Equalizing loudspeakers based on anechoic measurements (community project)

[Igor Kirkwood]

We could listen to how it sounds. Then, it would be interesting to look for musical content that is sensitive to ON/LW (drums ?) and musical content sensitive to ER/SP (organ, choir ?).

If it works, we could then test the hypothesis that some music would sound better if the speaker is equalized according to the LW curve, and some other style of music would sound better if the speaker is equalized according to the PIR curve.

Very interesting

 

[Maiky76]

First impression: now this is nice! An improvement over the stock 8030C and a lot better subjectively than your previous filterset. Very close to my own filters in terms of result, now this is worth doing blind.

Hi,

When I was playing with the directivity script I just published here:
https://www.audiosciencereview.com/...-cea2034-from-the-nfs-data.14141/#post-481709

I noticed that the DI was marginally smoother when the 8030C was listened at a 20deg angle.
The same EQ applied the scores are actually close to identical but with a steeper slope on the PIR.
How did you set up your speakers?
Can you try with a toe-in of 20deg in front of you and check if it is an improvement?

 

[TimVG]

Hi,

When I was playing with the directivity script I just published here:
https://www.audiosciencereview.com/...-cea2034-from-the-nfs-data.14141/#post-481709

I noticed that the DI was marginally smoother when the 8030C was listened at a 20deg angle.
The same EQ applied the scores are actually close to identical but with a steeper slope on the PIR.
How did you set up your speakers?
Can you try with a toe-in of 20deg in front of you and check if it is an improvement?

Couldn't really detect an obvious difference (sighted test) to be honest! I use them in the living room for general TV watching and background music.
Will post my KH80 impressions soon - it's been extremely hot the past week and locking myself in a small room with hot equipment en no A/C isn't really a pleasant environment

 

[aarons915]

Many know I've been experimenting with the KEF R3 the past month or so and everything I've done has been an improvement over no EQ but I had a bit of a breakthrough. My strategy has been to focus on the LW and ER but out of curiosity I decided to estimate my actual early reflections and use that to EQ and it has resulted in the smoothest sound yet. If you look at the curves that the ER curve is estimated from it's very general so I think most people could benefit from making it a bit more custom to their personal situation. I ignore the rear wall because I have about 15 feet of space behind me so that reflection is way down in amplitude when it reaches me, I focus on the sidewalls, floor and ceiling reflection and with my setup of no toe I noticed all my early reflections are a convenient 30 to 40 degrees. What this means is I can focus on the direct sound and a few angles for my ER instead of the many angles specified in the CTA-2034 ER curve, this is also a good reason to point your speakers straight ahead but I get that it is personal preference.

It's still a good idea to check the other curves to make sure they look ok but I've found this to be the smoothest sound so far and I haven't felt the need to tweak it much further. It also intuitively makes a lot of sense to customize your ER curve because of the many different setups out there, unless your room and setup fit the average very well there is a good chance you may be prioritizing a curve that isn't representative of your space.

Edit: Scratch that somewhat. I made some in room measurements and I took a bit too much out of the ER peak at 2800Hz which tells me that peaks in the ER curve aren't as bad as they seem, probably because of the further distance they travel and hence get attenuated compared to the direct sound. Adjusting that filter based on my in room MMM response makes the ER curve a very good approximation. When you have a situation where you have to compromise between a dip in the LW and a peak in ER curve, using an in room MMM measurement is a good tool to figure out that balance.

 

[Chromatischism]

Wow, this a really interesting thread, and it's something that I'll be doing on my JBL 308p Mkii once Amir reviews them....he's said he's bought them already, so they're in the pipeline somewhere! I don't fully understand all of the conversations in this thread, but I have a basic understanding of quite a lot of the graphs/points being discussed......my initial strategy would be to EQ my speaker above 500Hz using the on-axis spinorama (I toe them in), I'd EQ that flat, and I may also EQ the speaker in the same way for the area below 500Hz too, but I'd of course need to end up putting more filters on top of that sub 500Hz section to allow for roomEQ so it could be argued that you don't need to EQ the speaker below 500Hz and just EQ the room instead below 500Hz.....at this point I'm undecided. I'll definitely be posting my results in this thread once Amir does get his review done for the 308p Mkii.

I think you're on the right track. If you were to measure different speakers at the same location in your room, you would see where the bass response is almost the same curve despite the speakers not being the same. That is where the room is completely taking over. You'll notice a point where they diverge and form their own response, which is between 150 in large rooms, to 350 Hz in small rooms.

From there, there are two schools of thought.

1. Use EQ to adjust the full range of the speaker, correcting the in-room response
2. Use EQ to adjust only up to the in-room transition frequency, then add EQ above that based not on the in-room measurement, but on the anechoic measurements (when available)

And that is what is being tested and refined in this thread. Myself, I can actually see both sides. I am more drawn to the latter due to discussions I took part in with Dr. Floyd Toole, however I do see some merit in what the more advanced room EQ systems are doing. The thing is, they are a tool, and tools can get you worse results if the user isn't carefully guiding them. What I don't think makes much sense is equalizing based on a "predicted" in-room measurement. It would make much more sense to just EQ based on actual in-room measurements.

And I just realized that I replied to an old post!

 

[Robbo99999]

I think you're on the right track. If you were to measure different speakers at the same location in your room, you would see where the bass response is almost the same curve despite the speakers not being the same. That is where the room is completely taking over. You'll notice a point where they diverge and form their own response, which is between 150 in large rooms, to 350 Hz in small rooms.

From there, there are two schools of thought.

1. Use EQ to adjust the full range of the speaker, correcting the in-room response
2. Use EQ to adjust only up to the in-room transition frequency, then add EQ above that based not on the in-room measurement, but on the anechoic measurements (when available)

And that is what is being tested and refined in this thread. Myself, I can actually see both sides. I am more drawn to the latter due to discussions I took part in with Dr. Floyd Toole, however I do see some merit in what the more advanced room EQ systems are doing. The thing is, they are a tool, and tools can get you worse results if the user isn't carefully guiding them. What I don't think makes much sense is equalizing based on a "predicted" in-room measurement. It would make much more sense to just EQ based on actual in-room measurements.

And I just realized that I replied to an old post!

Yep, that was an old old post, ha! Once Amir has tested the JBL 308p I've decided I'm gonna EQ the listening window above the transition frequency, and I may experiment with EQ'ing below the transition frequency in the same way, but as you say the room takes over so I'm not expecting that to yield better results because I'd just end up with 2 sets of EQ filters on top of each other below the transition frequency. And yep I'd use in room measurements to EQ below the transition frequency.

 

[Chromatischism]

I like Audyssey XT32 or Dirac up to 300-400 Hz (corrects the in-room bass problems using high filter resolution), then EQ based on listening window above that using another device. That could be a MiniDSP or software EQ depending on your source. I think if you were to do any EQ in the bass region for the speaker response, you would include that in the chain first for room EQ to pick up on, but I'd like to see other opinions on that. My thinking is that since the post-room EQ response is largely what-you-see-is-what-you-get in the bass region, you'd mess it up if you applied EQ afterward.

 

[Robbo99999]

I like Audyssey XT32 or Dirac up to 300-400 Hz (corrects the in-room bass problems using high filter resolution), then EQ based on listening window above that using another device. That could be a MiniDSP or software EQ depending on your source. I think if you were to do any EQ in the bass region for the speaker response, you would include that in the chain first for room EQ to pick up on, but I'd like to see other opinions on that. My thinking is that since the post-room EQ response is largely what-you-see-is-what-you-get in the bass region, you'd mess it up if you applied EQ afterward.

Yes, I agree, apply the speaker EQ first from the spinorama before taking in room measurements. I'd use the spinorama from Amir to EQ the speaker after the transition frequency (& before the transition frequency if you like too), then you take in room measurements (I'd use UMIK-1 & REW), and then I'd EQ the bass region up to the transition frequency using the results of those measurements.

 

[daftcombo]

Nothing for Aria 906 or DBR-62 or 305p mk2 so far?

 

[Robbo99999]

Nothing for Aria 906 or DBR-62 or 305p mk2 so far?

305p Mkii has already been measured by Amir if that's what you mean, if you do a search you'll find the review.

 

[Chromatischism]

JBL 305P MkII was the very first to go in the Klippel NFS here (well, not including the test speaker included in the kit).

 

[daftcombo]

305p Mkii has already been measured by Amir if that's what you mean, if you do a search you'll find the review.

The three of them were. I meant: no EQ settings for them in this thread?

 

[Robbo99999]

The three of them were. I meant: no EQ settings for them in this thread?

I suppose you could do them if you want, download the spinorama data & import it into REW or something & do the EQ there, probably a number of programs you could use but I only familiar with REW.

 

[sweetchaos]

Nothing for Aria 906 or DBR-62 or 305p mk2 so far?

Don't know if you've seen this...

EQ's done by @ flipflop and put together on github by @ pierre
Raw data: https://github.com/pierreaubert/spinorama/tree/develop/datas/eq
or represented in a more user-friendly way on https://pierreaubert.github.io/spinorama/eqs.html

I'm currently using the EQ filters for my JBL 305P MKII and it sounds awesome!

 

[alont]

I think you're on the right track. If you were to measure different speakers at the same location in your room, you would see where the bass response is almost the same curve despite the speakers not being the same. That is where the room is completely taking over. You'll notice a point where they diverge and form their own response, which is between 150 in large rooms, to 350 Hz in small rooms.

From there, there are two schools of thought.

1. Use EQ to adjust the full range of the speaker, correcting the in-room response
2. Use EQ to adjust only up to the in-room transition frequency, then add EQ above that based not on the in-room measurement, but on the anechoic measurements (when available)

And that is what is being tested and refined in this thread. Myself, I can actually see both sides. I am more drawn to the latter due to discussions I took part in with Dr. Floyd Toole, however I do see some merit in what the more advanced room EQ systems are doing. The thing is, they are a tool, and tools can get you worse results if the user isn't carefully guiding them. What I don't think makes much sense is equalizing based on a "predicted" in-room measurement. It would make much more sense to just EQ based on actual in-room measurements.

And I just realized that I replied to an old post!

From my own experience EQing my Sonus Faber Olympica I speakers, I definitely got much better results using the second method.
Using just Dirac Live to EQ the whole frequency range, the results ranged from "meh" to slight improvements (depending on the target curve I used).
The best results by far were achieve by delegating Dirac Live only to frequencies up to the transition frequency, and using an EQ with filters that smoothed out the speaker's quasi-anechoic LW.

This is all based on my subjective opinion, of course, but I tried to be as objective as I could be during testing - I used blind A/B with level matching, with two of my friends participating as well.

 

[pierre]

Nothing for Aria 906 or DBR-62 or 305p mk2 so far?

flipflop did EQs for these speakers too. Available here.

 

[daftcombo]

Don't know if you've seen this...

EQ's done by @ flipflop and put together on github by @ pierre
Raw data: https://github.com/pierreaubert/spinorama/tree/develop/datas/eq
or represented in a more user-friendly way on https://pierreaubert.github.io/spinorama/eqs.html

I'm currently using the EQ filters for my JBL 305P MKII and it sounds awesome!

Is there a way to transform quickly those PEQ into EqualizerAPO files?
I did it manually for the Edifier but it could be done automatically.

 

[pierre]

Is there a way to transform quickly those PEQ into EqualizerAPO files?
I did it manually for the Edifier but it could be done automatically.

do you have a description of the syntax for EQapo?

 

[Robbo99999]

Is there a way to transform quickly those PEQ into EqualizerAPO files?
I did it manually for the Edifier but it could be done automatically.

I think so, I think you can copy & paste this into a text file created by notepad:
https://raw.githubusercontent.com/pierreaubert/spinorama/develop/datas/eq/JBL 305P Mark ii/iir.txt
That link is simply clicking on the "Raw" tab.
Then in Equaliser APO you click "Open" and open that file you've created. It looks like it's gonna work but not tried it.

 

[sweetchaos]

If anyone is using PEACE interface on top of EqualiserAPO, just hit "Import" which will import the .TXT file with the filters.