Freq Responses graph

[ciotime]

I've attached below a new Freq Response taken using REW with 3 subs via the miniDSP at the MLP. I was able to get a +- 3dB across 3 seats from 15Hz to 100Hz ( not posted ). The FR below in the sub region 15Hz-300Hz is with NO smoothing. The FR from 20Hz-3KHz is with 1/12th smoothing taken after calibration/integration with the mains and turning on Audyssey . I expected the peaks and nulls to come out. I'm just hoping for some input on what you guys think of the FR from 100hz .Room is treated with about 15% absorption. Thanks.

 

[PENG]

The FR for the 20-100 Hz looks quite good but I am sure you can improve it further if you wish. From there to 3 kHz looks quite typical of those posted by others on this and other forums. At higher than 200 Hz EQ won't help much as it has more to do with the speakers than the room. Even if you manage to flatten it at the mid to high range, it doesn't mean it would translate into audibly better (or worse) "sound quality" according to Dr. Toole, who suggested EQ above the room transition frequency are generally not necessary and the results may not be reliable anyway. So you probably should focus on perfecting the below 200 Hz range.

As I mentioned many times before for +/- 3 to 4 dB for that range is easily obtainable without using minidsp if you have Audyssey XT32 even without manual tweaking, or +/- 1 dB (1/12 smoothing) with tweaking using the Editor App. For users of YPAO, the minidsp may help but I really don't have experience with the latest YPAO version. I would imagine it would be the same with the also popular AARC and Dirac Live that many believe can do a better job.

So every time I see graphs like those I can't help but wonder why people would bother using minidsp unless there are other reasons than just a flat to customize FR in that range. Yours just confirm yet again my own experience. I really don't mean to pour cold water on REQ enthusiast projects, just tell the truth based on my experience. The use of 3 or 4 subs is not an issue really in practice, and I have been using between 1 and 5 subs in the past 10 or more years. The minidsp may be good to have in many applications when the device don't have build in REQ such as those 3 or 4 mentioned.

 

[ciotime]

The FR for the 20-100 Hz looks quite good but I am sure you can improve it further if you wish. From there to 3 kHz looks quite typical of those posted by others on this and other forums. At higher than 200 Hz EQ won't help much as it has more to do with the speakers than the room. Even if you manage to flatten it at the mid to high range, it doesn't mean it would translate into audibly better (or worse) "sound quality" according to Dr. Toole, who suggested EQ above the room transition frequency are generally not necessary and the results may not be reliable anyway. So you probably should focus on perfecting the below 200 Hz range.

As I mentioned many times before for +/- 3 to 4 dB for that range is easily obtainable without using minidsp if you have Audyssey XT32 even without manual tweaking, or +/- 1 dB (1/12 smoothing) with tweaking using the Editor App. For users of YPAO, the minidsp may help but I really don't have experience with the latest YPAO version. I would imagine it would be the same with the also popular AARC and Dirac Live that many believe can do a better job.

So every time I see graphs like those I can't help but wonder why people would bother using minidsp unless there are other reasons than just a flat to customize FR in that range. Yours just confirm yet again my own experience. I really don't mean to pour cold water on REQ enthusiast projects, just tell the truth based on my experience. The use of 3 or 4 subs is not an issue really in practice, and I have been using between 1 and 5 subs in the past 10 or more years. The minidsp may be good to have in many applications when the device don't have build in REQ such as those 3 or 4 mentioned.

Thanks for the response again Peng. I actually did try using just the Audyssey XT32 app without the minidsp and I could just not get the FR I wanted. There were differences in the bass region. I did limit the Freq Range of the Audyssey XT32 to EQ only until 300Hz as this is about when the Schroeder frequency is in my room. I've attached some graphs below that show the FR across 3 seats with and without the miniDSp.

 

[PENG]

Thanks for the response again Peng. I actually did try using just the Audyssey XT32 app without the minidsp and I could just not get the FR I wanted. There were differences in the bass region. I did limit the Freq Range of the Audyssey XT32 to EQ only until 300Hz as this is about when the Schroeder frequency is in my room. I've attached some graphs below that show the FR across 3 seats with and without the miniDSp.

View attachment 49027View attachment 49028

If the graphs were plotted with full range towers, try setting XO to 80Hz that in my experience would yield much better results than with XO 60Hz but obviously it depends a lot on the room too.

Question, I don't remember if yours is compatible with the app, if it is, have you tried using it with Ratbuddyssey yet?

 

[ciotime]

If the graphs were plotted with full range towers, try setting XO to 80Hz that in my experience would yield much better results than with XO 60Hz but obviously it depends a lot on the room too.

Question, I don't remember if yours is compatible with the app, if it is, have you tried using it with Ratbuddyssey yet?

Speakers are towers but set to small with an 80 Xover. After calibration Audyssey always sets the speakers to Large with a 40Hz Xover but I always change it to Small/80 Xover. Yes my system is compatible and yes I have used Ratbuddyssey to set my house curve but never tried it in the 100Hz-200Hz region.

 

[PENG]

Speakers are towers but set to small with an 80 Xover. After calibration Audyssey always sets the speakers to Large with a 40Hz Xover but I always change it to Small/80 Xover. Yes my system is compatible and yes I have used Ratbuddyssey to set my house curve but never tried it in the 100Hz-200Hz region.

That's good news. If you are willing to spend an hour on the App/Rat, you can try using 75 dB as reference and should be able to flatten the FR a lot just using cuts where those humps/bumps are from 20 Hz to 400 Hz, likely effective up to 600 Hz or even 1 kHz. Try not to use any boosts at all especially if the minidsp/REW has already made some boosts. After that you can easily tweak further to your taste.

 

[ciotime]

That's good news. If you are willing to spend an hour on the App/Rat, you can try using 75 dB as reference and should be able to flatten the FR a lot just using cuts where those humps/bumps are from 20 Hz to 400 Hz, likely effective up to 600 Hz or even 1 kHz. Try not to use any boosts at all especially if the minidsp/REW has already made some boosts. After that you can easily tweak further to your taste.

So say I have a peak at 200Hz that's 80dB...I go into the Rat app on my PC and enter 200Hz/-5dB? And I do the same for the other Freq that have peaks of +5dB or more?
When loading the calibration file Audyssey XT32 results into RAT on the PC you can see each individual FR of ALL the speakers...so I need to CUT and make adjustments in each individual speaker?

 

[PENG]

So say I have a peak at 200Hz that's 80dB...I go into the Rat app on my PC and enter 200Hz/-5dB? And I do the same for the other Freq that have peaks of +5dB or more?
When loading the calibration file Audyssey XT32 results into RAT on the PC you can see each individual FR of ALL the speakers...so I need to CUT and make adjustments in each individual speaker?

That is correct, and you can use as many anchor points as you see fit. If you enter a cut of 5 dB, that is -5, you may see exactly 5 dB drop but it will be close.

Since in your case the mini is apparently doing something good so just let it do its part but use Audyssey+App/Rat on top to help fine tuning. As you know, I don't use the mini anymore so for me I picked my anchor points based on the original REW plot post Audyssey calibration as reference and if I had to make a second Rat attempt, I would still go back to that original reference curve as I found that's the best way to perfect the final outcome. In your case, I suspect you can simply use your existing curve as reference and calculate the required cuts based on that curve. You existing curve is already as good as others posted on sites such as AVSF, but if you have time, it may be fun trying to make it look even better.

 

[ciotime]

That is correct, and you can use as many anchor points as you see fit. If you enter a cut of 5 dB, that is -5, you may see exactly 5 dB drop but it will be close.

Since in your case the mini is apparently doing something good so just let it do its part but use Audyssey+App/Rat on top to help fine tuning. As you know, I don't use the mini anymore so for me I picked my anchor points based on the original REW plot post Audyssey calibration as reference and if I had to make a second Rat attempt, I would still go back to that original reference curve as I found that's the best way to perfect the final outcome. In your case, I suspect you can simply use your existing curve as reference and calculate the required cuts based on that curve. You existing curve is already as good as others posted on sites such as AVSF, but if you have time, it may be fun trying to make it look even better.

The FR in REW is the FR of ALL the speakers correct? So I don't base my cuts on that but on the individual FR of each speaker as shown in the Graph View on the RAT app?
So ex I click on the Center speaker in the RAT app and take note of all the peaks that I wanna cut and apply that on the boxes and the same goes for all the other speakers?

 

[PENG]

The FR in REW is the FR of ALL the speakers correct? So I don't base my cuts on that but on the individual FR of each speaker as shown in the Graph View on the RAT app?
So ex I click on the Center speaker in the RAT app and take note of all the peaks that I wanna cut and apply that on the boxes and the same goes for all the other speakers?

That's what I would do first, that is on individual basis and see what happens, then go from there..
By the way, I should have mentioned that even for cuts, it is probably better to do it gradually. For example, if you think you need a 4 dB cut at 180 Hz, try cutting just 2 dB first and see the effect and then make another cut accordingly in your second attempt. The process works but still involve some trial and error, no hard and fast rules.

 

[3db]

I've attached below a new Freq Response taken using REW with 3 subs via the miniDSP at the MLP. I was able to get a +- 3dB across 3 seats from 15Hz to 100Hz ( not posted ). The FR below in the sub region 15Hz-300Hz is with NO smoothing. The FR from 20Hz-3KHz is with 1/12th smoothing taken after calibration/integration with the mains and turning on Audyssey . I expected the peaks and nulls to come out. I'm just hoping for some input on what you guys think of the FR from 100hz .Room is treated with about 15% absorption. Thanks.

View attachment 49025View attachment 49026

I'm using a miniDSP in two systems, both with 2 subs, adding a 3rd sub to the greatroom which is an acoustic nightmare to tame. Maybe Im wrong but it looks like you put in a house curve with your mini. I did the same thing and really enjoy it. Packs a really nice wallop for movies but music remains relatively untouched as there is not much music content below 30 Hz .

 

[ciotime]

I'm using a miniDSP in two systems, both with 2 subs, adding a 3rd sub to the greatroom which is an acoustic nightmare to tame. Maybe Im wrong but it looks like you put in a house curve with your mini. I did the same thing and really enjoy it. Packs a really nice wallop for movies but music remains relatively untouched as there is not much music content below 30 Hz .

Yep I did apply my own house curve via the miniDsp.

 

[3db]

Yep I did apply my own house curve via the miniDsp.

The miniDSP is more effective in cutting than boosting the dbs. I dont know if Auddessy can give you a house curve like the miniDSP and it appears to be a lot more involved in doing so. I did try a flat frequency response verses a house curve and the house curve sounds much better to my ears. YMMV.

 

[ciotime]

The miniDSP is more effective in cutting than boosting the dbs. I dont know if Auddessy can give you a house curve like the miniDSP and it appears to be a lot more involved in doing so. I did try a flat frequency response verses a house curve and the house curve sounds much better to my ears. YMMV.

Flat response is boring...most prefer an upward trend in the bass region. Even if you have a flat response turning on DEQ will also give you an upward tilt on the lower freq. I even have a setup with both using a house curve and the DEQ on. Felt like having mini buttkickers .

 

[PENG]

Flat response is boring...most prefer an upward trend in the bass region. Even if you have a flat response turning on DEQ will also give you an upward tilt on the lower freq. I even have a setup with both using a house curve and the DEQ on. Felt like having mini buttkickers .

I just thought of something.. Assuming the curves you posted were for the FL+FR+Sub(s), if you want to give the App+Rat tweak another chance you could/should leave things the way they are now and use the App to run Audyssey again. The App allows you to store a great number of curves so if things get worse you can always load the original one back to the AVR.

With the fresh run, you can then us the App/Rat to customize the curve from the new virgin post Audyssey curve. In fact, if you post the new REW curve, no smoothing, I can use it to set up a table for you to enter the values to the App. I am sure you already know how to do it, but since I have spent so many hours playing with it, my table may just help you save some time.

With those cuts/boosts values you will get a much flatter curve. From there you can easily set up your own "buttkickers" curve(s).

 

[ciotime]

I just thought of something.. Assuming the curves you posted were for the FL+FR+Sub(s), if you want to give the App+Rat tweak another chance you could/should leave things the way they are now and use the App to run Audyssey again. The App allows you to store a great number of curves so if things get worse you can always load the original one back to the AVR.

With the fresh run, you can then us the App/Rat to customize the curve from the new virgin post Audyssey curve. In fact, if you post the new REW curve, no smoothing, I can use it to set up a table for you to enter the values to the App. I am sure you already know how to do it, but since I have spent so many hours playing with it, my table may just help you save some time.

With those cuts/boosts values you will get a much flatter curve. From there you can easily set up your own "buttkickers" curve(s).

The curves I posted is the FR by REW. Im not sure if its both the FL/FR with subs or just the FL with subs.

Yes I can run Audyssey again. I don't quite understand what you mean by "enter the values to the App"...you mean enter the values in the Rat app? If so there are different values for each individual speakers.

 

[PENG]

The curves I posted is the FR by REW. Im not sure if its both the FL/FR with subs or just the FL with subs.

Yes I can run Audyssey again. I don't quite understand what you mean by "enter the values to the App"...you mean enter the values in the Rat app? If so there are different values for each individual speakers.

I meant enter the values to the App via Ratbuddyssey, so yes I should have said "in the Rat UI..", the Rat is not an App, its is just an user interface (UI) for use with the MultEQ Editor "App".

The link below should give you an idea of how I did it using the REW ref graphs and Excel (at least at that time):

The Audyssey MultEQ Editor app users thread (with facts and tips) | Page 8 | Audioholics Home Theater Forums

 

[ciotime]

I meant enter the values to the App via Ratbuddyssey, so yes I should have said "in the Rat UI..", the Rat is not an App, its is just an user interface (UI) for use with the MultEQ Editor "App".

The link below should give you an idea of how I did it using the REW ref graphs and Excel (at least at that time):

The Audyssey MultEQ Editor app users thread (with facts and tips) | Page 8 | Audioholics Home Theater Forums

Thanks...so when I run REW with Audyssey in reference mode I should choose L+R under the output master volume? Or just L?

 

[PENG]

Thanks...so when I run REW with Audyssey in reference mode I should choose L+R? Or just L?

For me, I would always run all of the following:

- Subwoofers only.
- FL only.
- FR only.
- FL+FR.
- FL+FR+Subs.

In each case I would run it for both Audyssey ref and Audyssey off.

I would save them all in a folder so that I can easily find the reference curves whenever I feel like to tweak with the App/Rat.

In the past I would even do it for the surrounds but its getting so ridiculous now that I have 11 channels to play with so I don't bother any more. As far as I am concerned the surround channels are for ambient effects, it makes little difference one way or the other.

 

[ciotime]

Ok so I'll run the XT32 app again and take a FR with REW after. So what's the REW curve, no smoothing, that you need from me to set up a table to enter the values to the Rat App? All of these?
- Subwoofers only.
- FL only.
- FR only.
- FL+FR.
- FL+FR+Subs.