Custom curve like dirac where you can customize it?

Sent from my SM-G900P using Tapatalk

Asere, graph does not seem to be a valid graph, so no conclusions can be made. (by me at least!)

Why not valid? It was taking from mlp with one sub placement. I wasn't asking for comparison for multiple locations just that one I took as a starter.

Sent from my SM-G900P using Tapatalk

Because that one point measuement graph does not show what our ears hear.

I'm confused with ear/rew. I thought you are suppose to use REW and place the sub were you get the fastest response and set it there.
In my case I realize I need to take more measurements at different areas but once you find that response you go with it.


Sent from my SM-G900P using Tapatalk

I remember reading about those KK subs. Yes, in my personal opinion the Audyssey graph is a bit better than the other one. But once you spread the subs apart, using the room sim as a starting point, I think that you should be able to achieve a much better native frequency response. Then, you should be able to get better results, and the comparison between the two forms of EQ might become even more interesting.

Incidentally, I do agree with Feri on the desirability of doing multiple measurements at least a few inches apart. At the moment, it's not so much that you are trying to measure the frequency response at multiple listening positions. It's more that a single point in the middle of our foreheads doesn't really give us complete information about the MLP. It's even worse if we are measuring speakers rather than subs, but even with lower frequencies, the same principle somewhat applies. Sometimes moving the mic even a few inches can completely change the measured frequency response.

What I'm doing is placing the umik mic on the couch using the little tripod that it came with. You are saying I should move the mic around a few inches too?

Sent from my SM-G900P using Tapatalk

It's a little bit of a complicated question for me. In order to position the subs, I would probably do just what you are doing and use a single point, although I would make sure that the mic is at ear level.

But then, when you get down to a comparison of which system of EQ is doing a better job, moving the mic around a few inches, and taking multiple measurements for both systems, would probably give you a fairer and more accurate picture.

For now, though, I would probably just use a single point measurement to find good general locations for the two subs.

I agree that the Audyssey curve is a little better, and it's good enough below about 95 Hz, even with the peak at about 75 Hz, which, while weird, might be heard as a little extra "punch." The curve above about 95 Hz is more erratic, though. The good news is that the greatest deviations cover narrow ranges of frequencies, so may not be noticed by ear (anomalies less than 1/3 octave wide tend to be ignored by most listeners -- the ability to ignore minor problems is important to happiness). How much smoothing did you use? 1/6th octave? 1/3 octave? The bad news is that some of the lesser deviations above 95 are larger than we would like to see. Others may give you specific advice, but I suspect that your speakers &/or MLP(s) need to be moved around some.

EDIT: I agree with Mike and Feri that using multiple positions for your REW mic would be better, at least in evaluating the final frequency response, after Audyssey, once you have found the best positions for your subs, and mains. Your mains may benefit from being moved nearer or farther away from the wall, regarding the uneven response above 95 Hz. Are you crossing over your subs at 80 Hz? I'd determine the best positions by ear, at normal ear height, from the single MLP (you could try all seats, if you want to go crazy). then confirm it by a cluster of REW mic positions in, or near, the MLP seat. Now that there is less work for Audyssey to do, run Audyssey using all of its 8 mic positions, with the final two where Feri suggested. Or, if you are going to listen alone, or with just one or two people, cluster the mics in the area of that one, or those few, seats, with most (including #1) at ear level, and the last two somewhat above &/or somewhat in front. There is a lot of alchemy to this; it's not that psychophysics wouldn't be better, it's that you won't get a complete model of your room, as heard from the seating positions, with 8 microphone positions, and the best use of information from even those 8 positions awaits the future. We are miles ahead of where we were in the 1980s with Crown Parametric equalization, or the 1970s with 1/4 octave Altec Acoustovoicette graphic EQ. When you're done, listen for about a month, and see what you think. Then use 8 REW mic positions in the same positions as your 8 Audyssey positions were in, ask REW to average the curves, and compare your subjective impressions to the graph, both of which I hope you will share with us.

In my case, two mic positions that are in front of the MLP (by a bit) acknowledge a disturbing reality -- the sound is clearly different if I lean forward. I sit about 16 feet away. Last time I ran Audyssey I took the leaning forward thing into account in positioning two of the mics. It was my best run, and I don't care to do another, unless there is an equipment change. The sound is good for all seats in our HT, is good closer, is good farther away. That wasn't true before Audyssey, with all other equipment the same. Although good in all those locations, it is different in all, as expected.

here are the results with 2 last measurements against the wall (before) and with 2 last measurements slightly before sofa to the left and to the right (after). didn't make a huge difference, but still, it looks like it was less aggressive with the surrounds this time.

Here's word document on the new audyssey smartphone app conversation with Chris K in his facebook tech forum June 16-22, 2016.
This will work with Denon 2016 models Oct-2016 timeframe.

to summarize the Facebook discussion .....in one response by Chris K:

Attached Files

File Type: doc

2016 New Audyssey smartphone App discussion with Chris K June 16-21-2016.doc (202.5 KB, 53 views)

Hi,

You may already know this, but those graphs that you are looking at don't actually show the specific corrections that Audyssey made. They are just a crude approximation of the corrections which Audyssey intended to make. They are crude approximations because Audyssey implements thousands of filters in an effort to smooth frequency response, and the graphs only show a few broad data points.

The key question is whether things sound better after your last calibration? Sometimes some experimentation is required to find the specific microphone pattern that will yield the best results. And, once you find a pattern that satisfies you, it is important to document it so that you will be able to repeat it, if you ever need to run Audyssey again.

Regards,
Mike

yep, i realize that audyssey is much more than an equalizer.

regarding how things actually sound - i didn't actually have any complaints about the sound even after the previous calibration run that was done with 2 measurements right againts the wall. but you probably know that audiophiliac paranoia that nags you with questions "is this really sounding right? can it sound better than it is now with a more proper calibration?" and such but it feels like it became slightly bassier this time, i don't know if it's the "right" sound but i'm okay with it. no idea why it sets my center speaker as large though (each time i ran a calibration, not just this last time), it has 2 small 4-inch woofers and according to the specs a frequency response of 60hz to 40khz. but i guess it knows better, right? (i got no sub atm)

"Audiophiliac Paranoia"! I like it! And yes, I know all about that. The good news is that, in my case at least, I actually did achieve better calibrations with practice.

If you don't have a sub in your system, your AVR will have no choice but to set your front speakers as Large. But you may be able to reset your center channel to Small, with a crossover of say, 60Hz or 80Hz. That would redirect bass below the crossover to your front speakers. If you are able to do so, it won't affect Audyssey's operation in any way.

I posted elsewhere but felt this might be a more appropriate place, apologies for the double post.

I had the bass in my relatively small room nearly perfect using MultiEQ XT32 and following guidance here. I re-ran Audyssey (long reason and not relative) under identical conditions and "poof" bass impact was gone. Nothing I can figure changed:

Optimized placement of both subs for best response; adjusted levels of each sub to 75-80 db with supplied Audyssey microphone using sub gain dials to get post calibration gain in a good post cal range (-4 to -5db); ran all eight positions using boom mic stand (including tape marking to ensure same placement). I have four seats along back wall with Mic positions focused on MLP (about 12" away from back wall along the same plane around MLP for positions 1-5 and just in front of the MLP for positions 6-8). Mic height is above seat back, just above ear level. After calibration I do raise sub trims equally by 4-5 db (as that is my preference) and raise crossovers on all five speakers to 80-90hz (5.2 system); LFE is set at 80hz. No other changes post cal and no Dynamic EQ or Volume; using Audyssey movie curve. No other room changes to furniture wall hangings, window blind position, or open doors.

Subs:
-SVS 25-31 PC+ (front left corner)
-Definitive Technology Supercube Reference (back right corner on same plane with my four seating positions)
-Placed in front/back opposing corners, internal crossovers bypassed and phase at 0 degrees

Processor:
-Integra DHC-80.3 with latest firmware

I've reran multiple times and cannot get it back. Raising the sub trims by 7-9 db (an additional 4-5 dB over previous successful calibrations) does increase bass but results in boomy excessive bass in some program material without the punch or mass air movement I had before. Even reset my processor to factory settings - no luck in getting it back. I'm completely stumped.

Any ideas?

Hi,

That's puzzling! Assuming that neither of the subs was moved at all, I can only think of two possibilities. Either something has changed with your microphone (based on having done microprocessor resets to test your AVR) or something has changed with at least one of the subs. (You could try a couple more AVR resets, just in case.) You might try testing the subs individually to try to determine whether both are performing properly, and you can buy a replacement microphone. Beyond that, I am stumped at the moment. Were your post-calibration results different lately from what they used to be? Were the trim levels, distances, and crossovers approximately the same. If your microphone were going bad, I would expect some calibration result to be different. If you had moved mic position 1 by even a few inches, I suppose you could get such a different result, but you say you didn't?

Here's one tip which may be tangential to your fundamental problem. But it's worth exploring. You would be better off doing your next calibration so that you get a trim level of at least -8 or -9 (raise the gain on your subs). That will give you plenty of room to adjust trim levels upward while staying in negative trim levels. It is barely possible that some of that boominess you heard as you raised your trim levels into positive numbers was the subs distorting. Try checking the subs individually, and then re-calibrating with lower trim levels and see what happens. I hope that works.

Regards,
Mike

Thanks Mike, was considering redoing with slightly higher sub gains. Subs were moved to test some isolation feet but placed back at nearly identical locations (within a inch or two). It is also possible I had my mic height off by a couple of inches (although I've adjusted and reran since). Hard to believe such a drastic difference with no apparent changes - multiple successful runs previous. I may try borrowing the Audyssey mic that came with my sons Marantz. Appreciate the ideas....

You are very welcome! It does seem strange, that such small changes can make a difference, but sometimes moving a sub even a few inches can excite room modes in different ways. It may be that you are getting some cancellation now at a frequency where you were experiencing really good mid-bass before. That is where you would probably notice it most--that chest punch range between about 50 and 100Hz.

So, you may need to experiment with moving the subs just a little. It's a lot of trouble, but you might want to consider doing a sub crawl if you can't get the subs back to their original positions. Based on what you said about moving the subs slightly, I am optimistic that you can get back to where you were before. I have also found that mic height can affect calibrations, although with me, it is more the upper mid-range and treble where I notice the differences, if I get the mic a little too high.

IMO, it's all about small changes making a difference. I found that the bass roar of the rapids in How the West was Won sounds very different after moving over by one narrow seat width in my room.

I considered deleting this post, or at least saving it for next April Fools Day. Am I all wrong about the following? Even though the sub will stay on the floor in almost all cases (some exceptions noted), the ear won't usually be on the floor when watching a movie, or listening to music (some exceptions noticed). So even if the sub is on the floor, it will be heard from the ear height of the MLP, where there might be a dip caused by the location of the ears relative to ceiling height. I assume there is little that can be done about such a dip, other than stacking the subs, or elevating them (I've seen quite a few pictures of these, lately), but we might as well compare room locations with each transducer (ear and sub) at the height they will be, unless it is decided to change their height. Height of the ears might, or might not, matter in reinforcing or cancelling "punch" or "boom", depending on how high the ceiling is. What is the wavelength of 80 Hz, where many of us crossover to the sub? Something like 13 feet? Of course, the sub will contribute something an octave or so higher -- at 150/160 Hz, where average ceiling height of 7 or 8 feet may closely match the wavelength. I've often wondered if a "sub scoot," or a "sub roll" -- gliding around on a desk chair with rollers, with ear height the same as ear height at the MLP might be better than a sub crawl. One could also periodically stand up and sit back down, as JFK did to make Gore Vidal do likewise. It would be good exercise, and comic, to say the least.

Hi Gary,

To me at least, what you are saying makes perfect sense. I don't honestly know to what extent height is a significant factor in subwoofer frequency response. And it's something I have wondered about too. In theory, low (long wave) frequencies can be anywhere in the room. In practice, I have always envisioned them as being more prevalent lower in a room. I do think that with respect to many long waves, if they come out fairly low in a room, they may tend to stay lower, where shorter waves would ricochet more vigorously.

As I have read sub location recommendations from experts such as Toole, Welti, Harman and others, the only one I remember specifically suggesting varying subwoofer height, for improved frequency response, is Earl Geddes. And even he only recommended varying the height of one of three subs by a few inches, as I recall. So, there is uniformity in recommending multiple subwoofers (3 by Geddes; 4 by the others) but no emphasis on subwoofer height as an important factor in improving frequency response.

But with respect to the relationship between sub height and ear level, there is clearly a connection. As you know, the sub crawl is ideally something like the way you described it, only in reverse. Instead of putting the sub at floor level and scooting around the floor on a roller chair (which actually sounds like fun), the idea is to try to get the sub close to ear height at the MLP, and then crawl around the floor at sub height. So, the same relationship between ear height and sub height should theoretically be preserved in either case, although it's always seemed to me that getting the sub to be right at ear height at the MLP could be problematical. If, as you speculate, inches matter, that might make a difference. And your method might be better, if one had access to a chair with wheels (it could be done without wheels, but scooting would be more fun).

I don't understand the subject well enough to have a clear explanation, or a strong independent viewpoint, but from all my reading, horizontal positioning of subs seems more significant than vertical positioning. So, moving a sub a few inches left or right, forward or backward, seems to have a greater impact on frequency response than raising or lowering the sub (or MLP) by a few inches. You may have confirmed that somewhat with your example of changing seats in "How the West Was Won". I wonder if that effect would have been as pronounced a little closer to floor level, or sitting up higher in your chair?

I can't really take the discussion any further, as I can't offer a good explanation for why the horizontal dimension would be more significant than the vertical dimension, but that is certainly my impression. And people are able to get awfully good frequency responses and sound quality with their subs on the floor. If I had to speculate, though, I would guess that as frequencies get shorter, and particularly above about 100Hz, the desirability of having drivers located higher rather than 12" or so off the floor would increase. It's certainly true with tweeters and mid-range drivers, but I suspect it is also true for bass drivers above about 100Hz, or so. But that's just speculation. Interesting subject! I'm glad you didn't delete your post.

Regards,
Mike

After thinking it over, it looks to me like neither ear height nor sub height would be an issue in a sub crawl. With a typical 8 foot ceiling, to get a null within the frequency range of a sub, with the crossover points we usually use (e.g., 80 Hz) either the sub or the ear would have to be about 4 feet off the floor. I think.

Ummm, that's too simplistic. There are axial, tangential and oblique modes. And there are a slew of online calculators.

Jeff, I have yet to see a room calc that requires the input of subwoofer height and ear hight measured from floor. Is there any?

lol, no the calculators identify where in the room modes occur by frequency. Room modes are independent of where the sound sources are located, but one would need to know where one's head it.

Jeff

I was considering axial only, between the floor and an 8 foot ceiling, only, since I was considering ear and sub height, only, that might be a factor during a sub crawl. I was using one of the online calculators, the multicolored one I think is Harmon's. I assumed the tangential and oblique would not be as much of a factor. Maybe that was wrong.

Axial modes are primary, and the strongest.. And then there are 1st, 2nd and 3rd order axial modes. For the floor to ceiling axis the 1st order mode is at half the distance, the 2nd order at 1/3 and 2/3 the distance and 3rd order at 1/4, 1/2 and 3/4 the distance.

Pro theater designers aim for "ear placement" at fifths, i.e. 20%, 40%, 60% and 80%. And they do focus only on axial. If one is starting from scratch with a large enough space, this is a piece of cake. But with most residential spaces ... repurposed rooms ... seating positions can be more affected by viewing distance, critical distance consideration, spacing for walkways and, in multipurpose spaces, other "non-A/V" furniture.

For my dedicated 13 X 21 space, my seating positions were determined by everything I mentioned EXCEPT modes. (I did avoid the 50% room length one.) But then I was also able to bass trap the snot out of it.

Sub placement will only determine how much coupling there is with the naturally occurring modes, but those modes will always be at the same locations regardless of placement.

Jeff

Jeff,

That was a good post on what is, for me at least, a very confusing subject. The room mode calculators can help us to predict where the modes will be, assuming rooms with regular geometry (which I don't have) but predicting how particular subs, placed in specific locations, will collectively excite room modes is much harder. From my reading, it appears that even the real experts such as Welti, Seaton, and Geddes, can spend several hours (sometimes a full day) situating and EQing subs in a room. But in a way, that lengthy and meticulous approach is dictated by their own perfectionism.

Fortunately, the good news is that, unless we are measuring with REW, and chasing the measurements and not just the SQ, most of us find that our room/sub placements can be fairly forgiving, and can result in very good sound. As with you, it took a combination of sub placement and extensive bass trapping for me to achieve the sound I wanted. But the irregular geometry of my room probably presented additional challenges.

Regards,
Mike

I am glad to have contributed! Slightly extending my post is that the 1/2's, 1/3's and 1/4's apply to all three axes, not just front-to-back. And the room modes are ALWAYS in the same physical locations; it is the frequencies that change with room dimensions. The mode at half the length of a 15 foot long room will be 2x (in frequency) of that same 1/2 room length axis in a 30 foot room. The biggest takeaway that someone pondering on this for a while should have is that they should never use any room dimension that is a multiple of another or, horrors of horrors, IDENTICAL. The modes will stack and produce black holes that not even NCC-1701Z could get out of.

I also have a slightly irregular room in that the left wall, the rear wall and the ceiling are all skewed by a degree. I did this on purpose when I designed the house. (I have a very funny story about the drywallers.) The purpose was to eliminate slap echo. The good news is that I have none of that while the bad news is that all the damn mode calculators are not accurate for my space. Nor can I afford software that would model it. So, really, for me it is all about the acoustical treatments.

Jeff

Okay, so... Audyssey made a quite a boost in midrange regions of my three front speakers...mostly center channel which by the looks of EQ Graph has a 10 dB boost somewhere in a wide 300 Hz area. Coincidentally, my midrange driver started buzzing in center and is waiting for replacement (no visual damage though). I don't know if main cause of damage was that boost, but I am wondering what can I do about it? I know I can bypass audyssey on left and right speaker...but if I turn it off completely then I will have no bass management.
Any workarounds I don't know off other then switching to something more flexible like Dirac?

Hi,

I don't recall if you have posted pictures of your system and a description of your speakers, but I don't know that I would try to read too much into the crude graphs. As has been stated many times on the thread, it is hard to draw meaningful conclusions from graphs that attempt to depict at such a broad level the corrections that Audyssey is trying to make. The actual corrections occur at thousands of locations across the frequency spectrum.

If your CC is buzzing, it is probably damaged alright, but I don't know that I would conclude that Audyssey caused the damage. If the CC were damaged, that might explain why Audyssey was trying to boost it. It would be very rare for Audyssey to damage a speaker by boosting it. It is simply trying to get all of the speakers to play at the same volume at the MLP, and trying to get the full frequency band within about +/- 3db. An already damaged driver might not be able to play at the MV you are using, and the buzzing might then become more audible. But the cause would likely be the defective driver to start with. The boost just makes the defect more noticeable.

I don't know of any way to protect a defective CC from playing with Audyssey, other than by removing it from the system, either physically, or virtually in the Speaker Configuration Menu, and running a phantom center. To be thorough with the options, I guess you could also just turn down the trim level on the CC to make it inaudible, with the same effect of creating a phantom center. I would do one of those things, in any case, until I could replace or repair the CC, regardless of the EQ system.

If you will post pictures, descriptions of your speakers, crossovers and trim levels, distance to MLP, and MV levels, you may be able to get some helpful trouble-shooting advice. I'm sorry to hear about your damaged CC.

Regards,
Mike