AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)
New Posts  All Forums:Forum Nav:

"Official" Audyssey thread (FAQ in post #1) - Page 1219

post #36541 of 62741
...within the limits of the existant equipment.
It might bring a sub to its limits, to have a dip filled in at the listening (or measurement) position. At minimum the peaks should be smoothed, because they are much more anoying...
post #36542 of 62741
Quote:
Originally Posted by gurkey View Post

...within the limits of the existant equipment.
It might bring a sub to its limits, to have a dip filled in at the listening (or measurement) position. At minimum the peaks should be smoothed, because they are much more anoying...

I'm not talking about boosting frequencies with an EQ. That creates more problems than it solves. Dips can be fought with active and/or passive absorption.
post #36543 of 62741
Quote:
Originally Posted by ReneV View Post

It's not easy to say, Jeff. There's something less good about that seat, and lack of deep bass certainly has something to do with it. I've convinced myself that bass guitar is suffering there, which certainly points to a suckout of the particular kind shown in the graph, but ...

It's 1/4 of an octave and very deep, so I would expect three or four bass guitar notes would be just plain missing.

Quote:
Originally Posted by markus767 View Post

Not Rene but dips are not easy to hear, especially when narrow. You'd need to A/B or listen to a known recording or use a sine sweep. Anyhow, dips represent inaccurate reproduction and should be fixed if they are wide.

I'm most concerned with what could be causing it. Determining that will go a long way to fixing it.

I hope Chris can offer his expertise ....

Jeff
post #36544 of 62741
Quote:
Originally Posted by pepar View Post

I'm most concerned with what could be causing it. Determining that will go a long way to fixing it.

MultEQ works in the time domain so it probably changed something there

Here's the group delay introduced by MultEQ (pre out of yesterday's Test A):

post #36545 of 62741
Hey guys this thread is so large and I am having trouble finding the answer to this simple question.

I do not have a subwoofer, nor do I have an avr.

As one goes up the Audyssey food chain, will a subwoofer sound more musical?

For example one of the knocks on the BIC F12 is that it is not musical.

I know the filter resolutions are different when looking at both MultEQ and MultEQ XT with 128 filters, versus 512 filters for MultEQ XT32. I just don't understand what it means though. So I guess I can only ask what does it mean, and how much better can it make a subwoofer sound. Mainly, though, I am asking how much better can it make a subwoofer sound (I am thinking along the lines of someone running 2 BIC F12's, for example. Then maybe getting nicer subwoofers in the future.)

How would what XT32 can do, with calibrating the subwoofers separately and then pinging them together, affect the musicality of subs like two BIC F12's when comparing this to MultEQ / MultEQ XT? Are MultEQ and MultEQ XT the exact same as far as the subwoofer goes (basing my presumption on the # of subwoofer filters), or am I missing an aspect there?

If there are any links that you guys would rather provide, I'm sure that would be helpful as well. I guess that's it for now. Thanks.

Edit- Do the different Audyssey's change from year to year. Example: is this years MultEQ XT the same as last years, which will be the same as next years?
post #36546 of 62741
Quote:
Originally Posted by markus767 View Post

MultEQ works in the time domain so it probably changed something there

Here's the group delay introduced by MultEQ (pre out of yesterday's Test A):

Any sign of group delay in your big room with the pre outs below 100Hz?
post #36547 of 62741
Quote:
Originally Posted by rickardl View Post

Any sign of group delay in your big room with the pre outs below 100Hz?

No.
post #36548 of 62741
Quote:
Originally Posted by markus767 View Post

Moved everything to another, smaller room with no acoustic treatments. One subwoofer, left and right speaker.
...
Sub pre out with and without MultEQ:
...

Since the smaller room, without treatments, is showing significant corrections in the sub preout, are you concluding that, in your original room, thanks to the treatments, Audyssey was not finding much to improve with the sub?

What model sub do you have?
post #36549 of 62741
Quote:
Originally Posted by Bill Mitchell View Post

Since the smaller room, without treatments, is showing significant corrections in the sub preout, are you concluding that, in your original room, thanks to the treatments, Audyssey was not finding much to improve with the sub?

Sub location, treatments (the whole front wall is covered with polyester batting) and size of the room (10m x 4.50m, open to the hallway and 3 other rooms) - that's my conclusion for now.

Quote:
Originally Posted by Bill Mitchell View Post

What model sub do you have?

Two DIY sealed enclosures with a downfiring Shiva-X and Shiva-X2.
post #36550 of 62741
Quote:
Originally Posted by streetsmart88 View Post
By the way, each time you show your charts, what really bugs me is the -3 db point of your left front speaker. In the latest charts, it is at 200 Hz. That's way too high. There must be something wrong in the speaker location or the woofer may be blown.

Mark
Thanks for the feedback Mark

Hmm, i never noticed that on the graphs tbh

I have done 3 seperate measurements and looking at all 3 shows the same as you mention.

The Right speaker fires across a stationary table and has a brick wall near the right hand side of the speaker and is in front of a stationary cabinet with glass doors.

The Left speaker however has no room furniture in front of it and no corner boundary on the left or right of the speaker.

So you are saying that one of the woofers is blown? How can i check for this please as it does not seem as though there is a problem with it when listening?

The calibrated mic also plugs into the Left channel phono socket on the SEQ when calibrating btw.

How do you know it's at 200Hz? The graph does not show any frequencies and i am guessing that the horizontal and vertical lines are frequency points. Is there any data on these?

The weird thing is, the system now sounds fantastic with the SEQ in place

I've attached the 3 graphs for the the 3 times i have done a calibration.

The current set up in measurement is with graph 3

 

Audyssey Final HF Roll Off 1 pdf.pdf 417.6357421875k . file

 

audyssey measurent 2 pdf.pdf 408.501953125k . file

 

Audyssey Measurement 3 HF Roll Off 1 Mid Range Comp ON.pdf 405.83203125k . file
post #36551 of 62741
Quote:
Originally Posted by markus767 View Post

... - that's my conclusion for now. ...

Thanks. When I measured my sub preout, and looking at the graphs of a few others, my first hypothesis was that Audyssey left a 9dB boost below the -3dB rolloff point, that this was a side-effect of boosting the sub as it approached the rolloff. Later, when there was a discussion of some measurements of mains below the recommended crossover, a second hypothesis developed, that Audyssey did not care what happened outside the -3dB rolloffs, that it optimized its filters to achieve the best results over the target range, and whatever happened outside that range was fine, be it a constant 9dB boost, a gradual return to zero, or even a slope to below zero. The example of your larger room, where the sub curve is relatively flat immediately above its rolloff, supports the second hypothesis.
post #36552 of 62741
mahm...I don't think it's a blown woofer, but to check: have you tried swapping the left and right speakers to see if that drop in bass is due to the speaker or the room? Most speakers would be flatter just standing in the middle of the room, so that left speaker just looks not so good. The right speaker close to the wall would explain why that bass is stronger. It's possible that the woofer in the left speaker has a torn cone or needs refoaming (not sure how old your speakers are but you could probably see this type of physical damage by just taking off the grill).

The dark vertical lines are 100hz 1000hz and 10,000hz and it's a log scale.

Your sub calibrates to flat up to 120hz, so if you have set your front crossovers to 120hz, you aren't loosing any content...but anything over 80hz can be directional (you hear where it's coming from. If the sub is in the front, you probably don't notice this.
post #36553 of 62741
Several pages back I had mentioned a problem I was having with Audyssey (Denon 3311CI) maxing out the trim levles on all five of my speakers to +12 dB when calibrating. What appears to be the most likely issue is a defective AVR and I am working with my retailer to address that since it is still under 30 days since purchase.

A couple of questions though, without having to get back into the whole troubleshooting analysis:

1) Is there any way that trim levels of +12 db all the way around would be accurate given reasonably competent mid-level speakers, 125 wpc and a 14’ x 18’ room?

2) Has anyone else experienced Audyssey maxing out the trim levels for their speakers when calibrating, either with a Denon 3311 or otherwise?

Thanks.

(cross-posted on both the Denon 3311 forum and the Audyssey forum for obvious reasons)
post #36554 of 62741
Quote:
Originally Posted by jakeroux View Post

Several pages back I had mentioned a problem I was having with Audyssey (Denon 3311CI) maxing out the trim levles on all five of my speakers to +12 dB when calibrating. What appears to be the most likely issue is a defective AVR and I am working with my retailer to address that since it is still under 30 days since purchase.

A couple of questions though, without having to get back into the whole troubleshooting analysis:

1) Is there any way that trim levels of +12 db all the way around would be accurate given reasonably competent mid-level speakers, 125 wpc and a 14' x 18' room?

2) Has anyone else experienced Audyssey maxing out the trim levels for their speakers when calibrating, either with a Denon 3311 or otherwise?

Thanks.

(cross-posted on both the Denon 3311 forum and the Audyssey forum for obvious reasons)

This sounds like a defective microphone....Have you asked your dealer to loan you (or swap) the one that came with his store demo unit???
post #36555 of 62741
I posted this in the Integra thread and someone thought it would be better answered over here.



I want to replace my Yamaha 663 and would really like to go totally with separates. I have been thinking hard about getting the 40.2 which has MultEQ. I wonder now if I would be better off just getting the TX-NR3008 which has MultEQ-X32. I have no experience with Audyssey and only have used YPAO. I also realize that Audyssey will be an improvement over my current YPAO but wonder if I will be disappointed that I didn't get a receiver with the new multEQ-X32. Help me clear my mind on this Is there that big of a difference between MultEQ and MultEQ-X32?
post #36556 of 62741
Quote:
Originally Posted by ghstudio View Post

This sounds like a defective microphone....Have you asked your dealer to loan you (or swap) the one that came with his store demo unit???

Thanks, but I have the identical model of mic that came with my Denon 1610. Both mics calibrate properly (albeit in a different room) on the 1610, and neither can calibrate without maxing out on the 3311.
post #36557 of 62741
Quote:
Originally Posted by Bill Mitchell View Post

Thanks. When I measured my sub preout, and looking at the graphs of a few others, my first hypothesis was that Audyssey left a 9dB boost below the -3dB rolloff point, that this was a side-effect of boosting the sub as it approached the rolloff. Later, when there was a discussion of some measurements of mains below the recommended crossover, a second hypothesis developed, that Audyssey did not care what happened outside the -3dB rolloffs, that it optimized its filters to achieve the best results over the target range, and whatever happened outside that range was fine, be it a constant 9dB boost, a gradual return to zero, or even a slope to below zero. The example of your larger room, where the sub curve is relatively flat immediately above its rolloff, supports the second hypothesis.

Sorry Bill, I'm not sure I understand what you're saying.
post #36558 of 62741
Quote:
Originally Posted by jakeroux View Post

Several pages back I had mentioned a problem I was having with Audyssey (Denon 3311CI) maxing out the trim levles on all five of my speakers to +12 dB when calibrating. What appears to be the most likely issue is a defective AVR and I am working with my retailer to address that since it is still under 30 days since purchase.

A couple of questions though, without having to get back into the whole troubleshooting analysis:

1) Is there any way that trim levels of +12 db all the way around would be accurate given reasonably competent mid-level speakers, 125 wpc and a 14’ x 18’ room?

2) Has anyone else experienced Audyssey maxing out the trim levels for their speakers when calibrating, either with a Denon 3311 or otherwise?

Thanks.

(cross-posted on both the Denon 3311 forum and the Audyssey forum for obvious reasons)

IIRC, there have been a few cases wherein the answer to both questions is yes. They involved speakers with very high sensitivities, particularly Klipsch and JTR. The typical solution was to use an attenuator.

Mark

Note: This is wrong. I mixed this up with the more common situation of Audyssey doing the maximum cut of -12 db.
post #36559 of 62741
Quote:
Originally Posted by mahm View Post

Thanks for the feedback Mark

Hmm, i never noticed that on the graphs tbh

I have done 3 seperate measurements and looking at all 3 shows the same as you mention.

The Right speaker fires across a stationary table and has a brick wall near the right hand side of the speaker and is in front of a stationary cabinet with glass doors.

The Left speaker however has no room furniture in front of it and no corner boundary on the left or right of the speaker.

So you are saying that one of the woofers is blown? How can i check for this please as it does not seem as though there is a problem with it when listening?

The calibrated mic also plugs into the Left channel phono socket on the SEQ when calibrating btw.

How do you know it's at 200Hz? The graph does not show any frequencies and i am guessing that the horizontal and vertical lines are frequency points. Is there any data on these?

The weird thing is, the system now sounds fantastic with the SEQ in place

I've attached the 3 graphs for the the 3 times i have done a calibration.

The current set up in measurement is with graph 3

Quote:
Originally Posted by ghstudio View Post

mahm...I don't think it's a blown woofer, but to check: have you tried swapping the left and right speakers to see if that drop in bass is due to the speaker or the room? Most speakers would be flatter just standing in the middle of the room, so that left speaker just looks not so good. The right speaker close to the wall would explain why that bass is stronger. It's possible that the woofer in the left speaker has a torn cone or needs refoaming (not sure how old your speakers are but you could probably see this type of physical damage by just taking off the grill).

The dark vertical lines are 100hz 1000hz and 10,000hz and it's a log scale.

Your sub calibrates to flat up to 120hz, so if you have set your front crossovers to 120hz, you aren't loosing any content...but anything over 80hz can be directional (you hear where it's coming from. If the sub is in the front, you probably don't notice this.

It's a logarithmic scale. Coming from the left, the first dark vertical line is 100 Hz. The light vertical line after that is 200 Hz. If you will notice, in your last reading, the left front speaker started a steep roll-off at 200 Hz. If you have a large tower speaker, your woofer could have a crossover with the mid-range driver at around 230 Hz, which would correspond well with the roll-off in the chart at 200 Hz.

As @ghstudio has suggested, try swapping your left and front speakers and see if the problem persists.

Mark
post #36560 of 62741
Quote:
Originally Posted by pepar View Post

It's 1/4 of an octave and very deep, so I would expect three or four bass guitar notes would be just plain missing.

I'm not an expert by any means but standard bass guitar tuning is roughly (E) 41Hz, (A) 55Hz, (D) 73Hz, (G) 98Hz.

That should mean that the suck-out significantly affects the first overtone of E and possibly A, as well as the fundamentals of D and probably G, along with a similar effect on the relevant modified notes, i.e., in no case is both the fundamental and the first overtone gone, i.e., no note would be "just plain missing", at least to my ear

EDIT: Wikipedia has something to say about this, it seems.
post #36561 of 62741
Quote:
Originally Posted by streetsmart88 View Post

IIRC, there have been a few cases wherein the answer to both questions is yes. They involved speakers with very high sensitivities, particularly Klipsch and JTR. The typical solution was to use an attenuator.

Mark

Wouldn't the trim levels maxed out mean that the speakers were producing way too little SPL's thus causing Audyssey to boost the trims to compensate?


Max
post #36562 of 62741
Quote:
Originally Posted by jakeroux View Post

Thanks, but I have the identical model of mic that came with my Denon 1610. Both mics calibrate properly (albeit in a different room) on the 1610, and neither can calibrate without maxing out on the 3311.

Hiya Jake,...IMHO, the +/-12 dB allowance of Audyssey in a Denon receiver should give you enough room to handle a small room with high efficiency speakers vs. a large room with low effinciency speakers, so maxing out this all together 24 dB range seems to be a rather peculiar case, so to speak. If micropocessor reset didn't bring about the expected results it seems to be a better/best option at the moment to contact your dealer for a replacement.

Chime back with further developments, please! Wish Ya good luck in advance!
post #36563 of 62741
Quote:
Originally Posted by djbluemax1 View Post

Wouldn't the trim levels maxed out mean that the speakers were producing way too little SPL's thus causing Audyssey to boost the trims to compensate?


Max

Yes, you're right. I mixed it up. The opposite is what is more common. Come to think of it, I've never heard of maxed out trims with Audyssey.

Mark
post #36564 of 62741
Quote:
Originally Posted by streetsmart88 View Post

Yes, you're right. I mixed it up. The opposite is what is more common. Come to think of it, I've never heard of maxed out trims with Audyssey.

Mark

Yeah, that's not common especially with a 14' x 18' room.

jakeroux, just out of curiosity, what speakers are you running? At 125wpc (even if the avr drops that output when all channels are run vs 2 channels which is where they get the power rating), unless the speakers are seriously difficult to drive, you definitely could be seeing a problem with the avr.

How does it sound to you when you get it up near Reference levels? If it was a problem with underpowering the speakers (either because the speakers are hard to drive and need more power or because the amp section is not putting out as much as it should), Reference level would be fairly soft and anything below Reference would be really soft. If Reference is pretty loud, then something's off with the avr's readings.


Max
post #36565 of 62741
Reference level at current calibration is VERY loud. Mains are Ascend Acoustics 340 SEs. Surrounds and center are leftover Boses.

As mentioned several pages back, I was able to properly calibrate these speakers in this room on my old Denon 987/2807, which leads me to believe the calibration signal is somehow being degraded between the mic and the processor, or simply "misprocessed" with the 3311. The sub channel will trim out down around +5 dB if I crank the sub (Hsu VTF-2) knob up to about the twelve or one o'clock position. Again, on the 987, it would calibrate well into the -'s, even when trimmed back to about ten o'clock on the knob.

Microprocessor reset had no impact on the calibration readings once re-run.

Thanks again for the input.

Edit - currently listening at a fairly quiet but still listenable "kids-are-in-bed-upstairs-so-keep-it-down" mode, and that puts the volume knob at -68 dB.
post #36566 of 62741
Quote:
Originally Posted by jakeroux View Post

Reference level at current calibration is VERY loud. Mains are Ascend Acoustics 340 SEs. Surrounds and center are leftover Boses.

As mentioned several pages back, I was able to properly calibrate these speakers in this room on my old Denon 987/2807, which leads me to believe the calibration signal is somehow being degraded between the mic and the processor, or simply "misprocessed" with the 3311. The sub channel will trim out down around +5 dB if I crank the sub (Hsu VTF-2) knob up to about the twelve or one o'clock position. Again, on the 987, it would calibrate well into the -'s, even when trimmed back to about ten o'clock on the knob.

Microprocessor reset had no impact on the calibration readings once re-run.

Thanks again for the input.

Edit - currently listening at a fairly quiet but still listenable "kids-are-in-bed-upstairs-so-keep-it-down" mode, and that puts the volume knob at -68 dB.

Definitely sounds like something is off in the avr. Maybe something is wrong with the mic input? If it's within the return window, I'd definitely return/exchange it.


Max
post #36567 of 62741
Quote:
Originally Posted by markus767 View Post

Looks like MultEQ is helpful in acoustically challenging situations but isn't good at improving an already decent starting position.

I think MultEQ finds that the lower end is already "decent" as you say and is simply concentrating on fixing the acoustical problems which were found to be more severe.
In your case, they are not in the lower frequencies. Instead, MultEQ may be spending its limited resources in the higher frequencies.
Quote:


The idea behind MultEQ is to capture multiple responses in the room in the time domain.
Then, it looks at the similarities among the these responses to determine which ones can be grouped (clustered) together.
In each cluster, MultEQ creates a representative response that contains the acoustical problems of the responses in that
cluster, but weighed depending on the severity of the problem.
From the representative responses, a "master" room response is created that is contains the weighted information from all the measurements.
Then the MultEQ correction filter is created by inverting that response with a proprietary FIR filter that distributes its coefficients (control points) nonlinearly in frequency (more at low frequencies).
post #36568 of 62741
Quote:
Originally Posted by rickardl View Post

I think MultEQ finds that the lower end is already "decent" as you say and is simply concentrating on fixing the acoustical problems which were found to be more severe.
In your case, they are not in the lower frequencies. Instead, MultEQ may be spending its limited resources in the higher frequencies.

The acoustical problems at higher frequencies are even less severe.

Left speaker:



Right speaker:



In the current state you won't even see the ceiling reflection at 4ms.
post #36569 of 62741
Quote:
Originally Posted by markus767 View Post

The acoustical problems at higher frequencies are even less severe.

In the current state you won't even see the ceiling reflection at 4ms.

Oh well, it was a theory...
Are those graphs available in REW somewhere?
post #36570 of 62741
Quote:
Originally Posted by rickardl View Post

Oh well, it was a theory...
Are those graphs available in REW somewhere?

Yes, see "Impulse Response Envelope" here:
http://www.hometheatershack.com/room...l.html#impulse
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Receivers, Amps, and Processors
AVS › AVS Forum › Audio › Receivers, Amps, and Processors › "Official" Audyssey thread (FAQ in post #1)