The Audyssey Pro Installer Kit Thread (FAQ in post #1) - Page 79

I think it was pepar who alerted us to the fact that you do not need a router or hub. You can connect Ethernet from a PC directly to the prepro with nothing in between. Apparently, it is self inverting. I just tried this last week and it works like a charm. Glad I paid attention!

Of course, you can do it via a network as well. But, best to have the PC close to the listening room because each new mike position sweep will require hitting a soft button in Pro to start it.

Thanks mate,

Yeah I tested it last night and it worked over wireless. I have an Ethernet switch in my HT room and will just run the ethernet cable from the laptop to the switch and all should be good . Thanks for telling which input the mic preamp plugs in! So PC input and left channel on the back. I will end up leaving the cable there too as I know I will be running a few cals.

Also when I run the first calibration. I have to make sure the top of the mic just clears the backrest of the seat correct?....or should I have the whole mic clear the back rest?

Like in this pic

http://www.hometheater.com/images/ar.../807Hook.1.jpg

I also cannot put the boom arm in between the mic and the speakers, yeah? *I read the auddyssey guide*lol...but can I have it to the side of it extended?...like in the pic above but from the front of the seats instead of the back like how that picture is?

Thanks a bunch mate.

Interesting. Self-inverting? I guess it must be or it wouldn't work. How does that work then - the self-inversion I mean?

So long as the mic can 'see' the rear surrounds (if you have them) you will be OK. If you don't have rear surrounds it's less critical. The thing then is to make sure that the mic is not so close to the back of the chair that it picks up reflections. I recline my seat and make sure the mic is a foot from the backrest, even if that puts the mic slightly in front of where my ears would be. I think it is easy to overthink this - we move our heads around a bit when we listen so it isn't critical to the inch IMO.


In the pic the boom arm is between the mic and the speakers if he has rear surrounds. Again, this is theory rather than fact IMO, but yes it is best to avoid any reflections from the mic arm. If you can, position the boom so that is not between the mic and speakers. Follow the Guide and/or the 101 in the FAQ and you should get a good cal first time.

Most devices now use Auto-MDIX Ethernet ports, which detect and switch to the correct config.

Ah, thanks mjf... I didn't know that. Makes life easier I guess. Especially for people like me who have a knack of entering the Twilight Zone whenever a router or an Ethernet cable is involved.

It certainly saves having to hunt around for the right cable. It's good to know it can be plugged in direct as I want to run my Pro calibration today and now I don't have to move my PC.

Good luck with the cal! I take it your new subs have arrived?

I am doing a new cal today, with mic positions all centred around the MLP and the speaker toe-in slightly adjusted. I was very happy with the last cal but want to see what, if any, difference in SQ and measurements I get when I concentrate on the MLP.

Yes, I got them last Friday.

I've been thinking about Mic patterns and I'm still not sure what to do for the best. I think I'll just have to pick a pattern, live with it for a week, then start experimenting.

I really need to get myself another Mic and start measuring with REW to see what is going on, but I could do without the extra cost at the moment.

Excellent. Let us know how they sound!


That's what I did/am doing. My last cal was very heavily biased to MLP and was excellent - the best I have ever heard it sound here.


My OmniMic paid for itself in terms of SQ by helping me first identify my big dip in the bass near the XO region, and then helping me eliminate it. My bass has never sounded so tight, deep and overall good. Without the OM I would have had to rely on Pro 100% and, while very good, that would not have been as good as I got it in the end by using the sub distance adjustment tweak.

EDIT: Finished a 12 point calibration around the MLP only. Superb sound. Too soon to say if it is any better than the last one (heavily biased to MLP). I have yet to do any sub distance tweaks - that is the next job.

2nd EDIT: Interesting. Audyssey set my sub distances to 12.2 and 12.0. This is *exactly* what I adjusted them to in the last calibration - they had been 12.8 and 11.2 and after OmniMic tweaking, they ended up at 12.2 and 12.0, which is where Audyssey itself set them today. The interesting thing is that after my OM session this afternoon, I cannot improve on the 12.2 and 12.0 that Audyssey set. Subjectively, the bass is the same as it was before. It graphs slightly differently in OM but it must be insufficient difference to be audible.

Differences between the two calibrations are 1) the mic positions were all centred around the MLP this time and 2) I have toed in the L and R speakers more, so they are now pointing exactly at the MLP, and the centre speaker is pointing much more directly at my head when seated at the MLP. The reason for messing with the toe-ins was that I believe I was getting a very slight 'edginess' on dialogue more often than not, and I decided to experiment with toe-in adjustments to see if I could, by more precise tweeter aiming, get Audyssey to do a little less work. This seems to have been successful as listening tests made later show no sign of the slight edginess on any material.

WRT to the mic pattern, I don't think concentrating around the MLP has made a huge difference - my last cal used a pattern predominately around the MLP and far less so around 2ndMLP. Of course, I have made the mic pattern change and the toe-in change at the same time so it is difficult to be sure which has had most influence on the result. I considered doing these two changes separately but in the end decided that that would take an entire day and I couldn't muster up the motivation.

I am extremely happy with the latest cal - as I was with the last one, other than this slight edginess on dialogue. I say 'slight' and I mean it - at times I think I may have been convincing myself. However, I now believe it is gone entirely, so placebo or not, it's a result. As a by-product, I also think that the extended treble is somewhat smoother - I attribute this to the precise toe-in of the speakers.

As usual, I will leave this cal alone for a week or so while I get used to it and play a variety of material. But happy bunny mode is definitely ON

He was talking about an Auto MDI-X port. It basically means the port automatically determines whether a connection requires a crossover or not and configures itself appropriately.

Thanks. And thanks for the link too. Great stuff.

1. Are you still running one of the subs at phase = 0 and the other at phase = 180?

2. Can you post the certificates for the old calibration vs. this new one? Would like to see how much less treble boost you're getting (if any) with the new calibration being tighter around the MLP w/greater speaker toe-in.

3. Strange about the sub distances now being set optimally when before they were not, and required you to do it manually. Actually disappointing too, because it means we're back to inconsistent calibration behavior. I thought we had concluded Audyssey does not optimize the x-over area because it never measures sub+mains together. Yet in this case, you get a seemingly optimized result. We're bouncing off guardrails!!

Congrats on this slightly better, less time-consuming-due-to-needing-external-measurement-and-additional-tweaking calibration!

Keith - if you're saying that Audyssey set your subs to exactly the same distances as you found after your tweaking via OmniMic measurement, that means one of two things, neither of which make sense vs. the 'conventional wisdsom':

1) Audyssey somehow didn't calibrate distance AT ALL for your subs, and kept your old settings. That would imply a "bug". Did any of the other speaker distances change?
2) By doing the 12 position measurement heavily concentrating on the MLP, you're getting to the same place that you did when you did a single measurement at the MLP when you got the final numbers in your tweaking odyssey. That implies to get the flattest response and the best results from Audyssey while minimizing tweaking, you want multiple, but only slightly varying from the MLP, measurements. How "tight" around the MLP were your measurements?

In a way, the latter makes sense to me as you're accounting for 'measurement error' (random variation in what's "truth") in the information going to Audyssey, and getting overall flatter response for the mains & sub (I assume), but OTOH, stabilizing the delay/distance for at least your subs. Whether that's a good thing depends on where you're sitting, of course.

It would be interesting to see how different your measurements away from the MLP are vs. the final ones you did a little while back. My guess is that you'll be doing worse.

Assuming there's no glaring error in what you did, the implications are profound...if this can repeated and measured independently to ID that the results are as "good" as before.

Of course, I could be wrong

Yes. It was that which made the biggest difference.

Attached.

I am assuming that sometimes Audyssey will get it right. If sometimes it gets it wrong, it's logical to assume that sometimes it will get it right.

Thanks. I will leave it for a week or so to acclimatise and then do another calibration using the old mic pattern but with the new speaker toe-ins.

 

Audyssey Pro 30 April 12 12pos.pdf 431.564453125k . file

 

Audyssey Pro 10 May 2012 12 pos MLP only.pdf 431.2744140625k . file

Heck of a coincidence isn’t it? Yes - the other speaker distances changed (for the speakers I moved to adjust the toe-in and slightly for some of the others - mic positioning). The 'bug' is only for reloaded calibrations AFAIK - this was of course a totally new cal.


Very. All 12 positions were within a foot of each other - less probably. The maximum spread would be about 3 feet in the forward direction and 2 feet in the lateral direction.


Much worse would be my guess. I didn't measure the other seat. But I don't care about the other seat anyway.


Subjectively as good. On the OM graphs, not quite as good. I will do another calibration next week using the 'old' mic positions but with the speakers toes-in as they are now. TBH they should always have been toed-in more than they were so that is a keeper now I have done it. I suspect that when I calibrate with the old mic positions, I will get something very like the last cal and that I will need to do the sub distance tweak.

Thanks for your observations and input (and to Tandy too).

Yes but "how" does it sometimes get it right, if a pre-condition of getting it right is pinging the subs and mains together which it never does? So if Audyssey "intelligently" got to those sub distances in this calibration, which you've already independently confirmed result in the best xover in your room, then it implies a combined pinging is not a requirement to that logic--and that it has enough information to optimize with the measurements it does take.

And that Audyssey is smarter about this than what some have been giving it credit for in the forums (me included). And then it becomes a quest of why does the crossover optimization sometimes work and sometimes (badly) fail? Do we have to give it a "seed value" prior to the calibration, much like we have to supply a best guess starting point when doing an iterative IRR calc?

Very bizarre (and somewhat unfortunate), as I'd rather know the exact capabilities of the system rather than the alternative which doesn't exactly inspire confidence: the xover integration logic will sometimes work, but there are no guidelines to predict when it will and when it won't, so always assume it won't and measure/fix accordingly.

Will review the certs when I get to a PC!

Thanks!

Agreed. What I meant is that if it is "guessing" sometimes it is going to "guess" right I think you are right on the money when you say "always assume it won't and measure/fix accordingly.". That way we are more certain. Trust but verify

I have a slightly different perspective. When I run a Pro calibration, my sub distances always come up slightly different, even though I carefully use the same measurement positions. The differences are small, usually less than .5 ft. Also, I am in the habit of always re-checking the sub distance tweaks after a new calibration. When the Audyssey-calculated sub distances are slightly different, I find that the optimum distance tweak changes slightly as well.

So, the point I am trying to make is, unless you re-measured with OM after the recent calibration, you can't be absolutely sure that your sub distances are indeed still optimized. I would hope this were the case, rather than believe that Audyssey "gets it right" sometimes, and not others.

Edit: I see you ran OM after the recent calibration and the distance was still optimized. Obviously different from my experience.

Hey guys,

I just did a 12 point Audyssey Calibration and to my ears it sounds pretty damn awesome!...Can hear stuff I couldnt hear before, its made it crystal like so far!. Dialog is a tiny bit low and might increase the center channel by 1db, if that is ok?.

But have a look and tell me what you think!. I dont have stuff like Omni Mic installed to tell you what really is going on in the room. But here is the certificate anyway

http://imageshack.us/photo/my-images...rtificate.jpg/

Cheers,
Kevin

I had done a 10-position calibration after confirming that loading an AMD file did not set sub distances (or configure the number of subs). I was not happy with the results and even noticed in the after graphs that I likely had not correctly applied my target curve. So last night I reloaded a previous calibration. As expected, it did not change the distances that had been detected with the 10-point cal. But it also apparently did not change the trim settings for the subs. The previous -3.5dB settings were there and NOT the -2.5dB indicated on the after Pro graphs.

I had watched an unfamiliar movie - Ghost Protocol - and it sounded great. Over the weekend I will watch something with a lot of bass and with which I am familiar. I am reasoning that the -3.5dB is correct (as opposed to the -2.5dB that had been detected) because the sub level matching process had me raising the subs' gains a bit to meet 75dB.

So, unless the sub level match screen is used and a fresh calibration is done, loading a calibration changes neither sub distances or trims ... ?

Jeff

It is OK, but you shouldn't have to do that. What and where is your center speaker? What is the surface of floor, ceiling and walls "between" you and the center channel speaker?

Jeff

Hey mate,

I did the best drawing as I could lol!....the floor is wooden though we have a rug in the room. There are two openings in the room as you can see. Curtains on the sidewalls L & R side where the subs and couches are. I didnt draw them in tho lol. Ceiling and walls are plaster and no acoustic panels at all. Here is the diagram

http://img528.imageshack.us/img528/133/roomlayout.jpg

Thanks mate

What happens if you put lots of messy papers or a blanket on the coffee table. Is dialogue better?

Might want to run another calibration with the blanket there as well.

Yeah but I thought it's better to calibrate the room the way its gonna be all the time?...If I put the blanket over the coffee table and then remove it after the calibration?....Wouldnt that just defeat the purpose? or does it still work?

Ive been trying to get the wife to let me put a table cloth over it atleast but she keeps saying "no" lol.

If the system now sounds "pretty damn awesome", my suggestion would be to first be sure you have the correct sound processing settings with the AVR and then, if that doesn't fix it, simply turn up the center channel a dB or two.

Then sit back with a drink and enjoy some movies.

Normally, surround speakers should be at 90-120 degrees. Yours seem to be directly behind the listening area. Any reason why they aren't placed on the sides? And the trim levels for the surrounds are quite low (-11 and -12) when compared with the fronts. Is there a reason why?

What is the height of the center compared to the coffee table? If the center and coffee table are about the same height, the coffee table should not be an issue.

I always boost the center a couple of db. It's used mostly for vocals so doing so doesn't distort the sound stage significantly for most content. My 70+ year old ears & brain don't pick vocals out of the background sound all that well and the improved Audio quality from Audyssey helped that a lot, but boosting the center helps further. This hearing issue begins in the 50's for most people and I think affects men more than women.

The blanket is a good idea, but would need to be put in place when doing any critical listening (if it is needed).

The big dip at just under 3 kHz on each of your surrounds is curious. It's corrected well enough to be a non-issue (for a surround) but is still interesting. The wavelength at 2800 Hz is about 5" so it's hard to pinpoint the problem. Seems unlikely its a speaker issue but it's hard to explain with reflections. Are the surrounds facing the couch? You might try having them face outward or even backward to make the sound more diffuse. But, given where it is frequency-wise and how shallow it is and what the surrounds are used for, it's surely not worth much effort, if any.