Audyssey XT32 vs Dirac Live

[TheZebraKilledDarwin]

All of these fidgeting is actually also trying to evaluate if I need to get the Dirac.
Since Dirac license is so expensive and it follows the receiver, not the owner.

Would it help, if an illiterate would buy a more expensive book, to enjoy reading more?

It must be known, how certain frequency ranges sound, before an EQ for the HT setup should even be touched.
This is not something we know, without learning it. We don't have a sense for "oh, that's the 500 range! That's the 1000 range, thats 100 Hz and that's 50 Hz."
We only know that, by hearing the filtered sound and looking at the frequency display of the filter.
It is the prerequisite to be able to make the necessary connection between a certain sound characteristic and the corresponding frequency range on an EQ.

Many media players allow to use parametric EQs. Enabling a bandpass filter and listen to it, means that it is learned, how the freqeuncy ranges are sounding.
Buying Dirac will not solve the problem of not knowing the frequency range.

For example: the KEF has a midrange very well suited to movies. It's 500 Hz range is lightly louder than the 1k range. This midrange balance is what helps to achieve intelligible, warm, pleasing dialogue sound, and not piercing the ears at higher levels with an upper midrange that is louder than the 500.

But because you don't know what the 500 range does to the sound, probably for some graphical or measurement optics, you are cutting the 500 and the 1k in relation becomes that much louder (if you cut 5 dB on the 500, then the 1000 becomes 5 dB louder in relation! Btw, that's why EQ changes at one range, always affect the whole sound. Most importantly ofcourse is the midrange. A really good sounding midrange is based on a very fine balance between 250-500-1000).
After making these - probably counterproductive cuts of the 500 - then you need all kinds of bass and treble boosts, because the mids are out of balance, dialogue intelligibility probably is reduced, and it probably even starts to hurt the ears, when things become loud.

Secondly to safe even more money:
Audyssey's XT32 auto correction can be switched off with a hack of the .ady MultEQ-app file and instead of the automatic correction with a user curve on top of it, a 100% individual target curve (more than 1000 filter points are possible) can be used (+-12 dB correction range) without any interfering frequency corrections done by Audyssey itself. It's even possible to export REW filter curves, convert the curves with Excel into the target-curve.
But first, it's necessary to learn, how different frequency ranges are sounding, to know which frequency ranges you really want reduced or emphasized...

 

[MiniC]

CinemaEQ rolls off high frequencies.

Googled high freq roll off. So that’s why.
I guess my ears are more sensitive to high pitch freq and needs them to be tamed.

 

[MiniC]

Would it help, if an illiterate would buy a more expensive book, to enjoy reading more?

It must be known, how certain frequency ranges sound, before an EQ for the HT setup should even be touched.
This is not something we know, without learning it. We don't have a sense for "oh, that's the 500 range! That's the 1000 range, thats 100 Hz and that's 50 Hz."
We only know that, by hearing the filtered sound and looking at the frequency display of the filter.
It is the prerequisite to be able to make the necessary connection between a certain sound characteristic and the corresponding frequency range on an EQ.

Many media players allow to use parametric EQs. Enabling a bandpass filter and listen to it, means that it is learned, how the freqeuncy ranges are sounding.
Buying Dirac will not solve the problem of not knowing the frequency range.

For example: the KEF has a midrange very well suited to movies. It's 500 Hz range is lightly louder than the 1k range. This midrange balance is what helps to achieve intelligible, warm, pleasing dialogue sound, and not piercing the ears at higher levels with an upper midrange that is louder than the 500.

But because you don't know what the 500 range does to the sound, probably for some graphical or measurement optics, you are cutting the 500 and the 1k in relation becomes that much louder (if you cut 5 dB on the 500, then the 1000 becomes 5 dB louder in relation! Btw, that's why EQ changes at one range, always affect the whole sound. Most importantly ofcourse is the midrange. A really good sounding midrange is based on a very fine balance between 250-500-1000).
After making these - probably counterproductive cuts of the 500 - then you need all kinds of bass and treble boosts, because the mids are out of balance, dialogue intelligibility probably is reduced, and it probably even starts to hurt the ears, when things become loud.

Secondly to safe even more money:
Audyssey's XT32 auto correction can be switched off with a hack of the .ady MultEQ-app file and instead of the automatic correction with a user curve on top of it, a 100% individual target curve (more than 1000 filter points are possible) can be used (+-12 dB correction range) without any interfering frequency corrections done by Audyssey itself. It's even possible to export REW filter curves, convert the curves with Excel into the target-curve.
But first, it's necessary to learn, how different frequency ranges are sounding, to know which frequency ranges you really want reduced or emphasized...

Yup. I totally agree, else I’m just trying to strike the mega ball lottery.

I probably need to get my umik + software to measure what the freq that I find it too harsh first.
Last night I played a video on YouTube that runs freq from 20hz & up. I find that from 2khz to 4khz it’s really piercing.

When we use softwares that does room corrections, a mic is being used.
However, individual ears are still susceptible to different freq response I suppose.
So can I say the room correction did the job well, but the user can hear some frequency due to his hearing issues. Then user need to know which frequency to bump up or down.
But when these manual changes to the freq response curve are made, will it affect the room correction? Or it becomes a chicken and egg thing?

 

[Andysu]

Yup. I totally agree, else I’m just trying to strike the mega ball lottery.

I probably need to get my umik + software to measure what the freq that I find it too harsh first.
Last night I played a video on YouTube that runs freq from 20hz & up. I find that from 2khz to 4khz it’s really piercing.

When we use softwares that does room corrections, a mic is being used.
However, individual ears are still susceptible to different freq response I suppose.
So can I say the room correction did the job well, but the user can hear some frequency due to his hearing issues. Then user need to know which frequency to bump up or down.
But when these manual changes to the freq response curve are made, will it affect the room correction? Or it becomes a chicken and egg thing?

canal its the canal yeah the canal is the key to harsh , i wonder now how my cats feel with their canal

get yourself a canal a silicon rubber ear placed over microphone in certain way it mimics an approximation of the elevated boots that regular common measurement won't take into account , its not big deal it needs a Peq bandwidth q filter and lowering it down

i did few videos

silicon ear testing with pink noise

 

[peng]

Would it help, if an illiterate would buy a more expensive book, to enjoy reading more?

It must be known, how certain frequency ranges sound, before an EQ for the HT setup should even be touched.
This is not something we know, without learning it. We don't have a sense for "oh, that's the 500 range! That's the 1000 range, thats 100 Hz and that's 50 Hz."
We only know that, by hearing the filtered sound and looking at the frequency display of the filter.
It is the prerequisite to be able to make the necessary connection between a certain sound characteristic and the corresponding frequency range on an EQ.

Many media players allow to use parametric EQs. Enabling a bandpass filter and listen to it, means that it is learned, how the freqeuncy ranges are sounding.
Buying Dirac will not solve the problem of not knowing the frequency range.

For example: the KEF has a midrange very well suited to movies. It's 500 Hz range is lightly louder than the 1k range. This midrange balance is what helps to achieve intelligible, warm, pleasing dialogue sound, and not piercing the ears at higher levels with an upper midrange that is louder than the 500.

But because you don't know what the 500 range does to the sound, probably for some graphical or measurement optics, you are cutting the 500 and the 1k in relation becomes that much louder (if you cut 5 dB on the 500, then the 1000 becomes 5 dB louder in relation! Btw, that's why EQ changes at one range, always affect the whole sound. Most importantly ofcourse is the midrange. A really good sounding midrange is based on a very fine balance between 250-500-1000).
After making these - probably counterproductive cuts of the 500 - then you need all kinds of bass and treble boosts, because the mids are out of balance, dialogue intelligibility probably is reduced, and it probably even starts to hurt the ears, when things become loud.

To do as you alluded to, one has to know a lot about speakers/room interaction, and how REQ/RC systems work in general. Also, many manual tweakers may or may not be aware that while their tweaked response may look great for the mmp, it may not sound any better (could be worse) than that achievable with auto with some minor tweaks post calibration, if the auto is done well using RC such as XT32 and Dirac Live because those software use their own algorithm to do spatial averaging that may result in better sound quality even for one listening seat.

Secondly to safe even more money:
Audyssey's XT32 auto correction can be switched off with a hack of the .ady MultEQ-app file and instead of the automatic correction with a user curve on top of it, a 100% individual target curve (more than 1000 filter points are possible) can be used (+-12 dB correction range) without any interfering frequency corrections done by Audyssey itself. It's even possible to export REW filter curves, convert the curves with Excel into the target-curve.
But first, it's necessary to learn, how different frequency ranges are sounding, to know which frequency ranges you really want reduced or emphasized...

That sounds like a great idea, but again, it is only going to work well if the users know enough to beat "auto". I think for most people, trust auto to a point, then do some minor tweaks with the apps post calibration would likely have a better chance to get better results than totally auto or totally manual.

 

[MiniC]

I decided to upgrade to Dirac live for my x4800h.

Just ran my 1st measurements, using focused imaging. Didn’t do any other tuning(since I got no idea how), just run and load into avr.

As I’ve been using saving private Ryan to tinkle my audyssey, I’m quite familiar with the Omaha beach scene.
The bass and treble are just right. Dialog is clear.(Been googling on Dirac and this is majority of people’s feedback).
The bullets raining on the soldiers when the carrier ship open its door is not harsh but just right. Not piercing to the ears.

I discovered that scene where Tom hanks was speaking to a medical battalion guy with a typewriter on the beach, taking shelter at the C beam. I heard a “ting” sound when the typewriter was ditched. I didn’t hear this with Audyssey. I replayed the scene again and it’s real.
I’m rewatching saving private Ryan and looking at my phone. And I was taken aback of hearing more details of background sound, that I didn’t notice when using Audyssey.

The volume is same for audyssey and dirac at 60db.

I would feel it’s really worth the upgrade as I can distinguish improvement on bass, treble and clarity.
Could spend my time enjoying my system now.