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

[OCA]

I've never understood this either, because if you've got crossovers _anywhere_, then you're getting more than LFE anyway ... so what is "main" beyond what is crossed over? Surely it's not doubling up your bass

If your front speakers have a crossover, you need to be in LFE mode. LFE + Main removes crossover from fronts. Sub and fronts play all signals altogether. You need to adjust lowpass filter frequency for the sub though. In fact, with all bar the latest Denon/Marantz AVRs, when you select LFE + Main, front speakers' crossover frequency setting becomes subwoofer low pass filter frequency setting. Newest models have a separate setting for that named "bass extraction freq.". This was known as double bass in the past and was avoided because noone knew how to time align the sub for that and it sounds quite bad without correct sub time and lpf alignment.

 

[Sagnet]

Evo compares both and picks the one that tracks the curve curve best and to some extent the one with less group delay in the bass region. More often than not, LFE + Main is selected. LFE is required if you opt for aligning the sub to centre speaker.

Thanks! So v3x set fronts as small at 90Hz. Maestro sets them as small at 120Hz. Maestro with "Force large fronts" sets fronts as large (obviously) at 90Hz and Bass to LFE + Mains. Do you have any idea what difference between v3x and Maestro is the reason behind this?

 

[PHD]

If your front speakers have a crossover, you need to be in LFE mode. LFE + Main removes crossover from fronts. Sub and fronts play all signals altogether. You need to adjust lowpass filter frequency for the sub though. In fact, with all bar the latest Denon/Marantz AVRs, when you select LFE + Main, front speakers' crossover frequency setting becomes subwoofer low pass filter frequency setting. Newest models have a separate setting for that named "bass extraction freq.". This was known as double bass in the past and was avoided because noone knew how to time align the sub for that and it sounds quite bad without correct sub time and lpf alignment.

In my new X4800H. I've set front to small XO at 40Hz and in subwoofer settings I set to LFE+Main. Does this mean that the 40H XO for the fronts is only for the LPF to the subs (fronts bass extraction sent to subs) not also for the HPF to the fronts?

So, to HPF the fronts I must choose LFE? This makes no sense. Why would you want to extract sound below 40Hs from fronts and send to subs w/o HPF to fronts? If the sub can do 80Hz then it would be better to set LFE+main at 80Hz.

Does the EVO maestro account for the new bass management in the X800/Cinema series?

So, LFE+main at 40Hz XO for the fronts does not protect the fronts from producing say a 10Hz signal. At least the LFE channel is not sent to the fronts unless you choose to do so using the bass distribution. This is new for me. I'm just starting to learn about the X4800H before I do any calibration using EVO.

 

[OCA]

Thanks! So v3x set fronts as small at 90Hz. Maestro sets them as small at 120Hz. Maestro with "Force large fronts" sets fronts as large (obviously) at 90Hz and Bass to LFE + Mains. Do you have any idea what difference between v3x and Maestro is the reason behind this?

I have some idea because I wrote the script Sub alignment is done at a particular frequency to deal with very limited time delays available to Audyssey. Depending on the chosen target curve, averaging method, frequency range filtered, minimal changes can lead to different sub alignment and a different sub alignment may result in different best crossover frequencies for target tracking. Maestro introduced "mic position distance to MLP" weighted measurement averaging which I guess is the main factor to be more specific but even with force MLP option which ignores measurements outside of MLP, different results are indeed possible. They are consistent in themselves and will all sound good though.

 

[OCA]

Does this mean that the 40H XO for the fronts is only for the LPF to the subs

Yes!

So, to HPF the fronts I must choose LFE?

Yes!

Why would you want to extract sound below 40Hs from fronts and send to subs w/o HPF to fronts?

So you have 2 more subs in the system!

Does the EVO maestro account for the new bass management in the X800/Cinema series?

Yes!

At least the LFE channel is not sent to the fronts unless you choose to do so using the bass distribution

Correct, don't do that, LFE signals are also boosted by 10dB. No harm in sending 10Hz to fronts though. During Audyssey sweeps they get 0-24kHz about 80 times in a row at very loud volume. All speakers are designed to work full range on their own unless they are specifically created to play with a sub (ie satellites). You only risk distortion at high volumes and that's often the amp giving up.

 

[PHD]

@OCA


I think it would be best if you could start a new thread here regarding the Denon/Marantz frequency sweep wobbling. I'd also start similar threads at AVS (separated from the A1 thread) as well as in the Audiohollics forum

Audioholics Home Theater Forums

Home theater discussion forums powered by Audioholics.com Free registration and guests can post.

forums.audioholics.com

We need to bring more attention to this issue and get some feedback from other audio experts. It is very strange that Amir's measurements of Denon/Marantz AVRs never show any wobbling when sweeps are carried out using the Audio Precision analyzer. Perhaps it's because the problem is absent in Pure Direct mode (the only mode Amir tests for AVRs)?

 

[OCA]

@OCA


I think it would be best if you could start a new thread here regarding the Denon/Marantz frequency sweep wobbling. I'd also start similar threads at AVS (separated from the A1 thread) as well as in the Audiohollics forum

Audioholics Home Theater Forums

Home theater discussion forums powered by Audioholics.com Free registration and guests can post.

forums.audioholics.com

We need to bring more attention to this issue and get some feedback from other audio experts. It is very strange that Amir's measurements of Denon/Marantz AVRs never show any wobbling when sweeps are carried out using the Audio Precision analyzer. Perhaps it's because the problem is absent in Pure Direct mode (the only mode Amir tests for AVRs)?

I haven't seen them in my many measurements from receiver preouts either. It looks like and is quite strange that problem exists only when playing through speakers. There's said to be already a thread here about 4800 DC/DC converter Heos chip noise problem which I could not find yet.

 

[Sagnet]

I have some idea because I wrote the script Sub alignment is done at a particular frequency to deal with very limited time delays available to Audyssey. Depending on the chosen target curve, averaging method, frequency range filtered, minimal changes can lead to different sub alignment and a different sub alignment may result in different best crossover frequencies for target tracking. Maestro introduced "mic position distance to MLP" weighted measurement averaging which I guess is the main factor to be more specific but even with force MLP option which ignores measurements outside of MLP, different results are indeed possible. They are consistent in themselves and will all sound good though.

Excellent explanation, thanks! I know you are the mastermind behind all this. I just discovered it (v3x was my first version), and it has really improved the sound quality of my system.

I will try running the fronts as Large, hoping my Denon X3600H (with no external amplification) will cope!

 

[PHD]

@OCA


I've just rerun a 3-point measurement on my X4800H with two subwoofers. Auddyssey app, as well as JSON editor, show 2 subwoofers being detected (11 channels - 9.1 + 2 subbs) and calibrated but EVO MJ (latest) only extracts measurements for one subwoofer. I can see SW10.txt to SW12.txt but no SW2*.txt in REW/extracted measurements.

I tried repeating the measurements multiple times and even tried using older measurements from my older AVR, the X4400H, and still SW2*.txt are absent.

My ady files for both X4400H and X4800H can be downloaded from these links:

Dropbox

www.dropbox.com

Dropbox

www.dropbox.com

Also see AVS posts:

https://www.avsforum.com/threads/audyssey-one-oca-does-it-again-and-again.3297198/page-706?post_id=63498030#post-63497793

Can you (or anyone else with dual subs) please check if you get the same behavior? Or, does Evo combine all subs into a single one after applying time alignment?

Thanks

 

[ooheadsoo]

Did you take the measurements with directional bass turned on?

 

[PHD]

Did you take the measurements with directional bass turned on?

No. Because my subs are placed asymmetrically so I don't think directional bass will be beneficial. If I understand correctly, in directional bass, each sub will output a lower extension of different speakers provided that the subs are placed on the left and right. With such a configuration, it will be impossible to apply proper bass management to compensate for room modes. But, I may be wrong here.

 

[Hyacin]

@OCA


I've just rerun a 3-point measurement on my X4800H with two subwoofers. Auddyssey app, as well as JSON editor, show 2 subwoofers being detected (11 channels - 9.1 + 2 subbs) and calibrated but EVO MJ (latest) only extracts measurements for one subwoofer. I can see SW10.txt to SW12.txt but no SW2*.txt in REW/extracted measurements.

I tried repeating the measurements multiple times and even tried using older measurements from my older AVR, the X4400H, and still SW2*.txt are absent.

My ady files for both X4400H and X4800H can be downloaded from these links:

Dropbox

www.dropbox.com

Dropbox

www.dropbox.com

Also see AVS posts:

https://www.avsforum.com/threads/audyssey-one-oca-does-it-again-and-again.3297198/page-706?post_id=63498030#post-63497793

Can you (or anyone else with dual subs) please check if you get the same behavior? Or, does Evo combine all subs into a single one after applying time alignment?

Thanks

View attachment 387839

I had this same problem but for the life of me, I can't remember how I fixed it.

The receiver was rejecting my uploads and when I checked there was only one sub.

I think I just went right back to the beginning and made a new 'dummy' file to zero out, and it went away after that.

Wait, that's the process for the other method ... so now I'm actually not even sure if I was having the issue with Evo, or the original method :-/

 

[PHD]

OK. It appears that with standard subwoofer config, Audysey will always merge the sub freq responses. So you need to take measurements in Directional mode or use OCA's standard to the directional conversion script. EVo will convert the final result back to standard. I will give it a try later.

 

[OCA]

OK. It appears that with standard subwoofer config, Audysey will always merge the sub freq responses. So you need to take measurements in Directional mode or use OCA's standard to the directional conversion script. EVo will convert the final result back to standard. I will give it a try later.

In "standard bass" mode which is the only available mode in receivers bar the latest models, Audyssey will apply just one set of filters for the combined output of the two subs. SW2 will only be given a distance and volume offset relative to SW1 and the EQ filters will be produced under SW1 in the calibration file. Evo extracts that and this is what you see in REW. The ioptimization process will run correctly for your configuration.

Now, with the latest models, you can switch to directional bass mode and run Audyssey calibration in that mode only to force Audyssey measure all subs individually. WIth that data Evo will do the alignment between subwoofers itself and produce a better optimized calibration. Evo can do this for up to 4 subs connected directly to the receiver. Note that, the final calibration is going to be in standard bass mode as directional mode is not really a useful mode for Atmos set ups.

There's also a converter script which can generate fake directional bass mode calibrations for the older models. This requires a tricky measurement process where you take 3 measurements all at the same central mic position and switch subs at the receiver preout between the second and third measurments. New models don't need that procedure.

 

[Mike Lima]

In "standard bass" mode which is the only available mode in receivers bar the latest models, Audyssey will apply just one set of filters for the combined output of the two subs. SW2 will only be given a distance and volume offset relative to SW1 and the EQ filters will be produced under SW1 in the calibration file. Evo extracts that and this is what you see in REW. The ioptimization process will run correctly for your configuration.

Now, with the latest models, you can switch to directional bass mode and run Audyssey calibration in that mode only to force Audyssey measure all subs individually. WIth that data Evo will do the alignment between subwoofers itself and produce a better optimized calibration. Evo can do this for up to 4 subs connected directly to the receiver. Note that, the final calibration is going to be in standard bass mode as directional mode is not really a useful mode for Atmos set ups.

There's also a converter script which can generate fake directional bass mode calibrations for the older models. This requires a tricky measurement process where you take 3 measurements all at the same central mic position and switch subs at the receiver preout between the second and third measurments. New models don't need that procedure.

Sorry if my understanding is lacking. I have an older lower level Denon that only has MultiEq. So if I have 2 identical subs (in the front)... combined with 2 different ones (in the rear). Would I have better optimized calibration if I disconnect the rears so that Evo calibrates the 2 subs better... (then maybe just add the rear 2 and adjust to taste?) Thank you for all you do.

 

[OCA]

I have no idea how the calibration will end up with 4 subs connected by Y-cables (I assume). You have to try but it's not very likely to yield optimal results. Identical subs will not have identical responses in different parts of the room.

 

[Mike Lima]

I have no idea how the calibration will end up with 4 subs connected by Y-cables (I assume). You have to try but it's not very likely to yield optimal results. Identical subs will not have identical responses in different parts of the room.

I have no idea how the calibration will end up with 4 subs connected by Y-cables (I assume). You have to try but it's not very likely to yield optimal results. Identical subs will not have identical responses in different parts of the room.

Thank you for your response. I was encouraged to disconnect my MiniDSP using EVO... Maybe I should bring it back into the chain.

 

[Dovydas]

There's also a converter script which can generate fake directional bass mode calibrations for the older models. This requires a tricky measurement process where you take 3 measurements all at the same central mic position and switch subs at the receiver preout between the second and third measurments. New models don't need that procedure.

Is this step necessary? I own x6700h. Just ran A1 Evo Maestro and bass freq. response looks much worse than MultEQ-X result. I measured it with Umik-1 and REW. Could it be a reason?

 

[Hyacin]

Thank you for your response. I was encouraged to disconnect my MiniDSP using EVO... Maybe I should bring it back into the chain.

Personally, I'd -

* Absolutely 100% put the MiniDSP back in
* Put a "perfect response file" from the Supreme method on the receiver, calibrate the subs with REW+MSO as per typical MiniDSP process
* Run the Audyssey multi-point with the MiniDSP in, on, and calibrated
* Run that through A1

Very curious to see what OCA says though

 

[Smitty2k1]

I have a Denon x3500h running 3.1 with a SVS SB-1000 PRO as my sub. I'm considering adding a second SB-1000. I'm under the impression I need a Minidsp to properly integrate a 2nd sub running this script, even though my AVR has two sub outputs because it isn't the newest Gen with the "directional" bass.

However, I also know the SB-1000 PRO has DSP built in - am I able to skip the Minidsp if I get a second SB-1000 PRO instead of a non-pro?