Full range target curves - Page 4 - Home Theater Forum and Systems - HomeTheaterShack.com

Reply
 
Thread Tools
post #31 of 39 Old 08-03-16, 07:02 PM
Elite Shackster
Quenten
 
Join Date: Jun 2007
Location: Tallahassee, Florida
Posts: 2,453
My System
Re: Full range target curves

As long as you are happy with the sound, that's all you can ask for. A couple of thoughts.

Quote:
I'm interested in any info. I've just received my MiniDSP, so I could easily perform a measurement with the correction loaded. Here is the result :

Full range target curves-19-final-ls-3.png

A huge bump at 300 Hz. I then went into new trials to see if I could get something more balanced, but if I try to remove the bump, it doesn't sound right. I just rose the low shelf from -3 dB to -2 dB to get a bit more bass below 200 Hz.
1) It was my understanding that phase needed to be checked without any correction applied.

Quote:
Name: Acoustic1.png Views: 24 Size: 9.3 KB
2) I don't see any mention of speaker "toe-in." The diagram has them pointing straight ahead. I suggest determining the best angle first, as the starting point. It will affect SS&I + room response. Should be optimized before attempting filters.

Quote:
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.
3) The mic should never be moved during the entire set of measurements. If anything, you could play around with small changes in mic position with each set of measurements to get a feel for needed filtering.

4) Use this information to apply absorption/diffusion to get the best room response (speaker postion is first). Then start applying filters.

I am by no means an expert on this subject, but I think these are basic principles. Please correct me if I'm wrong!

Nice thread!

Last edited by Tonto; 08-03-16 at 07:08 PM.
Tonto is offline  
Sponsored Links
Advertisement
 
post #32 of 39 Old 08-04-16, 06:15 AM
Elite Shackster
 
Join Date: Mar 2009
Location: Michigan, USA
Posts: 1,478
My System
Re: Full range target curves

Just my thoughts on a couple of points:

Quote:
Tonto wrote: View Post
1) It was my understanding that phase needed to be checked without any correction applied.
My experience is that the phase tracking in the XO range can be done prior to EQ. The impact of judicious EQ is not a big impact to the phase of a driver. It does have some effect however. After EQ is applied and the SPL response is closer to target it is easier to get a cleaner look at the final phase response of the 2 drivers. At that point we may find that a further adjustment to the timing will optimize the phase tracking. I would expect that small differences in timing in an SW XO range will not have a significant impact on the SPL or sound quality, so this is more just to fine tune the setup for the sake of it.

In many (most?) cases of SW XO there are room modes in the XO range of at least one of the drivers and the modes often fall at different frequencies for the different main channels. This complicates the situation and often the timing selection is thus a compromise to achieve the best overall SPL response. I would think the SPL response is the main focus here and addressing phase is more to provide confidence that the best solution has been found.

Quote:
3) The mic should never be moved during the entire set of measurements. If anything, you could play around with small changes in mic position with each set of measurements to get a feel for needed filtering.
Averaging in the LP area is a good way to helps smooth the midrange and HF SPL response in a given listening area and thus help with choosing an effective EQ solution. Whether a single LP measurement or an average measurement is used, measurements should be done one channel at a time. An SPL average of the 2 channels can be calculated in REW to address common SPL issues with a common EQ solution for both channels if that is the plan. The reason we don't measure with L+R active using sweep measurements is that the HF response is strongly impacted if the mic is even slightly off center due to phase differences. If a single measurement is taken on each channel and summed in REW it will closely agree with a measurement of L+R if the mic is exactly equal distance from the 2 channels. If the mic is off center the HF will be suppressed artificially.

If the RTA function is used for measuring with L+R operating it is likely the same or very similar to averaging the 2 channels independently and then summing them. The RTA only captures the SPL and thus the phase differences are not an issue. The RTA MMM method can be used it achieve quick and efficient averaging in an area and it provides very high repeatability.
jtalden is offline  
post #33 of 39 Old 08-04-16, 11:22 AM
Elite Shackster
Quenten
 
Join Date: Jun 2007
Location: Tallahassee, Florida
Posts: 2,453
My System
Re: Full range target curves

jtalden wrote:

Quote:
Averaging in the LP area is a good way to helps smooth the midrange and HF SPL response in a given listening area and thus help with choosing an effective EQ solution. Whether a single LP measurement or an average measurement is used, measurements should be done one channel at a time. An SPL average of the 2 channels can be calculated in REW to address common SPL issues with a common EQ solution for both channels if that is the plan.
Agreed, and all these things should be done first before EQ. I get the feeling our thread starter is missing these kind of basic set up priorities & jumping right into EQ. We really need to maximize his in room response 1st. What kind of treatments is he willing to get? I can't tell if he wants the room to be an average of several LP's or just one. Need to know that in order to proceed since they both have their own & different process.

Maximizing his room will save power for headroom when applying filters later.
Tonto is offline  
 
post #34 of 39 Old 08-06-16, 11:53 AM
HTS Moderator
Reviewer
 
AudiocRaver's Avatar
Wayne Myers
 
Join Date: Jun 2012
Location: Lincoln, NE
Posts: 4,836
Re: Full range target curves

Fun thread! Lots to think about.

My 2-cents worth, most of this has already been covered:
  • Never be afraid to try EQ whatever the state of your room and treatments.-Room treatments will almost always improve the sound, often dramatically, do as much as you can.
  • Do not over-do filtering. A little goes a long ways. I have been very happy with 3 to 6 bands total per side with some setups.
  • Do not sweat the sharp dips. You will want to push them up, until you hear how awful the result can sound. Leave them alone. I am not saying that you cannot hear them, only that you cannot hear them MUCH, and they sound less objectionably than the alternative.
  • All target curves are wrong. To be more accurate, no target curve is "right." Some have reasons, some depend on someone's research, all are based on assumptions which do not apply to you or your situation. Here is my own favorite target curve, which will also not apply, but here it is anyway. Try a number of "favorite target curves," then take what you think is best in them and create your own.
  • Speaker position, including toe-in, can make a big difference in soundstage & imaging (SS&I). Set speakers for best SS&I, which will probably cause some HF loss, then correct the high end with EQ.
Here is mine:
Hz Level
20 0
24 +3
60 +2
90 +1
120 0
1000 0
2000 +1
3000 +1
5000 0
12000 -2

Reasons:
  1. I am not a monster bass fan. I like the sound pretty flat. But the Equal Loudness Contours certainly justify SOME bass boost, although it is compensated for already in the mix in many/most cases. A little lift below 120 Hz adds a pleasant amount of solidness to the low end without being distracting.
  2. Any boost above 120 Hz contributes to a "boxy" sound very quickly, so the bass boost of the curve is over and down to zero by that frequency.
  3. Having auditioned many models, most speakers have a little emphasis in the region around 1.5 to 3 kHz. The few models I remember hearing that were completely flat through that range sounded drab and lifeless. I little lift, only a dB or so, really livens the sound, yet is small enough to still consider the sound very accurate. Note that this much HF energy will sound harsh with tweeters that are not super clean, super fast, and super smooth in their response. I believe it is tweeter designs rather than acoustics that have listeners shying from extended high frequencies most of the time. If you do not have great tweeters, ditch your speakers and get some that do. You will be glad for many years to come.
  4. The 2 dB drop from 5 kHz to 12 kHz takes the edge off of the highs, reducing listener fatigue, yet still sounds very flat.
  5. The result still falls within a +/- 3 dB tolerance, so can still be considered flat by all but mastering lab standards.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 signatures.
AudiocRaver is offline  
post #35 of 39 Old 08-24-16, 05:59 PM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Sorry, I didn't see this page of the discussion !

Quote:
Tonto wrote: View Post
2) I don't see any mention of speaker "toe-in." The diagram has them pointing straight ahead. I suggest determining the best angle first, as the starting point. It will affect SS&I + room response. Should be optimized before attempting filters.
The diagram is wrong. I setup the toe-in quite early in the process so as to get a large stereo image, although the speakers are closer to each other than they should be if the rule of the equilateral triangle would have been followed.
All trials were done after the toe-in was fixed. I can't remember if the whole set of measurements was done after, or before. All that I know is that if the measurements were done with a different toe-in, then I have checked the effect with additional measurements.
I did some additional measurements, at least to test the effect of the shallow table standing before the sofa (mostly nothing, just a little peak in the mids, among dozens of other ones), of a different speaker position, of a different setup of the acoustic controls on the monitor, and to compare left+right vs left alone and right alone, but i don't remember if I tested the toe-in.

Quote:
Tonto wrote: View Post
3) The mic should never be moved during the entire set of measurements. If anything, you could play around with small changes in mic position with each set of measurements to get a feel for needed filtering.
Quote:
Tonto wrote: View Post
We really need to maximize his in room response 1st. What kind of treatments is he willing to get? I can't tell if he wants the room to be an average of several LP's or just one. Need to know that in order to proceed since they both have their own & different process.
Following the advice of a french professional, I made 6 sets of 3 measurements (left speaker, right speaker, both speakers), from 6 different locations : at the listening position, laid back (above the back of the sofa), at the listening position, bent forward (above the front side of the sofa), 40 cm on the left (back and front), 40cm on the right (back and front). Then I made two additional control measurements, one 15 dB SPL below, to check the effect of background noise on the measurement (none, the result was nearly the same), and another one identical to the first one, to check the stability of the measurement chain over time (none, the last measurement was mostly identical to the first one).
From these 20 measurements, I made an average of the 6 ones from the left speaker alone, the 6 ones from the right speaker alone, and the 6 ones with both speakers.
I made two separate sets of filters, a stereo one from the left/right averages, and a mono one, from the average of both speakers. And listened to the result. The difference was small compared to the adjustments that could be made to the target curve.
I decided to go for a mono correction because, listening to sweep tones, the stereo filters introduced localized imbalances in the sound at some given frequencies, but before all because I've read that because of phase issues, in low frequencies, the total SPL of both speakers is by no means equal to the sum of the SPL of each speaker playing alone. Since low frequencies are more often mixed in mono than in stereo, starting from the left+right measurements should lead to a better result in low frequencies.

Quote:
jtalden wrote: View Post
My experience is that the phase tracking in the XO range can be done prior to EQ. The impact of judicious EQ is not a big impact to the phase of a driver. It does have some effect however. After EQ is applied and the SPL response is closer to target it is easier to get a cleaner look at the final phase response of the 2 drivers.
Wow, I didn't think about that. Fortunately, I did no correction around 2 kHz, where my crossover is working.

Quote:
jtalden wrote: View Post
In many (most?) cases of SW XO there are room modes in the XO range of at least one of the drivers and the modes often fall at different frequencies for the different main channels. This complicates the situation and often the timing selection is thus a compromise to achieve the best overall SPL response.
Since I have no sub, I can avoid this problem too

Quote:
jtalden wrote: View Post
The reason we don't measure with L+R active using sweep measurements is that the HF response is strongly impacted if the mic is even slightly off center due to phase differences. If a single measurement is taken on each channel and summed in REW it will closely agree with a measurement of L+R if the mic is exactly equal distance from the 2 channels. If the mic is off center the HF will be suppressed artificially.
Yes, I experienced this, and didn't understand what was happening. It was you who gave the answer, above in this discussion.

Although the measured response in high frequencies is not very smooth, I decided not to correct it. The speaker, measured by two independent reviewers, are supposed to be flat from 50 Hz to 20 kHz, and I rather trust these measurements, done with calibrated mics, than mine, done with the Umik.

But here is an interesting point : if the room itself causes accidents in the mids or highs (measured by myself at the listening location, but not by the reviewers), should they be corrected ?
It all depends on the way we define "high fidelity". If the goal is to have the feeling that the musician have come into the living room, then the acoustics of the room should not be corrected. Only the native speaker response (and its development in the 3d space all around) should be, except in the lows, where even the musicians would protest about the sound quality of the room before starting to play.
If on the other hand the goal is to have the feeling that we left the living room to be transported to the concert hall, then the room acoustics should be corrected together with the speakers response.

Quote:
AudiocRaver wrote: View Post
[*]Do not sweat the sharp dips. You will want to push them up, until you hear how awful the result can sound. Leave them alone. I am not saying that you cannot hear them, only that you cannot hear them MUCH, and they sound less objectionably than the alternative.
Good advice. For shallow dips in low frequencies, I found a trick : I set a low shelf filter that boosts all the lows, and adjust the nearby peaks equalization in accordance. The result is that all the frequency zone between the peaks is raised in a way that can't be done with a parametric filter.

Quote:
AudiocRaver wrote: View Post
[*]All target curves are wrong. To be more accurate, no target curve is "right." Some have reasons, some depend on someone's research, all are based on assumptions which do not apply to you or your situation. Here is my own favorite target curve, which will also not apply, but here it is anyway. Try a number of "favorite target curves," then take what you think is best in them and create your own.
Thanks. Eventually, I lost the idea of a "target curve". All I did was
1-Adjust the treble setting on the speakers themselves (affects the frequencies above 8 kHz, the manufacturer says that this control is useful if the acoustics of the rooms are overly damped, or bright).
2-Cancel the two first room modes at 55 and 70 Hz (according to the SPL measured level at 1 kHz for the time being).
3-Equalize carefully the acoustic resonance with the wall behind the speakers : reduce the level at 200, 330 and 580 Hz, boost the level at 100 Hz.
4-Adjust the subjective bass / treble balance by ear, using a low shelf at 1000 Hz (my set of filters act over the whole 50-1000 Hz range).
5-Adjust the 55 and 70 Hz filters again in the continuity of the rest.

In the end, the "target curve" was decided during step 3, when I began to try the two filters proposed by jtLaden, and a third one, and fixed their values listening to a playlist of various recordings with voices.

Full range target curves-25-harmonic4.png

Filters 1, 2 : room modes
Filters 3, 4, 5, 6, 7, 8 : acoustic interferences between speakers and wall
Filter 9 : personal house curve
Pio2001 is offline  
post #36 of 39 Old 08-24-16, 09:45 PM
Elite Shackster
Quenten
 
Join Date: Jun 2007
Location: Tallahassee, Florida
Posts: 2,453
My System
Re: Full range target curves

Sorry, I kinda lost track of this conversation. Are there any kind of treatments in the room (don't remember & haven't gone back to verify). It would be appropriate to treat the room before any eq (absorption & diffusion). Play with this until you get it the best you can & then eq off the final measurement. Sounds like you are getting a handle on it. And I think we all feel that treating/eq'ing the room such that it minimizes the affect it has on the speakers frequency response is the goal. Nice thread.
Tonto is offline  
post #37 of 39 Old 08-25-16, 06:03 AM Thread Starter
Shackster
 
Pio2001's Avatar
 
Join Date: Jul 2016
Posts: 30
Re: Full range target curves

Sorry. No there are no treatments in the room.

The speaker placement was decided before setting up the furniture, long before I started equalizing, when I moved in this flat (I had different speakers at that time, a pair of modified Dynaudio Gemini).
There were only the empty shelves, tables, chairs, a single armchair, and a lot of unpacked boxes all around. The first thing I did was setting up the CD player, amplifier, and speakers on their stands, and try all the dispositions in all the rooms.

The best sound was in the living room, by far. But in this room, anything, speaker or listener, placed in a given half, produced terrible resonances in the low frequencies, so both speakers and listener positions were chosen to be in the other half of the room. The direction was decided so that people sitting in front of the speakers could see the window and the trees outside. The good thing is that this way, there is no wall behind the listeners.

The first position for the speakers was 25 cm ahead of the current one, farther away (about 1 meter) from the wall... that is not a wall, but a large window, in fact. The 55 and 70 Hz resonances were weaker.
But when I got the Neumann speakers, the manual stated that they should not be positioned more than 80 cm away from the wall, to avoid acoustic interferences. With a native response of 0 dB at 50 Hz, and this close to the wall, the 55 / 70 Hz peaks were too strong, so I decided to equalize.
I managed to manually cancel them without a measurement microphone, using a sine generator to find the right frequencies to correct, and creating a convolution file with Rephase to load in Foobar2000. But the dip at 100 Hz didn't sound right, and other audio sources (Youtube, Blu-rays, CD Player) could not be equalized. That's where I discovered the world of digital equalization, and went for the Umik and the MiniDSP.

Now that I can equalize, I could move the speakers ahead again, but all the filters from 3 to 8 would have to be redone, and that's too much work. Maybe the peaks number 7 and 8 would just have to be translated to the left, but the peak 5 is more complicated. It is an average of two different peaks at two different frequencies on the left and right channels, because the left speaker is near a lateral solid wall, and not the right one, and the 100-200 Hz zone is a complete mess anyway.

Setting up the 55 / 70 Hz cancellation is a piece of cake. REW does it automatically and it does it right too. But finding a proper, naturally balanced correction above 100 Hz is an awful lot of work ! It took me 40 days, and around 100 hours of measurements, trials and critical listening to get to this point, and I won't get through the whole process again this year !

About acoustic treatments on the walls, as long as the speakers don't move, there should be no need of resetting completely the low-mid filters. But I'm a tenant, and the physical modifications I can make to the walls are very limited.
Besides, I'm just allergic to manual work

Last edited by Pio2001; 08-25-16 at 06:12 AM.
Pio2001 is offline  
post #38 of 39 Old 08-25-16, 10:23 AM
HTS Moderator
Reviewer
 
AudiocRaver's Avatar
Wayne Myers
 
Join Date: Jun 2012
Location: Lincoln, NE
Posts: 4,836
Midrange response problems are often the result of reflections. EQ will not help, only reflection control.
AudiocRaver is offline  
post #39 of 39 Old 08-25-16, 11:33 AM
HTS Moderator
 
ajinfla's Avatar
AJ
 
Join Date: May 2009
Location: Tampa FL
Posts: 1,604
My System
Re: Full range target curves

Quote:
Pio2001 wrote: View Post
But here is an interesting point : if the room itself causes accidents in the mids or highs (measured by myself at the listening location, but not by the reviewers), should they be corrected ?
Only if you trust single mic pressure readings by your eyes over your 2 ears/brain.
Science has found that the latter is not the same as the former. There is a great deal of "listen through" and subconscious adaptation involved that doesn't show up in pressure readings.
It has also found that if your onset (free field response) is smooth over a wide arc, the "room" diminishes as a factor. So just like the sound of a relative or a Steinway sounds like a Steinway in your living room, a small lounge or a concert hall, a good loudspeaker remains "good" in varying environments (outside reductio absurdum). At least to ears. YMMV.

Quote:
Pio2001 wrote: View Post
It all depends on the way we define "high fidelity".
Or "preference".

cheers

AJ

To view links or images in signatures your post count must be 10 or greater. You currently have 0 signatures.
ajinfla 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