Freq Responses graph

[TLS Guy]

@TLS Guy why on earth would you use the double bass settings? You just like that reinforcement?

Double bass is a misnomer. What it is, a high pass acoustic crossover with an electronic low pass one. That is often the best solution. Double bass is, and has been, a total misnomer, as it is nothing of the sort.

So my midlines have a second order acoustic crossover at 43 HZ, and the low pass is fourth order at 40 Hz. Asymmetrical crossovers are common in speaker design and more the rule than not. So you just need to see which gives the best response. Often the answer will be LFE plus main, as it is in my case. I think it probably is also for most with capable towers, and certainly for ported designs. If you had a ported enclosure with an f3 of 40 Hz, and you use LFE plus main then you have fourth order high and low pass crossover. If you set the crossover then the high pass is sixth order, and the low pass fourth order. This could be correct but also very likely not, so you need to experiment and measure when you can.

This whole receiver/pre pro crossover arena is awash in misunderstanding. People need to be able to work out their crossover orders and should. The operational order will be the sum of the acoustic responses of their speakers, and the electrical crossover orders.

 

[ciotime]

Where's the bump and what are its parameters?

I'm not quite sure anymore coz I did this way back before but I think it's 20Hz +6dB / 55Hz +6dB / 160Hz +0dB in the RAT app.

 

[lovinthehd]

Double bass is a misnomer. What it is, a high pass acoustic crossover with an electronic low pass one. That is often the best solution. Double bass is, and has been, a total misnomer, as it is nothing of the sort.

So my midlines have a second order acoustic crossover at 43 HZ, and the low pass is fourth order at 40 Hz. Asymmetrical crossovers are common in speaker design and more the rule than not. So you just need to see which gives the best response. Often the answer will be LFE plus main, as it is in my case. I think it probably is also for most with capable towers, and certainly for ported designs. If you had a ported enclosure with an f3 of 40 Hz, and you use LFE plus main then you have fourth order high and low pass crossover. If you set the crossover then the high pass is sixth order, and the low pass fourth order. This could be correct but also very likely not, so you need to experiment and measure when you can.

This whole receiver/pre pro crossover arena is awash in misunderstanding. People need to be able to work out their crossover orders and should. The operational order will be the sum of the acoustic responses of their speakers, and the electrical crossover orders.

So only specific to your particular arrangement via measurements with your limited setup?

ps since double bass does tend to duplicate range, how does it not?

 

[TLS Guy]

So only specific to your particular arrangement via measurements with your limited setup?

ps since double bass does tend to duplicate range, how does it not?

Because it is no different to any other crossover. They are all acoustic plus electrical. It is always the sum of both. So if a ported speaker has an F3 of 40 and you set the crossover LFE + main to the sub at 40 Hz, you have a symmetrical fourth order crossover at 40 Hz. There is no duplicate range. On the other hand if you set the sub crossover to 40 Hz, then you have a sixth order high pass filter, (2 electrical and four acoustic) and a fourth order electrical low pass. Now it could be either one is better or something else. That is why measurements really are required to see what is going on. Pulling any number and arrangement out of thin air has far more chances than not of being suboptimal. I suspect many tower systems could benefit from LFE + main arrangements. Again LFE + main does not have to mean duplicate range although if used improperly it can.

In my arrangement the center has a 45 Hz f3 and so that is fed full range the same as the mains. My surrounds have very robust drivers, and are sealed speakers second order roll off with f3 of 52 Hz. The rear backs are essentially full range, f3 measured at 27 Hz second order roll off. The other point is that the LFE is only on the sub outs, so that LFE signal only goes to the four 10" drivers in the bass lines in the right and left fronts. This arrangement greatly evens out the room bass response. There is no duplication. The arrangement works very well as the room curves show. The system is in no way bass heavy and very agile in the bass with not a trace of boom or overhang. In fact the bass is literally as "tight as a drum".

 

[PENG]

So only specific to your particular arrangement via measurements with your limited setup?

ps since double bass does tend to duplicate range, how does it not?

I don't know about his, but my REW graphs clearly showed my little Energy V2.3i towers would interfere with my subwoofers all the way down to 20 Hz. So setting the LF filter of the subs to 40 Hz would help but still not good as there will be duplication from 40 Hz down. We all know TLSG is TLSG, what applies to him may indeed not to others.

 

[3db]

Here is the before and after EQ graph.

View attachment 49057

I think you got better looking results with the mini.

 

[TLS Guy]

I don't know about his, but my REW graphs clearly showed my little Energy V2.3i towers would interfere with my subwoofers all the way down to 20 Hz. So setting the LF filter of the subs to 40 Hz would help but still not good as there will be duplication from 40 Hz down. We all know TLSG is TLSG, what applies to him may indeed not to others.

I'm wondering if my good results are because the sub is part of the speaker. This system has extreme integration, with the LF from all speakers timed.
You can see from the impulse response that all seven of the speakers are well integrated at the MLP. All crossovers overlap the lower the order the more overlap. However the higher the orders the greater the time shifts. I have never been convinced that spacing subs far from the fronts is a good idea. I know Ted Jordan thought that approach was a terrible idea. The result as I have said before is massive violence by separating fundamentals from harmonics on a grand scale. Intuitively I feel that can not be a good practice.

This is the room curve of my 2.1 system and impulse response. This system has the two subs well separated from the speakers. Look what happens to the impulse response.

I intend to fix that by building a sub to go between the speakers under the TV.

 

[PENG]

I'm wondering if my good results are because the sub is part of the speaker. This system has extreme integration, with the LF from all speakers timed.

I would think so.

Look what happens to the impulse response. I intend to fix that by building a sub to go between the speakers under the TV.

That doesn't look good at all, but your plan to fix it makes sense.

 

[ciotime]

If the center channel is full range or near full range, then it is should be included, to see how it integrates with the other FL/FR and subs. In my case I tried that in the beginning, found insignificant difference either way. So I think it depends, and should be investigated first. I wouldn't worry about center-sub, but if one has time, then may as well do that too. It's just one more graph to plot and the results should tell us what to do next.

Hi Peng...I just finished running Audyssey again and I've attached below the graph's with and without Audyssey on. Xover is set at 80Hz with the miniDsp being used. I limited the XT32 Freq Range to EQ only until 400Hz but limited the Sub EQ to 20Hz so it won't alter my EQ on the minidsp. Measured are the FL/FR/FR+FL/Subs/FR+FL+Subs.

Audyssey turned OFF

Audyssey turned ON

 

[PENG]

Hi Peng...I just finished running Audyssey again and I've attached below the graph's with and without Audyssey on. Xover is set at 80Hz with the miniDsp being used. I limited the XT32 Freq Range to EQ only until 400Hz but limited the Sub EQ to 20Hz so it won't alter my EQ on the minidsp. Measured are the FL/FR/FR+FL/Subs/FR+FL+Subs.

Audyssey turned OFF

View attachment 49105

Audyssey turned ON

View attachment 49106

Thank you, but can you identify for me which one is for FL+FR+Subs? I can assume but rather not. I will start a cuts/boosts data table for you based on the FL+FR+Subs. If that works then there is no point doing the others.

Also, when you you limited the Sub EQ to 20 Hz, did you mean so that Audyssey won't touch anything above that?
If that's the case it is not good and please include the REW Plot for one that is:
FL+FR+Subs and limited the range to 15 to 400 Hz. That is so that I can see what the effects are with the mini and Audyssey both on, and then figure out the customize curve accordingly.

You don't have to worry about the effects of the mini being nullified because if the result is good why not, and if it is not as good then you can limit the range again any time. That's the beauty of using the Apps, you can save as many target curves as you wish.

 

[ciotime]

Thank you, but can you identify for me which one is for FL+FR+Subs? I can assume but rather not. I will start a cuts/boosts data table for you based on the FL+FR+Subs. If that works then there is no point doing the others.

Also, when you you limited the Sub EQ to 20 Hz, did you mean so that Audyssey won't touch anything above that?
If that's the case it is not good and please include the REW Plot for one that is:
FL+FR+Subs and limited the range to 15 to 400 Hz. That is so that I can see what the effects are with the mini and Audyssey both on, and then figure out the customize curve accordingly.

You don't have to worry about the effects of the mini being nullified because if the result is good why not, and if it is not as good then you can limit the range again any time. That's the beauty of using the Apps, you can save as many target curves as you wish.

Thanks...below is the graph with both Audyssey On and Off for the FL+FR+Subs. The one with a huge bump fr 75Hz-130Hz is the one with Audyssey OFF. So with Audyssey ON it has a more linear downward slope fr 20Hz-130Hz.
"Also, when you you limited the Sub EQ to 20 Hz, did you mean so that Audyssey won't touch anything above that?"-YES
The Filter Freq Range on the sub in the app is fr 20Hz-250Hz...you want me to leave it untouched?
I'll try to take another measurement tomorrow without limiting the Sub Freq Range.

 

[PENG]

Thanks...below is the graph with both Audyssey On and Off for the FL+FR+Subs. The one with a huge bump fr 75Hz-130Hz is the one with Audyssey OFF. So with Audyssey ON it has a more linear downward slope fr 20Hz-130Hz.
"Also, when you you limited the Sub EQ to 20 Hz, did you mean so that Audyssey won't touch anything above that?"-YES
The Filter Freq Range on the sub in the app is fr 20Hz-250Hz...you want me to leave it untouched?
I'll try to take another measurement tomorrow without limiting the Sub Freq Range.

View attachment 49109

Yes please leave the default EQ range untouched for the sub because I want to see the combined effects of the minidsp and Audyssey. It is important if we want to find out whether the App/Rat can improve what you have so far a little further. By the way, as far as I know Audyssey would EQ down to 10 Hz, setting the EQ limit really should apply to the upper limit only. Again, simply leaving it at default would be good for this exercise, so in this case 250 Hz.

You can leave the EQ range for the towers untouched (that is default) as well for the purpose of taking the REW plots. After that you have done your REW plots you can put the limits back on the way you like it and send the curves to the AVR again for use while waiting for a new custom curve to play with, or you can simply use the App to send you original target curve (before re-running Audyssey) to the AVR so nothing should have changed, everything will be back to the way it was yesterday or the day before.

 

[ciotime]

Yes please leave the default EQ range untouched for the sub because I want to see the combined effects of the minidsp and Audyssey. It is important if we want to find out whether the App/Rat can improve what you have so far a little further. By the way, as far as I know Audyssey would EQ down to 10 Hz, setting the EQ limit really should apply to the upper limit only. Again, simply leaving it at default would be good for this exercise, so in this case 250 Hz.

You can leave the EQ range for the towers untouched (that is default) as well for the purpose of taking the REW plots. After that you have done your REW plots you can put the limits back on the way you like it and send the curves to the AVR again for use while waiting for a new custom curve to play with, or you can simply use the App to send you original target curve (before re-running Audys"sey) to the AVR so nothing should have changed, everything will be back to the way it was yesterday or the day before.

"You can leave the EQ range for the towers untouched (that is default) as well for the purpose of taking the REW plots."-So just the Towers right? Center not included? Or maybe you want the original where ALL the Filter Freq Range is untouched for ALL speakers ie Audyssey is EQ'ing all the way up to 20K Hz?
If I'm not mistaken the Filter Freq Range goes all the way up to 20K Hz.
So with the towers untouched you want me to post the REW FR from 15Hz-400Hz or 15Hz-20kHz?

 

[PENG]

"You can leave the EQ range for the towers untouched (that is default) as well for the purpose of taking the REW plots."-So just the Towers right? Center not included? Or maybe you want the original where ALL the Filter Freq Range is untouched for ALL speakers ie Audyssey is EQ'ing all the way up to 20K Hz?
If I'm not mistaken the Filter Freq Range goes all the way up to 20K Hz.
So with the towers untouched you want me to post the REW FR from 15Hz-400Hz or 15Hz-20kHz?

Let's be clear about the following when you use the Editor app to run Audyssey:

1). Audyssey will EQ all speakers in the range that it determines the speakers are capable of reproducing.
2) The App would create target curves based on the filters created (or to be created by the AVR when it receives the file/target curves).

The parts I am not sure is whether the App created the filters or it only sends the target curve and then the AVR/AVP would create the filters based on the target curve information, I am going to ask Audyssey and see if they would answer the question. Regardless, it doesn't matter either way for what we are trying to achieve here, so for now let's assume the Editor app itself would create the filters.

About setting the EQ frequency range:

Again, Audyssey would create filters for the range that it determines the speaker is capable of. So if it found FL can do 30 to 20,000 Hz, say based on the -3 dB point, it would create filters for that range. Now if you set the limit to below 300 Hz, then when you send the curve to the AVR, it will only send the filter set that EQs up to 300 Hz for that speaker/channel. Note: now I am assuming the App itself created the filter.

Back to what I am trying to do and what I need:

For now, let's focus on FL, FR, and subs only, I am asking you to leave the minidsp as is so it is doing what it has been doing to your satisfaction. So when you re-run Audyssey, Audyssey would create filters to EQ what has already been EQ by minidsp. When you run REW, you will then capture the overall/combined results of the minidsp's effects and Audyssey's. And this curve, the combined effects of the mini and Audyssey is the one that I want to use as my "reference" curve for calculating the cuts and boosts to further flatten your original curve (by original I mean the one you are getting, and seem happy about it, but you asked for input).

I have to go out for a few hour now, if you have more questions I will try to answer later.

 

[ciotime]

Let's be clear about the following when you use the Editor app to run Audyssey:

1). Audyssey will EQ all speakers in the range that it determines the speakers are capable of reproducing.
2) The App would create target curves based on the filters created (or to be created by the AVR when it receives the file/target curves).

The parts I am not sure is whether the App created the filters or it only sends the target curve and then the AVR/AVP would create the filters based on the target curve information, I am going to ask Audyssey and see if they would answer the question. Regardless, it doesn't matter either way for what we are trying to achieve here, so for now let's assume the Editor app itself would create the filters.

About setting the EQ frequency range:

Again, Audyssey would create filters for the range that it determines the speaker is capable of. So if it found FL can do 30 to 20,000 Hz, say based on the -3 dB point, it would create filters for that range. Now if you set the limit to below 300 Hz, then when you send the curve to the AVR, it will only send the filter set that EQs up to 300 Hz for that speaker/channel. Note: now I am assuming the App itself created the filter.

Back to what I am trying to do and what I need:

For now, let's focus on FL, FR, and subs only, I am asking you to leave the minidsp as is so it is doing what it has been doing to your satisfaction. So when you re-run Audyssey, Audyssey would create filters to EQ what has already been EQ by minidsp. When you run REW, you will then capture the overall/combined results of the minidsp's effects and Audyssey's. And this curve, the combined effects of the mini and Audyssey is the one that I want to use as my "reference" curve for calculating the cuts and boosts to further flatten your original curve (by original I mean the one you are getting, and seem happy about it, but you asked for input).

I have to go out for a few hour now, if you have more questions I will try to answer later.

Thanks for the thorough explanation.
"Now if you set the limit to below 300 Hz, then when you send the curve to the AVR, it will only send the filter set that EQs up to 300 Hz for that speaker/channel."-Yes...the XT32 app will only EQ up until where you set the limit. If I don't set a limit then I suppose that the app will apply the necessary EQ up to 20KHz. The reason I always limit the EQ range only until around 300Hz-400Hz is coz all of the Audyssey app tutorials I've seen tell you NOT to EQ above the Schroeder Frequency. But if you wanna see what Audyssey does for the full range then I can take a REW measurement with NO filters applied.

 

[PENG]

Thanks for the thorough explanation.
"Now if you set the limit to below 300 Hz, then when you send the curve to the AVR, it will only send the filter set that EQs up to 300 Hz for that speaker/channel."-Yes...the XT32 app will only EQ up until where you set the limit. If I don't set a limit then I suppose that the app will apply the necessary EQ up to 20KHz. The reason I always limit the EQ range only until around 300Hz-400Hz is coz all of the Audyssey app tutorials I've seen tell you NOT to EQ above the Schroeder Frequency. But if you wanna see what Audyssey does for the full range then I can take a REW measurement with NO filters applied.

Okay, it works like this:

Running Audyssey and setting the EQ freq range limit are two different thing.

Audyssey will do the filters for the whole range it sees fit without being affected by the limit. In fact, you can only se the limit after fact (you know that already obviously).

Setting the limit is after the fact, so by setting it, when you send the file/curve to your AVR/AVP, it will only include the filters for the range according to the limit you set.

If later on you remove the limit, that is, revert back to "full range", you can just do it in the App and send the file again, there is no need to re-run Audyssey as the filters are always there, its only a question of sending which part of the filter set according to the settings you make with the App.

For the purpose of our little project, it doesn't matter if you limit the range to below Schroeder or not before you send the file/target curve to the AVR, but since you prefer the effects of limiting EQ to 300 or 400 Hz, it is better if you do just that, before sending the file.

I would like to see the REW plots for the FL+FR+Subs for the whole range, and also for the range limited to 300 or 400 Hz. That's a function of REW, where you choose the range, scales etc., it has nothing to do with Audyssey. My focus will be to add cuts and boost (mainly just cuts) to the range below 200 Hz anyway, the reason I want to see one REW plot for full range, or at least up to 3,000 Hz is just so once we are done with the below 200 Hz range, we know what to do to restore overall balance between the low and high frequencies.

So again, to be clear, when you run Audyssey, it will EQ full range (the range it sees fit..), limiting it to the Schroeder frequency of your room is something you do manually with the App after the fact, and you can change it to anything any time after without having to run Audyssey again.

 

[ciotime]

Okay, it works like this:

Running Audyssey and setting the EQ freq range limit are two different thing.

Audyssey will do the filters for the whole range it sees fit without being affected by the limit. In fact, you can only se the limit after fact (you know that already obviously).

Setting the limit is after the fact, so by setting it, when you send the file/curve to your AVR/AVP, it will only include the filters for the range according to the limit you set.

If later on you remove the limit, that is, revert back to "full range", you can just do it in the App and send the file again, there is no need to re-run Audyssey as the filters are always there, its only a question of sending which part of the filter set according to the settings you make with the App.

For the purpose of our little project, it doesn't matter if you limit the range to below Schroeder or not before you send the file/target curve to the AVR, but since you prefer the effects of limiting EQ to 300 or 400 Hz, it is better if you do just that, before sending the file.

I would like to see the REW plots for the FL+FR+Subs for the whole range, and also for the range limited to 300 or 400 Hz. That's a function of REW, where you choose the range, scales etc., it has nothing to do with Audyssey. My focus will be to add cuts and boost (mainly just cuts) to the range below 200 Hz anyway, the reason I want to see one REW plot for full range, or at least up to 3,000 Hz is just so once we are done with the below 200 Hz range, we know what to do to restore overall balance between the low and high frequencies.

So again, to be clear, when you run Audyssey, it will EQ full range (the range it sees fit..), limiting it to the Schroeder frequency of your room is something you do manually with the App after the fact, and you can change it to anything any time after without having to run Audyssey again.

Ok...anyway's on the last graph I posted it was the FL+FR+Subs for the range limited to 400Hz...both with Audyssey ON and OFF but with the Filter Freq Range set to the lowest (20Hz) on the sub. I'll send one WITHOUT the limit's for the sub. I've posted it again below.

 

[ciotime]

Okay, it works like this:

Running Audyssey and setting the EQ freq range limit are two different thing.

Audyssey will do the filters for the whole range it sees fit without being affected by the limit. In fact, you can only se the limit after fact (you know that already obviously).

Setting the limit is after the fact, so by setting it, when you send the file/curve to your AVR/AVP, it will only include the filters for the range according to the limit you set.

If later on you remove the limit, that is, revert back to "full range", you can just do it in the App and send the file again, there is no need to re-run Audyssey as the filters are always there, its only a question of sending which part of the filter set according to the settings you make with the App.

For the purpose of our little project, it doesn't matter if you limit the range to below Schroeder or not before you send the file/target curve to the AVR, but since you prefer the effects of limiting EQ to 300 or 400 Hz, it is better if you do just that, before sending the file.

I would like to see the REW plots for the FL+FR+Subs for the whole range, and also for the range limited to 300 or 400 Hz. That's a function of REW, where you choose the range, scales etc., it has nothing to do with Audyssey. My focus will be to add cuts and boost (mainly just cuts) to the range below 200 Hz anyway, the reason I want to see one REW plot for full range, or at least up to 3,000 Hz is just so once we are done with the below 200 Hz range, we know what to do to restore overall balance between the low and high frequencies.

So again, to be clear, when you run Audyssey, it will EQ full range (the range it sees fit..), limiting it to the Schroeder frequency of your room is something you do manually with the App after the fact, and you can change it to anything any time after without having to run Audyssey again.

Hi Peng..just took the FR. This is with Audyssey ON and OFF. FL+FR+Subs. All 3 have NO limits set on the Filter Freq Range. No smoothing applied as well.
I noticed that compared to the FR with Audyssey ON and with the SUBs limited to NO EQ ( Filter Freq Range set to 20Hz ) this one with NO LIMITS ( SUBs EQ up to 250Hz ) this one has a huge dip at about 58Hz. Plus Bass starts to go down at 25Hz. So I guess the miniDsp was doing it's job much better?

 

[PENG]

I noticed that compared to the FR with Audyssey ON and with the SUBs limited to NO EQ ( Filter Freq Range set to 20Hz ) this one with NO LIMITS ( SUBs EQ up to 250Hz ) this one has a huge dip at about 58Hz. Plus Bass starts to go down at 25Hz. So I guess the miniDsp was doing it's job much better?

Not really, consider the following:

1) You are now looking at the combined effects so you can't logically compare which one by itself is doing a better job, but of course the mini could have done some good in some ways, such as time aligning 3 subwoofers.

2) You can't go by that 58 Hz dip, because as you may know, we are not sensitive to such sharp dips, SVS will tell that for sure, as well as authoritative experts such as Dr. Toole.

So now look at this anyway, with Audyssey on, between 18 Hz and 120 Hz, you are within 5 dB peak to peak, or arguable +/- 2.5 dB. Compared that to with Audyssey limited to only 20 Hz, you are getting a 14 dB peak to peak variance, but a lot of that is obviously due to the boost you made in order to get that custom curve, but even without that, you would likely still have more than 5 dB peak to peak within the same 18 to 120 Hz.

With these curves, I already know you can use the Editor/Rat to improve what you already have by a couple dB for sure. Let me play with it this weekend and get you something hopefully useful.

 

[ciotime]

Not really, consider the following:

1) You are now looking at the combined effects so you can't logically compare which one by itself is doing a better job, but of course the mini could have done some good in some ways, such as time aligning 3 subwoofers.

2) You can't go by that 58 Hz dip, because as you may know, we are not sensitive to such sharp dips, SVS will tell that for sure, as well as authoritative experts such as Dr. Toole.

So now look at this anyway, with Audyssey on, between 18 Hz and 120 Hz, you are within 5 dB peak to peak, or arguable +/- 2.5 dB. Compared that to with Audyssey limited to only 20 Hz, you are getting a 14 dB peak to peak variance, but a lot of that is obviously due to the boost you made in order to get that custom curve, but even without that, you would likely still have more than 5 dB peak to peak within the same 18 to 120 Hz.

With these curves, I already know you can use the Editor/Rat to improve what you already have by a couple dB for sure. Let me play with it this weekend and get you something hopefully useful.

Oh yeah forgot that sharp dips are not really that problematic as compared to big wide dips/nulls. That wide dip/null from 50hz-80hz with Audyssey OFF is a good example. Thanks again...will be looking forward to that.