Custom room fit target curve

[gasolin75]

Is it possible to make a custom room fit target curve or download additional filters
Can i have 2 or 3 profiles like having flat,harman and B&K roomcorretion profiles i can switch between


How big is the difference between B&K and Harman

 

[gasolin75]

why this difference (both BK target curve) why the diffrence in the midrange ?

With sub (down to 20hz) No sub (only down to 30hz)

 

[dominikz]

why this difference (both BK target curve) why the diffrence in the midrange ?

With sub (down to 20hz) No sub (only down to 30hz)
View attachment 26629View attachment 26628

RoomFit has only 10-bands of PEQ per channel, and each PEQ filter is limited to +/-12dB gain. This means that RoomFit cannot fix *every* irregularity in the measured response, and therefore it has to prioritize some corrections over others.

Therefore if you change the range for correction (20Hz vs 30Hz bottom limit), or change the response shape somewhere (like how bass response changes when you remove the sub) the RoomFit filter assignments will change accordingly as well.

We don't have any details of how RoomFit algorithm prioritizes specific response irregularities when assigning PEQ filters, however.
Though in my opinion highest weight must be given to fix irregularities at low frequencies.

 

[dtc]

RoomFit has only 10-bands of PEQ per channel, and each PEQ filter is limited to +/-12dB gain. This means that RoomFit cannot fix *every* irregularity in the measured response, and therefore it has to prioritize some corrections over others.

Yes - that is why many equilizers use 20 bands rather than 10. 10 is a nice start, but the number really should be raised.

 

[gasolin75]

Yes - that is why many equilizers use 20 bands rather than 10. 10 is a nice start, but the number really should be raised.

Mabye ask for that if theres is a thread of what you would like to see in a future update (pro equalizers often have 2x30 band adjustments)

 

[dominikz]

Yes - that is why many equilizers use 20 bands rather than 10. 10 is a nice start, but the number really should be raised.

IME 10-bands PEQ per channel is usually quite enough for the purpose of room correction (which implies low-frequency EQ only).

EQ at higher frequencies is not really room correction anymore - we should think about it as loudspeaker correction and/or tonality tuning. For this purpose we can now use additional 10 bands of PEQ per channel (in addition to RoomFit).

So I don't really see the number of PEQ bands as a practical limitation anymore.

 

[gasolin75]

Peq is not limited to 10 fixed frequencies like the eq is on a wiim amp pro

 

[dtc]

IME 10-bands PEQ per channel is usually quite enough for the purpose of room correction (which implies low-frequency EQ only).

If you only use it for low frequency, 10 is probably OK. For a larger frequency range, like in the graphs above, more bands would do a better job. If it is only to be used for low frequency then the tool should be limited to that. If you can do 20 to 20K it should be designed to work over the entire range. IMO

 

[dominikz]

Peq is not limited to 10 fixed frequencies like the eq is on a wiim amp pro

Both WiiM Amp Pro and WiiM Amp Ultra EQ function supports either Parametric EQ (configurable frequencies, gain and Q) or Graphic EQ (fixed frequencies with only gain adjustment).
You can see both "tabs" in this screenshot below:

Either type of EQ (GEQ or PEQ) can be enabled in addition to RoomFit.

 

[dominikz]

If you only use it for low frequency, 10 is probably OK. For a larger frequency range, like in the graphs above, more bands would do a better job. If it is only to be used for low frequency then the tool should be limited to that. If you can do 20 to 20K it should be designed to work over the entire range. IMO

Well if WiiM asked me, I'd suggest to limit the RoomFit correction range to a maximum of about 500Hz anyway instead.

Full-range (20Hz to 20kHz) matching of an in-room steady-state response towards a generic target curve is simply not a very good approach to room correction. Quite the contrary, this can actually be quite detrimental to sound quality because it compensates for audibly benign response deviations at higher frequencies (caused by reflections), creating new and audible resonances in the direct sound of a loudspeaker in the process.

This effectively means that by allowing full-range correction in RoomFit WiiM run the risk of making people's systems sound worse with RoomFit than without it - which should never really be the case (if appropriate configuration is used).

A lot more detail and explanations around this topic can be found in dr. Floyd Toole's book "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms" (which I'd highly recommend reading to anyone interested in HiFi).

 

[gasolin75]

Yes - that is why many equilizers use 20 bands rather than 10. 10 is a nice start, but the number really should be raised.

15-band graphic EQ + 10-band parametric EQ on a eversolo play which is a more expensive alternative to a wiim amp ultra

 

[gasolin75]

Well if WiiM asked me, I'd suggest to limit the RoomFit correction range to a maximum of about 500Hz anyway instead.

Full-range (20Hz to 20kHz) matching of an in-room steady-state response towards a generic target curve is simply not a very good approach to room correction. Quite the contrary, this can actually be quite detrimental to sound quality because it compensates for audibly benign response deviations at higher frequencies (caused by reflections), creating new and audible resonances in the direct sound of a loudspeaker in the process.

This effectively means that by allowing full-range correction in RoomFit WiiM run the risk of making people's systems sound worse with RoomFit than without it - which should never really be the case (if appropriate configuration is used).

A lot more detail and explanations around this topic can be found in dr. Floyd Toole's book "Sound Reproduction: The Acoustics and Psychoacoustics of Loudspeakers and Rooms" (which I'd highly recommend reading to anyone interested in HiFi).

Room correction i also find to be a tool for getting a flatter sound (not to bright sound)

With a sub, flat target curve is very bright but B&K gives me more bass and a less bright sound then flat and is what i use

Do to the 3 target curve you not limited to a fixed target curve so you can have a little flexibility to how you wanna have it adjusted

If you use spinorama setting https://www.spinorama.org/ im shure in most case it wont be good, because there is almost no chance the speakers will have the same frequency responce do to the difference of the room compared to measured FR in the link spinorama link

Meaning you adjust it according to the spinorma link BUT it won't be flat or 98-100% flat, do to how the room effects the sound (your room vs the room used in the spinorama link)

So use a mic to get it adjusted aut THEN you can adjust it using the eq if you want it brighter,darker or more bass then the target curve flat,BK or harman

 

[dominikz]

Room correction i also find to be a tool for getting a flatter sound (not to bright sound)

With a sub, flat target curve is very bright but B&K gives me more bass and a less bright sound then flat and is what i use

Do to the 3 target curve you not limited to a fixed target curve so you can have a little flexibility to how you wanna have it adjusted

If you use spinorama setting https://www.spinorama.org/ im shure in most case it wont be good, because there is almost no chance the speakers will have the same frequency responce do to the difference of the room compared to measured FR in the link spinorama link

Meaning you adjust it according to the spinorma link BUT it won't be flat or 98-100% flat, do to how the room effects the sound (your room vs the room used in the spinorama link)

So use a mic to get it adjusted aut THEN you can adjust it using the eq if you want it brighter,darker or more bass then the target curve flat,BK or harman

Spinorama.org contains anechoic (i.e. just the loudspeaker without any room interference) loudspeaker measurements.

The EQ presets on spinorama.org are intended for use *above* the room transition frequency. This is "loudspeaker response correction".

RoomFit (or any other room correction EQ) should be used *below* the room transition frequency. This is "room correction".

So the two EQ types are intended for completely different purposes - from the quoted post it seems you may have misunderstood this.

If you run RoomFit full range you're effectively trying to make it do both room correction (which it does well) and loudspeaker correction (which it can't do well). The reasons why I've covered in the previous post, but they are covered much better in dr. Toole's book.

Of course anyone is free to use RoomFit however they like, and it is great if you're happy with your results.
My intention is just to clarify the nuances to those who might be interested.

 

[gasolin75]

You have to try alot of things to find what you like

What you or you like, might not be what I like

 

[harkpabst]

You have to try alot of things to find what you like

What you or you like, might not be what I like

One more time:

RoomFit is not about finding a pleasing overall EQ setting.

 

[gasolin75]

Theres only 3 target curve and belive it or not it can change i in the highs with and without a subwoofer

So fare i use my sub

 

[harkpabst]

I'd really like to understand what you are saying.

 

[gasolin75]

Flat sounds much bright with my subwoofer than without my subwoofer

 

[harkpabst]

Flat sounds much bright with my subwoofer than without my subwoofer

1. Why use a flat target curve at all?
2. A subwoofer is intended to increase the low frequency extension of your system, not the amount of bass output. It can still be abused to perform the latter. No matter which way you go, any form of additional low level content can change the perception of highs, but so what?
3. Independent of which target curve you use, there is an option to include subwoofer calibration into RoomFit or not. This will change the behaviour, of course.
4. All this doesn't influence the need (or lack of) of custom target curves. It has been explained in detail why this is not the best approach.

It's still a valid request, of course, but I don't second it.

 

[slartibartfast]

Theres only 3 target curve and belive it or not it can change i in the highs with and without a subwoofer

So fare i use my sub

There are 3 target curves but also the option for stereo or independent channel room correction which effectively gives you 3 more target curves.
The final result depends on the target level chosen by the algorithm. That might be different with or without the sub. That should be evident in the frequency response curves captured during the RoomFit process.

 

[gasolin75]

Flat as someone menstioned isn't flat

I just feel with a subwoofer flat is much brighter than without a sub

I opted to use BK

A subwoofer is intended to increase the low frequency extension of your system

I beet no live event with a big pa system has close to a flat bass,sound