Audyssey Room EQ Review

[Chromatischism]

That is a good result. It's worth it to put in the time to get that straightened out.

 

[Sal1950]

I recently replaced the plate amps in both my SVS S2000 subs and needed to run XT32 again to re-calibrate.

Curious as to why you replaced the plate amps, did one or both fail? How old are your subs.
I have 2 SB2000's here and wondering about possible issues?

 

[Steve Dallas]

Curious as to why you replaced the plate amps, did one or both fail? How old are your subs.
I have 2 SB2000's here and wondering about possible issues?

Both amps developed a hiss that I could hear from 9' away. SVS replaced both under warranty in very trouble-free fashion. The replacement amps are quiet.

 

[Sal1950]

Both amps developed a hiss that I could hear from 9' away. SVS replaced both under warranty in very trouble-free fashion. The replacement amps are quiet.

Appox how old were they?

 

[Steve Dallas]

Appox how old were they?

Around 2 years. They always had some hiss, and it became progressively worse over time.

 

[Sal1950]

Around 2 years. They always had some hiss, and it became progressively worse over time.

OK, thanks for the info.
I just got mine back in Jan 2021 and don't notice any noise at all.
I'll keep my fingers crossed.

 

[GalZohar]

The most important manual tweaks I've done (from most to least important, at least to my understanding):

1. Set crossover to 80Hz. As far as I know and as was mentioned earlier, Audyssey measures speakers performance with a specific signal, and apparently in real usage the speakers usually won't be able to play as well as Audyssey measures. Also, Audyssey only checks what your speakers are capable off, and not if your subwoofer might be even more capable for said frequency range. That's why Audyssey crossovers are a minimum value and not the optimal value.

2. Tweak subwoofer distance for most flat response in REW (separately for left and right speakers and also for L+R signal, ended up with pretty much the same distance of +2 meters compared to what Audyssey set). This was a noticeable difference even a beginner like me could hear.

*Trick I used for subwoofer distance by ear (before getting a mic for REW) was to play a 80Hz test tone and change the subwoofer distance to get the minimum volume, and then add 1/2 of a 80Hz wavelength (~2.15m) back to get the maximum volume. Finding a minimum by ear is much easier than finding the maximum, not to mention hearing the entire frequency response, and it seems like the result isn't all too different than what you would get with using REW to find the best distance. I bet this is much better than just listening tests with various distance settings.

3. Limit eq frequency in the app. While I only have XT and not XT32 which has limited EQ, bass (drums etc) sounded better with Audyssey enabled, and some videos with speech sounded worse. Limiting the eq frequency as recommended in many places gave the best of both worlds.

While many recommend making some changes to the target curve (or just boosting the subwoofer level), I found that for me Dynamic EQ performs well for restoring the lack of bass you will probably notice when listening at reasonable volumes (normally -20 to -30) with Dynamic EQ disabled. I tried adding a few db to the subwoofer but it overall sounded exaggerated when Dynamic EQ was also enabled, although I have no real way of knowing how it really is supposed to sound.

I'm still trying to find if there are more tweaks I'm missing, but those are the ones I found that made a difference that even a beginner like me can hear.

 

[srrxr71]

Was ready to pull the trigger on the Denon x3700h. Then I learned that no matter how migh in the range you go the Multeq XT32 handles only 2 subs. Audyessy offers a quad sub product which no AVR manufacturer has taken up to date.

So back to the drawing board I guess. Would have paid the premium for ease of use and future multi channel upgradability. Maybe wait 2 years before someone releases such a product?

 

[GalZohar]

Was ready to pull the trigger on the Denon x3700h. Then I learned that no matter how migh in the range you go the Multeq XT32 handles only 2 subs. Audyessy offers a quad sub product which no AVR manufacturer has taken up to date.

So back to the drawing board I guess. Would have paid the premium for ease of use and future multi channel upgradability. Maybe wait 2 years before someone releases such a product?

I thought the best way to handle multiple subs (at least without super expensive solutions) is the mindsp 2x4 hd anyway, which is external and independent of your AVR anyway?
Seems like the way Audyssey time-aligns sub is flawed and manual alignment is required (although it would be nice if someone could confirm to for the alignment of 2 subs, as I only checked the speaker-sub alignment myself), which as far as I'm aware is not possible to apply pre-calibration with Audyssey. When there's only a single sub then manually adjusting distance for integration with main speakers is possible, but it doesn't seem like you can manually align the subs to each other even with XT32 (which again should be done pre-calibration, so EQ is applied to the summed result of the time/phase-aligned subs).

 

[srrxr71]

I thought the best way to handle multiple subs (at least without super expensive solutions) is the mindsp 2x4 hd anyway, which is external and independent of your AVR anyway?
Seems like the way Audyssey time-aligns sub is flawed and manual alignment is required (although it would be nice if someone could confirm to for the alignment of 2 subs, as I only checked the speaker-sub alignment myself), which as far as I'm aware is not possible to apply pre-calibration with Audyssey. When there's only a single sub then manually adjusting distance for integration with main speakers is possible, but it doesn't seem like you can manually align the subs to each other even with XT32 (which again should be done pre-calibration, so EQ is applied to the summed result of the time/phase-aligned subs).

Wow that’s a real waste of a technology.

Yes the all things I read and even the 2 hour MSO video I watched used a mini dsp 2x4HD.

Once the 2x4HD handles the multi sub optimization I guess I could treat that whole cluster as a single sub.

Now how do I integrate that cluster with my Genelec Mains? GLM could high pass the mains I guess (manually) and the mini dsp would low pass the subs. But I would still have to manually integrate them right?

It seems I would need 6 channels of Mini DSP to make this work.

 

[AudioJester]

How many subs are you planning to use?
I have 3 and use an RCA Y splitter for 2 which my AVR treats as 1 sub. Works pretty well.

 

[srrxr71]

How many subs are you planning to use?
I have 3 and use an RCA Y splitter for 2 which my AVR treats as 1 sub. Works pretty well.

Where do you place them? And how do you time/phase align them? Are you doing it old school where you play with levels and phase on each sub and keep iterating with REW?

The point of multi sub is to dsp each sub to hit a target combined response for which each sub would need it’s own channel and it’s own DSP (channel).

 

[Robbo99999]

How many subs are you planning to use?
I have 3 and use an RCA Y splitter for 2 which my AVR treats as 1 sub. Works pretty well.

I've briefly thought about getting subs and had imagined that as one way of doing it, but good to confirm you can split the signal. Do you lose input signal power to each sub when you use a Y-splitter, is it halved?

 

[srrxr71]

I've briefly thought about getting subs and had imagined that as one way of doing it, but good to confirm you can split the signal. Do you lose input signal power to each sub when you use a Y-splitter, is it halved?

Yes possibly but that’s only 3dB and you can get it back with gain adjustment.

What I intend is to have multiple subs in which each sub is individually peq’d so that the sum at the listening position is as flat as possible. Since each sub is dealing with a different acoustic environment each would need its own measurement and eq. Nothing wrong with copying the same signal and adjusting positions,phase, delay and level to find what is flatest at your listening position but we have more sophisticated ways of doing this now.

 

[Robbo99999]

Yes possibly but that’s only 3dB and you can get it back with gain adjustment.

What I intend is to have multiple subs in which each sub is individually peq’d so that the sum at the listening position is as flat as possible. Since each sub is dealing with a different acoustic environment each would need its own measurement and eq. Nothing wrong with copying the same signal and adjusting positions,phase, delay and level to find what is flatest at your listening position but we have more sophisticated ways of doing this now.

Yes, I can see how using a Y-splitter on inputs to the subs can reduce your configuration/tweaking options because 2 subs are getting the same signal (inclusive of any EQ within that signal, so they can't be individually EQ'd, but as you say there are probably other means of tweaking the subs rather than just on the input signal to them - sub positional placement and any kind of EQ options built into the sub itself)..

 

[GalZohar]

Many sources claim that you shouldn't individually EQ them, but rather time/phase align them, set the level for total SPL (with both having same gain) and then EQ both the same. I suppose there is some logic to it (for example, if 1 sub has a peak where another has a dip, cutting that peak for both subs would result in a flat anyway), even though I don't really undrestand if the advantages outweight the advantages of individual EQ. In any case Audyssey XT32 time aligns the subs (although might not be in the best way and seems like you can't manually adjust it) and then EQs them both as one sub anyway. So if you time align with minidsp you get the same results as XT32 would give except you can manually time-align.

 

[Robbo99999]

Many sources claim that you shouldn't individually EQ them, but rather time/phase align them, set the level for total SPL (with both having same gain) and then EQ both the same. I suppose there is some logic to it (for example, if 1 sub has a peak where another has a dip, cutting that peak for both subs would result in a flat anyway), even though I don't really undrestand if the advantages outweight the advantages of individual EQ. In any case Audyssey XT32 time aligns the subs (although might not be in the best way and seems like you can't manually adjust it) and then EQs them both as one sub anyway. So if you time align with minidsp you get the same results as XT32 would give except you can manually time-align.

Interesting to know the different approaches, if I get some subs then I'll look into it more and choose what I think is the most valid method....preliminarily I think it would be a manual approach rather than using paid room eq software (eg Audyssey).

 

[srrxr71]

Many sources claim that you shouldn't individually EQ them, but rather time/phase align them, set the level for total SPL (with both having same gain) and then EQ both the same. I suppose there is some logic to it (for example, if 1 sub has a peak where another has a dip, cutting that peak for both subs would result in a flat anyway), even though I don't really undrestand if the advantages outweight the advantages of individual EQ. In any case Audyssey XT32 time aligns the subs (although might not be in the best way and seems like you can't manually adjust it) and then EQs them both as one sub anyway. So if you time align with minidsp you get the same results as XT32 would give except you can manually time-align.

I suspect the reason is that eq-ing these subs is too complex for anyone to do “by hand”.

There is a 2 hour video of someone using multi sub optimizer program for windows on YouTube.
What it seems to be doing is running something like Monte Carlo simulations. So you can have it run for a few minutes or a few hours just iterating on the combinations. This is something if you tried by hand your life would be over before you get to say 1dB coherence.

Also the webpage for the MSO program links to some great Harman/O’Toole articles. It really opened my eyes to the basics subwoofer placement. I realized for my purposes probably 2 subs is enough. 4 is better but with trade offs. They simulated even hundreds of subs. Basically 2 and 4 are optimal. Anything more actually lowers room output. Of course the assumption being a domestic sized room. I bet it doesn’t apply so much for large theaters and if it does then I suppose they will use stacked subs in the same 2 or 4 locations. So basically I suppose I should say 2 or 4 locations instead of subs.

 

[srrxr71]

Interesting to know the different approaches, if I get some subs then I'll look into it more and choose what I think is the most valid method....preliminarily I think it would be a manual approach rather than using paid room eq software (eg Audyssey).

I agree. Check this and scroll to “before you begin” the first article from 2002 is all about manual placement. The second is more recent and talks about some of the experiments they did. It doesn’t get into phase and delay management but I’m sure there are good articles and videos for that. One thing that is necessary is that your subs have an adjustable phase some only have a 0/180 switch. Even so all is not lost you just have fewer options with regard to distances I suppose.

 

[AudioJester]

I use the splitter in my HT system, main use is for movies and not music. Two of the subs are on front wall: left-sub-centre-sub-right. So both subs equidistant to main listening spot. I have no issue with gain. Third sub back corner.