"Official" Audyssey thread (FAQ in post #1) - Page 1213

Markus, a full band Low Frequency Effect (LFE) channel doesn't sound too good to me, nevertheless, can you provide a link to such new(er) standards?

I think I remember Chris kinda saying that the recommended measurement pattern and distances gave Audyssey info about the room that would improve SQ at the MLP.

I would be interested in Audyssey's take on EQ'ing for an area versus for two seats right next to each other. Common sense says you would tighten up the measurements considerably when EQ'ing for two seats. But for some reason I came away with a different impression the last time this was visited.

Yep, I watched Ironman II set at 80hz(THX) and 120hz, could not hear any changes for better or worse. Most important is that it didn't boom at me, just solid low end. I'm just going to leave at 120hz, wouldn't want to miss anything. Maybe only things in the LFE at 120 is when the fat lady closes the refrigerator door, I hate when sounds like that are blown way out of wack.

It's possible that that person's sub is placed so it excites room modes that create, say, a peak between 80Hz and 120Hz in the listening position. My point is, I don't think we have sufficient data to conclude that any of what's been reported is silly. On balance, perhaps, but factual issues cannot be ruled out. Of course, if the issue is room modes, changing the LPF shouldn't be plan A.

I've seen many people (all much more knowledgeable on this than I) say that there is VERY little material in the .1 channel above 80 Hz, so that in almost all cases it matters not whether one sets it at 80 or 120 Hz. Still, there may sometimes be relevant signal above 80, so why not just set it at 120 and forget it (as several have suggested above).

If the two seats are in the middle of the room straddling the first mic position and all measurements worked off this point and were placed 18-24 inches from there, wouldn't that be the ideal situation and give you the most accurate calibration?

And acoustical treatments.

My understanding is you are supposed to calibrate the room, not the position.

As I said , I can't remember Audyssey's exact responses, but if memory serves right the take is that spreading measurements out somewhat gives a better picture of what is happening in the room acoustically. I have had success by going about 3 feet to the left and right of the MLP, 2 feet in front of the MLP, and about a foot and a half behind the MLP. That is pretty close to what Audyssey recommends. I think the best thing with Audyssey Pro is that you can try a few measurement patterns and then compare. From my experience so far, there isn't anything to be gained by straying too far from the general recommendations presented in the first post here and illustrated in the official Audyssey mic pattern doc.

I still have a hard time understanding why one would need to place the mic outside a smaller acoustic bubble box where your heads would be (assuming one or two listeners). Because anything picked up by a measurement outside that box that would effect what you hear at the MLP would surely be picked up within that smaller box. This is something I don't know enough about, and I have accepted that Audyssey knows much more than me, and that is just works great the way they recommend.

I have exactly that setup - two seats next to each other. I do fairly tight measurements - one on each seat at ear height, one in front of each seat, about 2 feet in front of the first three, one in between each seat, front and back as the two seats, and the last two to make a sort of diamond around the MLP (my seat) to bias the EQ towards me rather than my wife, who doesn't really care that much. This seems to give me a good result but I too have often wondered, and may yet try, a wider area setup to see if it improves the sound any. Logic tells me that a tighter mic placement is good in my situation - why would I care to EQ for parts of the room where nobody sits?

Can any of the Audyssey gurus on here (Jeff?) comment on that for me? I have a window of opportunity this weekend when my wife is out riding all day.

Kind Regards,

Keith

And my understanding is you are supposed to calibrated for an "acoustic bubble" around seats , ...not the entire room ...and not only for 1 strict position. Always avoid going off-axis the L and R speakers and don't measure near to side walls (nobody will be sitting there, anyway!)

The Genelec info is pretty good, but not totally accurate. Since they mention SDDS, they may also be talking about film systems, and DTS Digital film LFE is limited to 80 Hz. DTS Digital consumer is (or was) limited to 90 Hz. Dolby Digital does have a filter that brickwalls at 120 Hz, but that filter can be turned off, in which case the LFE channel goes out to 600 Hz.

PCM and lossless are correctly stated as wideband pipes, but when they deliver LFE, the content maker is responsible for constraining the bandwidth.

I was recently noticing that my well tuned room sounded great on 2-ch programs but occasionally had excessive/plump bass on some 5.1 music discs. Turns out many music discs do not have well filtered LFE tracks--easily seen using REW's spectrum analyzer. It also turned out that my SSP did not filter the LFE at 120 Hz or the like.

I did some experiments comparing SACD/DVD-A music recordings with the LFE unfiltered, or filtered at 120 or 80 Hz, and compared the results with the 2-ch mixes on those discs. It was pretty obvious that the mixers were listening with a monitor system using an LFE sub filtered at 80 Hz. The match was obviously right, whereas at 120 Hz it was not even close, and not very pleasant. They filtered the LFE sound in the room rather than the signal feeding the recorder.

I did a similar survey of movie soundtracks, and REW showed all were well filtered near 120 Hz at max. Some DTS movies were rolled off lower, like 90 Hz. In listening to these movies with 80 and 120 Hz LFE filters, it was possible in direct A/B to sometimes hear a difference only with the 120 Hz LFE tracks, but using either the 80 or 120 Hz filters sounded great and sounded correct. The impression was that the 80 Hz setting yielded "deeper, tighter" bass than the 120 Hz, and this has been a major reported difference between Dolby and DTS soundtracks since the days of laser discs. Interestingly, DTS HDMA does not employ the 90 Hz filter, so that "advantage" is now gone, even for the core lossy DTS track.

I have my SSP's LFE set for 80 Hz all the time (F/W updated!). It makes a huge benefit for 5.1 music, and a small benefit for movies, so it all sounds great now.

Roger, how does your room "look" wrt decay/waterfall .. in the 80Hz to 200Hz region?

Jeff

Hi Roger, I read your above post with great interest, but now I'm a bit concerned. As a conclusion, are you saying the ground rule of setting the LPF of the LFE channel in the AVR to 120 Hz (and forget about it!) is no ground rule anymore?

It is a bit confusing. People wandering in here that take all measurements in one spot are told to sample more of the room even though they only use one seat. Yet, we know to not measure too close to walls or "outside" the L&R speakers, so clearly there are places we shouldn't measure. And it doesn't make any sense to measure in a completely different part of the room than where anyone sits.

So, is .5m/24" the optimal measurement density, and should smaller seating areas be sampled correspondingly less than larger ones? Even with Pro, should a two-seat theater be sampled only six or so places?

IOW, once we've gridded only our seating are per the .5m/24" spec, are we done? C H R I S ?

(BTW, this is exactly what I do with Pro. I grid my seating area and then add two measurements behind my second row of seats. I could measure 32 spots with Pro, but I only use 13 for my two rows of three seats.)

Good question, Feri. He seems to be saying that nothing sounds worse with LPF at 80Hz, but hi-res multichannel audio sounds a lot better. I wonder if he has room modes that are at the root of that?

That rule was based on a society where everybody played by the rules. As we have found, there are bad apples among us. We can no longer leave our doors unlocked, or our LFE filters at 120 Hz.

No. If I did, stereo music would have the same problem.

How does that matter?

I guess the takeaway is that the boominess is in the content due to the content mastered on a system where the filter was on the monitoring system and not in the signal chain of the content itself?

It doesn't if the above is the case.

Care to expand on that Roger?

Yes, The room around the seats. Not one foot apart around the head. Unless you have a big head. If that is the case, it doesn't matter. Having a headless body in your MP would be better.

Certainly is.


I'd also certainly welcome a definitive answer from Chris if he's around. I've seen plenty of advice for larger theatres but not specifically for a small two seater room like mine. In the UK our room sizes tend to be much smaller than the US rooms and so our HTs are correspondingly smaller too. Mine is even at the small end for the UK - just two seats, and the backs of those are close up against the rear wall. That's why I have always done my measurements in a tight grid. Same with my EQ1 - as you know it can do 32 positions but I would find it impossible to do more than 10, and even that is pushing it.


Kind Regards,

Keith

What I can't quite get my (normal sized) head around is why I would ever want to measure in places where nobody will ever sit? If those parts of the room where nobody sits have an influence on the sound at the MLP, surely measuring around the MLP will include those issues too? I can see the point of taking wide area measurements in a larger room, where one wants to get good sound across a fairly wide area, for a larger number of seats - but in a small two seater room, I am only interested in how the sound sounds at a very small area around my two seats - aren't I?

Kind Regards,

Keith

Yes, I think you got it right. Only you need a dummy in one of the seats. Which one is up to you.

No, you can't calibrate the whole room. What you can (and are, with MultEQ) calibrate is the room + the source position + listening position(s).

Wow, you mean your wife has a Harley and you have to stay home!

Who said the whole room? The room around the listening positions. Not the MP listening position as some are attempting.