Multi-Sub Optimizer

[baMarek]

I recently came across a new software called Multi-Sub Optimizer (MSO) and I believe it deserves some attention. Basically it is designed to help with setting up multiple subwoofers not only to make SPL vs frequency graph as flat as possible but it also allows for making it flatter in multiple listening positions. All this is possible because MSO unlike other known to me programs makes multiple subs work independently. It is not a surprise that highs and lows generated by one source, reflected or not, can be tamed by other sources. MSO uses some funky formulas to come up with different filters for each subwoofer which played together make the sound free from spikes and dips caused by speaker design and room modes. Well, this is theory at least but I hope we will soon be able to verify if it works as claimed.

 

The software is free and can be downloaded together with a great tutorial from here: http://andyc.diy-audio-engineering.org/mso/html/index.html

 

To make it even sweeter, the guy who created this software, @andyc56 found out we were talking about it in a different thread and signed up to SNA just to answer our questions. Respect Andy! As you are from Texas I am sure communication will not be a problem...we are also in the '80s here 

 

I don't claim to be an expert of advanced acoustic measurements or laws of physics used by MSO but I am willing to give it a try and see if I can get better results. I will post my measurements, subjective listening impressions and share any issues and thoughts. Please note, I am totally opposed to using just one sub for stereo so my views may be biased.

 

Marek

Edited September 8, 2016 by baMarek

[baMarek]

My trial system consists of ESL 57 electrostatic speakers and two dipole 15" subwoofers. All fed from one multichannel DAC with Jriver based crossover at 200Hz/24dB. Without going crazy with EQ and with my work-in-progress DAC setup I get the results below. This will be my benchmark. For the purpose of testing MSO I will start without any filters applied to the subs except for the crossover at 200Hz.

 

I have just started making my first measurements and came across issues with time alignment. I noticed a while ago that HOLMimpulse doesn't give consistent impulse response data and cannot be used for time alignment at least when a measuring microphone is connected via USB. The only information I found about it (mentioned in MSO's tutorial) is that it is impossible for a system where both the record and the output use USB to give time aligned response. Why? I have no idea, but I suspect the problem is not related to USB mic only as I have already seen reports of odd behaviour of systems using more than one USB output. All I can confirm at the moment is that using asynchronous DACs in multiple DACs system will always create time alignment problems regardless of the driver you use.

 

Here are 3 measurements done using HOLMimpulse with time zero lock function activated, USB mic and USB output to DAC. As you can see the results are horribly inconsistent and for now I'm gonna say bye bye to Holmimpulse:

 

Supposedly REW software can handle time alignment better. It sends a high frequency sweep before doing a custom sweep to let software know when to expect the actual signal (time delay between test sweep and actual sweep is always the same). I tested it and I see a problem with time alignment again, although at much lower scale. My test procedure is as follows:

- After the measurements I click Impulse, go to Controls and click Estimate IR delay. No measurement gives me the same value and I get anything between 0.307ms to 0.323ms. It is only 1cm distance but it is not perfect. I don't know if the problem is with the software, USB drivers, my PC, DAC or anything else. 

 

Second problem is with using REW itself. Because REW uses very high frequency sweep to create a time reference for the main sweep I can't see how one can align the timing of subs vs mains. Measuring the time delay specifically for subs seems impossible in REW, at least when Acoustic Time Reference is used for USB mic. @andyc56, I can give MSO some data with consistent time delay between subs and mains but will I not get better results if I first do the time alignment and only then produce the sweeps to be used by MSO?

 

Edited September 8, 2016 by baMarek

[andyc56]

8 hours ago, baMarek said:

Please note, I am totally opposed to using just one sub for stereo so my views may be biased.

Does this mean you are using stereo subs?  If so, MSO does not apply to that situation, as it assumes the signal to the subs is a mono signal derived from the sum of the main channel signals.  I've updated the hardware requirements paragraph on the main page to clarify that.  MSO is geared toward HT-style bass management, where the crossover frequency is not much more than 100 Hz.

 

If your subs are indeed stereo and you were to change them to mono while retaining the 200 Hz crossover frequency, that would cause problems with sub localization.  I would not recommend the use of MSO in that situation.  In addition, you'd start to run into MSO's absolute maximum limits on things like PEQ center frequency.  MSO is only meant to alter response in the modal region, and a crossover frequency of 200 Hz is in a transitional area between the modal and statistical regions of a typical room in a home.

 

Edited September 8, 2016 by andyc56
Original information was incomplete

[andyc56]

5 hours ago, baMarek said:

Supposedly REW software can handle time alignment better. It sends a high frequency sweep before doing a custom sweep to let software know when to expect the actual signal (time delay between test sweep and actual sweep is always the same). I tested it and I see a problem with time alignment again, although at much lower scale. My test procedure is as follows:

- After the measurements I click Impulse, go to Controls and click Estimate IR delay. No measurement gives me the same value and I get anything between 0.307ms to 0.323ms. It is only 1cm distance but it is not perfect. I don't know if the problem is with the software, USB drivers, my PC, DAC or anything else. 

 

If I understand what you're doing correctly, the acoustic timing reference can't be used to measure absolute delay, only delay relative to the speaker used as a reference.

 

Quote

Second problem is with using REW itself. Because REW uses very high frequency sweep to create a time reference for the main sweep I can't see how one can align the timing of subs vs mains. Measuring the time delay specifically for subs seems impossible in REW, at least when Acoustic Time Reference is used for USB mic. @andyc56, I can give MSO some data with consistent time delay between subs and mains but will I not get better results if I first do the time alignment and only then produce the sweeps to be used by MSO?

 

I'm not privy to the specifics of REW's internal implementation of the acoustic timing reference.  All that matters for MSO is that the relative delays of all the speakers measured, both mains and subs, be preserved with respect to one another at a given listening position.  There is nothing to be gained in MSO from using measurements whose absolute delays are preserved.

Edited September 8, 2016 by andyc56

[baMarek]

13 hours ago, andyc56 said:

Does this mean you are using stereo subs?  If so, MSO does not apply to that situation, as it assumes the signal to the subs is a mono signal derived from the sum of the main channel signals.  I've updated the hardware requirements paragraph on the main page to clarify that.  MSO is geared toward HT-style bass management, where the crossover frequency is not much more than 100 Hz.

 

If your subs are indeed stereo and you were to change them to mono while retaining the 200 Hz crossover frequency, that would cause problems with sub localization.  I would not recommend the use of MSO in that situation.  In addition, you'd start to run into MSO's absolute maximum limits on things like PEQ center frequency.  MSO is only meant to alter response in the modal region, and a crossover frequency of 200 Hz is in a transitional area between the modal and statistical regions of a typical room in a home.

 

Yes I was planning on using stereo subs as I thought MSO could handle both mono and stereo. With Jriver as my crossover I can obviously change the signal to mono but I can't keep the crossover frequency at 200Hz. It is a compromise but even in mono I prefer two subs over one so I will give it a go and drop the crossover frequency to 90Hz to start with.

[baMarek]

14 hours ago, andyc56 said:

 

If I understand what you're doing correctly, the acoustic timing reference can't be used to measure absolute delay, only delay relative to the speaker used as a reference.

 

I'm not privy to the specifics of REW's internal implementation of the acoustic timing reference.  All that matters for MSO is that the relative delays of all the speakers measured, both mains and subs, be preserved with respect to one another at a given listening position.  There is nothing to be gained in MSO from using measurements whose absolute delays are preserved.

 

The value I measured in REW, i.e. about 0.3ms, had no significance, but the fact that I get different results each time I measure is worrying. Acoustic timing reference is used to ignore the delay caused by the system (which can vary and can be tested using loopback feature). The delay between initial high frequency sweep and the actual sweep is the same, it's like start and finish of a song, so if the values measured by REW differ then I am not sure I can fully trust it. Anyway, I will keep the timing reference on and see where it takes me.

Edited September 9, 2016 by baMarek

[BradC]

I have also been experimenting with MSO recently.

My aim is to optimise a pair of mono subs at the front and rear wall midpoints. I have inverted and delayed the rear sub with the aim to cancel the longitudinal modes. Manual measurements in REW show that I can get effective reduction in the 2,0,0 mode amplitude and decay time.

Running MSO with the same data gives a somewhat different result. Ie it's flatter but has more ringing.

So @andyc56 have you considered including the amount of ringing in your metric? Ie build a fitness function that calculates frequency 'flatness' at multiple delays

 

Brad

[andyc56]

It looks to me like you're trying out a variant of the so-called "double bass array" (DBA) idea as described in this post by Todd Welti.  I'm not sure how well DBA will work with only two subs though, as you can't get the plane wave that a true DBA can come close to achieving.

 

At any rate, whenever you reverse the polarity of some subs and not others in a non-DBA setup, you run a greater risk of non-minimum phase behavior than with a more conventional setup in which they are all in the same polarity.  This gives a greater risk of degraded time domain behavior.

 

I have no plans to incorporate time-domain behavior into MSO's optimization.  Doing so would require not just a change in the fitness function, but also a change from a single-objective to a multi-objective optimizer.  The best way to minimize the risk of strange time-domain behavior is to keep all the subs in the same polarity and not use any boost in the parametric EQ.  The latter approach ensures that MSO is constrained to fill in any notches in response by controlling the interaction of the subs rather than boosting the response.

[Snoopy8]

New to this forum, refugee from DTVForum (long time user) which is shutting down soon... 

 

Am surprised that there has not been more interest in MSO here.  Found MSO a week ago as part of research to put in dual dissimilar subs.  Have an open plan family room with some curtains, soft furnishings and very limited choices for placement of speakers and subs.  WAF considerations preclude further room treatments.

 

The results have been excellent with  stereo 2.2 setup showing tighter and more controlled bass than old 2.1 setup.  The home theatre 5.2 setup running with Anthem Room Correction (ARC) + MSO optimised subs showed similar improvements, and much better than ARC with non-optimised subs.  And without doubt, the variation between seating positions has dramatically improved for both stereo and home theatre.

 

What was most impressive was the easy-to-use software and excellent documentation.   Had no prior experience with dual subs, and within a week, had them integrated and running well.  Thank you Andy... 

[andyc56]

38 minutes ago, Snoopy8 said:

New to this forum, refugee from DTVForum (long time user) which is shutting down soon... 

 

Am surprised that there has not been more interest in MSO here.  Found MSO a week ago as part of research to put in dual dissimilar subs.  Have an open plan family room with some curtains, soft furnishings and very limited choices for placement of speakers and subs.  WAF considerations preclude further room treatments.

 

The results have been excellent with  stereo 2.2 setup showing tighter and more controlled bass than old 2.1 setup.  The home theatre 5.2 setup running with Anthem Room Correction (ARC) + MSO optimised subs showed similar improvements, and much better than ARC with non-optimised subs.  And without doubt, the variation between seating positions has dramatically improved for both stereo and home theatre.

 

What was most impressive was the easy-to-use software and excellent documentation.   Had no prior experience with dual subs, and within a week, had them integrated and running well.  Thank you Andy... 

 

You're quite welcome!  I've included a link to your review on the MSO home page as a second opinion about whether using MSO might be worthwhile with less than three subs.  I've been conservative about my recommendation for the minimum number of subs, as I don't wish to build up high expectations, only for users having less than three to find that MSO might not provide a substantial improvement. That's because there's a lot of work involved to perform all the necessary measurements.

[Snoopy8]

4 hours ago, andyc56 said:

 

You're quite welcome!  I've included a link to your review on the MSO home page as a second opinion about whether using MSO might be worthwhile with less than three subs.

Andy, that forum is shutting down next month (hence I am taking refuge here!) and your link will disappear.  What I will do is take that review, modify it eg. remove references to software etc and re-post it here.  Then you can change your link to point here.

 

While I have your attention, suggest you add a sentence or 2 describing how MSO works with room correction software within your documentation.  There is some discussion within your AVSForum MSO thread and within the MiniDSP DDRC-88A apps note about using Sub-Only with room correction.  I used the DDRC-88A info to guide me on what to do for ARC (Anthem Room Correction).  I may have missed it, but I did not read any info on how to use room correction with MSO in your documentation.  Perhaps add it to your Optimising Sub-Only with a something similar to this: Sub-only optimisation is strongly (?) recommended when you use room correction software eg. Dirac, Audyssey, Anthem etc.  This is because room correction could potentially cancel out MSO's optmisation...

[Snoopy8]

The following review was posted on Dtvforum.info just before its closure (on 1 July 2017) was announced (**Edit: no longer closing, now taken over by this forum!) and has been reproduced here.  Additional comments added at the end.

------------------------------------------------------------------

Multi-Sub Optimiser (MSO) is a free software tool to optimize multiple sub woofers by integrating the subs with main speakers and improving the evenness of bass response across the whole listening area. The program was written by AndyC on AVS Forum (discussion thread hosts all discussions, support, bugs etc.)

 

The documentation states that MSO gives the best results with three or more subs but with two subs, the improvement may be limited. However, good results were obtained in my environment using 2 dissimilar subs.

 

Ideally, you should be familiar with REW and have a calibrated mic. And a digital signal processor (DSP) is required.

 

Room

An open plan family room, with limited room treatments (curtains, as much soft material as can be sneaked in!) and very limited choices for placement of speakers and subs. WAF considerations preclude further room treatments.

 

MSO for Stereo

Stereo was previously running as a 2.1 system using a SVS SB13U sub. After adding a second smaller SVS SB2000 sub and applying the MSO filters, the bass came across tighter and more controlled in the main listening position (MLP).  Even positions to the left and right of MLP are much better than before.

 

The stereo setup has improved significantly; am (re)discovering the bass in my music library! 

 

MSO for Home Theatre

Currently using Anthem Room Correction (ARC) for HT audio and multi-channel music.  However, ARC cannot compensate for spatial variation between seating.

 

The previous 5.1 setup used the SB13U.   The new 5.2 setup, with the addition of the SB2000 and MSO, gave a noticeable better bass response for movies. MSO+ARC did an excellent job of integrating the subs, better than ARC alone. What was impressive was the evenness of bass response across the listening area.

 

For multi-channel music, tracks which had a significant amount of bass came across better with MSO.  And more uniform bass across the different listening positions.

 

Conclusion

MSO is a free, easy-to-use tool to optimise multiple subs. MSO with 2 subs has significantly improved my stereo setup and movies, and to a lesser extent multi-channel music.  What is impressive is the evenness of bass response across the listening area.  Delighted with this upgrade of a second sub and MSO.

------------------------------------------------------------------

MSO is an easy-to-use software tool with excellent documentation.   Used MSO to help place the 2 subs and found that having both in the front, inside the left and right mains, gave the best results (and improved WAF!).  Had no prior experience with dual subs, and within a week, MSO helped integrate them and made a marked improvement to the stereo and home theatre. 

------------------------------------------------------------------

***Edit2: AndyC (MSO author) referenced this post as an example of successfully using MSO with 2 subs.  To get the best of 2 subs, placement was important. Used quick MSO runs on various sub location combinations to help identify better locations, i.e. response curves with less pronounced troughs.

 

MSO did reasonable job of flattening the response curve with 2 subs.  For stereo, best result using mains+subs config, optimisation option as flat as possible.  For HT, used subs only config, optimisation option of best match of MLP to other positions. Tried various optimisation tweaks which sometimes created pronounced troughs beyond 150 hz.

 

With 2 subs, MSO did produce good results, with the following caveats.  Care should be put into sub placement and need to understand that MSO can only do so much.

 

Edited June 19, 2017 by Snoopy8
DTVForum no longer closing; more comments on 2 subs

[mcb]

I also use Multi-Sub Optimizer (MSO) with good results and have posted some comments at

For these contemplating using MSO, I strongly recommend trying it.  As with Snoopy, I use REW for the measurements, however I use an un-calibrated UMIK mic from Mini-DSP.

[Snoopy8]

1 hour ago, mcb said:

I also use Multi-Sub Optimizer (MSO) with good results and have posted some comments at

For these contemplating using MSO, I strongly recommend trying it.  As with Snoopy, I use REW for the measurements, however I use an un-calibrated UMIK mic from Mini-DSP.

Thank you.  New to forum, so missed that thread. 

 

Like you, my large family room with high ceiling is far from ideal.  Front is blank wall, right is large windows with blinds, rear is large glass sliding doors with heavy curtains, left is open to kitchen.  Was a nightmare to get better audio.  MSO plus a second sub has made a significant difference.  A third sub maybe in the longer term, but WAF a major factor.  So far used MSO Mains+subs for stereo, and subs only for HT. Using a Behringer DCX2496 to connect subs to stereo and HT.  May try some of the other optimisation strategies at a later time.

 

Used a Dayton Audio calibrated UMM-6 mic with REW.

[andyc56]

On 6/4/2017 at 3:36 AM, Snoopy8 said:

While I have your attention, suggest you add a sentence or 2 describing how MSO works with room correction software within your documentation.  There is some discussion within your AVSForum MSO thread and within the MiniDSP DDRC-88A apps note about using Sub-Only with room correction.  I used the DDRC-88A info to guide me on what to do for ARC (Anthem Room Correction).  I may have missed it, but I did not read any info on how to use room correction with MSO in your documentation.  Perhaps add it to your Optimising Sub-Only with a something similar to this: Sub-only optimisation is strongly (?) recommended when you use room correction software eg. Dirac, Audyssey, Anthem etc.  This is because room correction could potentially cancel out MSO's optmisation...

 

I'll put something in there about room correction.  The reason there isn't any help on that topic at present is that I'm a two-channel guy and have never actually used any commercial room correction at all, only manual PEQ.  So I can't write about it with any basis in experience.  I'm dependent on the experience of others.  Over at AVS forum, the only user who's discussed this in detail is AV_mike.  He has written a description of what he does, but he uses a "main+subs" configuration and runs MSO both before and after room correction.  I gather he's the only one using this procedure.

 

It's my understanding that systems like Audyssey can mess up the integration of main speakers and subs by getting the sub distance wrong, requiring the "sub distance tweak".  How do you deal with this problem when using commercial room correction and MSO together?

[Snoopy8]

3 hours ago, andyc56 said:

 

I'll put something in there about room correction.  The reason there isn't any help on that topic at present is that I'm a two-channel guy and have never actually used any commercial room correction at all, only manual PEQ.  So I can't write about it with any basis in experience.  I'm dependent on the experience of others. 

Have a look at video from Home Theatre Lounge which gives a good ARC Overview. Then read  Interview with ARC developers (Audioholics). Audioholics also did similar articles for Dirac and Audyssey. Hopefully this will give you more background when you write.

 

3 hours ago, andyc56 said:

It's my understanding that systems like Audyssey can mess up the integration of main speakers and subs by getting the sub distance wrong, requiring the "sub distance tweak".  How do you deal with this problem when using commercial room correction and MSO together?

For some reason, ARC does not calibrate the distances (others do).  User has to enter the distances of all the speakers. For multiple subs, use average distance. No Audyssey experience, so unable to comment on sub distance tweak. 

 

4 hours ago, andyc56 said:

 Over at AVS forum, the only user who's discussed this in detail is AV_mike.  He has written a description of what he does, but he uses a "main+subs" configuration and runs MSO both before and after room correction.  I gather he's the only one using this procedure.

AV_mike's approach puzzles me.  It over complicates things, a bit like chasing your own tail. 

 

When I first read it, the subs-only method suggested in the DDRC-88BM application note seemed sensible.  ARC treats multiple subs as a group and running MSO subs-only creates this integrated group.  I did a quick test of having the subs calibrated manually, re-ran ARC, but did not sound as good as MSO+ARC.  So for ARC, the subs-only approach is a valid one.

[andyc56]

1 hour ago, Snoopy8 said:

When I first read it, the subs-only method suggested in the DDRC-88BM application note seemed sensible.  ARC treats multiple subs as a group and running MSO subs-only creates this integrated group.  I did a quick test of having the subs calibrated manually, re-ran ARC, but did not sound as good as MSO+ARC.  So for ARC, the subs-only approach is a valid one.

 

I'll have a look at those references, thanks.

 

The sub distance tweak after using Audyssey (PDF) is done by performing repeated measurements at the MLP of combined main speakers and subs, trying out different sub distances in order to "optimize the splice" between main speakers and subs.

 

Are you saying that after running ARC, you just measure the average physical distance from the subs to the MLP and use that number without any further measurements with REW?  If that's the case, you might be able to get better integration with something like the Audyssey sub distance tweak.  Often, because of the additional electrical delay introduced by the LPF of the AVR's bass management, the optimum sub distance setting can be rather different from the sub's physical distance.

[Snoopy8]

24 minutes ago, andyc56 said:

The sub distance tweak after using Audyssey (PDF) is done by performing repeated measurements at the MLP of combined main speakers and subs, trying out different sub distances in order to "optimize the splice" between main speakers and subs.

 

Are you saying that after running ARC, you just measure the average physical distance from the subs to the MLP and use that number without any further measurements with REW?  If that's the case, you might be able to get better integration with something like the Audyssey sub distance tweak.

Correct, using a tape measure, measure the distance and enter it into the Anthem AVR.  Rather than use the average, probably a better way is to enter the distance to the sub that does not have the delay within the MSO sub-only configuration.

 

30 minutes ago, andyc56 said:

Often, because of the additional electrical delay introduced by the LPF of the AVR's bass management, the optimum sub distance setting can be rather different from the sub's physical distance.

Maybe that is the reason why Anthem decided to use physical distance?

[Snoopy8]

Andy, unable to apply Application Options once a Config has been set up.  While they can be changed, they are ignored by optimisation.  Feature or bug? 

[Snoopy8]

Been experimenting with adding, removing PEQs. Sometimes, PEQs end up with similar frequencies e.g. 90.3, 90.5 (in dcx2496, entered once as 91) . When I remove one of these, rerun optimisation, the curve is less flat. Am I correct in assuming that the curve with redundant PEQs was a false representation in the first place?

*** Edit: Can't say I understand how the optmisation works, but it appears to cluster some of the PEQs. Previously, removing the redundant PEQs resulted in a larger frequency response error. By keeping and constraining the PEQs to prevent them moving too close to one another, the frequency response error was kept the same.

 

Also, have been adding and removing PEQs on the fly. Does it create any problems?

*** Edit: adding/removing PEQs does not appear to create any problems.  Typically using between 4 to 6 PEQs.

Edited June 19, 2017 by Snoopy8
Answered own questions!

[mcb]

@Snoopy8 , my apologies for the delay in responding. 

 

If I have understood your post correctly, the similar PEQ's are not redundant, but are intended to work with each other, to combine to give the best result.  Most of my filters include some that are at the same or similar frequency.  Usually, the Q value or cut/boost value will be different.

 

You can actually see the impact of turning off each PEQ within MSO.  Make sure you save the file before you start playing around, then open both the config view and also the properties view (pin the properties view open).  Then with the graph of the projected results (not filters) open, selected the desired filter in the config view and go across to the properties view and zero out the cut/boost of the PEQ.  This will then show the effect of the change on the graph.  You will need to note the value of the cut/boost before altering it, if you want to reinstate it afterwards.  Of course, if you lose track of the values, close the file (without saving the changes) and then re-open it.

[Snoopy8]

39 minutes ago, mcb said:

You can actually see the impact of turning off each PEQ within MSO.  Make sure you save the file before you start playing around, then open both the config view and also the properties view (pin the properties view open).  Then with the graph of the projected results (not filters) open, selected the desired filter in the config view and go across to the properties view and zero out the cut/boost of the PEQ.  This will then show the effect of the change on the graph.  You will need to note the value of the cut/boost before altering it, if you want to reinstate it afterwards.  Of course, if you lose track of the values, close the file (without saving the changes) and then re-open it.

Good tip, thank you. 

[andyc56]

On 6/8/2017 at 8:33 AM, Snoopy8 said:

Andy, unable to apply Application Options once a Config has been set up.  While they can be changed, they are ignored by optimisation.  Feature or bug? 

The filter parameter limits in Application Options only affect newly-created filters.  These values never change the parameters of existing filters.

 

Edit: See my post two posts down for additional information.

Edited June 24, 2017 by andyc56
Left out some important information

[andyc56]

14 hours ago, mcb said:

You can actually see the impact of turning off each PEQ within MSO.  Make sure you save the file before you start playing around, then open both the config view and also the properties view (pin the properties view open).  Then with the graph of the projected results (not filters) open, selected the desired filter in the config view and go across to the properties view and zero out the cut/boost of the PEQ.  This will then show the effect of the change on the graph.  You will need to note the value of the cut/boost before altering it, if you want to reinstate it afterwards.  Of course, if you lose track of the values, close the file (without saving the changes) and then re-open it.

 

Another way of doing this without disturbing your existing configuration is to clone the configuration you want to experiment with.  In the Config View, select the root node of the configuration (the one with the configuration name), then right-click and choose "Clone Configuration".  Or you can choose "Config, Clone" from the main menu.  This opens up a dialog box asking you for the new config name.  By default, "Clone associated graphs" is checked, meaning all the graphs associated with the original configuration will also be cloned, but their traces will be set to refer to the newly-created copy.  Then go over to the Data View and show the newly-created graph.  It will have an auto-generated name which you'll be able to recognize and rename if you wish.  You can then do all the experiments you like with the filter parameters of the newly-created copy without disturbing the original.

[andyc56]

I forgot to mention that if you want to adjust the limits of any parameter of an existing filter or gain or delay block, you do this by selecting the filter or gain or delay block in the Config View on the left, then in the Properties Window on the right, you set the minimum and maximum values of the parameter.  For example, if you want a specific PEQ to have a center frequency between 75 Hz and 125 Hz, you'd select that PEQ filter, then go to the "Center freq (Hz)" parameter in the Properties Window on the right, and set "Minimum Value" to 75 and "Maximum Value" to 125.  In this way, the optimizer won't be allowed to set the center frequency of that particular PEQ outside those limits.

 

This is discussed in more detail in the reference manual.

Edited June 24, 2017 by andyc56
fixed typo