Audyssey App results

[TB Rich]

I've just finished doing a second play through now I have XT32 and the Audyssey App - I've got 3 different curves, interested to know which one most people may choose? - Is this broadly how other people are setting up their systems too?

Red line is the as-is XT32 output, so mains were set as large/full range, sub level reported at -10.5dB, reference curve, etc (although I did turn off midrange compensation and all the dynamic eq/volume malarky).

Purple line was using the same XT32 results, but changed from reference to flat, plus I set the mains to small. Sub was increased from -10.5dB to -6.5dB.
I experimented with different crossover points, noting that 80hz created a massive hole in the response, whereas 90hz and 100hz fixed that. - The sweep above is at 100hz. (I'm hoping it won't be noticeable/directional this high).

Green line, this was copied from the Purple with the adjusted settings - and then started to mess with the Audyssey app curves. Added in a slight lift at 30hz and drop at 65hz - to create more of a slope. My reasoning being that as bass gets lower you need progressively more volume to hear it. Lastly I also pulled the Audyssey curtain back to 500hz for all speakers.

Cheers!

 

[Deleted member 901590]

What are your settings for Dynamic EQ (relevant to people's answers)?

 

[TB Rich]

What are your settings for Dynamic EQ (relevant to people's answers)?

All those widgets like dynamic EQ, volume and LRC - all off.

 

[Deleted member 901590]

I've made my choice.

As you're asking for opinions, the boosts in the green and purple go far too high up the FR for me. It would sound too muddy and ruined, especially music in soundtracks.

I would have put the other fixes into the red for crossovers and changed it to FLAT, then put a peak around 4dB at 20Hz sloping down to no boost at 40Hz.

But that's what I prefer. We'll call it the yellow line.

 

[TB Rich]

The Red, although I've not listened to it, I have had very similar curves when using my miniDSP, always sounds completely anaemic. I listen around -20dB MV, so I've always needed a good slug of bass on top to make it sound good.

(The purple didn't have any specific curve boosts - just took the -10.5dB sub level up to -6.5dB).

-I suppose, this is where the dynamic EQ (or volume?) would help out?

 

[TB Rich]

then put a peak around 4dB at 20Hz sloping down to no boost at 40Hz.

I can’t seem to actually do that?

Here‘s the default curve on the sub:

And if I put a 4dB boost at 20hz it lists the whole range:

Interestingly/annoyingly, the first run through of Audyssey gave me a flat curve on the sub:

Which does allow me to do as you say, but I can’t seem to get the second run through to behave that way?

 

[roscopervis]

What sounds the best - blind test?

 

[TB Rich]

What sounds the best - blind test?

You can't blind test an Audyssey upload! But sure I've now tested a few scenes from BR2049 - the Green sounds the best.

Purple has a hint of bloom - wouldn't say muddiness, the KEF's and in-room response I have I don't think would ever be even close to sounding muddy!
Green just had that little extra low end depth and cleaner elsewhere.
Red, lack lustre - I think people don't appreciate the fact I listen at -20-25dB MV region (-22dB felt more than loud enough earlier!). If I was up at -10dB then no doubt Red, or a much closer variation of would be ok.

I think I need to get the XO down from 100hz though, it's not that I can fully localise it, but there's just a perception of some volume/weighting being to the side the sub is on.
Setting to 80hz creates a massive null, however I've read now that could be sorted with adjusting the distance (i.e delay) setting on the sub.
I'm not entirely sure how to use REW to work it out though?

I've found a big thread on ASR (marmite site granted) with peoples FR sweeps, tbh many people are running way hotter bass than me! So I'm happy Green is totally within the normal realms, it certainly doesn't sound exaggerated in anyway (especially in context of my listening volume). This is also HT only, not Music use doesn't touch any of this.

 

[WozzaB]

I can’t seem to actually do that?

Here‘s the default curve on the sub:
View attachment 1671568

And if I put a 4dB boost at 20hz it lists the whole range:

View attachment 1671569


Interestingly/annoyingly, the first run through of Audyssey gave me a flat curve on the sub:
View attachment 1671578


Which does allow me to do as you say, but I can’t seem to get the second run through to behave that way?

Where you've added the 20hz +4db you should be able to bring it back down further up the FR, have you tried touching 50 or 60hz and dragging I down a touch? I can be fiddly on a tablet

 

[WozzaB]

I can’t seem to actually do that?

Here‘s the default curve on the sub:
View attachment 1671568

And if I put a 4dB boost at 20hz it lists the whole range:

View attachment 1671569


Interestingly/annoyingly, the first run through of Audyssey gave me a flat curve on the sub:
View attachment 1671578


Which does allow me to do as you say, but I can’t seem to get the second run through to behave that way?

Just had a quick play with the app, you should be able to boost 20hz without lifting everything.

 

[TB Rich]

Where you've added the 20hz +4db you should be able to bring it back down further up the FR, have you tried touching 50 or 60hz and dragging I down a touch? I can be fiddly on a tablet

Yeah that previous screenshot was just to show what was happening, the actual one that supports the Green line, I did as you you said:

So if I wanted to bring the bass down a bit I could drop each those a touch, and the overall bass from say -6.5dB to -7.5dB. But tbh I actually now I’ve listened to it haven’t any complaints. Will watch a whole film in a bit and see!

It’s just that the first time I ran through the XT32 setup, it gave me a straightline on the sub curve and I was able to put it to:

However, not that it actually made any difference, I’ve just been going over pretty much all my sweeps and I’m always losing about 10dB from 30hz to 20hz, which echoes the RAW un-EQ’d sub response.

I think this is all boiling down to needing a ported sub tuned to a lower frequncey. I’ve turned my nose up at ported subs, but it’s dawned on me my main speakers which I adore…are ported!!

 

[TB Rich]

Just had a quick play with the app, you should be able to boost 20hz without lifting everything.

View attachment 1671735

Yep - not possible with the current XT32 curve it’s given me on the flat one (as per my post a moment ago) - I could do this.

 

[WozzaB]

Yep - not possible with the current XT32 curve it’s given me on the flat one (as per my post a moment ago) - I could do this.

Ah OK, I thought it wasn't letting you do it all!

 

[TB Rich]

Ah OK, I thought it wasn't letting you do it all!

I guess I would need to run Audyssey again, and hope that it would be a flat curve rather than the sloping drop off towards 20hz it’s given me this most recent time, but there’s no guarantee it will be.
It actually made next to no difference in practice when I was messing about with the flat curve, as I tried boosting 20hz by 9dB just to see how it’d respond - but it didn’t even budge tbh. It’s either room response/BK not man enough/or just some weirdness I am yet to explain!

 

[WozzaB]

I guess I would need to run Audyssey again, and hope that it would be a flat curve rather than the sloping drop off towards 20hz it’s given me this most recent time, but there’s no guarantee it will be.
It actually made next to no difference in practice when I was messing about with the flat curve, as I tried boosting 20hz by 9dB just to see how it’d respond - but it didn’t even budge tbh. It’s either room response/BK not man enough/or just some weirdness I am yet to explain!

I'm not sure how accurate this is either?
I had dual XXLS400's previously and really liked them, I only changed them as I wanted a bit more output for the really low stuff, so maybe your sub does tail off where the graph suggests?

 

[TB Rich]

I'm not sure how accurate this is either?
I had dual XXLS400's previously and really liked them, I only changed them as I wanted a bit more output for the really low stuff, so maybe your sub does tail off where the graph suggests?

Same, just want some more low stuff, not using it for music now as the mains cover to just below 30hz on their own.

I expect I’d have to get the miniDSP cabled back up to push the BK into providing something approaching usable 20hz, however part of buying this AVR was to declutter/simplify everything! I think I’m happy with the result anyway, probably not worth much more messing around until I get a bigger sub I feel.

 

[roscopervis]

Where is the sub situated in the room? You may have said, but I haven’t seen. Also, looking at your various traces in this and the other thread, I think you’re getting some good SBIR, and I’m thinking it’s your mains.

 

[TB Rich]

It's in the front left corner which is the best place in the room having measured it all over. I dragged it around yesterday yet again (before doing compression testing with it) and found that moving it even several feet away from the corner, made zero difference to the curve beyond very minor deviations to the largest peak and null in the response. The other corner (front right) produces essentially the same curve too, so the null at 40hz can not be addressed by a 2nd sub

The only likely good place I have not tested the sub is the opposite corner (right rear) behind the sofa. The issue with that location is of course getting a signal to it. There's wireless but that will create a delay and add additional complication to timing them. Fixable I'm sure, but as I do not currently know how to use REW to even fix the single sub problem I have when the crossover is set to 80hz..., multiple subs are not happening!

I watched Drive last night as I had not seen it in a while, not a mega bass heavy film it seems - but awesome sound track - and tbh everything sounded very well balanced. However I would have liked more bass in truth - for me I want clean/accurate in my 2ch setup and that's what I have, but for movies... well I want an exaggerated impact, it's meant to be a spectacle after all!!
But it might well be the film compared to say a big sci-fi bass heavy film, so I'll watch a few more before redoing any curves!

Also note that none of the sweeps and frequency responses on this forum, other than the one at the top of this post (using Audyssey and the sub), are relevant or representative anymore.
I very recently moved the speakers (front baffle distance) 1m out from the wall, and so all previous sweeps prior do not show this, and I haven't done any dedicated 2ch sweeps with no sub in this new position.
SBIR did used to plague me around 100-120hz, I don't think it's as much of a problem now, but with how SBIR works it'll have moved to a different frequency, and at a slightly reduced prominence given the speakers are further out from the boundary. (Try find me a room (5.5x3.3m) that doesn't exhibit SBIR though!).
Overall the speaker move has been transformational though, it's fixed the slight right side volume bias that was present back when at the seating position, and it's been probably the single biggest upgrade - that said I did it almost immediately after adding the new pre-amp! But regardless listening to 2ch now is even more phenomenal with the detail and imaging. This was also clearly noticeable in last nights movie watching, some of it could be attributed to upgrading to the 6012 (but like the pre-amp change that's probably minor in the scheme of things), the speaker move has been the ticket here I think.

As said I think some more watching of different films is needed before drawing to any conclusions or settling on a Audyssey curve.

 

[Deleted member 901590]

It actually made next to no difference in practice when I was messing about with the flat curve, as I tried boosting 20hz by 9dB just to see how it’d respond - but it didn’t even budge tbh. It’s either room response/BK not man enough/or just some weirdness I am yet to explain!

I believe that there is a maximum EQ plus and minus that Audyssey can apply. So, when you put a manual boost in, sometimes it can't do it if it already using the EQ scope that it needs. People have proven this on AVF by doing REW measurements after trying different manual curve changes.

In your case though, it wouldn't surprise me if it didn't allow 9dB! However, I'd be surprised if you could do less than this. So, maybe there isn't something right here.

On the "how to do it". Sorry, AVF didn't ping me your question, weirdly. So, came back myself today to see your thread as it was interesting and oh lots of comments etc. Anyway, yeah, you need to put the boost in then put another point to bring it down again. To do this on a phone screen TOOK ME AGES!

So well done. Imagine if you were the giant from The Princess Bride trying to do it.

 

[roscopervis]

It’s a nice journey. The way I would look at checking the delay (distance) and crossovers in REW is by changing one thing at a time and taking a measurement. So set the crossover to 80hz and measure with the current settings and get the null (I assume). Then adjust the distance setting for the sub, probably start increasing the distance by a metre and see how it measures. You can go more granular if you like, in my room, there is a 30cm wave where it comes in and out of being good and there are really good measurements at 4.43m, 6.19m and 7.59m. I’ve played with this a lot!

In my room, the 60hz crossover works best, but that is with my Acoustic Energy Aegis Evo 3’s and BK Monolith. You can try various crossovers, though I wouldn’t go lower than 30hz (ideally 40hz) above the lowest level your main speakers can play.

The 40hz null can be targeted with a second sub, you just have to put it in a place where it would peak at 40hz - it might not be great elsewhere in the frequency range but if it has a peak at 40hz, it will sum the frequency range and Audessy will flatten the response. You can use the REW room sim to figure out where that location might be.

 

[TB Rich]

In your case though, it wouldn't surprise me if it didn't allow 9dB! However, I'd be surprised if you could do less than this. So, maybe there isn't something right here.

It allows you to do it, but the response as measured with REW after it made zero difference in terms of actually giving 20hz an uplift - it drops off from 30hz by 10dB still - so either the room is sucking 20hz or the sub is not man enough. No doubt both of those things are true.
The only way I ever got it to do vaguely what I wanted was using a narrow Q on the miniDSP to pull down the 30hz peak, suppose I can put it back in the mix to flatten the response and run Audyssey after.

Then adjust the distance setting for the sub, probably start increasing the distance by a metre and see how it measures.

Haha yeah that’s the exact process I was hoping to avoid! I believe its possible to use REW to sweep the L and R mains, and the sub - and use another speaker as a timing reference; which I think allows you to work out the m/s delay required - and then that can be converted to a Audyssey distance measurement based on speed of sound… but thus far I’ve not found an idiots guide!
The irony is I could have done the manual process 10 times over by now, but still, I’m already sick of running sweeps so perhaps on the weekend!

The 40hz null can be targeted with a second sub, you just have to put it in a place where it would peak at 40hz - it might not be great elsewhere in the frequency range but if it has a peak at 40hz, it will sum the frequency range and Audessy will flatten the response. You can use the REW room sim to figure out where that location might be.

This is the crux of the problem - I’ve measured extensivley now the 2 front corners of the room and they both have the same null. The left corner it is in now was better than the right corner at every point.
I’ve not measured the very middle of the front wall, but that’s where the kit is, so going to take something fairly drastic for that to ever be feasible!
REW’s room sim doesn’t work fantastically in my room, it’s ok, but I’ve enclosed stairs across much of the front wall in the lounge and I think this is what is throwing things out. Typical mixed purpose room problems I guess, but have to deal with what we have of course :|

 

[pmar4015]

Not sure where to post my query (in terms of the specific forum) therefore apologies in advance if this is the wrong forum

Quick question, I am using MULTEQ-X to EQ three subwoofers for use on my Denon 8500HA.

I am using the following three subwoofers

Paradigm DSP-3100 (1 of these)
SVS PB-1000 (2 of these)

Connections:
The two SVS PB1000 subs are connected via a ‘Y-Connector” to the 8500HA whereas the Paradigm DSP-3100 is connected directly (i.e. by itself) to the AVR8500HA (thus both of the LFE independent outputs located on the back of the AVR are being utilized)


Physical Distances of Subs:

[*]One SVS PB-1000 is roughly 4 feet from the MLP
[*]One SVS PB-1000 is roughly]9 feet from the MLP
[*]The Paradigm DSP-3100 is roughly 6 feet from the MLP


ALL SUBS ARE CONNECTED to the AVR via the single LFE input located on the back of the subs to the the two single LFE outputs located on the back of the AVR

I realize Audyssey recommends that any two subs of identical make/model be placed equidistant from the MLP but I seem to achieve the smoothest response using the configuration noted above


In any event, at the end of the calibration everything appears fine; however, I am receiving the error message that the subs are “out of phase” (see pic below (red circle with the strikethrough))

Is this anything I should be concerned about? Should I play the phase settings dial located on the back of each of the respective subs?

Thanks in advance for the help!

 

[TB Rich]

Another monologue but I've had another bash at getting a response I was happy with, including trying the miniDSP back in, and happened by chance that inverting the sub signal on the miniDSP filled in the hole I was getting when the crossover was set to 80hz.

I assumed therefore it would be worthwhile to do the whole Audyssey process again with the sub signal inverted - however the end result was once again a null in the crossover region, this I thought was very odd and unexpected?
I then set the signal back to normal and tried manually messing with the distance setting of the sub iteratively from about 3m up to 12m. However the response barley changed regardless of distance, and certainly never came close to filling the crossover null. - Again I thought this was odd, I would have expected to have been able to find a distance that would have filled the XO hole?

So in the end I went back to my 2nd Audyssey run, removing the miniDSP and setting the BK phase knob at 180deg (note it needs to be out of LFE mode for the phase knob to work!). Doing this was producing an identical result to the miniDSP inverted signal, and so basically a long process followed to simply tweak many (many!) times the Audyssey curves.
There's no doubt the miniDSP is a better system for tweaking, the Audyssey app curve editor is truly odd in how the end result actually measures in REW vs what you set in the curve screen - a lot of the time you can seemingly make a change on the Audyssey curve but yet it will have zero measured effect in REW. But by eventual fluke I eventually got what I was after:

Essentially a mix of the Purple and Green responses from before, but flatter, with a lower XO, and more balanced to the overall response (10dB rise from 10khz to 30hz - adjustable of course with the sub volume which is now at -7.0dB so lots of scope to raise or lower to taste; noting my typical -20dB MV warrants a few dB hotter than reference would require).

I've accepted defeat though I can't get 20hz anywhere near the sort of levels I 'want'.
Unless I put a massive amount of low end boost at 20hz with the miniDSP, but of course distortion starts to massively rise doing this so isn't feasible to actually leave in place.