I guess you could also topple a downward-firing sub on its side if needed...A benefit of forward firing subwoofers
Wiim Amp with Polk XT15 & Polk PSW10
- Thread starter Vicious
- Start date
-
- Tags
- discussion question wiim amp
All thanks to you man - Thanks for the support!You did a very good thing indeed!
Great job!
Perfect! Moving the sub around is well worth it, usually it is possible to find a spot that works better than others. It is icing on the cake if that spot turns out to be this convenient.
BTW, by turning the sub 90° you moved the driver closer to the back wall, and a bit further away from the left wall. This changed the SBIR frequencies (I explained the theory recently in this post) and in your case removed the most severe null. I also have my sub rotated 90° in my desktop system for the same reasons.
The best way to learn these effects is to experiment as you did, kudos for taking the time to do it!
I'd personally set Min Gain to -12 dB because a couple of the filters obviously need more negative gain (e.g. at 66 Hz in one channel and 64Hz in the other).
I'd personally set the lower bound to 30Hz since your sub doesn't have much output below it. Boosting below 30Hz is IMHO just wasting headroom and might result in increased distortion.
You can keep the upper bound at 1kHz if you like, but the problem areas in your room seem to all be below 300Hz.
Really happy to hear this!
Absolutely, placement is the first this to sort out - that way you also maximize the potential with RC/EQ that you run later.
All in all, great work!
Now my Indie/Rock and Metal genres are a treat, drums are catching my attention, as well as acoustic guitar strumming
I'll tweak the above-mentioned (Min Gain, Frequency) tomorrow, I think my wife is a bit "Over" hearing the RC test haha.
No problem, it was my pleasure!
Great!
I think you can also keep it as is - these changes might not be very significant at all. Given that there's a level of unpredictability in the WiiM RC algorithm it might also end up being worse than what you have already.
So let's keep it as is for now until WiiM updates the RC with future improvements?No problem, it was my pleasure!
Great!
I think you can also keep it as is - these changes might not be very significant at all. Given that there's a level of unpredictability in the WiiM RC algorithm it might also end up being worse than what you have already.
Regarding the RC, the Q, what's ideal, or what can you set it and work with, does it make such a difference, or is it rather minor from an audible standpoint? Had it as 6 when the sub was ideal, then tweaked it back to 10 and crossover to 80Hz. But, I'm not that sure that it makes such an extreme difference at this level
That makes sense.
Just note that your response now is anyway about 99% there. I'd suggest to now spend a few weeks at least listening to this setup and then see it you feel any need to tweak. It usually takes a bit of time to get accustomed to a new tuning.
Any tweaking going forward would either be more academic/educational, or should focus on fitting your personal preference (i.e. unrelated to RC). I personally wouldn't expect any dramatic improvements in objective quality of the correction itself - i.e. if you'd like you could keep it like this regardless of what WiiM do to RC in the future.
Note that a new feature is coming that should simplify tuning the response to personal preference: Beta Testing: Separate Room Correction and EQ (Available only on WiiM Ultra). See also the related help article containing a bit more information. This is not (yet) available on most WiiM devices. Once this becomes available on the WiiM Amp you would e.g. no longer need to repurpose L/R RC PEQ #10 for the high-shelf; you could instead implement the high-shelf on a separate stereo input PEQ and run it in parallel to L/R RC PEQ profile (which will be running on the output).
"Max Q" controls the maximum PEQ filter sharpness (pointiness?
) that RC is allowed to use when calculating the correction.High Q value means sharper filters are allowed, and sharp filters are more precise so they can fit the response closer to the desired target.
Note that human hearing doesn't have infinite resolution when detecting peaks and dips in the frequency domain, and distinct spectral deviations with Q significantly above 5 might not be very audible at all. In addition, we are better at detecting peaks than we are at detecting dips in the spectrum. I.e. matching of a response to a target doesn't really need to be perfect to sound "perfect" - luckily for us. A good match doesn't hurt, though (at least not at low frequencies)!
So there is not a lot of drawback to allowing relatively sharp negative-gain PEQ filters, but sharp positive-gain PEQ filters with a lot of gain could result in audible resonances and ringing which could be a problem. In WiiM a single parameter controls maximum Q of both kinds of PEQ filters which is why some basic care is needed.
Normally I keep Max Q at 10 because it gives me the nicest-looking curve. But if I notice that RC returns PEQ bands with positive-gain filters with high-Q values I may choose a lower Max Q (like 5) instead. In most cases I wouldn't use anything below Max Q = 5, as that is no longer precise enough to deal with distinct room resonances when using 1/12 smoothing.
In your case most PEQ filters are anyway calculated with relatively low-Q, maximum gain is limited to a low value (3,5dB), and the sharpest filters in your correction anyway have only negative gain so you should be OK either way.
This is all to say that I'm not surprised that there's not a lot of audible difference between Max Q set to 6 or 10 in RC.
Good explanation and makes sense.That makes sense.
Just note that your response now is anyway about 99% there. I'd suggest to now spend a few weeks at least listening to this setup and then see it you feel any need to tweak. It usually takes a bit of time to get accustomed to a new tuning.
Any tweaking going forward would either be more academic/educational, or should focus on fitting your personal preference (i.e. unrelated to RC). I personally wouldn't expect any dramatic improvements in objective quality of the correction itself - i.e. if you'd like you could keep it like this regardless of what WiiM do to RC in the future.
Note that a new feature is coming that should simplify tuning the response to personal preference: Beta Testing: Separate Room Correction and EQ (Available only on WiiM Ultra). See also the related help article containing a bit more information. This is not (yet) available on most WiiM devices. Once this becomes available on the WiiM Amp you would e.g. no longer need to repurpose L/R RC PEQ #10 for the high-shelf; you could instead implement the high-shelf on a separate stereo input PEQ and run it in parallel to L/R RC PEQ profile (which will be running on the output).
"Max Q" controls the maximum PEQ filter sharpness (pointiness?
High Q value means sharper filters are allowed, and sharp filters are more precise so they can fit the response closer to the desired target.
Note that human hearing doesn't have infinite resolution when detecting peaks and dips in the frequency domain, and distinct spectral deviations with Q significantly above 5 might not be very audible at all. In addition, we are better at detecting peaks than we are at detecting dips in the spectrum. I.e. matching of a response to a target doesn't really need to be perfect to sound "perfect" - luckily for us. A good match doesn't hurt, though (at least not at low frequencies)!
So there is not a lot of drawback to allowing relatively sharp negative-gain PEQ filters, but sharp positive-gain PEQ filters with a lot of gain could result in audible resonances and ringing which could be a problem. In WiiM a single parameter controls maximum Q of both kinds of PEQ filters which is why some basic care is needed.
Normally I keep Max Q at 10 because it gives me the nicest-looking curve. But if I notice that RC returns PEQ bands with positive-gain filters with high-Q values I may choose a lower Max Q (like 5) instead. In most cases I wouldn't use anything below Max Q = 5, as that is no longer precise enough to deal with distinct room resonances when using 1/12 smoothing.
In your case most PEQ filters are anyway calculated with relatively low-Q, maximum gain is limited to a low value (3,5dB), and the sharpest filters in your correction anyway have only negative gain so you should be OK either way.
This is all to say that I'm not surprised that there's not a lot of audible difference between Max Q set to 6 or 10 in RC.
Would a max gain of 3-3.5db be ideal for future tweaking?
Ex.
Max gain 5db (Or higher)
Min gain -12db
Freq 30-1000Hz
Or is the current setup I have set ideal?
With the current WiiM RC algorithm I'd say what you have done already is very good. I see no need to increase Max Gain.
What will be best in the future will depend on how WiiM tweak the RC algorithm. No way to know for sure at the moment, I'm afraid.
Re-run at the above-mentioned settings that you’ve suggestedYou did a very good thing indeed!
Perfect! Moving the sub around is well worth it, usually it is possible to find a spot that works better than others. It is icing on the cake if that spot turns out to be this convenient.
BTW, by turning the sub 90° you moved the driver closer to the back wall, and a bit further away from the left wall. This changed the SBIR frequencies (I explained the theory recently in this post) and in your case removed the most severe null. I also have my sub rotated 90° in my desktop system for the same reasons.
The best way to learn these effects is to experiment as you did, kudos for taking the time to do it!
I'd personally set Min Gain to -12 dB because a couple of the filters obviously need more negative gain (e.g. at 66 Hz in one channel and 64Hz in the other).
I'd personally set the lower bound to 30Hz since your sub doesn't have much output below it. Boosting below 30Hz is IMHO just wasting headroom and might result in increased distortion.
You can keep the upper bound at 1kHz if you like, but the problem areas in your room seem to all be below 300Hz.
Really happy to hear this!
Absolutely, placement is the first this to sort out - that way you also maximize the potential with RC/EQ that you run later.
All in all, great work!
Attachments
That looks pretty much perfect now, as far as room correction goes.
Time to get yourself a little Dayton Audio iMM-6C and import its calibration file into the WiiM Home app, so you can see the real frequency response, in particular in the higher range.
