A collection of speaker target responses in .csv/.txt format

[staticV3]

Your Toole curve is based on the typically reflective room curve. But there’s also the curve mentioned in this post. It rolls of more on the high end. Shouldn’t this actually be the Toole target curve?

Might be. From personal experience I can say that I prefer a more linear treble roll off. Do you want me to convert it?

 

[DeLub]

It stops at about 60-70Hz. Perhaps I’ll try the other Toole curve first.

 

[digitalfrost]

Interesting. I tried to scan these curves with various tools but the results were always unsatisfying to me. I ended up writing some python code to create these curves myself. It's a biquad shelving filter around 100hz for the bass level and then a linear drop from 400-500hz towards 20khz.

Looking at the graphs, seems close to rtings for me. Advantage is, you can parametrize bass and treble level.

 

[milus]

Interesting. I tried to scan these curves with various tools but the results were always unsatisfying to me. I ended up writing some python code to create these curves myself. It's a biquad shelving filter around 100hz for the bass level and then a linear drop from 400-500hz towards 20khz.

Looking at the graphs, seems close to rtings for me. Advantage is, you can parametrize bass and treble level.
View attachment 89092

Maybe this can help you.

There is a chrome extension, too: https://chrome.google.com/webstore/detail/webplotdigitizer/blomnnpciekdhecimpnjfcchdicopdii?hl=en

Here too: https://github.com/ankitrohatgi/WebPlotDigitizer

 

[gorman]

I've used this target and matched target based on my 8 measurements average from slightly different positions around my listening position (my head, sitting at a desk, I use iLoud Micro loudspeakers). Result below. It says no smoothing because the average is calculated based on 8 var smoothed measurements. So I though further smoothing was not needed. If I'm wrong in this let me know.

Anybody here knowing how best to proceed on the above?

I have 8 measurements, taken with VAR smoothing in REW. I average them to then proceed to calculate filters. The average is not smoothed. This leads to the EQ curve not being smoothed as well. Is this correct or should I smoothen the average and then calculate the filters?

I EQ all frequencies as these are studio monitors, extremely nearfield. About 60/70 cm from tweeters to my ears.

This is the result without VAR smoothing:

This with smoothing:

This is the equalization curve that Peace (EqualizerAPO) shows as applied for the filters with no VAR smoothing in calculations:

This is the equalization curve that Peace (EqualizerAPO) shows as applied for the filters with VAR smooth in calculations:

Is one approach preferable for some reason? Generally speaking the sound is very similar but I don't know if I'm creating ringing or whatever mistake I might be doing in not smoothing. That's harder to "tell by ear" (for me at least). If this question would be better asked in another discussion, I would greatly appreciate it if you could point me to it.

 

[ernestcarl]

Anybody here knowing how best to proceed on the above?

I have 8 measurements, taken with VAR smoothing in REW. I average them to then proceed to calculate filters. The average is not smoothed. This leads to the EQ curve not being smoothed as well. Is this correct or should I smoothen the average and then calculate the filters?

I EQ all frequencies as these are studio monitors, extremely nearfield. About 60/70 cm from tweeters to my ears.

This is the result without VAR smoothing:

View attachment 89359

This with smoothing:

View attachment 89360

This is the equalization curve that Peace (EqualizerAPO) shows as applied for the filters with no VAR smoothing in calculations:

View attachment 89364

This is the equalization curve that Peace (EqualizerAPO) shows as applied for the filters with VAR smooth in calculations:

View attachment 89363

Is one approach preferable for some reason? Generally speaking the sound is very similar but I don't know if I'm creating ringing or whatever mistake I might be doing in not smoothing. That's harder to "tell by ear" (for me at least). If this question would be better asked in another discussion, I would greatly appreciate it if you could point me to it.

I would use no smoothing prior to averaging, and then only very minimal like 1/24 or var afterwards. There hardly looks like any difference in your graphs anyway. I’d be more interested in the actual results post EQ. It’s unfortunate though that these appear to need so much correction above the transition zone.

 

[ernestcarl]

Interesting. I tried to scan these curves with various tools but the results were always unsatisfying to me. I ended up writing some python code to create these curves myself. It's a biquad shelving filter around 100hz for the bass level and then a linear drop from 400-500hz towards 20khz.

Looking at the graphs, seems close to rtings for me. Advantage is, you can parametrize bass and treble level.
View attachment 89092

Yeah I’m kind of surprised there’s so much variability to these target curves. Just pick or create one that’s able to match your speakers’ natural curve rather than forcing it to fit would be my preference.

 

[richard12511]

I had also collected and digitised Harman loudspeaker target curves from different publications and averaged them (black line):

View attachment 86030

Attached below is the average as a txt file to use in REW or other programs. Mind you though, target curves depend on loudspeaker directivity and room reverberation and thus equalizing a loudspeaker with unknown directivity to a predefined target doesn't guarantee anything:
https://www.audiosciencereview.com/...ut-room-curve-targets-room-eq-and-more.10950/

Is this the curve you use? I’m a little confused. This is the curve I was using(“ski slope”), but in the other thread you recommended I use a more steadily decline 10db slope. It seems there are different “Harman curves” out there?

 

[thewas]

Is this the curve you use? I’m a little confused. This is the curve I was using(“ski slope”), but in the other thread you recommended I use a more steadily decline 10db slope. It seems there are different “Harman curves” out there?

Yes, there are several "Harman curves" out there, which also isn't surprising as they depend also on the used loudspeakers and rooms.
I often experiment and change them, my current preferred average tends to converge to approximately 0.8dB/oct with the loudspeakers I currently use and the room I have them.

 

[Le Concombre]

I have come to use 2 different Target curves in order to create 1 single set of convolution filters.
Ever read that the Room is in control below Schroeder frequency, the Speaker above ?
Above Schroeder I align L & R DIRECT SOUND responses, assuming that the Direct Sound is correctly represented by REW's Vector Average
Below, I align the STEADY STATE responses as given by "Average the Responses" in REW (RMS Average).

 

[Le Concombre]

I agree with the opinion promoted by ie Barry Diament that correcting the Room response by affecting the Direct sound of the speaker is not... sound (pun intended of course). Hence my Right Speaker's response is left (almost) untouched below Schroeder. The surroundings of each speaker is very different in my room ; unfortunate but I have to deal with it... Please note that there is then no sense to place the target at the same level for both channels. I my case the target for Left is set about 6 dB lower. I then simply compensate the Volume difference in RePhase (O yes btw, it's mandatory to correct Time Domain (phase) in RePhase to get satisfying results).

 

[Le Concombre]

The asymmetry of surroundings causes that the fit is not that good below Schroeder for the Left Channel. OTH the Direct Sound response is smoother.

 

[Le Concombre]

Consequently the fit with the chosen target below Schroeder (Synthesis Arcos applied on RMS Average) is better for the Right channel

 

[Le Concombre]

And the fit with the chosen target above Schroeder (Dr Toole applied on Vector Average) is better for the Left channel.

What really really matter is the alignement : see in #30 how L and R match below Schroeder on the convolved RMS Averages (top curves) and how L and R match above Schroeder on the convolved Vector Averages (lower curves)

 

[Le Concombre]

The L+R convolved Steady State Response aims at Synthesis Arcos and the fit is well within a +- 3 dB tolerance

 

[Le Concombre]

This might look bass heavy but just sounds right. I once advocated a change of convolution filters based on different targets depending on the program. I'm just happy with the one set presented here. Maybe because it respects the most my room's natural bass response? Please note it's only 3 dB above the much touted RR1 bElow 50 Hz while being much leaner in the mud region, especially between 100 et 250 Hz. RR1 might be suited for pop and such or to act as a loudness compensator but is improper for acoustic sounds reproduction IMO while Synthesis Arcos, while beefier below 50 Hz, sounds considerably more natural most probably because of its profiling of the mud region. Works for me with any kind of music.

 

[richard12511]

I have come to use 2 different Target curves in order to create 1 single set of convolution filters.
Ever read that the Room is in control below Schroeder frequency, the Speaker above ?
Above Schroeder I align L & R DIRECT SOUND responses, assuming that the Direct Sound is correctly represented by REW's Vector Average
Below, I align the STEADY STATE responses as given by "Average the Responses" in REW (RMS Average).

Can you add legends? I’m having trouble seeing which curve is the target and which are the real responses.

 

[gorman]

I agree with the opinion promoted by ie Barry Diament that correcting the Room response by affecting the Direct sound of the speaker is not... sound (pun intended of course). Hence my Right Speaker's response is left (almost) untouched below Schroeder. The surroundings of each speaker is very different in my room ; unfortunate but I have to deal with it... Please note that there is then no sense to place the target at the same level for both channels. I my case the target for Left is set about 6 dB lower. I then simply compensate the Volume difference in RePhase (O yes btw, it's mandatory to correct Time Domain (phase) in RePhase to get satisfying results).

All this sounds very interesting but, due to my ignorance, I am sure I am missing something... are you saying Diament suggests not to correct the bass response of speakers? I know Schroeder frequency varies according to room size, but for "normal" sized rooms we talk about bass, mid-bass frequencies at most, right?

 

[gorman]

I would use no smoothing prior to averaging, and then only very minimal like 1/24 or var afterwards. There hardly looks like any difference in your graphs anyway. I’d be more interested in the actual results post EQ. It’s unfortunate though that these appear to need so much correction above the transition zone.

Did as you suggested and retook all measurements after buying phonoabsorbing bases for the speaker stands. Also found out that previous measurements had been taken with an high frequency boost activated in the speakers.
This is the end result (Var smoothing applied only after averaging). Don't know if it's good, wrong, right... it sounds good to my ears. Tried Toole trained listeners tool curve too. I have both EQ settings saved, sometimes one sounds better, sometimes the other does (mainly genre dependent I'd say).

 

[Le Concombre]

All this sounds very interesting but, due to my ignorance, I am sure I am missing something... are you saying Diament suggests not to correct the bass response of speakers? I know Schroeder frequency varies according to room size, but for "normal" sized rooms we talk about bass, mid-bass frequencies at most, right?

https://audiophilestyle.com/forums/...io-truth/page/43/?tab=comments#comment-248537 and he repeated this credo several times