"Official" Audyssey thread (FAQ in post #1) - Page 448

I would love to cut out the frequecy's above 80Hz going to the sub...but then I think there would be a gap in the sound range from 80hz-100hz because I doubt the speakers can properly produce sound below 120hz. If someone else thinks that is a good idea to change the sub cross frequency to 80hz, I will. Again, I thought I have heard people advising me to keep the sub cross at 120??

Art,

I also have owned the 9.8 and now the 9.9 since it's inception. I have run MultEQ XT and the PRO calibration probably 30 different times on my own system! (I'm a PRO installer) and as a test try running the minimum 3 position's in your 1st main listening position without moving the mic (making sure the mic is at Grazing incidence with your tweeters and at the same "bottom" height as your tweeters {if the tweeters are 39" from the ground make sure the mic is not 40" with your mains} ) READ: THIS IS "NOT" THE CORRECT WAY TO MEASURE YOUR ROOM AND DO "NOT" USE IT AS A "VALID" ROOM CALIBRATION !!! The reason I suggest this "TEST... THIS IS ONLY A TEST...." is the high freq's coming from the tweeters are just like a flashlight! I'm not familiar with your system but this is especially true with planar tweeters like Electrostatic or Infinity's Emit's (That I have..) Make sure your mains are toed in toward the main listening position. "IF" your measurement pattern within your listening area is too large (outside the listening parimeter of the tweeters) the high freq fall-off will be detected by MultEQ XT and will compensate accordingly. This "may" cause the "sizzle" you are experiencing (I know, it's happened to me... ) This is "ALL" system, speaker, room dependent but of course this is a "TEST... THIS IS ONLY A TEST....!!" When listening don't pay attention to the bottom end (it probably will be completely messed up...) but see if the "Sizzle" is diminished. If so then try closing up the measurement pattern a bit paying CLOSE attention to mic height (I personally don't vary from the 8 position Right/Left/Front/Rear 2' offset from the EXCELLENT SETUP GUIDE {Thanks Giomania...} on my personal system) and see if that helps.

As far as your back channels being too hot this is an inexpensive "TEST... THIS IS ONLY A TEST....!!" The Harrison Lab's Passive (3,db, 6db, 12db) attenuator's have helped me on several systems available here:

http://www.parts-express.com/pe/show...TOKEN=39608546

Hope this helps and of course this is IMOP....

Bob..

There are more knowledgable members here on this than me. But I always thought the higher you have the sub crossover set the easier it will be to locate. I am not sure what front/center speakers you have though. But you can try them at 80Hz to see how it sounds.

Bill

The crossover being discussed is often mis-named the sub crossover but it is really the LFE crossover.

Paraphasing Chris: the setting on your sub should be set to max (usually120Hz) or defeated if possible. In the receiver the sub should also be set at 120Hz.

I would run a sweep through your sub - say 10 to 300Hz. The sound should be clean all the way through (you may not hear the 10 - 20Hz). If you hear some vibration or distortion from the sub at any point, then your sub's driver is shot - probably a loose coil.

Don't worry too much about the speakers being recorded as "Large". That's probably just the speakers coupling with the wall/corner. In the settings, just take them off Large, and set them to 80Hz (for starters). But I think you need to confirm if you have a sub problem first.

BTW, after running Audyssey, my mains were set at 150Hz. I left them there. This meant that any frequency up to 150Hz that was meant for the mains would be sent to the sub. However, I haven't heard any "localising" of the sub, although I must admit that my sub sits right underneath my centre.

Bill, this is a common point of confusion. The crossover for the speakers determines the content that goes to the sub from each of the channels except the LFE channel (the.1 of 5.1). When set at 80, all freq<80 go to the sub, all above 80 "crossover" (go) to the speaker. The higher above 80 Hz it is set, the higher the freqs going to the sub and the more chance of localization to the sub. If the sub is up near the fronts there is much less potential for problem localization.

The 80/120 setting, which Kal just clarified for you as the "LFE crossover" is most properly referred to as the LFE LPF (low pass filter). If set to 80, nothing above 80 in the LFE channel will go to the sub (and you might miss the higher freq part of the LFE rumble) but it will never get crossed over to any speaker (it is a filter after all, not a crossover). It is recommended that it be kept at 120 as that is the max freq content allowed in that channel and AFAIK has not been accused of any localization-related crimes.

SoundofMind,

Thanks for the detailed explanation, much appreciated. I was right on one thing there are more knowledgable members here than me.

Bill

Ready to pull the trigger on the 3808 and had a question on the Audyssey. I have NHT U2 Subwoofers which includes a separate X1 Active Crossover, how does the Denon 3808 Audyssey handle this when the X1 is a separate component?

I posted this in MCACC thread and no reply so far, so I'm reposting it here...

"...

What's the consensus on the above two calibration methods and which one is considered "better"?

Pros and cons of each?

Thanks"

If the subs are self powered and have RCA connections, you should be able to leave the crossover out of the system completely, unless you end up needing the gain control on the crossover to make Audyssey report approx 0 db level for the sub.

Lots of really great ideas here. I'm learning so much about audio by reading this thread, from practical results rather than theoretical ideas.

The question that comes up is if the treble is like an incandescent flashlight, with a small strong spot roughly in the middle of a much wider spot, do we want to calibrate in the "hot spot" or try to find an area away from the "hot spot" that will have roughly the amount of output that's average for the wider spot beam.

I'm thinking the latter. Our ears are unlikely to always be in the narrow hot spot. It's probably best to toe the speakers in enough that the hot spots converge in front of the listener/mike by say a couple of feet, just enough to keep it away from the listener, but not too far away.

So perhaps an "advanced setup guide" would have the listener first calibrate a single position until the distances to the two speakers come out exactly the same (because some have expressed a concern that different distances can introduce delays that mess up imaging). Use that mike spot as the sweet spot. Then calibrate three positions, listening and repeating moving forward/back or adjusting toe in until the imaging/"sizzle" is optimized. Only then go ahead and calibrate all 6 or 8 positions. I think this will be my strategy for my next calibration.

Here is a link to this thread with MCACC being searched:

http://www.avsforum.com/avs-vb/searc...rchid=12735595

EDIT: as pointed out by krabapple below, the link above didn't transfer but it was with the word "MCACC" in the "search this thread" in the upper right corner.

Thanks for the advice. It seems that the fingers are pointing at the sub for the odd tune and localization. Is it possible for me to download a test CD for my sub anywhere?? What is the most affordable test CD that I can obtain for testing the sub...and maybe the whole system?

Would it be better to move the speakers? That way you could keep the sweet spot where you want it to be, like centered in front of your screen.

I do not have that setup but I have separate crossovers for my mains and my subs (DCX2496). I let Audyssey run whatever it has too then I can tweak anything still with my active crossovers.

The Active crossovers do not effect what audyssey does but they are great for after the fact tweaking.

Because Audyssey does apply EQ above the transition band.*

To put it another way, what does Audyssey do that addresses the concerns of those authorities who believe that frequencies above the transition band are best treated by other means than EQ, if at all? If such EQ is problematic , how has Audyssey solved the problem(s)?



(Of course , Audyssey is not alone in this among room EQ packages)

The link's broken. If it was a search of this thread, you can't save those as links.

The original poster can replicate it using the 'Search This THread' tool in the upper right of the AVSF window.

I am certain that everyone agrees that treatment of all frequencies is best done physically and that electronic EQ, of any kind, is necessary only when one cannot or will not employ physical treatments. I think that even the proponents and manufacturers of electronic EQ would give, at least, tacit agreement. The problems are two-fold.

First is that treatment for lower frequencies requires much larger objects than many will accept for domestic situations. This is why electronic EQ has made its greatest impact and market acceptance here. The fact that many sub EQs are offered for this is a matter of convenience, unfortunately, since one really wants EQ up to the transition (Schroeder) frequency and that is almost always much higher than the upper limit of a subwoofer. Nonetheless, working even within the bandwidth of the sub, important improvements can be effected.

Second is that treatment for higher frequencies is relatively easy, although still resisted by many. It would be great if everyone paid attention to it but, since most do not, EQ for these non-modal aberrations can be done. One uses wider bandwidth, lower Q filters, almost like a "tone control."

Note that Anthem's ARC provides an option for setting the upper limit for its EQ in recognition of these issues but that most of the users (see that thread) seem to prefer correction over a much wider bandwidth. I do, too, as, when it is successful, it measures and sounds better. Same for Audyssey.

So, it is not a matter of "never do any EQ above the transition frequency." It is that one should attend to that range in a different way and only use EQ if absolutely necessary. Unfortunately, I'd bet that, for the majority of systems, even here on AVS, it is absolutely necessary. If your system does not need it, here's pat on the back for you.

Grazing is necessary with the Pro mic?

Please don't read too much into it. This was JUST A TEST.... for Art to try to see if "HIS" system measurement pattern "MAY" be too wide causing a HF roll-off that MultEQ "MAY" be compensating for. Nothing more.

I realize the Audioholic in us gets going but please don't vary from the setup guide! Your analogy suggests a "hot spot" of a few inch's in the main listening position. In "most" systems this is not the case and these extremes are not necessary.

The idea is to sample the listening area as per the guide without going too far outside the dispersion pattern of a particular speaker/room/system combination. The 2 foot 8 pattern offset in the guide works great in most rooms. Some rooms have wide couches and the desire to measure at those extreme outer positions takes over and that's where potential problems arise.

Another possible "Sizzle" issue is running MultEQ with the speaker grills on. The argument that most listen with the grills "ON" is completely valid. However, speaker grills/cloth do effect the very top octave and MultEQ will compensate for it. If "YOUR" system seems a bit too hot in the top end it's worth a try.

Of course all is IMOP..

Bob..

THIS WAS A TEST, ONLY A TEST.... "DO NOT VARY FROM THE SETUP GUIDE"

Sorry guys, I think I opened a can of worms...

http://www.realtraps.com./test-cd.htm

Absolutely.. the PRO mic is no different other than it's calibrated with the matching phantom power pre-amp to within .5db tolerance vs 2db for the production mic's.

Bob..

Hi Criss,
I'm trying to decide where abouts to move my front speakers, at the moment my sub (SVS PB10) is about 1.5" -2.0" away from my right speaker, should I move this to another part of the room. (will it matter where in the room the sub goes) I can probaly only move the Celestion A2's very slightly as my cinema room doubles up as the sitting room.

Many thanks.

Mike.

Yes. Proper measurement of reverberant rooms requires pressure-field calibration and grazing incidence. Anechoic measurements of speakers are done with a free-field calibrated mic pointing to the speaker.

Kal already summarized this very well. I assume that by "transition band" you are referring to the Schroeder frequency, right? What is often missed when quoting Floyd's work is that his statement about EQ above the Schroeder frequency has some caveats: "A properly designed loudspeaker in a properly treated room should not require EQ above that frequency". This is very true.

Starting with these two design requirements (room and speaker) is critical.

The way that the Audyssey algorithms deal with this is to use different weighting (and different methods) for the two regimes. The decision on what to do and how much to do is based on information that is gathered during the measurement process and the subsequent deconvolution of the impulse response. There are many components to this including frequency warping, early and late energy weighting, selective smoothing, and of course spatial clustering that is critical in understanding the distribution of the problems across the listening area.

Dr. Geddes did qualify his statement in saying that a great room/great placement /great speakers should have no need for EQing....which is true but meaningless for 99% of all audio owners out there

Hi Mike,

Moving the sub will certainly change its response. However, I think you are trying to find a lower roll off point for your main speakers, right? I would try to get them out of the corners as much as possible. If there are any cabinets or other big furniture items near them these can also create "cavities" around the speakers that produce big bumps in the 100-200 Hz region.

Thanks, looking back I should have known this from previous posts.

Sooo, it would see the hump and consider the -3dB point from that as the speaker's -3dB point even though w/o the hump, the flat response continues lower?

Thanks, I'll move the sub as it probaly blocks some response out of the right speaker, then I'll move both speakers out slightly and stop them pointing in to the room so much.

Many thanks again.

Mike.