet and I can only assume that the crossover frequency is set to ~80 Hz.

Once again, make absolutely sure to use the Left (LFE) input and turn the LOW PASS CROSSOVER FREQUENCY knob to the BYPASS position, as recommended by @DanielT. Leave the PHASE knob at 0⁰, the WiiM Home app is the much better place to deal with absolute phase and delay/latency! This is what Subwoofer Calibration is for. Start with the VOLUME knob in the 12 o'clock position and the subwoofer level set to 0 dB in the WiiM Home app. If you find that there is too much or too little bass (with RoomFit disabled) adjust the volume, preferably in the WiiM Home app. Remember that the goal is a seamless integration with the speakers, so the sub should not stand out. If in doubt, a slightly higher level is better, as this gives RoomFit something "to play with" in non-boost mode. However, the current sub level is too high.

You can also tell this by comparing the "Flat" and the "Harman" target curves. Harman aims for an overall higher bass level than Flat, so in turn, the Flat correction tries to tame your subwoofer even more. Since each individual peak filter is limited to -12 dB, which is not enough in this case, RoomFit starts to pile upmmultiple filters at very close frequencies to be able to reach the required amount of negative gain. As a result, the 10 available filters are eaten up just by the bass region.

The dip in the ~4.2 kHz region with the Harman target curve would have been in the Flat correction as well, if not all available PK filters would have been already used up. It also tells us that you have set the upper frequency limit for RoomFit to something higher than 4 kHz, although you've been advised to lower the upper limit. In this case it makes just a tiny difference. But it's all easier if you stay focused and don't make changes we're not aware of.

RoomFit has undergone some massive improvements in recent months. Switching to an external tool without having exhausted all the possibilities it brings would be a bit of a hasty move.

Sticking to your phone's mic for now can be a good idea, but this also depends on the make and model of the phone. I phone mics are generally consistent and their properties are known to software vendors. Some models don't work well below 60 Hz. Android phones can be good enough for bass correction, too, but especially MediaTek SoCs are known to have almost unusable mic implementations. They will often measure much too low levels below 80 Hz and there's no way for the app to work around this.

But most importantly we need to see the frequency response before and after correction as displayed in the WiiM Home app before you save your RF profile!

Those are my sub settings and I've attatched a pic of what it looks like on the back of the sub. I think you were correct the sub was a little to high since I had it set to 5dB in the Wiim app. Ive changed it back down to 0 now.

These are the settings I am using for roomfit, I thought it was advised to do 20 - 20,000hz? If not what do you suggest? As for my phone I have an iphone 15.

What do you mean by "we need to see the frequency response before and after correction as displayed in the WiiM Home app" what frequency response are you referring to ?

 

[tkdj123]

The hardware looks alright.

You could add screens of your sub pages. And from your measurement the PEQ list.

Hope that helps!

 

[ShadowBoxer]

Why are filters 4&5 almost identical?

 

[tkdj123]

I mean setting it up initially. Not sure what you might have changed. Best to set up sub as best as possible on its own before applying dsp.....for example, why is the phase set that way? Level?

Not sure if I understand but that is the way I set up the sub initially. The phase is at that 0 because that is what is recommended? See how the speakers and sub are placed in my first pic.

 

[tkdj123]

Why are filters 4&5 almost identical?

I am not sure, that is what the room correction came up with

 

[harkpabst]

We're getting much closer now.

Those are my sub settings and I've attatched a pic of what it looks like on the back of the sub. I think you were correct the sub was a little to high since I had it set to 5dB in the Wiim app. Ive changed it back down to 0 now.

I've seen the backplate of your sub in your previous posting. The level of your sub is controlled by the combined settings on the back of your sub and the settings in the WiiM Home app. The sub is not too loud because it was set to +5 dB in the WiiM Home app (this can be required in other setups), it is too loud because of what we see in the RoomFit correction filters. So, it's a good thing you did reduce the level.

Looking at your latest screenshot ...

I'd reduce the level even more, maybe to -2 dB. RoomFit pulls the level of your sub down by way more than -12 dB anyway.

Do you have "Main speakers output bass" disabled as well? It should be disabled.

These are the settings I am using for roomfit, I thought it was advised to do 20 - 20,000hz? If not what do you suggest?

No, this is absolutely not recommended. The primary intention of RoomFit (and room correction systems in general) is to mitigate the negative effects of the room, not to change the sound of your speakers in general. Room modes and standing waves dominate the behavior of your system up to the Schrödinger frequency, which is different from room to room, but can be estimated from its dimensions. For now you can go by a reasonable rule of thumb and set the upper limit to 500 Hz.

I would also increase the lower limit again, set it to 30 Hz or at least 25 Hz. RoomFit tries to boost your subwoofer's level at 21 Hz ...

... but your sub cannot handle 21 Hz even at just slightly elevated levels anyway. This boost puts unnecessary stress on your sub and occupies one of the 10 PK filters, which might be put to better use at some higher frequency.

As for my phone I have an iphone 15.

Should be good enough for room correction, but don't expect it to be sufficiently accurate at higher frequencies. But even with a UMIK-1 or comparable quality microphone it is generally not recommended to run RoomFit up to 20 kHz. In some cases it might be beneficial to correct up to 4 kHz or 5 kHz, but this is the exception, not the rule. Stick to the rule for now.

Why did you pick the "Flat" target curve?

A flat frequency response is desirable when measuring speakers in an anechoic environment. It's not the desired in-room response. If you put a speaker with a flat anechoic response on-axis and smoothly falling highs off-axis in a typical living room, you will usually observe a downwards sloped frequency response and this is usually the preferred target.

You started off with "Auto" and have been advised to try "Harman" or "B&K" instead. Both are usually a better fit than "Flat". Both target curves are not too dissimilar, with "Harman" putting more emphasis on the bass.

Unfortunately, we still don't know what exactly you don't like regarding your RoomFit results and how you like the sound with RoomFit disabled (and for what reasons. This is important!

Also make sure to enable the beta feature "Moving Microphone Method" in the advanced RoomFit settings. You will see on-screen instructions on how to perform the measurement. If not already selected, make also sure that the smoothin algorithm is set to "Variable".

What do you mean by "we need to see the frequency response before and after correction as displayed in the WiiM Home app" what frequency response are you referring to ?

We need to see the screen that looks somewhat like this one:

It shows the target curve (Target) the measured frequency response with no correction applied (Raw) and the measured frequency response with RoomFit correction applied (Actual).

Your screen will look a little different, but you will recognize the correct one immediately.

Remember that you have to re-run RoomFit every time you change your RoomFit settings (trivially), your subwoofer settings (hardware or WiiM Home app) or change the position of any of your loudspeakers (including the sub).

 

[harkpabst]

Why are filters 4&5 almost identical?

This would indicate that the maximum gain setting is too low. RoomFit will apply multiple filters to still reach the required boost.

It's a bit strange, though. If maximum gain was really set to +6 dB when performing the room calibration, a single filter should have been sufficient to apply a total gain of 4.9 dB. The Q factor is almost identical, too.

I still advise to re-set the max. gain to +10 dB.

 

[tkdj123]

Using BK curve here is before room fit correction

Here it is after the room correction. Not using moving mic measurement.

These were the new settings I used like you mentioned. Smoothing is variable.

As for the moving mic measurements I am still not sure what it is asking for so I didn't use it but I would love to if I could figure it out. Is it arcs in front of my head or around my head horizontally? Its confusing.

 

[Mnyb]

Using BK curve here is before room fit correction

View attachment 562408



Here it is after the room correction. Not using moving mic measurement.

View attachment 562409View attachment 562411


These were the new settings I used like you mentioned. Smoothing is variable.

View attachment 562410



As for the moving mic measurements I am still not sure what it is asking for so I didn't use it but I would love to if I could figure it out. Is it arcs in front of my head or around my head horizontally? Its confusing.

MMM is meant to give you an average I think so I swept basically my position and generous parts of the sofa I’m sitting in

It makes good sound for more people and removes very localised peaks and dips , you want to be able to move your head .

I use it only under 200Hz I don’t boost dips to much it’s not advisable due to the phenomenon if the cancellation is perfect nothing would happen in measured positions but possibly be to much elsewhere in the room and put quite a stress on the system .

My tactic has been use the umik mostly fight peaks below 200hz , and if the result is to lean for taste used a shelf filter in the normal PEQ setting to have some bass boost ( 1-2dB ) in the whole bass region ?

I also compared to what REW measured . And experimented with subwoofer settings and levels and crossovers before applying any correction to have a good starting point.

 

[tkdj123]

MMM is meant to give you an average I think so I swept basically my position and generous parts of the sofa I’m sitting in

It makes good sound for more people and removes very localised peaks and dips , you want to be able to move your head .

I use it only under 200Hz I don’t boost dips to much it’s not advisable due to the phenomenon if the cancellation is perfect nothing would happen in measured positions but possibly be to much elsewhere in the room and put quite a stress on the system .

My tactic has been use the umik mostly fight peaks below 200hz , and if the result is to lean for taste used a shelf filter in the normal PEQ setting to have some bass boost ( 1-2dB ) in the whole bass region ?

I also compared to what REW measured . And experimented with subwoofer settings and levels and crossovers before applying any correction to have a good starting point.

Here is the curve using MMM before correction with the same settings as my last post.

Here is the curve after correction

Here is MMM using the Harman curve

Also tagging @harkpabst for their input as well

 

[DanielT]

This is an interesting thread. I’m learning new things, so thanks to the knowledgeable people contributing here.

@tkdj123 , have you tried moving your subwoofer around and measuring the results? Try tucking it into the right-hand corner, for example. The advantage of corner placement (corner loading) is the extra acoustic boost it provides; it gives you more headroom to work with during EQ adjustments. That said, there can be downsides (a bit boomy, but you do have EQ, after all) too—it depends on your room's layout and dimensions.

Just don't forget to apply the appropriate time delay for that placement—or for whatever position you choose for your sub.

Here is a Gemini AI summary regarding corner placement:


Corner loading—a concept known since, well, probably around the time the loudspeaker was invented. It likely didn't take long for that effect to be identified. It was popular back when amplifier power was expensive. Perhaps it was also aesthetically pleasing—fitting the interior design style of the era? Here are screenshots of some vintage corner speakers:

 

[Ruediger]

The last measurement makes sense to me. Filters 1 and 2 tackle the bass booming through the room.

You now have it within filter capacity to work things out.

Many profiles can be created and freely edited. I'd still put a room gain comp manually as filter 1 and try how that goes. Should replace the first 3 filters and maybe make the graph above a bit smoother.

Typically I don't have access to a WiiM atm. Maybe someone else can chip in what to enter for highpass, 12db slope down, start at 40hz - where you change the frequency and see what this does to your bass in the room.

 

[harkpabst]

Here is the curve using MMM before correction with the same settings as my last post.

I think you're really making great progress here!

Some general comments for a start. @RandomEar has published a quick guide (plus references) for using MMM here:

The moving microphone method (MMM) for Dummies using REW

After getting a couple of weird results on my first MMM measurements, I went searching around the forums & YouTube for what went wrong. To save others from gathering various bits and pieces of info from a dozen sources, I decided to publish this quick write-up here. You can use it as a rough...

www.audiosciencereview.com

It's mainly written with REW in mind but the basics apply, no matter which tool you use. My very short summary: a) Don't move the mic/phone too fast and cover the area around your main listening position with your movement. An arm's length distance is certainly enough, your room isn't that big anyway. b) Make sure to use an appropriately high level when using MMM, because MMM is much more prone to disturbances by external noise. I'd probably use a volume setting of at least 35 or 40 on the WiiM Amp. Your measurements do look good and consistent, though.

Now on to your measurements. Did you notice the graph legends?

The "Results Overview" shows the target curve, the measurement before correction, the actual equalisation applied (this is the same as the graph you find in the RoomFit profile, just in a slightly different presentation) and the estimated response after correction.

The "Final Assessment" shows the target curve (same as above), the measurement before correction (same as above) and the frequency response that was actually measured after applying the EQ.

The important bit here is that the estimated curve and the measured curve match almost perfectly! This proves that your measurements are reliable and if high quality.

The one thing that stands otu is the small but deep dip slightly above 100 Hz. This dip remains the same, no matter if you're using the swept sine measurement or the moving microphone method. If the dip remains even with the spatial averaging of the MMM, this is strong evident for the dip not being cause by room modes!

The most likely reasons are either inverted polarity between main speakers and sub (this could be fixed using the phase switch in the WiiM Home app) or by SBIR (speaker boundary interference response). My personal guess is SBIR. If there was a polarity Inversion I would expect the dip to be right at the crossover frequency between sun and mains. However, the acoustical crossover frequency could differ somewhat from the electrical crossover frequency set in the WHA.

There's one easy way of testing this: Set the phase switch in the WiiM's sub management to 180⁰ and repeat the RoomFit run (with all other settings being the same as before).
If the dip goes away or becomes less pronounced, this is it.
If the dip stays the same, becomes more pronounced, moves or a second dip shows up, the the phase setting of 0⁰ was correct and the dip is caused by SBIR, indeed.

Fighting SBIR is less trivial. You basically have two options: a) Move the main speakers further away from all walls. It looks to me like your option are limited in your room concerning this approach. b) Move the speakers (and sub) as close as possibe to the front wall (but still as far as possible from the side walls) and increase the crossover frequency to 100 Hz or even 110-120 Hz.

The latter might sound counterintuitive, but it's really purely science based. With a crossover frequency above 80 Hz you should not try corner placement, though. Keep the sub rather symmetrically between the speakers, but also as.clise.to the front wall as possible.

Breakfast time for me. You've got some more things to try.