Full range target curves - Page 3 - Home Theater Forum and Systems - HomeTheaterShack.com

Reply
 
Thread Tools
post #21 of 39 Old 07-11-16, 05:13 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

PS : third hypoithesis: the bass sounds sooo bad in my living room (very long resonances) that the only thing that sounds right is to remove it completely.
Pio2001 is offline  
Sponsored Links
Advertisement
 
post #22 of 39 Old 07-12-16, 06:39 AM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Never mind. I was just using too narrow filters or unsuited low shelfs to adress the 400 Hz bump. With a parametric filter of q=0.77 it is good.

Yes, I should stop spamming with every minor modification in my filters. I'm just in the part of the process where I'm drowning in a sea of filters
Pio2001 is offline  
post #23 of 39 Old 07-12-16, 09:47 AM
Elite Shackster
 
Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,400
My System
Re: Full range target curves

Thanks for the file.

I didn't find anything particularly noteworthy relative to phase. Nothing beyond what is normal for a live room.

I did look at to see what I would first try in regard to EQ. It appears that a common filter for both channels may work out pretty well. Most all the major discrepancy between channels is <100Hz where many have said that the bass is often mixed as mono anyway. I don't know how true this is, but that would help with the smoothness of the bass in this case.

There may be better progress to be made with different speaker/LP positions in this room, or with some acoustic treatment. Setting those options that aside, you may want to try the attached filter set. HF filter (#5) just brings down highs a little to meet the house curve I prefer. If this is too much roll-off of the highs for your preference just reduce the level or eliminate this filter. Attached is the modified .mdat file. Note that, I summed the L, R and use that to create this filter set. Your L+R measurement was done with the mic a little off-center so the HF was artificially rolled off.

Full range target curves-eq-1.png

Pio2001-ja-1.mdat
jtalden is offline  
post #24 of 39 Old 07-12-16, 02:32 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Quote:
jtalden wrote: View Post
Your L+R measurement was done with the mic a little off-center so the HF was artificially rolled off.
Oh, so this is the cause of all these variations that I get from a measurement to another ! I was beginning to wonder if my speakers had a different amount of treble everyday.

Thank you for your curve. It's interesting.
I've tried your filter on my test tracks. First, I had to reduce the amount of bass for the two peaks at 55 and 70 Hz. I don't know what's going on here. I suspect that there is so much reverberation in the room at these frequencies that this part of the sound seems to play completely apart from the rest of the music. The new predicted curve is

Full range target curves-jtalden-new.png

Now the result is interesting. Better on metal (Nightwish, Whithin Temptation). It doesn't have the cavernous effect of my equalization. But on the other hand, there is a harshness higher in the midrange that affects the voices.
With your curve, the music seem well-balanced, but with harsh vocals. With the one I currently use, the vocals seem better, but the overall balance is affected by an excess of low-mid frequencies.

For the time being, I think I prefer my curve, but I keep yours at hand just in case I get tired of listening to mine.
Pio2001 is offline  
post #25 of 39 Old 07-12-16, 03:07 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

I'm so bad at evaluating frequecies !
The harshness was not in the mids, but at 8000 Hz. I removed the treble filter, as you proposed, and now the result is better. The vocals are good and the overall balance is good too !

Thank you very much. I'm going to load the new correction in the MiniDSP
Pio2001 is offline  
post #26 of 39 Old 07-12-16, 03:36 PM
Elite Shackster
 
Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,400
My System
Re: Full range target curves

There are strong reflection nulls, both at 275 Hz and also near 400 Hz. These are possible contributors to the harshness. There is little PEQ can do about that. If the offensive frequencies can be identified then they can be suppressed a little in level via a house curve adjustment to reduce the emphasis on the problem. Here, I tried to softening the adjoining peaks with filters 3 and 4, but that didn't work out. It may well be better to just remove those filters, as is often advised. PEQ filters often tend to be counterproductive in this frequency range. Room placement and room acoustic changes are the far better approach in this range. Good Luck.
jtalden is offline  
post #27 of 39 Old 07-12-16, 04:25 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Yes, yes, it worked. Look at my second message. Your eq did the job just right.
Pio2001 is offline  
post #28 of 39 Old 07-12-16, 04:36 PM
Elite Shackster
 
Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,400
My System
Re: Full range target curves

Oh, I missed that! Thanks for pointing it out.
I had no real expectation that this would be that close your preference. It was just an alternate starting point in case it differed significantly from what you were already evaluating. I'm glad to hear you are getting close.
jtalden is offline  
post #29 of 39 Old 08-02-16, 04:54 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Hi,
My filters have not changed for two weeks. I think they are good now. I'd like to share an interesting finding with you.

Thanks to jtladen, I was able to have a good correction in the low-mid range with two filters at 327 and 580 Hz. Later, I found that a minor peak at 200 Hz was causing a major audible annoyance. So I added a third filter at that frequency. The three filters, isolated from the rest of the correction, look like this :

Full range target curves-20-harmonic.png

What's interesting is that they match the effect of the reflections of the acoustic waves on the wall behind the speakers :

Name:  Acoustic1.png
Views: 127
Size:  9.3 KB

Since the speakers are 80 cm away from the rear wall, there should be a cancellation at 100, 300, 500 etc, together with peaks at 200, 400, 600 etc.
However, this is true in the direction perpendicular to the wall. In reality, we must take into account the energy radiated around this direction :

Full range target curves-acoustic2.png

Which means that the result should be dips just below 100 Hz, below 300 Hz, below 500 Hz... and peaks below 200 Hz, below 400 Hz etc.
...Which is exactly what can be seen on the measurement, except for the peak at 200 Hz, that is weak, and not really below 200. Maybe it is partially masked by some other reflections.

Here are the 6 raw measurements (both speakers active) with 1/24 octave smoothing, which show exactly what's happening.

Full range target curves-24-harmonicoverlay.png

Below 100 Hz, the two main peaks are perfectly dealt with using the auto correction, provided that we know exactly what should be the target level. I used trial and error to find this, and, to get a better overall balance, I added two low-shelf filters. One at 200 Hz in order to restore the level around 100 Hz, and another one at 950 Hz to adjust the overall bass / treble balance.
Trying to apply a continuous slope rising from 1000 Hz to 20 Hz didn't do any good. The result is better with +1 dB from 950 Hz, then 2 more from 200 Hz.
Here are all the filters together. The interesting thing with the 200 Hz low shelf is that is allows to boost the 100 Hz zone without creating any annoying peak. It needs to be applied before adjusting the amplitude of the parametric filters, if we want to match their level with a given reference.

Full range target curves-21-harmonic.png

And here is the final measured result, with the correction loaded in the miniDSP, in variable smoothing and psychoacoustic smoothing.

Full range target curves-23-harmonicmvar.png

Full range target curves-22-harmonicmpsy.png

I didn't manage to correct completely the dip at 100 Hz. Things sound worse if I rise the level more.

What's interesting is that dealing with much higher frequencies, until 580 Hz, worked well, while we could expect that it would be much more risky to change anything so high in frequency.

A possible explanation is that the reflections coming from the wall behind the speakers are affecting the sound coming from the direction of the speakers, which would mean that they can be safely corrected.
Another possible explanation is that it is always better to cancel peaks than to rise dips.

Overall, the final curve doesn't look much like the usual house curves seen here and there. I don't know what to think about it. But at the same time, I'm not sure that I can trust the Umik measurement microphone to this point. It can surely spot local peaks and dips, but is it really capable of reading a 2 dB imbalance between bass and treble ?
Pio2001 is offline  
post #30 of 39 Old 08-02-16, 07:17 PM
Elite Shackster
 
Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,400
My System
Re: Full range target curves

Thanks for sharing.
The common measuring mics are all pretty flat from maybe 40-4k Hz. I would also expect this to be the easiest range to calibrate also. So, if the MiniDSP cal data is fairly smooth and flat in this range then that would be normal and reassuring.
If there is a problem with the mic or the calibration then it is more likely to occur at the upper or lower frequencies.
I wouldn't be too concerned with the bass level you chose. While most prefer a stronger bass level there is large variability in preference there. I would think the room, program material and other factors would influence this so there is no right answer. The good news is you now have the facility adjust the house curve easily if you ever do decide to try something different.
Enjoy!
jtalden is offline  
Reply

Bookmarks

Tags
curves , full , range , target

Quick Reply
Message:
Options

Register Now




PLEASE COMPLETE ALL REQUIRED FIELDS BELOW... THANKS!

REQUIRED FIELDS ON THIS PAGE
YOU MUST COMPLETE ALL OF THESE

Username
Password
Confirm Password
Email Address
Confirm Email Address
Random Question
Random Question #2




User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.

Password:


Confirm Password:
Email Address
PLEASE READ BELOW PRIOR TO ENTERING AN EMAIL ADDRESS!

ATTENTION!

YOU MUST ACTIVATE YOUR ACCOUNT!

Activation requires you reply to an email we will send you after you register... if you do not reply to this email, you will not be able to view certain areas of the forum or certain images... nor will you be able download software.

AN INVALID EMAIL ADDRESS WILL CAUSE YOUR ACCOUNT TO BE DELETED!

See our banned email list here: Banned Email List

We DO NOT respond to spamcop, boxtrapper and spamblocker emails... please add @hometheatershack DOT com to your whitelist prior to registering or you will get nowhere on your registration.


Email Address:
OR

Log-in









Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.



Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page



Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML is not allowed!
Trackbacks are Off
Pingbacks are Off
Refbacks are Off

 


For the best viewing experience please update your browser to Google Chrome