"Official" Audyssey thread (FAQ in post #1) - Page 1115

Also, to show how "little" I know, (But know just enough to realize I need to ask this question) I am ready to place an order for my lumber and in reading this thread have realized having a square room is not good (but need a better understanding as to why) so in building my new HT I have the option of "up to" a 20' 8" w x 23-24' l x 7' 6" h room and can even taper the walls so they're not as wide in front (Can't with ceiling since I won't have quite 8' ceilings though, right?) but if I build the room this size, which is the max, it is pretty close to square so I thought about making it less wide. Is this considered a large or medium room?

I know this isn't specifically about Audyssey so feel free to PM me and/or redirect me to a proper thread(s) to learn more about this and to get more input.

The end goal of course is to build as "good" of a room as possible so Audyssey will have less work to do and what it can/does do (Will have MultEQ XT32 w/ new Integra 80.2) will have more effect and get me closer to a flat response without any major peaks or dips but right now I'm feeling like quite the dip!

Thanks,

--J

the master handbook of acoustics

Thanks Roger,

I just wanted to make sure that I wasn't getting a different crossover slope/ splice than what Audyssey is indicating for it's crossover optimization.

No; since Audyssey measures the actual speaker responses and knows the XO slopes it should be OK.



Why can't it just calculate the complex sum of the speakers' responses?

I guess one complication is that a given freq near XO may appear in any/all of the satellites.

Not Chris ...

The "better resolution" reason disappears with XT 32 in re raising crossovers, but others could still apply. As for the smoothness of the transition, I don't know for certain that Pro looks at all measurement positions and recommends the smoothest splice, but that's my guess. It seems unlikely (to me) that any variety of MultEQ varies the sub distance to smooth the splice, and from craig john's tweaks that seems like a valid tactic. Tonight I will be at his house and will get to hear the difference between the detected distance and the one that produced a nearly ruler straight splice when measured. If you haven't seen his graphs on the Seaton thread, the difference in the measurements were dramatic.

Jeff

This is why it takes me about 4 hrs.to run a 12 point calibration because I try to get the mic in the same exact measurement locations by measuring exactly too every spot that I have marked on the position chart that I print out.This probably still doesn't get me the same exact measurement results,but it's as close to repeatability as I can get and the before and after charts are almost exactly the same from measurement to measurement session if there aren't any changes to the room.


Hi jevansoh,

Just so you know the LR4 LP/LR2 HP means Linkwitz-Rily 4th order low pass and Linkwitz-Rily 2nd order high pass which is one of the most common crossover slope types.

Here is a link you can read about different crossover slopes.

http://www.rane.com/note160.html

But according to you, the basis upon which it knows the XO shape is an overly simplistic assumption. As seen below, the actual crossover responses are never measured.

BTW, I'm not alleging any problem due to this. But it looked like you were. If not, I guess I do not understand the point you were making in the top quote. It seems to me what StevenLansing is doing is just following standard Audyssey practice.

Seemingly missing ... and somebody correct me if my terminology (or thinking) is faulty ... is a "time corrected propagation plane" in that the sub (or subs in craig john's case) are not physically in the same location as the mains ... any of them. Absent that, how can any calculation be made to pick the smoothest splice?

Craig varied the sub distance to smooth out a very bad cancellation that was present using the "detected" distances. I will get to hear the two settings tonight.

I was referring to THX's assumption that a speaker designed with LR2 HP @80 Hz would exhibit that in everyones's rooms.

Even if they all got the same signal, they wouldn't sum the same if the proportionate levels varied over time.

Yes, it would be ideal to be able to measure through the satellite/subwoofer crossover and have the filters correct for any anomalies that happen at the blend region. However, to do that would require a huge shift in the way AVR makers are implementing their block diagram. We have proposed this nearly 2 years ago and were turned down...

Given that Audyssey does not always get the sub distance/time calculation right, as seen by Craig John's experience among others, taking the complex sum of the satellites against the sub will be problematic if the sub distance is off.


Certainly being able to correct each speaker to a target taking into account the filter makes sense, and still allows the user to correct the distance if necessary. Correcting one speaker against another works only if the distance/timing and levels are already right. I imagine the Audyssey Pro work flow could be designed to accommodate this, but it might be a difficult work flow to introduce in the retail version.

Absolutely happy to hear thoughts like this. I don't totally understand why the swivel needs to be >90 degrees. What am I missing?

If that assumption is false, then this impacts Audyssey's method which assumes LR4 is in effect. So we're back to where we started.

In a receiver such as the Denon 4311CI with MultEQ XT32 I see that it can output to (2) separate subwoofers.

My question is, can the built in XT32 EQ time align and effectively integrate two separated NON-Colocated subs with the same EXACT functionality/effectiveness that the outboard Audyssey Sub-EQ and SVS AS-EQ1 can?

Thanks!


...Glenn

So, what is the best method today to find the optimal crossover for us with MultEQ XT after Audyssey setup?
Run sweeps (REW, XTZ RA, etc) for each of the possible crossover settings to see which one produces the flattest blend between the sub and the other speakers?

Hello all! New to Audyssey and a 3311CI. I've searched, read and read and I know most dont want another question on mic placement but please be patient with me.

It seems to go back and forth from its critical to don't get to caught up in placement vs seating area.

- my primary seationg area due to room and sectional couch is almost straight out from the FR speaker
- this is the corner of the sectional couch and is against a wall on the right side (4' from the back wall)
- this area has bass... 3 feet to the left closer to the center channel axis is a dead spot for bass
- the primary seating area is about 5' off the center channel axis

Am I confused in believing I should calibrate the mic closer to the center instead of against the right wall in the primary seating area?

THANK YOU for any help.... and sorry for being a audyssey rook!

I was wondering if anybody knew if you could see what Audyssey did to my speakers during setup? I know I can see crossover frequencies, distance, etc.., but I was wondering if you could actually see the eq settings. maybe through some hidden menu etc.. I have a Onkyo 876

Sorry, no.

No, not really. You can look at a crude representation, but it is so crude that it is worthless.

Not clear to me that Craig John's sub "acoustic distance" results from Audyssey were wrong, in the sense that the did not accurately measure how long it takes from electronic signal impulse until the sound reaches the measurement location. What I've assumed CJ is fixing is an anomaly resulting from phase shift (likely in mains as the "group delay" of the sub at the crossover region ought to be equal to the acoustic "distance" while it might be longer at lower frequencies).

Seems to me that if the mains were to have a 90 -to-180 degree shift at the crossover frequency, the combined sound based strictly on distance could result in cancellations. [I think that Brian at Rythmik attempted to explain this concept on the Rythmik thread some weeks ago. Made my head spin.] Correcting via sub distance certainly could help. I would not advocate "correcting" via changes in the delays on the left, right, center or surround speakers, because I think that the appropriate imaging, which is improved via the use of delays to account for non-equidistant speakers, is too imortant to trash via corrections intended to assist only in the bass region.

In addition to the books above you might PM Chris/Audyssey.
He could email you some papers and direct you to some links on Audio theory.

Other info:
- HTS - Room Accoustics FAQ (some links - still a work in progress)
- Rives Audio for info
- consider the "AVS/Dennis Erskine Pro Theater Layout" service
$750 Retail
$600 now


Mike

I heard the difference last night between the "detected" sub distance and the one that Craig dialed in to smooth the splice. It is not simply a measurement thing; it was night and day when we listened to content, too. To recap, Craig has three subs at varying distances from the main listening position/first measurement position ... all on one sub channel.

Jeff

yes, that's the whole point of SubEQ HT! It effectively puts the AS-EQ1 in the box -- the subs are separately time aligned and level matched, and THEN Audyssey EQ's them as a "summed" response. So it is specifically meant to deal with non-colocated or asymmetrical dual sub placements.

Hmmmm.... I don't think the Audyssey/Integra Distance result was "wrong" either. I have one sub that is closer than the other two, and closer than any of the speakers. It's about 8.8' from the primary listening position/first measurement position. Audyssey measured it at 10.8', which is just about right to account for the on-board DSP's latency.

However, the speakers, which are 10.8' away from the LP, are closer than the other 2 subs, (12.2' and 14.4'), and the speakers need to be timed to the combined output of all 3 subs. (Remember that I have my 3 subs hooked up to just ONE subwoofer output, so there is no timing correction for the 3 subs.)

IOW, Audyssey measured the arrival time of the wavefront from the *closest* sub, but the optimal Distance setting is the one that accounts for the timing of the *combined* 3 subs.

Again, I don't think this is an implication of a fault in Audyssey/Integra's system. It is an anomaly of the way I have my system setup. For more conventionally set up systems, I would fully expect A/I to get the Distance setting spot-on. However, I don't think there is any harm in measuring the result to verify this.

Jeff, (pepar), and Dennis, (DMark1), were here last night and we measured and listened to the system with the different Distance settings. They can verify that the more optimally measuring Distance setting also sounded appreciably better.
Edit: I see that Jeff did comment on this above.


Phase shift in my mains? They are Atlantic Technology 8200e's. Please explain why or how the mains would shift the phase at the crossover frequency.


Honestly, I doubt this is the reason I had cancellation at the crossover, or why changing the Distance setting corrected it. However, maybe I just don't understand what your saying. Could you please provide a link to the post by Brian at Rythmic? Thanks.

Craig