Wiim room correction advice

[davmol]

I've run room correction multiple times with the WiiM and settled on a preference for flat, over Harmon /bk. My current setup:
- Wiim Ultra
- Fosi Audio V3 Mono amps
- Wharfedale Linton 85th speakers
- Wharfedale SW-150 subwoofer

I set parameters of flat correction, from 35hz (low spec of the sub) to 400hz, with min of -12db / max of +0.5db and max Q=10, along with 1/6th octave (just chose the middle) and using multiple measurements. I'm using a Hyperx Quadcast S mic, set Omni directional, to take the readings as it provided results that were much less erratic than using my phone!

Here are the results of the correction (sub included):

Question 1:
Overall it's a big improvement on the playback without correction, but I'm not sure about the structure of the results. Seems odd to have corrections so close together, so can they be cleaned up or improved by consolidating or removing unnecessary bands?

Question 2:
This might be a silly question, but do I need different room correction profiles for different amps? If transparent amps have no audible differences, then I assume this means the room correction profile from one transparent amp should be valid for any other transparent amp. Is that assumption correct (and if not, why not)?

 

[Multicore]

Q1: At the low frequencies the problems in the room are from standing waves that come from the geometry of the room. Often the length and breadth of the room are similar and as a result peaks and troughs are close. The correction filters correspond.

Q2: That's correct. It is a silly question :P You're doing room compensation, not amp compensation.

 

[Nicoff]

On the subject of WiiM room correction, I noticed one can choose stereo mode or L/R. I was wondering which one should one choose and why?

 

[Multicore]

On the subject of WiiM room correction, I noticed one can choose stereo mode or L/R. I was wondering which one should one choose and why?

I don't understand the difference in the WiiM context.

For a couple of reasons I think it is in general better to develop a correction for each loudspeaker separately. So this would mean you hear a sweep or other test signal on each speaker one at a time.

 

[Nicoff]

I don't understand the difference in the WiiM context.

For a couple of reasons I think it is in general better to develop a correction for each loudspeaker separately. So this would mean you hear a sweep or other test signal on each speaker one at a time.

Agree, I used L/R option. But wondering why WiiM provided both options and when to use one over the other. Note: The OP used the Stereo option.

 

[Taiga]

When WiiM added the L/R option, I expected the result to be better than selecting stereo. In my case, the L/R sounded awful. Not sure if they have made any changes to it after the initial release.

 

[Multicore]

But wondering why WiiM provided both options and when to use one over the other.

I suspect WiiM is learning how to do this with a bit of good ol' trial 'n' error.

 

[Multicore]

When WiiM added the L/R option, I expected the result to be better than selecting stereo. In my case, the L/R sounded awful. Not sure if they have made any changes to it after the initial release.

Interesting.

The thing to do would be use REW to get measurements with L/R option, with stereo option and with room correction turned off. We actually have a WiiM but do correction in the Plex. And in any case I find putzing around with the UMIK-1, mic stands on the sofa, long USB cables, laptop on the coffee table and REW and all that complicated config a total drag. And I can only do it when the lady of the house and the wife and dogs are all out.

 

[davmol]

I have a very odd L-shaped room with different floor levels and ceiling heights. The left speaker is towards a corner where the ceiling height is 2m ceiling and the right speaker has no adjacent to the right, and the ceiling height transitions to 3m at that point.

I've found that the L/R option produces two quite different / distinct measurements, which I suppose is to be expected given the positioning, and maybe that is the use case - for odd room shapes / positioning. I prefer the end result using L/R, vs using a combined stereo correction.

I also found the measurements were much improved when I used an external usb mic (not calibrated, but still had far fewer "wild fluctuations" vs using my phone). I'm very pleased with the results of the WiiM auto corrections and I'm just glad I don't have to buy a umik and learn to use REW ;-)

 

[Connor1a]

Has anyone done an actual objective + subjective comparison between WiiM’s RC and Dirac Live? I mean run the graphics / looked at the different PEQS etc. I keep staring at a MiniDSP + Dirac, but by the time you factor in HW + SW, a MiniDSP Flex + SW starts at over 1000$ (if you include bass control). Compared to a WiiM Ultra at < 350$, that’s a tough nut to swallow. I have to ask if the end result really justifies the increase in cost (and yes, I get it, that’s a purely subjective question). YMMV and all that. Still, would be interesting to see something quantitative in terms of what the 2 sets of SW recommended as settings + how those settings then measured against the same target (curve)…

 

[Petrushka]

I just installed a Wiim Amp today, so I not going to start giving advice.

But they seem to have added multiple measurements and something called moving mic.

And a bit off topic, but the Amp has Airplay2.The Ultra and Amp Pro do not.

 

[davmol]

I tried using multiple measurements for a while but I found that the algorithm doesn't manage it very well and I would always end up with multiple corrections in or around the same frequency band, and on occasion it would be the same exact band corrected 2 or 3 times. I'm sure they will in time improve the aggregation but for now I have gone back to single measurements, as the corrections make much more sense.

 

[Connor1a]

WiiM seem to be rolling out enhances / fixes for their RC every couple of weeks. As @Multicore said, they’re learning + they’re clearly investing in the product. They just added another set of user definable PEQs to the box such that you can run RC, let the box define its RC PEQs and then come back with another set of 10 PEQs to further tune the sound. Neat trick.

 

[Bernard23]

I have a very odd L-shaped room with different floor levels and ceiling heights. The left speaker is towards a corner where the ceiling height is 2m ceiling and the right speaker has no adjacent to the right, and the ceiling height transitions to 3m at that point.

I've found that the L/R option produces two quite different / distinct measurements, which I suppose is to be expected given the positioning, and maybe that is the use case - for odd room shapes / positioning. I prefer the end result using L/R, vs using a combined stereo correction.

I also found the measurements were much improved when I used an external usb mic (not calibrated, but still had far fewer "wild fluctuations" vs using my phone). I'm very pleased with the results of the WiiM auto corrections and I'm just glad I don't have to buy a umik and learn to use REW ;-)

I have a similar shaped room to you, with some pretty grim standing waves.
I played around with RC with a Dayton mic and Android phone when it was first released and tbh struggled to find a consistent correction curve.
Now the app is updated I've tried again and I'm still not entirely sure how to run a profile test with a sub. I've set the sub to max crossover frequency and mid volume. On the Ultra I've set the crossover f to 65Hz and unchecked the box to play full range through either the sub or the main speakers.
The run the roomfit process and notice that the sweep test creates some clipping on the sub or maybe some furniture is resonating. The resultant profile sounds anaemic so I adjust the sub volume both on the sub and using the PEQ with a bass shelf.
I suspect I'm not doing the sweep test correctly, the volume of the system isn't particularly high when I'm running the sweep test.
Any suggestions as to what if anything I'm doing wrong here?

 

[Connor1a]

I suspect I'm not doing the sweep test correctly, the volume of the system isn't particularly high when I'm running the sweep test.
Any suggestions as to what if anything I'm doing wrong here?

What kind of sub(s) do you have? Do they have any active settings?

 

[Bernard23]

It's a, wharfedale SW150 and yes it does have active controls as in a built in power amp with crossover frequency and gain controls.

 

[Connor1a]

Full disclosure, I am not in the least an expert here. I “tinker”... There isn’t a lot of “science” in my approach at times. Take my ideas with a grain (large grain) of salt.

1) Have you tried running the RC with “Non-Boost Mode” turned on? It’s under the RoomFit menu. It supposedly helps avoid sub distortion / clipping by keeping each correction to a net 0 dB change.

2) Have you tried WiiM’s moving mic RC? It’s in beta at the moment, but available under the RoomFit menu. I’ve only done it a couple of times so far with mixed results. It’s similar to what Sonos does in their RC. WiiM has you move your mic around the room as part of the RC measurement process. That may give you a different curve.

3) Speaking of curves, have you tried one of the other “advanced” curves? B&K is default, but you can also try Harman and Flat. They may give you a bettter result.

4) You may want to play around with the smoothing setting. By default it’s set to 1/6th octave. You can change the value to 1/3rd and 1/12th. You may prefer 1/12th.

5) Finally, you can try limiting WiiM’s RC to a frequency range. I read another post where it was stated that limiting WiiM to below 300Hz may deliver a better result (implication being that correction in the higher ranges delivered lesser results). You can can set that up in the RoomFit menu as well…

Maybe one of these gives you something to go on? Hope they help….

 

[Bernard23]

Full disclosure, I am not in the least an expert here. I “tinker”... There isn’t a lot of “science” in my approach at times. Take my ideas with a grain (large grain) of salt.

1) Have you tried running the RC with “Non-Boost Mode” turned on? It’s under the RoomFit menu. It supposedly helps avoid sub distortion / clipping by keeping each correction to a net 0 dB change.

No I haven't, but suspect that works once you've activated the correction profile, rather than during the measurement (as there's no reference) so essentially applying PEQ auto gain.

2) Have you tried WiiM’s moving mic RC? It’s in beta at the moment, but available under the RoomFit menu. I’ve only done it a couple of times so far with mixed results. It’s similar to what Sonos does in their RC. WiiM has you move your mic around the room as part of the RC measurement process. That may give you a different curve.

No, and on my list! Sonos have a similar process for RC but is apple only. I have to say it turned 2 x Play5 +Sub into a, system that sounded way better than it's price tag would suggest.

3) Speaking of curves, have you tried one of the other “advanced” curves? B&K is default, but you can also try Harman and Flat. They may give you a bettter result.

Yup and I always tweak the output to suit my preference so I use flat.

4) You may want to play around with the smoothing setting. By default it’s set to 1/6th octave. You can change the value to 1/3rd and 1/12th. You may prefer 1/12th.

Agreed

5) Finally, you can try limiting WiiM’s RC to a frequency range. I read another post where it was stated that limiting WiiM to below 300Hz may deliver a better result (implication being that correction in the higher ranges delivered lesser results). You can can set that up in the RoomFit menu as well…

I've set mine to 500Hz because there's some odd stuff going on in my room up to that point.

Maybe one of these gives you something to go on? Hope they help….

 

[Bernard23]

Right, success!

I reran Roomfit but using two significant changes, I changed to the B&K target instead of flat and used the beta moving mic method.
The difference between the two profiles is huge. The new profile has a much more appropriate bass response. The original profile sounds like it's being played through a cheap soundbar in comparison. That's more like it!

 

[Bernard23]

Should add that the sub integration is working as I'd expected. I switch between sub on and sub off and no change. The only difference is the bass drivers in my elacs go into warp drive with the sub off. Sub on and the SQ doesn't noticeably change so that tells me the wharfedale sub is properly integrated, and the x over frequency at the right point. And that it's a banger for £150! I reckon my system has never sounded this good, and it's better than the one I used to own that cost 5x more without any form of RC