Topic: Automated Audio Calibration

[barrister]

Automated audio calibration: What are the limits to what they can do?  They're generally good, but can they also worsen the sound of your setup? 

Discuss the different types of auto audio calibration systems here:

Audyssey,
YPAO (Yamaha),
MCACC (Pioneer),
MRAC (Marantz),
DCAC (Sony),
ADAPTiQ (Bose),
EzSet (Harman Kardon), etc.

[ninjababez®]

Audyssey, somewhat near my preference.  minor adjustment after running the setup.
adjustment made, +2db for center (sometimes +4 depending on the movie) and + 2-4db for the sub. 

[Conan]

Audyssey, somewhat near my preference.  minor adjustment after running the setup.
adjustment made, +2db for center (sometimes +4 depending on the movie) and + 2-4db for the sub. 

Same here! 

[barrister]

YPAO yung akin.  

Dalawa lang ang tweak ko after running YPAO: (a) large front speakers changed to small; and (b) 80Hz crossover changed to 60Hz.

My YPAO is capable of up to 8 points mic position, but it can also do a single-point calibration.  

I decided to use only a single-point calibration.

The way I see it, there can only be one sweet spot.  To calibrate correctly at the sweet spot, the mic must be positioned only at the sweet spot and nowhere else.  



================



For example, let's say your HT has 3 seats: a, b and c, with "b" being the sweet spot.  Let's also say that "100%" is the best on a sound quality scale of 1% to 100%.

- When calibrated using only 1 position (seat "b") --- Let's say that sound quality is 100% at seat "b"; 80% at seat "a"; and 60% at seat "c".

The advantage is that you get the best possible calibration at the sweet spot "b"; the disadvatage is that you get bad sound at seat "c".

- When calibrated using all 3 positions --- You won't get a 100% for all three seats, which I think is impossible.  What you'll probably get is an 80% for each of the three seats.

The advantage is that you get a sound quality of 80% wherever you sit.  

The disadvantage is that you'll never get a sound quality of 100% no matter where you sit.  

[Conan]

With my last calibration I did all 3 seating positions at the same spot. It seems to sound better than moving the mic around. I also use the 80hz as per THX recommendations.

[barrister]

With my last calibration I did all 3 seating positions at the same spot.

That's why I started this thread.  Para makakuha tayo ng kakaibang opinions, hindi yung parang pare-pareho na lang ang sinasabi.  

I also use the 80hz as per THX recommendations.

The THX recommendation is the reason why the 80Hz crossover is popular.  

But my fronts are big floorstanders.  80Hz is too high for my set-up.  Mas maganda talaga sa akin ang 60Hz.  

With an 80Hz crossover, I lose some bass "thump".  With a 60Hz crossover, the bass thump comes back and it sounds like there's more cooperation between front bass and subwoofer.

If you have small front speakers, use 80Hz crossover.  If you have big, bass-capable fronts, also try a crossover setting below 80Hz.


==============



Gene DellaSala thinks the purpose of THX's 80Hz crossover spec is licensing revenue.  Lower the crossover spec, and smaller systems will not be capable of meeting it.  Keep the crossover high at 80Hz so that both small and large systems can handle it.  Here's his audioholics article:

THX has become a recognized industry standard. Aside from THX's proprietary post processing, many of the requirements set by THX can easily achieved by quality electronics but several of these electronics manufacturers choose not to pay the premium to advertise this.

My biggest issue with THX is:

No channel shall have bass info below 80 Hz other than the subwoofer channel.

While this may serve movie theater purposes, it is not always acceptable for home theater in my opinion for my previously mentioned reasons that many of us utilize bass capable speakers in our systems that produce linear bass extension below 80Hz and that the pressure waves of 80 Hz bass frequencies can be localizable to the ear if the subwoofer is placed too closely to the listening position. I believe THX is configured like this for ease of speaker integration and to preserve dynamic range of all channels by limiting their received bandwidth. This in turn may allow for easier certification of less capable power amplifiers and speakers since they are not tasked to produce the low frequencies that require ample power and driver excursions. Therefore this helps manufacturers certify a greater amount of their products resulting in guaranteed system compatibility and consistency. THX carried over much of its philosophies of the movie theater environment to home theater, especially relating to the satellite subwoofer speaker systems. While these systems don't always offer the pinnacle of performance, they usually are easier to integrate and set-up in most home theaters. For those who don't want to fuss with the acoustical tribulations of multiple bass sources, this type of system with a fixed 80Hz crossover may be the right choice. However, for those who enjoy bass capable speakers supplemented with their subs, choosing a receiver/processor with multiple crossover settings and flexible bass management can offer more setup versatility.  

THX is about two things in my opinion:

Their spec.
Compatibility.
 
The reason they stick to their spec is to promote compatibility. Then, regardless of which manufacturer you choose, if you go with a product with the THX certified good housekeeping seal of approval, you know they will work well together. This is turn allows them to attract more manufacturers, which increases the generation of licensing revenue.

http://www.audioholics.com/tweaks/get-good-bass/bass-management-the-right-stuff

[ninjababez®]

YPAO yung akin.  

Dalawa lang ang tweak ko after running YPAO: (a) large front speakers changed to small; and (b) 80Hz crossover changed to 60Hz.

My YPAO is capable of up to 8 points mic position, but it can also do a single-point calibration.  

I decided to use only a single-point calibration.

The way I see it, there can only be one sweet spot.  To calibrate correctly at the sweet spot, the mic must be positioned only at the sweet spot and nowhere else.  

Bro, im not familiar with how yamaha's calibration works, but with audyssey, they always advice to use all mic positions available.  they also have a pattern if in-case you only have 1 sweetspot.  

================



For example, let's say your HT has 3 seats: a, b and c, with "b" being the sweet spot.  Let's also say that "100%" is the best on a sound quality scale of 1% to 100%.

- When calibrated using only 1 position (seat "b") --- Let's say that sound quality is 100% at seat "b"; 80% at seat "a"; and 60% at seat "c".

The advantage is that you get the best possible calibration at the sweet spot "b"; the disadvatage is that you get bad sound at seat "c".

- When calibrated using all 3 positions --- You won't get a 100% for all three seats, which I think is impossible.  What you'll probably get is an 80% for each of the three seats.

The advantage is that you get a sound quality of 80% wherever you sit.  

The disadvantage is that you'll never get a sound quality of 100% no matter where you sit.  


Bro, im not familiar with how yamaha's calibration works, but with audyssey, they always advice to use all mic positions available.  they also have a pattern if in-case you only have 1 sweetspot.  

With my last calibration I did all 3 seating positions at the same spot. It seems to sound better than moving the mic around. I also use the 80hz as per THX recommendations.

bro the 3 position for audyssey is there to measure the room.  it wont be accurate if you dont use all available positions (on ideal locations), IIRC the minimum is 3.
dont worry i used to do that (single mic position) whenever im in a hurry to calibrate the sub and get 0db.  

btw like what barrister replied, we have different fronts (LCR) speakers.  di masamang di THX standard ang gamit mo kung ikagaganda naman sa pandinig mo yung resulta.  

let me add that i was using to do 80hz when i was using a-audio genesis (this BS sounds like it is a floorstander).  now that i replaced it with an a-audio tablette, it is now set to 100hz cutoff.  the new floorstander kasi is more on clarity. (i could be wrong, but it sounded better @ 100hz)  

on the subwoofer part, instead of using the auto settings via SVS bash amp settings, i did manual and set it to 80hz.  im still in the process of tweaking my sub and i think frequency between 81hz - 99hz is not coming out sa speakers ko  
but it sounded better .. according to my preference   (pero maaring pseudo effect lang)  

OT:
my target is still to do a little under reference "standard", but i need to make sure that my room can do that.  next project before tweaking the audyssey settings again is room treatment.  

[barrister]

btw like what barrister replied, we have different fronts (LCR) speakers.  di masamang di THX standard ang gamit mo kung ikagaganda naman sa pandinig mo yung resulta.  

Korek!    

First, know what the experts recommend and why.  Then you can follow your personal preference.  

That way, there's rhyme and reason to your personal tweaks, hindi yung kahit papaanong kapa-kapa na lang.

OT:
my target is still to do a little under reference "standard", but i need to make sure that my room can do that.  next project before tweaking the audyssey settings again is room treatment. 

The best pa rin ang room treatment, IMO.

I don't think any electronic equalization, no matter how sophisticated, can beat proper room treatment.

[praktikal]

YPAO yung akin.  

Dalawa lang ang tweak ko after running YPAO: (a) large front speakers changed to small; and (b) 80Hz crossover changed to 60Hz.

am using 659 at same lang din ang nagiging tweak ko after the auto calibration. may nabasa ako na ok daw ang crossover ng 1 step higher sa lowest setting ng avr. since 40Hz ang sa 659 kaya 60Hz ang pinipili ko. beside, mas gusto ko ang malambot. although kadalasan na auto setting e nasa 120Hz.

pero last week may nagpahiram sa akin ng spl meter, after the auto calibration na set ko sa 75db lahat ng speakers from 1 sweetspot. mas nabuhay ang surround speakers. pero sa tingin ko madami pang pwedeng maimprove pero ok na ako sa setting ko sa ngayon.

Happy New Year po sir barrister at sa lahat. 

[barrister]

No, the crossover does not depend on the settings allowed by the receiver.  It depends on the capabilities of your speakers.  Anyway, yung lambot naman talaga ng bass ang ultimate basis mo sir, kaya OK rin ang set-up.

Using an SPL meter is also good.  It's best to use automated caibration as a starting point, then tweak after auto set-up.

Malakas ang Yammy 659, sir.  Same level yan sa binenta kong 657 (1 year older than 659), kayang-kaya ang malaking floorstander, makapal ang tunog.

Happy New Year!

[iiinas]

i think the main purpose for the auto calibration (specifically audyssey) imho, is not the setting of the levels, crossovers and distances by the program for the user but more on the room correction algorithms that is being calculated by the program to apply on your room to tame it from the ringing and nodes that each of our ht room's will definitely have (even the treated ones). the setting of the levels, distances and crossovers can be easily done with the help of an spl meter, ruler and reading the capability (specs) of your speakers as sir barrister has rightly stated.  

that is why for people using audyssey, it is encouraged to use all the measuring points available, so that the program could really get an accurate reading of the rooms nuances and correctly calculate the correct eq to tame the ht room being measured.

so for me the gem of audyssey is really the equalization it will apply to a room. the levels and crossovers definitely should be tweaked after to match the capabilities of equipment and personal audio preference of the user.

in addition to audyssey, i was lucky enough to also have the audyssey as-eq1 sub auto equalization module to also help equalized the bass response of my ht room. and the result is simply amazing.

here is what i posted at the equalization thread:




this is the graph generated by the program for the svs 12plus, as you can see, there are definite drops in magnitude (db) for my room @ 30, 40, 60 and 100hz signals before calibration and it was flattened by  the eq1 after calibration. drop of magnitude (db) shown after 16hz which is what was the tuning of the 12plus with one port sealed.



this graph is the one generated for my calibration with the svs 10nsd, basically the same before graph, showing same drops on same freq range (accuracy of eq1 is very good) and the after which the eq1 tamed the room but signal started to drop at around the 19-20hz range, which is the tuning of the 10nsd.

feeling more of the sub-sonics now.  

[streetsmart]

i think the main purpose for the auto calibration (specifically audyssey) imho, is not the setting of the levels, crossovers and distances by the program for the user but more on the room correction algorithms that is being calculated by the program to apply on your room to tame it from the ringing and nodes that each of our ht room's will definitely have (even the treated ones). the setting of the levels, distances and crossovers can be easily done with the help of an spl meter, ruler and reading the capability (specs) of your speakers as sir barrister has rightly stated.  

I substantially agree with this. Electronic room correction is mainly for:

1) Smoothening the frequency response

2) Reducing the effects of unwanted reflections (basically "ringing")

In addition, however, it is also important for:

1) Setting the correct "acoustic" distance. Some speakers (and this is very common with subwoofers) have electronics which cause a delay in the sound. A good room correction tech should detect this and compensate for it by setting a longer distance for the speaker.

2) Determining the -3 db cut-off point of the speaker, or the minimum crossover, below which no room correction will be applied. Generally, it is appropriate to raise the crossover and give more work to the subs (thus providing more headroom for the power amp) while it is not correct to lower the crossover below the minimum crossover. Please note that the -3 db point of a speaker inside a room is not necessarily what its specs say. The room has a big influence on the actual performance of the speaker. For example, if you locate the speaker near a boundary, its crossover may go lower.

I'd say these are the main things. Here's the graphs of my speakers after an Audyssey MultEQ Pro calibration:



There is a huge difference in the flatness of the frequency response. Keep in mind as well that my room already has a lot of acoustic treatment but in the region up to about 200 Hz, the room modes dictate and sometimes, no matter how much treatment you put in, you will still need electronic room correction. These charts don't show the difference in "ringing". Maybe next time, I'll show that chart. It's pretty impressive.

By the way, note how low the subwoofer goes? That's 2 x SVS PB-13 Ultras and 2 x Paradigm Servo-15's.

[barrister]

Sir iiinas and streetsmart, please give us some primer about those graphs ... you guys are talking way over my head ...  

Is that the REW of Home Theater Shack?  How do you get the graphs and how do you read them?  Sa Audyssey lang ba puwedeng magkaroon ng graph?

I want to know if this is easy to do, or if it might be too complicated for me to put up with .   

[streetsmart]

Sir iiinas and streetsmart, please give us some primer about those graphs ... you guys are talking way over my head ... 

Is that the REW of Home Theater Shack?  How do you get the graphs and how do you read them?  Sa Audyssey lang ba puwedeng magkaroon ng graph?

I want to know if this is easy to do, or if it might be too complicated for me to put up with .   

The graphs aren't complicated. If you take a look at my charts first, for each speaker, the black chart is the "Before" and the red chart is "After". The horizontal axis is the frequency. The chart is logarithmic so the first solid black vertical line is 100 Hz, the 2nd is 1000 Hz, the 3rd is 10,000 Hz.

The vertical axis is the amplitude in db and the solid horizontal line is 0 db.

If the curve is perfectly flat, it should be right on top of this solid horizontal line. However, in the "Before", you can see that the curve is jumping up and down, especially in the left part, or the frequencies up to around 200 Hz, which is where room modes dominate. In the "After" curve, you can see that it is now almost perfectly flat, which is the result of the room correction filter.

The charts of Iiinas are the same but they only go up to 500 Hz because they are meant only for the subwoofers.

My charts come from Audyssey MultEQ Pro. The charts of Iiinas come from the SVS AS-EQ1. They aren't free. However, they give a graphical representation of what Audyssey MultEQ does, and MultEQ is in a lot of receivers, including Denon, Onkyo, Integra, Marantz and NAD.

For me, the graphs are pretty to look at but the more important thing is what the graphs represent, which is a filter that flattens the frequency response in a near perfect manner. That's very difficult to do manually. Apparently, it's possible to do it with good parametric equalizers but it's extremely complex and will take a long time.

[ninjababez®]

bro kailangan ba ng software and mic to get those graph?

[barrister]

Thanks for the quick reply, sir streetsmart.

You're right, it's relatively easy to read the graphs.  But producing those graphs might be a problem for me --- I don't think I'd be willing to spend for them ...  

Sayang, I think it would be satisfying to have a graphical confirmation of the improvement you got with parametric equalization.

[ninjababez®]

Thanks for the quick reply, sir streetsmart.

You're right, it's relatively easy to read the graphs.  But producing those graphs might be a problem for me --- I don't think I'd be willing to spend for them ... 

Sayang, I think it would be satisfying to have a graphical confirmation of the improvement you got with parametric equalization.

sana software and mic lang ang katapat noh! 

With my last calibration I did all 3 seating positions at the same spot. It seems to sound better than moving the mic around. I also use the 80hz as per THX recommendations.

hi bro,  this was given by bro stickfighter  
HTH

[Stagea]

Very nice plots Sir Mark! The post-EQ response for all the channels are astonishingly flat, smooth and consistent!  This would make so many people envious.

I noticed that the calibration applied what seems to be a small gundry dip and a very gradual rolloff for the last two octaves. Out of curiosity, would this happen to be Denon's (or Audyssey's) ideal curve? Was the gundry dip applied to make the system less fatiguing (being there by design)? Would the mildly tapered HF response happen to be derived from the smoothening of the pre-EQ upper octave response (because I noticed that a rolloff is also present with the "before" graph for most of the channels)? Thanks in advance.

Again, a very impressive system sir. What a perfect way to start the year! Happy New Year to you and your family!

[Conan]

sana software and mic lang ang katapat noh! 



hi bro,  this was given by bro stickfighter  
HTH

I know all about the seating locations. I just read what Mike of SVS advised someone in one of the threads about taking all the measurements in one location so I wanted to try it out.

[ninjababez®]

I know all about the seating locations. I just read what Mike of SVS advised someone in one of the threads about taking all the measurements in one location so I wanted to try it out.

ah ako yata yun or was it martin, regarding sub calibration.  for room correction naman yung pic sa taas.  sorry for assuming you didnt know about the mic positioning.

[streetsmart]

Very nice plots Sir Mark! The post-EQ response for all the channels are astonishingly flat, smooth and consistent!  This would make so many people envious.

I noticed that the calibration applied what seems to be a small gundry dip and a very gradual rolloff for the last two octaves. Out of curiosity, would this happen to be Denon's (or Audyssey's) ideal curve? Was the gundry dip applied to make the system less fatiguing (being there by design)? Would the mildly tapered HF response happen to be derived from the smoothening of the pre-EQ upper octave response (because I noticed that a rolloff is also present with the "before" graph for most of the channels)? Thanks in advance.

Again, a very impressive system sir. What a perfect way to start the year! Happy New Year to you and your family!

The target curves come from Audyssey, not Denon.

Yes, there is a "gundry" or "BBC" dip at 2 kHz. What I recall is that it is based on the experience of the BBC and removes some harshness at that frequency, which is apparently common to most loudspeakers. The dip is on by default in all Audyssey MultEQ implementations but in the Pro model, you can remove it. I tried to remove it in the past but it didn't sound as good. Come to think of it, with my new speakers, I need to experiment again.

The tapering of the high frequency is designed to remove the excessive shrillness in small rooms, as compared to large theaters. My understanding is that this is mainly caused by the imbalance between high frequency waves which are more directional vs. low frequency which are more omni-directional. This causes the listener to hear more direct sound from the high frequencies. The consumer version of Audyssey MultEQ has a fixed target curve. The Pro version has selectable and customizable curves. Basically, the larger the room, the steeper the roll-off.

[Stagea]

The target curves come from Audyssey, not Denon.

Yes, there is a "gundry" or "BBC" dip at 2 kHz. What I recall is that it is based on the experience of the BBC and removes some harshness at that frequency, which is apparently common to most loudspeakers. The dip is on by default in all Audyssey MultEQ implementations but in the Pro model, you can remove it. I tried to remove it in the past but it didn't sound as good. Come to think of it, with my new speakers, I need to experiment again.

The tapering of the high frequency is designed to remove the excessive shrillness in small rooms, as compared to large theaters. My understanding is that this is mainly caused by the imbalance between high frequency waves which are more directional vs. low frequency which are more omni-directional. This causes the listener to hear more direct sound from the high frequencies. The consumer version of Audyssey MultEQ has a fixed target curve. The Pro version has selectable and customizable curves. Basically, the larger the room, the steeper the roll-off.

Thanks for the answers and clarifications sir Mark! These are very helpful.

I get their point for adding the 2kHz dip. Many speakers (especially older ones) integrate the same "Gundry Dip" to control harshness, as our ears are most sensitive to these frequencies. Let us know how the Electras perform without the dip.

As for the HF rolloff, it's much clearer to me now. That probably makes acoustic images more consistent too (images along a speaker's axis will sound less different from an image between speakers), helping complex scenes and panning tremendously.

I appreciate your help.

[jerix]

Based on barristers comment - if you have a bigger speaker for your fronts you can go lower than 80Hz? Ill try this one 

[barrister]

I hope Pioneer users can comment about MCACC, because I hear MCACC is also good.

OK lang yung hindi masyadong high-tech ang comments.  Hindi ako makasunod pag sobrang high-tech na ang discussion, e ...   

[iiinas]

I hope Pioneer users can comment about MCACC, because I hear MCACC is also good.

OK lang yung hindi masyadong high-tech ang comments.  Hindi ako makasunod pag sobrang high-tech na ang discussion, e ...   

hehe sir barrister. imho think the important things really for any automated calibration is:

1. use all available mic positions during measurement, result will be better computation of eq settings by the program to flatten room response.

2. as for crossovers, levels and distance measurements, using the program's measurement is ok, but we can always change these to a more accurate setting with the help of rulers, spl meter and of course personal preference.

it is also very helpful that you can toggle between using the program's eq or not and really see if there is audible improvement. in my case. huge improvement was really noted specially in the bass dept.

[barrister]

Madali lang ang distance, phase and crossover.  The auto set-up can save time, but it won't be much harder to do manually.

Yung parameteric equalization talaga ang tricky.

I think I'm going to have to try multiple positions for the mic.  Mas magaling talaga ang result ng computation by the software based on its own algotrithms, ha?  Hmmm ... sige, baka nga mag-improve pa.

Pero listening lang ako ... I'm not willing to make graphs for it  .

Sabi daw, Pioneer's MCACC can set different calibrations on the receiver's memory.  Konti lang yata ang Pioneer users, walang nagco-comment sa auto set-up. 

[Nelson de Leon]

Madali lang ang distance, phase and crossover.  The auto set-up can save time, but it won't be much harder to do manually.

Yung parameteric equalization talaga ang tricky.

I think I'm going to have to try multiple positions for the mic.  Mas magaling talaga ang result ng computation by the software based on its own algotrithms, ha?  Hmmm ... sige, baka nga mag-improve pa.

Pero listening lang ako ... I'm not willing to make graphs for it  .

Sabi daw, Pioneer's MCACC can set different calibrations on the receiver's memory.  Konti lang yata ang Pioneer users, walang nagco-comment sa auto set-up. 

Yes sir. The EQ is very tricky. I don't have auto calibration in my system pero if ever i would have one in the future, i would initially use the auto calibration and fine tune it later to my preference kapag i feel the need for improvement.

[ninjababez®]

Madali lang ang distance, phase and crossover.  The auto set-up can save time, but it won't be much harder to do manually.

Yung parameteric equalization talaga ang tricky.

I think I'm going to have to try multiple positions for the mic.  Mas magaling talaga ang result ng computation by the software based on its own algotrithms, ha?  Hmmm ... sige, baka nga mag-improve pa.

Pero listening lang ako ... I'm not willing to make graphs for it  .

Sabi daw, Pioneer's MCACC can set different calibrations on the receiver's memory.  Konti lang yata ang Pioneer users, walang nagco-comment sa auto set-up.  

updated mo kami bro with the results    

[iiinas]

Madali lang ang distance, phase and crossover.  The auto set-up can save time, but it won't be much harder to do manually.

Yung parameteric equalization talaga ang tricky.

I think I'm going to have to try multiple positions for the mic.  Mas magaling talaga ang result ng computation by the software based on its own algotrithms, ha?  Hmmm ... sige, baka nga mag-improve pa.

Pero listening lang ako ... I'm not willing to make graphs for it  .

Sabi daw, Pioneer's MCACC can set different calibrations on the receiver's memory.  Konti lang yata ang Pioneer users, walang nagco-comment sa auto set-up. 

mahirap talaga yang parametric equalization, im for automation. 

the graphs are just for confirmation. but tama ka, its still the listening result that is more important.

kung ganun yung sa pioneer (able to save multiple settings) ganda din yun. but in theory, if the program is good and the receiver is use only in one room, running multiple eq sessions should not make any difference,, kasi the program should more or less have the same eq algorithms applied to the same room. exceptions will be:

1. user have two speaker set ups in one room.
2. or user moves the receiver in different room environments.

[barrister]

Dalawa lang ang tweak ko after running YPAO: (a) large front speakers changed to small; and (b) 80Hz crossover changed to 60Hz.

I forgot about the subwoofer.  

I do a manual set-up for the sub.  Laging mali ang auto set-up for the sub, e.

kung ganun yung sa pioneer (able to save multiple settings) ganda din yun. but in theory, if the program is good and the receiver is use only in one room, running multiple eq sessions should not make any difference,, kasi the program should more or less have the same eq algorithms applied to the same room. exceptions will be:

1. user have two speaker set ups in one room.
2. or user moves the receiver in different room environments.

May silbi rin yon sir.

Puwedeng memory 1 for movies, then memory 2 for gaming when you sit up closer.  Or maybe memory 3, using a higher crossover for certain types of movies.