My AVR-3806 Audyssey experience! (measurements inside) - Page 10

120Hz

To expand on the 120Hz:

The LFE track goes, in theory, up to 120Hz, though the majority of it is under 80Hz in most movies. So, if you want to be sure of getting the full LFE channel use 120.

When you set your speakers to small, and the crossover at 60, anything less than 60 is going to be sent to the sub irespective of what you set the sub crossover to.

If somehow you find your sub generating too much "noise" you could drop the 120 down to 100 or even 80 and probably not miss much LFE and you will not loose out on the base being redirected from your mains unless you lower the sub to less than the crossover you set for the mains, in this case 60.

I run mine at 120 various crossovers between 90 and 40 on the other 5 speakers (based on Audyssey recomendations) and I don't feel that I am missing any frequencies. Once I got used to it, I also realized that Audyssey setting my subs at -6db actually sounds better, especially with music, and I can alwasy run it a coupld of db hot on movies if I wan't to.

Ruin

Are you saying you don't think I fleshed out the topic enough with my post?

Nick

All filters operate in the time and frequency domains; in linear systems theory, they are just two ways of looking at the same thing: the complex impulse response of the system.

The Audioholics review claims that the Audyssey filters are minimum-phase. This means that their magnitude response entirely determines their phase response, and that they cannot independently correct magnitude and phase.

Is the Audioholics review correct? Are the filters indeed minimum phase?

Please note that I am not saying the minimum phase is a bad idea. Most of the literature of which I am aware concludes that trying to correct room acoustics via non-minimum phase filters adds unpleasant audible artifacts (such as preechoes), and minimum-phase is probably the safest choice. However, I am confused by a claim that implies that Audicy's filter "minimizes reflections occuring after the main peak" unless Audioholics was misinformed.

Chris,

Do you know if the Denon version of the Audyssey PC application will be for the new 5805MKII or also for the original 5805 receiver?
Will it also be available for the EU AVC-A1XV model?

Regards,

Adri.

TESTING AUDYSSEY ON A BUDGET SYSTEM


I have an AVR-2807 and I wanted to test and see how Audyssey would calibrate a very cheap HTIB system. My normal system is a Paradigm Monitor 7 setup, CC370, ADP370, and PW-2200.

I hooked up my old Kenwood HTB-504 system that I had in storage and listened to it without any EQ for a week. As you may know, this 5.1 system uses paper cones and is in no way comparable to a real audio setup. However, in terms of HTIB, it's probably middle of the road for its time. The sub is an active Kenwood sub.

Listening to music on this system was painful -- everything sounded extremely dull and 2D. You could not make out background vocals or individual instruments at all. Movie soundtracks were OK with a standard "boomy" bass that is standard with these types of systems. Dialogue was still very muddled and unclear.

I ran the AVIA frequency sweeps -- there is a huge hole from 80-100 Hz with a huge peak around 55-60Hz on the sub. Overall, fluctuations were about +/- 10 db on a full frequency sweep with a -20dB dip at around 85 Hz. Using the circling tone that is supposed go smoothly speaker to speaker around the room, I found that the sound jumped from speaker to speaker and that it sounded completely different when coming from the fronts, centre, and surrounds.

I then ran the Audyssey setup to see if it could inject any life into this system. As a note, the crossover frequencies were:

Fronts - 90Hz
Centre - 120 Hz
Surrounds - 200 Hz (yes, it's terrrible)

I took 6 measurement around the room to ensure a wide soundfield. Looking at the graphs (yes, I know they are approximate), there were large compensations of +/- 12 db at the various frequencies. It was obvious where Audyssey was trying to fill in the holes.


RESULTS

I ran the frequency sweep and amazingly there was only a +/- 3 db variation along the frequencies! Of couse, this is bad when compared to my Paradigm setup, but compared to the previous results, it was quite remarkable. There was a 10 db dip around 80 Hz (below where the fronts could produce sound and above where the sub was effective), however it was not as acoustically noticable as before.

Using the panning tone around the room, it actually sounded very similar from all the speakers! Also, the panning seemed to be smoother, likely due to the tonality being evened out. Putting my ear against the sub, you could hear it kick in when the tone when to the surrounds compensating for their poor response.

Finally, turning on music was unbelievable. Suddenly, you could clearly hear the background vocals and instruments. Everything was in 3D instead of flat localized response. Also, the sub now reproduced different notes of bass instead of a single "boom-boom".

The system doesn't compete with Paradigm, but the effect Audyssey had on a budget system was remarkable. Of course, most people spending $1000 on an amp would never use these speakers, but it speaks to the power of Audyssey. The before and after sound like I am switching speakers altogether, not just activating the EQ on the amp.

Thank goodness I can now throw out my HTIB and never have to do this test again!


Sumster


P.S. If people are really interested, I can take photos of the Audyssey EQ graphs and post them before I disassemble the setup.

Very interesting test. It is nice to know that it can have that dramatic an effect when needed. Of course mostly Audyssey is being applied on systems that sound pretty decent even without it so the change is more subtle.

Thanks for posting your results!

Ruin

anyone know if there is a way of upgrading the Audyssey version on a 3806? I just got the 3806, but reading thru the review at Audioholics and at Stereosound, seems like there is new "improved" version of Audyssey....can anyone shed light on this?

I was able to download a new version for the 4806 when I had it. Go to the Denon website and register for the updates. It will tell you what is available.

i already did, and nothing shows up.

on my 3910, I have done several, up to DL3, but nothing on the 3806..........I dont even know how you would upgrade it on the unit, other than thru the serial port?

Isn't the Audssey software on the Denon AVR-3806 already the latest version? I thought the update was intended for the '4806, or one of the others?

Guess so, based on the above-noted experiences.

BTW, updating is done with serial or ethernet cable after burning files to CDR.

Kal

Chris from Audyssey can confirm, but, I don't think you can upgrade. The 4806 has more memory available to Audyssey than the 3806 and there is no Ethernet connection on the 3806 in any case.

Nick

This is a dumb question, but how do you know if MultEQ has been disengaged? Is this indicated by the light turning from green to red? Is this 512 point filter active in the "flat" setting as well? I ask since I tend to prefer this setting over audyssey.
Thanks,
Rob

The red light does not mean it is disengaged, but means that you have changed something from the original setup. It is still working otherwise.

The filter is still active in the "flat" setting. It is indeed the filter that tries to achieve the flatness.

When the Audyssey light is completely out then it is disengaged/bypassed. Green means you are using the Audyssey filtering with no user changes. Red means you are either using the Flat or Front settings, or have made a user change to the Audyssey settings.

Ruin

Set room eq to "off" to disengage MultiEQ for sure.

Nick

ok...is there ANY way of disengaging Audyssey from the LFE signals? I love what is doing from 80hz+, but my subwoofer sounds muffled with it engaged....I turn it off (audyssey), and the subwoofer comes alive like its supposed to.

I've double checked distances (fixed those), levels, etc....

any ideas if I can disengaging it from the Low freq stuff?

You can't. It's either on and EQ'ing everything or off and EQ'ing nothing...

that sucks

I suppose you could disconnect the sub after it's first been detected by the auto setup and before it takes the various measurements. Might end up with some strange results though...

I know what you mean about the sub output. Audyssey appears to be cutting back on a lot of my output below 20hz (if I use my Velodyne's OSD as a guide).

You could also trying keeping the sub turned off for the entire auto-setup and audyssey eq routine. Afterwards add the sub to the speaker config, set voume and distance as desired, and start with setting the speakers x-over to 80 Hz. I think you'll still get some equalization of the sub from Audyssey, but it should be minimized.

You could also try re-running the Audyssey set-up, this time with the microphone moved a couple of inches from where you previously had it. It could be that the microphone was in room node (peak or valley) and Audyssey tried to compensate for it.

In my case I've re-run Audyssey about six times now. Taking measurements at all eight positions (Amp is the A1XV - 5805) and throughout the room. In each case the Velodyne's OSD shows the output as starting to roll off from around 20hz or so. Switch off Audyssey and the output is restored. I think Audyssey works very well but you end up with a rather curtailed sub 20hz output, which is odd. The effect is almost like THX Ultra2 Boundary Gain Compensation is switched on but I have that switched off in the Amp's setup menu.

Exactly!!!

I have a sub that is pretty flat to 15hz, and it is the nicest piece in my setup....with movies like War of The WOrlds, Nemo, etc, the walls come out

Today I put Nemo for a friend, and I was like....HUH?! What the hell happened to my Sub?! turned OFF Audyssey and viola! there it was.

Your sub in your room probably has a rising frequency response in the bottom which you prefer and which Audyssey is trying to flatten. I would try disconnecting the sub first, then running the setup without the sub, then reconnect the sub, set up the crossovers manually, and dial in the sub level to suit your tastes. Audyssey would then have no effect on the curve or level of your sub.

yeah...that's the next I was thinking of doing in my head. I'm gonna do that and see what happens.

Audyssey is trying to straighten this out:


but as I've posted before, none of these EQs (Audyssey, MCACC, YPAO, Logic7, etc) will do a good job as a BFD will, specially in the sub 35hz region.

Is there an "Audyssey for Dummies" book I could pick up to figure out how to optimize the sound from my Denon 2807 without first getting a degree in sonic engineering? I too am frustrated that Audyssey keeps minimizing the sub -- in fact, I've got my sub on the same plane as my front speakers, but Audyssey keeps reading it as though it's 15 feet farther back (which would put it outside of my house approximately in the middle of my neighbors driveway.) Can you do better using the Avia test CD and a radio shack sound meter and making all the settings manually?

two different things ginso: Audyssey is supposed to not only check your distances/levels/phase/etc, but also apply their highly algorithmic EQ....two things going on at once.

You can have Audyssey still do the EQ for you, and then go and manually adjust your levels/distances/size/phase/etc manually with AVIA and an SPL meter.

my only thing with this is the horrendous job it does with LFE levels....wish Audyssey would really look harder in the .1 channel, as in many cases it is the MOST important channel for many people

I usually just let Audyssey do it's thing, then boost the sub's amp gain 3-4 db's to my taste. I also switch any speakers that the setup calls "large" to "small" and then choose the crossovers.

Interesting. I am the only person where Audyssey sets my sub about 6 db hot. I have tried on two different subs, same thing. No big deal, as its an easy fix. Just to be sure, you guys have the mic dead level and pointing at the ceiling, right? No offense intended I just happened to missed that my first time through and naturally got an very odd result.

Nick