Audyssey Manual Calibration “OCA’s REW + Audyssey Awesomeness”

[Chromatischism]

You can upload any target curve to REW during A1 optimization and it will replace Toole curve. A1 creates the minimum phase version of the curve on the go. Calculated crossover frequencies might deviate though but you can adjust them manually anyway.

First I just want to say, brilliant work.

I'm one of the lucky ones that has been using the app for years and get great results with minimal effort because my placement of everything is nearly perfect and very little tweaking afterward is needed (sub distance tweak and sub levels to taste is all I do). I normally correct to 300-400 Hz on the main speakers and 500-600 Hz on the surrounds since they are wall-mounted and that creates issues higher up that need to be fixed.

However I will try your script anyway because as an amateur coder and tweaker, I'm intrigued by your approach. Who knows, maybe it will be better.

Second, I have a question about the above-quoted comment. I normally achieve my perfect bass quantity + quality with Dynamic EQ (I get just the right curve with the rooms I've used it in with my 2 subs). I also listen at very different volume levels. I would of course test your stock result to see what I think, but if I wanted to "undo" the Toole curve and use DEQ, what curve would I load into REW? It wouldn't be a straight line, would it? Maybe some toggle-able options could be implemented in a future version. Proceed with or without Toole curve, DEQ surround boost defeat on/off, things like that.

 

[oupee]

Thanks. Started again from scratch and still end up with the following
error: speaker has no bass response according to your measurement file. Please check! - program execution has stopped

@OCA
Same problem and the speakers are big

 

[OCA]

@OCA
Same problem and the speakers are big

That's often caused by a buggy Android version of the Multeq app. There must be a download link for the correct app version in the pinned video comments.

 

[OCA]

First I just want to say, brilliant work.

I'm one of the lucky ones that has been using the app for years and get great results with minimal effort because my placement of everything is nearly perfect and very little tweaking afterward is needed (sub distance tweak and sub levels to taste is all I do). I normally correct to 300-400 Hz on the main speakers and 500-600 Hz on the surrounds since they are wall-mounted and that creates issues higher up that need to be fixed.

However I will try your script anyway because as an amateur coder and tweaker, I'm intrigued by your approach. Who knows, maybe it will be better.

Second, I have a question about the above-quoted comment. I normally achieve my perfect bass quantity + quality with Dynamic EQ (I get just the right curve with the rooms I've used it in with my 2 subs). I also listen at very different volume levels. I would of course test your stock result to see what I think, but if I wanted to "undo" the Toole curve and use DEQ, what curve would I load into REW? It wouldn't be a straight line, would it? Maybe some toggle-able options could be implemented in a future version. Proceed with or without Toole curve, DEQ surround boost defeat on/off, things like that.

from video comments:

"Audyssey One will provide perfect Dr Toole target curve response at 75dB (MutlEQ Editor app measured level, -30dB on the receiver scale) at Audyssey Reference setting with Dynamic EQ, Dynamic volume and LFC all off (default). But if you are somehow not satisfied with the outcome or desire certain changes including the ability to use Dynamic EQ, here are some tips to manipulate the sound. * If you want to take advantage of Dynamic EQ and prefer the higher bass output of Harman Kardon target curve, the settings below will get you the closest system response to Harman curve at 75dB: Audyssey settings¦ MultEQ XT32: Reference, Dynamic EQ: On, -Reference Level Offset: 5dB, Dynamic Volume: Off, Audyssey LFC: On, -Containment Amount: 1 Surround Parameters¦ Cinema EQ: On * You find the sound too bright: "Cinema EQ: On" setting will dim high frequencies gradually from 1kHz onwards by about 4dB at 75dB volume level * You want more bass but not Dynamic EQ or Harmon curve in particular: Increase subwoofer volume level Try inverting subwoofer polarity (there have been cases where Audyssey got it wrong in the first place and A1 doesn't check for that at this time) "Dynamic Volume: Heavy" setting will tilt the sound response to provide relatively more bass PS: The effects of the above parameters have been tested directly from preamp outputs of a Marantz SR6015 and may not fully reflect your speakers' in room response."

 

[Zedly]

I don't have the app, but I do have MultiEQ-X. Can Audyssey One be used with MultiEQ-X?

 

[OCA]

I don't have the app, but I do have MultiEQ-X. Can Audyssey One be used with MultiEQ-X?

No, mic calibration file is encrypted in mqx files which makes it impossible to import the actual measurements to REW.

 

[oupee]

That's often caused by a buggy Android version of the Multeq app. There must be a download link for the correct app version in the pinned video comments.

Thank you for your time. But I have the IOS version 1.11.1 (3.20.0)

 

[Bucking]

So the new 1.3x version reported, that I should change the phase of my sub and see if that improves things.
I reversed the phase to 180 and did measure again. After that I didn't get any recommendations for my sub in the log.
Does that mean, that having my sub at 180 is the way to go?

 

[Chromatischism]

So the new 1.3x version reported, that I should change the phase of my sub and see if that improves things.
I reversed the phase to 180 and did measure again. After that I didn't get any recommendations for my sub in the log.
Does that mean, that having my sub at 180 is the way to go?

It probably just means it had fewer issues getting things aligned. But post-correction measurements (and listening) is always the best practice.

 

[oupee]

@OCA
Same problem and the speakers are big

Solved, but it cuts my center at 200Hz, while it's big. I manually set it back to 80Hz.

 

[oupee]

Can someone try the movie Argyle, just the beginning. I have an SR8015 and can toggle present1/2 instantly. If I listen to the original Audyssey Reference, the deep vocals can be heard from the front but wide, maybe a little to the sides. Audyssey edited via OCA, the voice sounds completely separate in the middle. It sounds like a completely different sound mix. I don't know what is right. The ensuing short firefight sounds better through the OCA.

 

[Chromatischism]

from video comments:

"Audyssey One will provide perfect Dr Toole target curve response at 75dB (MutlEQ Editor app measured level, -30dB on the receiver scale) at Audyssey Reference setting with Dynamic EQ, Dynamic volume and LFC all off (default). But if you are somehow not satisfied with the outcome or desire certain changes including the ability to use Dynamic EQ, here are some tips to manipulate the sound. * If you want to take advantage of Dynamic EQ and prefer the higher bass output of Harman Kardon target curve, the settings below will get you the closest system response to Harman curve at 75dB: Audyssey settings¦ MultEQ XT32: Reference, Dynamic EQ: On, -Reference Level Offset: 5dB, Dynamic Volume: Off, Audyssey LFC: On, -Containment Amount: 1 Surround Parameters¦ Cinema EQ: On * You find the sound too bright: "Cinema EQ: On" setting will dim high frequencies gradually from 1kHz onwards by about 4dB at 75dB volume level * You want more bass but not Dynamic EQ or Harmon curve in particular: Increase subwoofer volume level Try inverting subwoofer polarity (there have been cases where Audyssey got it wrong in the first place and A1 doesn't check for that at this time) "Dynamic Volume: Heavy" setting will tilt the sound response to provide relatively more bass PS: The effects of the above parameters have been tested directly from preamp outputs of a Marantz SR6015 and may not fully reflect your speakers' in room response."

I saw this and appreciate it. However, it doesn't offer a "zero curve applied" outcome. My room curve comes from Audyssey equalizing to flat and then using Dynamic EQ, nothing else. No smoothing 10-300 Hz:

 

[OCA]

At Audyssey's measured 75dB (-30 dBfS scale) level, this is Flat vs Dynamic EQ default (0dB):

DEQ effect is far from an application of the slight tilt a speaker's steady state response is supposed to fit. It tries to fix the absurd HFR1 reference curve at the high end and boosts the bass massively in the low frequency band for all practical volume levels. IMO, Audyssey has confused reference level peak with the reference level itself (105dB vs 85dB - Reference level is mastered in studios at 85dB with highest peak to hit 105dB) in the original design and they continued to stick with that all these years generating false excuses instead of a fix. DEQ 0dB has 0 effect at 105dB (which should have been 85dB), DEQ 5dB has 0 effect at 100dB (105 - 5), .. DEQ 15dB has 0 effect at 90dB (105 - 15). So, the DEQ strength also has nothing to do with music or movie listening as claimed. Add to these another 6dB for surrounds and 3dB for the heights!!!

There's a high pass 6dB/octave Butterworth filter at 15,75Hz embedded in the units causing the tilt at the left end of Flat. When DEQ is on, no filters are applied below 24Hz. Subwoofers are applied either one of these two filters by default: Linkwitz Riley 24dB/oct lowpass at 120Hz or at 250Hz distinguished by "isReversePolarity" property in the calibration file. I can go on forever

 

[Chromatischism]

DEQ effect is far from an application of the slight tilt a speaker's steady state response is supposed to fit. It tries to fix the absurd HFR1 reference curve at the high end and boosts the bass massively in the low frequency band for all practical volume levels. IMO, Audyssey has confused reference level peak with the reference level itself (105dB vs 85dB - Reference level is mastered in studios at 85dB with highest peak to hit 105dB) in the original design and they continued to stick with that all these years generating false excuses instead of a fix. DEQ 0dB has 0 effect at 105dB (which should have been 85dB), DEQ 5dB has 0 effect at 100dB (105 - 5), .. DEQ 15dB has 0 effect at 90dB (105 - 15).

Ok. But if you follow that through to its end result, you'd have very little bass. In fact, nearly everyone who has used Audyssey says they did not have nearly enough bass post-calibration with DEQ off. I'd be interested to hear your thoughts on why that is. Because with DEQ on, bass levels are much closer to perfect. And it's not just the overall level of bass, but the shape of it that sounds great to me--and that it's consistent at all volume levels, something I can't say about systems without DEQ. I've always thought that they really spent some of their R&D on getting that right.

 

[oupee]

...nearly everyone who has used Audyssey says they did not have nearly enough bass post-calibration with DEQ off...

It is true. For me, Audyssey set the LCR speakers to fullband and there was not enough bass. It was enough to reset the LCR to small 80Hz and the bass was almost like DEQ. DEQ itself is terrible.

 

[antcollinet]

OK, late to the party. I've been struggling to get a good calibration for music on my Denon 3800 for a few months (movies have been fine). Now I've just done a default Audyssey 1 process, and finally have something I am happy with.

So thanks @OCA (I'm the person who replied last night on youtube about Safari not saving the file)

I've yet to run any measurements on the result - going to have to re-learn REW for that first. But for now, I'm finally happy with my new AVR

 

[Chromatischism]

It is true. For me, Audyssey set the LCR speakers to fullband and there was not enough bass. It was enough to reset the LCR to small 80Hz and the bass was almost like DEQ. DEQ itself is terrible.

Don't mean to turn this into yet another DEQ thread, but I can't agree with that blanket statement. I think people get bad results if it is misused or used with equipment that isn't right for it. My results however, are excellent. And it's one of the reasons I switched back from Dirac.

 

[antcollinet]

I've just done a default Audyssey 1 process, and finally have something I am happy with.

So things have moved along a little.

After more than just a little casual listening that resulted in the quoted statement, I realised that I was really lacking some (a lot of) bass.

So I tried the suggestions above in @OCA s post above to achieve something resembling the Harmon Kardon curve - in particular enabling Dynamic EQ. Bass came back - in spades. Unfortunately much too much. It felt like it was thumping me in the eardrums (like the beating effect of having the rear windows of a car partially rolled down)

So now I've gone into options on the AVR, and turned down the sub level about 7dB. This sounds about right now. However, I'm guessing I am not getting the ideal smooth response by doing this, since just adjusting the sub level is to coarse a band.


So I'd like to try adjusting the Toole target curve to increase the slope. I'm thinking if I multiply all the values by a factor to do this, then subtract an offset to keep the maximum boost at the peak of the bass the same, this should achieve what I want - like the picture.

What do you think?

(Green is the Toole curve, Blue is the modified curve)

Or do I just need to add a bass "hump"?

 

[EasyC]

So things have moved along a little.

After more than just a little casual listening that resulted in the quoted statement, I realised that I was really lacking some (a lot of) bass.

So I tried the suggestions above in @OCA s post above to achieve something resembling the Harmon Kardon curve - in particular enabling Dynamic EQ. Bass came back - in spades. Unfortunately much too much. It felt like it was thumping me in the eardrums (like the beating effect of having the rear windows of a car partially rolled down)

So now I've gone into options on the AVR, and turned down the sub level about 7dB. This sounds about right now. However, I'm guessing I am not getting the ideal smooth response by doing this, since just adjusting the sub level is to coarse a band.


So I'd like to try adjusting the Toole target curve to increase the slope. I'm thinking if I multiply all the values by a factor to do this, then subtract an offset to keep the maximum boost at the peak of the bass the same, this should achieve what I want - like the picture.

What do you think?

(Green is the Toole curve, Blue is the modified curve)

Or do I just need to add a bass "hump"?

View attachment 367879

What you can also do, that's what I have done and it gives me better bass, is use the Harman Curve instead of Toole's. This adjusts the bass nicely.

 

[antcollinet]

What you can also do, that's what I have done and it gives me better bass, is use the Harman Curve instead of Toole's. This adjusts the bass nicely.

Where do I get that from?

Thanks