"Official" Audyssey thread (FAQ in post #1) - Page 101

200Hz seems high, did you try 80hz, or maybe 120hz? I wouldn't have touched the tone control.

thank you for the response! make sense. i triple checked everything and it all checks out.

I find it utterly absurd that there's no way to back up an Audyssey Pro calibration. This falls entirely in the lap of Audyssey. There's no excuse for not being able to simply re-download your calibration. Is there anyone who thinks this is OK?

The built in Audyssey software usually does a really good job and is a great feature to have in a receiver, but sorry, there's no chance I would ever pay for an Audyssey Pro calibration or ever even suggest to anyone else to do one.

Thanks for your suggestion.

80 hz was not enough so I went to the extremes. These settings did clear up the problem. I wll do bring it down bit by to the lowest point and do the same with the bass.

Avia II low bass circulating sounds really show thhe problem along with lack of "clearness" in normal TV using any 7.1 (like Ultra 2 THX playback) decoding.

I am looking for another way to solve the issue.

Which NHTs are these? How much distance are we talking about to your boundaries? Have you considered acoustical treatments? Here's one place you could try:

http://www.atsacoustics.com/

I do have bass corner foam in both rear corners. The speakers are abut 6-10 inches from the side and read wall and in the ceiling. I will try to find their name-but they were bought about 7 years ago and were then new from NHT but (not their best) for the ceiling.

My NHT Dipoles are the same vintage-but were then the top of the line.

I recently upgraded to ML Vantage and Cinema for the front three speakers. They of course sound very good when properly oriented. I use a Bryston 4ssb for the 2 Vantages and an Onkyo 7500 for the Center, rear, and surround speakers.


The rear two speakers are not to important-but since I have them I would like to have them in use for those special movies.

Joel

"One important thing to note: the number of measurements is not the same as the number of listening positions."

Thanks. Next time I run Audyssey, as I'm sure I'll be running it again, I'll run it as many times as I can.

Derek,

Subwoofers are perfectly capable of reproducing frequencies up to 1 kHz or so. Not that you would ever want to do that, but crossing a speaker over at 120 Hz and expecting the sub to play up to that frequency is pretty standard.

Chris

Had the same issue, see here Post #2718:

http://www.avsforum.com/avs-vb/showt...t#post13511067

Audyssey (Chris) response is down the page at #2730. Re-running the calibration at all 8 points chosen within 2 ft of the three front row seats in my theater (and therefore staying away from walls, corners and boundaries) made a world of difference. It's the best SQ I've ever had in my room, treatments included...

John

Chris,

Two more questions:

1) Is the first measurement position ONLY use for calculating crossover points, delay, levels, etc.?

2) From what I read you recommend the 3-3-2 pattern when setting up the mic. Three measurement position across the main listening seat, 3 more 2 ft in front of the main listening seat and if possible 2 more in the back of the main seat if not back wall. Here is my question, is there is something wrong with setting up the mic in 2-2-2 or a 2-3-2 pattern? The reason for me doing this is that that my love seat is a little bit off-center. The left side is centered with the TV. If I place the mic two feet off that the mic will be too close to a lamp that I have on top of a end table.

Thanks.

Some very useful advice on getting a good Audyssey calibration suggested by maxdb over the the Onkyo 875 thread. Sorry if it's already been posted.

http://www.avsforum.com/avs-vb/showt...1#post13562221

Not all of it is necessarily good advice. The poster recommends taking as few measurements as possible as close to the main listening position as possible. That will work fine if you're setting things up for a one or 2 person setup but if you've got seating for several people and are trying to get a good result for all of them it's bad advice.

What Audyssey does is try to deliver good results over an area, not just a single point. Since bass response in a room can change considerably if you move your head by 4-6" in some cases, that's a good thing unless you want to listen with your head in a vice so that you can't move it. The problem with trying to deliver good results over an area is that some 'averaging'for want of a better termis going to be involved. The greater the area, the greater the room variation that needs to be compensated for. You can get better results for the main position if you take your measurements in a limited area but you will also get worse results for other positions if you have them. Take the measurements over a larger area and other positions will get better results but the main position may end up slightly worse off than if you use a smaller area.

It's your choice about how you do things but I live alone and take measurements over the whole area of my sofa. I don't always sit in the centre and I do have friends over regularly. Measuring across the whole area of the couch gives me good results wherever I choose to sit, or lie, and lets others get a good sound when they're there. I can get slightly better results for me when I'm on my own, but only at the cost of restricting where I sit or lie and that's not an option as far as I'm concerned because I don't want to be limited to sitting in one spot. There are times when I want to lie down and watch a movie, or sit leaning in a corner of the sofa, rather than sitting erect in the centre. If I never had company and used a chair instead of a sofa it would be a different matter.

Basically, space the measurements out to cover the area where you are going to have people sitting. Make the measurement area as big or small as the seating area because you'll get good results in all spots then. If viewing/listening is a social activity for you then you want others to get good sound too. If it's not a social activity for you, then space the measurement locations out to cover all of the areas where you sit or lie so that you get good results whatever your chosen position is.

And do take as many measurements as your receiver allows. You don't sit with your head in a vice and small changes in location can produce big differences in sound at bass frequencies. More measurement locations gives you good results when you move your head. The difference between being able to hear a low bass note clearly and distinctly and not being able to hear it at all can be a matter of less than 1 foot's difference in your head's location as I well know from experience.

Jim, eight different locations is the max. Measurement locations may not be the same as listening positions, though each and every listening positions should be a measurement position. Of course, if you have eight seats, then that would be the eight measurement positions. More than eight seats, and somebody's gotta go!

edit: I just realized that I am coming from the 885/9.8 and I don't know if the other manufacturers have the same eight location limit . . .

Delays and levels, yes. Not 100% sure on the x-overs, but my guess is yes.

Chris,

"MultEQ will calculate filters down to the -3 dB point that it finds for each loudspeaker (where it starts to roll off). It won't calculate lower than that because that would mean it would start boosting the speaker beyond its capability at low frequencies."

I wish Audyssey would rethink this feature, and make it defeatable and/or adjustable.

First, it can't always assure its aim of protecting the speakers, as most any system can be overdriven simply by turning the volume up high enough.

Second, it can unnecessarily restrict the performance of the most capable systems.

As I've pointed out before, many subs these days, particularly large drivers in small sealed boxes, can handle a lot of boost well below F3.

Also, an EBS (extended bass shelf)vented system can have Fb and max power handling an octave or more below F3.

pepar,

That's why at this point I said as many as I can. I am not where I can look and not going to search for the manual at this moment. Its like 6 or 8, again not critical. What I am going to do thought is the measure twice for the same listening position, one at the listening position, one 2 feet in front. I can see how measuring those reflections, echos, etc., will create a more robust data base from which Audyssey will be crunching numbers.

Am guessing that the more information the better the results. Less variables, etc., in this regard probably a good thing.

Still getting to know it of course. I'm not dissing it in any way as the results speak for themselves, but clearly more complicated than MACC(Pioneer) or YPAO(Yamaha) that I've used. Its not exactly your plug n play. Not if you want to use it for what's its fully worth. Again NOT dissing it. Just stating the obvious.

Now that I know the graph/freq map in the parameters review section is the mirror image of my speaker response, I can also figure out what my speaker response is in the real world of my front room. Despite the occasional sand worm that pops through my front room it is in the real world by the way. While I'm probably not going to settle on manual settings simply because of all what Audyssey does, I will at least be playing with the manual eq after writing down the results Audyssey gets on those graphs.

I have on more question on the graph chart that Audyssey creates for the frequency testing part. Are the manual EQ settings in play when Audyssey does its thing IF you have previously adjusted and selected the Manaual EQ in setup?

Or does all that get bypassed along with the xover settings each time we run Audyssey?

Is it a better idea to set the manual eq all to flat prior to running Audyssey, make sure no manual eq is selected, etc.?

As pepar says, levels and delay, yes. I think I remember that the crossover algorithm depends on all the measurements, but I'm not entirely sure either.

No problem, the 3-3-2 pattern is one way of covering the space. In your case, you could take a measurement in front of the lamp and behind it. The important thing is to cover the area (if it's not too large - you want to stay between the speakers in sort of a single blob), and use all the measurements you can, so if you can take 8, use 8.

I think that's more or less what Chris would tell you.

No. the first measurement position is not ONLY used for calculating crossover points, delay, levels, etc.. It is also used for calculating EQ.

However, ONLY the first measurement is used for calculating crossover points, delay, levels, etc..

Dear Chris:


I too am grateful for all your advice and explanations in this thread. I hope you will comment on a question I have.

I have a Denon 4806 set up in 7.1. My subwoofer is an Axiom EP-500. This sub has a brickwall filter at 100 Hz. (See post #2989 in this thread). When I run the Audyssey setup it consistently sets my surround speakers to crossover at 120 Hz. Should I keep the crossover at 120, 100 or maybe even 80?

Thanks for your help!

Huh?

made sense to me. All measurements are used for EQ calc including the first one. But only the first one is used for levels and distance and XO.

measurement 1: determine levels, distance, xo, eq for master position
measurement 2: eq for position 2
measurement 3: eq for position 3
etc
etc

The graph is only a coarse representation of the filters set in the frequency domain and does not indicate filters in the time domain. The manual filters do not have the resolution of the Audyssey-set filters and using them defeats the Audyssey-set filters. Running Audyssey setup after setting manual filters defeats those manual filters.

I just would have expected more clarity from a writer. But then that's why people have jobs as editors I guess.

Hello everybody.

I did a search of this thread for "speaker switches" and "boundary" but I haven't found exactly what I'm looking for, so I apologize if this question is intolerably noobistic.

I'm looking at bookshelf speakers for my front LCR, and there is a subset of bookshelf speakers that are labelled "in-cabinet". The in-cabinet speakers have "boundary switches" (which are -3 dB woofer adjustment) and also sometimes a tweeter switch +3dB, 0 dB, -3 dB.

If I have an AVR that has Audyssey (looking at Onkyo 875) do I need to consider these "in-cabinet" speakers, or will Audyssey do all the necessary adjustments for boundary effects?

Until I started following this thread a few days ago, I felt like I was stuck with finding speakers with these switches (at least the boundary switch), but after reading some of this thread I'm thinking it isn't necessary. Maybe not even desirable.

Thanks. I was trying to be more precise with my wording than were the preceding posts.

I think my words were very clear as long as you parse the sentences carefully.

No. The first measurement also captures the frequency response there and that information is combined with the other measurements to produce the MultEQ filters.

2) From what I read you recommend the 3-3-2 pattern when setting up the mic. Three measurement position across the main listening seat, 3 more 2 ft in front of the main listening seat and if possible 2 more in the back of the main seat if not back wall. Here is my question, is there is something wrong with setting up the mic in 2-2-2 or a 2-3-2 pattern? The reason for me doing this is that that my love seat is a little bit off-center. The left side is centered with the TV. If I place the mic two feet off that the mic will be too close to a lamp that I have on top of a end table.

The "2 ft." recommendation is approximate and will vary in individual cases. I just use it as a guideline because some people early on were moving the mic 5-6 ft every time and that wasn't capturing the needed information in the main listening area.

Chris

Noah,

As I have responded before, this only applies to satellite speakers and not subs. MultEQ does not find the -3 dB point of subs. There we use a different method that does not restrict the response below F3. I am unable to give the details of that method, but it can be verified by simple measurements.

Chris