"Official" Audyssey thread (FAQ in post #1) - Page 1763

I'm pretty sure the thunderstorm I mentioned was recorded HOT. There's no way the LFE channel was restricted to 115db peaks! LOL

As far as weird weather goes, we had 70f degree weather in February (when it's usually between the teens to 30's), 80f + degree weather in March, then it snowed this week in April. Maybe the world is coming to an end in 2012? Enjoy those HT's while you can people!


Al, although the general consensus is that bass below 80Hz is non-directional, I've discerned that as with many things, YMMV.

All my speakers currently play flat in-room to the 30's Hz. I tried a simple test of switching between crossovers of 40Hz (the Audyssey XT32 default in my system) to 80Hz (the common recommendation) after the 'stereo bass' discussion came up.

I listened to both stereo music tracks as well as multi-channel movie clips. I heard the slight improvement in stereo music, but the difference was more easily heard/felt with surround sound. Specifically, one of the clips that I used for testing was the 'Star Wars: The Phantom Menace' pod race scene.

In that scene, I could hear the difference between the crossovers. I realized that I could hear the difference between my normal 40Hz crossover as the pods rumbled/roared by, vs the 80Hz crossover where I could actually hear the difference between, "these frequencies are panning across the speakers, THESE frequencies are not". The effect was a difference between the sensation of an actual real large object/vehicle zooming past vs the reproduced sounds of an object panning past.

It appears that I can localize below the regularly mentioned 80Hz. The papers that were posted also indicate that many folks can hear the differences and suggest that the real point at where folks cannot determine directionality or any envelopment differences is about 45Hz and below.

As I mentioned though, YMMV, but if you upgrade the speakers (and if you can get a great deal on them, why not?), you can try these tests for yourself.




Bold and underline, together, makes text really stand out, especially if you prefer not going through the hassle of separately quoting passages/paragraphs from a person's post to answer several different questions.


Max

....evidently has MultiEQ XT. Does anyone know why the US version does not? And, does anyone know if it's possible to upgrade via a firmware update from Denon?

I thought so, Keith. Although many of us here in "the colonies" still refer to them under the older "Jack Russell" name. Old habits die hard. One of my neighbors has had one for the past decade. My wife calls Sparky "a dog on springs". Nothing deters him when he's after porquepines, field mice, rabbits, ground squirrels, raccoons, foxes, even the occasional coyote! He's fearless and determined to a fault.

On the other hand, our dog Missy (a 14yr.old Australian Shepherd/Border Collie cross) has always been afraid of loud noises. Thunder storms, gunshots, and the Fourth of July fireworks scare the s**t out of her. She won't even stay in the room during close to reference level action scenes in my HT. The train wreck scene in Super 8 really does a job on her.

See, I was able to squeeze in a reference to HT in an attempt to keep this post OT

Cheers,
SB

Hi, welcome. Here is an accurate guide to the features of the current Denon models. That AVRxx12 thread would probably be a better place to post your question. Edit: but see jdsmoothie's post below

There is no FW upgrade of MultEQ to XT available on any model I know of. The 1712 is the budget audiophile model with no networking but it has XT. The 1912 adds other features but has MultEQ. The 2112, with XT, is really the sweet spot inexpensive-but-loaded value Denon AVR model.

I don't know about in the States, but here the Parson is somewhat different to the regular Jack. The Parson is a Jack but taller, basically. So they think they are even bigger dogs than the regular Jack - and that is *big*


Yes, that describes mine to a T. The other adjective that springs to mind is 'stubborn' But I like 'determined' more...


There's a lot of HF in that scene (in fact many people complain about the lack of deep bass in it - me included) and maybe that is what freaks her out. Mine is pushing 6 years old - he isn't scared of the SPLs in my HT room but he regards it with disdain - he looks in, looks at me like I am nuts, scratches himself and goes off to look for more interesting passtimes.


Dogs and HF - that's OT isn't it?

There is no 1712 or 2112 in Europe, XT starts with the 1912 there.

Was the shorter or longer distance the Audyssey setup distance? And is that both of the fronts together with the sub? Or what does the other one look like?

K no more non audyssey measurements. That is the flattest I could get it being in a reasonable range from what audyssey set. I don't know how extreme I can go without effecting timing.


The longer one was Audyssey. And that is both the fronts together.

I have purchased a real mike stand and aud. mike adaptor, but the overly bright sound is still there ,I am now going to do outside of the box measurements any one have any ideas.

Worth remembering that these REW (or equivalent) measurements are interesting, but only valid in a general sense since they probably do not make the same multi-measurement and weighting calculations that Audyssey does. In a sense it is a difference of opinion between two "algorithms," not a means to verify the exact performance.

Your comparison is quite convincing but something about the post-EQ adjustment of sub distance has been bothering me.

When the sub distance is changed, the phase or "time" of all sub frequencies is shifted relative to that of all higher frequencies. Is that okay? Might Audyssey be taking that into account when coming up with what looks to be an imperfect blend between sub and satellite?

We are taking a very complex sound wave that is broken down into components (sub frequencies and higher frequencies) and outputting it via two different drivers (sub and satellites). If those two sound components are re-mixed in the air between the speakers and our head with a different phase/time relationship, haven't we changed the complex sound wave?

Adjusting the sub phase/time so that the SPL magnitude at the blend or XO point no longer has a dip suggests we are bringing the two into phase at that frequency. However, ported subs usually have a large group delay meaning that the phase/time changes rather significantly at progressively lower frequencies. Matching the phase/time at the XO may mean a greater mis-match at progressively lower frequencies and maybe a greater distortion of that complex sound wave.

Has Chris ever weighed in on this? I wonder if we are missing something and maybe correcting something that is already good from a perceptual standpoint.

Since Audyssey EQ's for a good blend, it just doesn't seem like something that would be gotten wrong. And if a good blend involves more than just the narrow frequency band around the XO, why can't the amplitudes at the XO be adjusted to cover this dip at the XO.

Is this issue common only with ported subs? Is it common only with XT and not XT32 (higher resolution of XT32 takes care of it)? Does the pro kit's more complex blend algorithm take care of it?

Just tossing this out there looking for peace of mind. I don't have the issue but it bothers the engineer in me all the same.

I agree with the above post. I think it is fine tweaking it maybe a foot or two, but something like a 5ft change usually starts messing with how things "sound" because of the difference in delay time.

I think Keith's comment about the potential for new satellites to not change the 100 Hz crossover point is a good one.

Notice that Audyssey is finding your satellite outputs dropping at around 100 Hz even though your current satellites have capability considerably below that. This will not change with the more capable satellites. The room characteristics have diminished the output below 100 Hz and will do so for the new satellites regardless of how low their -3db point is. Hence the likelihood that the crossover point will not change.

If the room characteristic that diminishes SPL below 100 Hz in the listening area is only 10 or 20 Hz wide, the higher output of the new satellites at lower frequencies might turn a roll-off into a dip and one that Audyssey can correct. In this case Audyssey might choose a lower XO. However, given that the B3 only gets you 7 HZ more I think it is very unlikely to help. The added 21 Hz improvement from the S3 just might get you a lower XO.

You might try different mic positions, maybe raise the mic a bit or make the MLP a different seat (or move it a bit if it's between seats) to give Audyssey a different basis on which to identify the low-end cut-off of the satellites.

I'm not saying I wouldn't recommend the upgrade, but I think there's a good chance the XO won't change.

From the distance experiment I did with my room/sub, going from 12.5' to 16' made a world of difference (that along with x-over setting).

Hi Harrison, while I fully agree with you on the engineering side of the story, I tend to belive that a test microphone surely behaves differently than our ear/brain combo. Clear cut graphs can be taken showing delays (or even early arrivals when sub distance is set less to what Audyssey set it), but I think our hearing system will be able to take care of that or maybe pay not too much attention to delay shifts rather than paying attention to variations in the frequency range (limited around the splice).

Just my 2c.

Harrison, your questions seem valid to me, and I too had reservations about messing with what Audyssey had set.

To add a little more background information, I have three sealed subs (Hsu ULS-15's), and an AVR with MultEQ XT32. It is always a challenge to run an Audyssey calibration when more than two subs are involved. My approach is to connect two subs to the first sub channel with a Y-cable, and place these subs equidistant from the MLP. The third sub is on sub channel 2, and placed considerably closer to the MLP than the other two subs. After calibrating in this manner, I listen closely to the bass response first with the default Audyssey distances, and then after adjusting the distances for flattest response using REW. Based only on my ears, the bass sounds better with the adjusted distances. I don't know how to measure the effect on phasing. However, if the effect were bad, I suspect I could hear it.

There are a number of us here on the Audyssey thread that have used this approach when calibrating three or more subs. I believe Craig John was the first to suggest this to me. I have not discussed this on Ask Audyssey, so I don't know what the official answer is.

+1. The relative phases of the frequencies also change whenever you move whether your sound source is natural or from speakers. I am pretty convinced re are not wired to react to relative phase per se. I lost the link but somewhere a site shows and lets you hear the phase inaccuracies of a number of headphones. A square wave sounds the same yo me even when the phases are so misaligned that the eave beats no visible resemblance to a squsre wave. I was a bit surprised

I do not believe this is the case. Source?

Here's an interesting discussion on Group Delay of Loudspeakers and the audibility thresholds of group delay of the human ear.

AFAIK, actually Audyssey EQs each speaker and sub(s) inedependent of each other, then reports the results to the avr and the avr sets the crossovers with HP and LP filters. The slope of those filters will be responsible for the flatness of the dip at the splice at and around the c/o point and as wee see it's far from being perfectly flat. So, in other words, Audyssey is least responsible for the splice.

In my case after audyssey the mains are set full range. Depending if I choose 40 80 or 120hz crossover the correct distance setting based on flattest response is different. And all of those distances are different than the distance set by the receiver setup. And it varies from 6ft to 18ft.

Changing the sub distance this much does sound different, but it doesn't seem to be the freq response rather the timing of the bass with other freq. Could be just placebo since that is what I'm "trying to here" but don't know ho to prove it.

First why not settle with the proper c/o setting (like 80 Hz), then tweak the distance for a flat(ish) splice, then forget about it and enjoy! How is that?

Well, because testing a month or so ago showed 80hz is too high as surround pans are much better with 40-60hz crossover. If I run them full range I don't have to worry about the distance as much because there is no integration with the mains and that would be the best setting. But knowing full range is not optimal, I wanted to use 40hz for the front crossover but that gives the best sub integration with a sub distance of 18.5ft vs the 13ft set by the autosetup. But music sounds "off" with that distance setting.

So OCD has won and there is no perfect setting.

Very good read but I didn't see any answer in there.

INTEGRA DHC80.3

Ran the room correction 3x, each time during speaker set up LCR RSB all show up, LSB pings but the Audessy doesn't highlight it and I get an error.

Ideas?

BTW 5.2 config.

No particular source other than Chris' comments over the years. Also, as I recall, a key feature of the pro kit is a more elaborate algorithm for developing the blend; hence my question about whether XT32 does a better job of achieving a smooth gain curve at the XO.

What we don't know is what all goes into the blend calculations. If it's just a matching of the phase of the sub and satellites at the XO (and a corresponding gain adjustment) then it's a simple matter and it's hard to believe that Audyssey is ever getting it wrong.

I'm assuming that all satellites are adjusted for similar phase at the XO and when the dip occurs, it occurs with all satellites. I don't know that we know that either.

But, there seems to be a consensus that adjusting the sub phase (distance) improves the sound and I can't argue with that.

I've been assuming that Audyssey knows what the XO filter characteristics are and takes those into account when setting the speaker distances and the Audyssey filters. I wonder if either that isn't the case or the receiver manufacturer is not giving Audyssey accurate information to work from. Is this problem common to only certain receivers?

A very interesting read indeed. I recall reading similar material when I compared characteristics of vented and sealed subs a few years ago and decided to go with sealed subs.

I wonder though, with the advent of Audyssey and improved speakers and HD and SACD sound, might we be more sensitive to group delay than the authors referenced found (I didn't check their publish dates but I assume they pre-date the latest technology)? Maybe it's the OCD but we are fussy bunch with keen ears and some excellent equipment. I'd bet that group delay is a bigger issue today (for us) than it was 5 or 10 years ago. This suggests that if adjusting the sub distance improves sound, it's the right thing to do. If the sub distance is increased to smooth out the blend, then a byproduct is effectively a reduction of the group delay.

Have those that adjusted the sub distance found that increasing it is the solution?

In a 5.2 setup you don't have a Left Surround Back and a Right Surround Back. You have a Left Surround and a Right Surround. Re-connect your surround speakers to the proper connections on the AVR and run Audyssey again.

1998 is mentioned at the top of the article. So, ancient in tech terms. A lot will have happened in 14 years I'm guessing.