REW-Good Looking Freq Response Graph Doesn't Tell The Whole Story - Random Thoughts/Findings & Lessions Learned

[cjf]

Hello,

 

I've recently went thru a rather long and painful "Retune/Reconfigure" process of my system due to the addition of a 3rd Sub (Rythmik F18SE) to compliment the two existing Subs (Rythmik E15HP2 SE's) I was using previously and as a result I needed to basically reconfigure the entire system. During this process I came to find out that adding an additional Sub wasn't as easy or straightforward as I originally expected.

 

My thought going into this was that I would just drop the new Sub into one of the remaining available spots in my room and leave the two previous Subs where they were. In theory this sounds all well and good but in practice it didn't work out that way at all. I came to find out that doing this ended up making a dogs breakfast of my nice looking original REW Freq Response Graph. No matter where I tried to place the new Sub things just kept getting worse so I had to clean the slate and start over from scratch.

 

I'll admit up front that I am no expert in doing this stuff and its certainly not part of my day job so this info is just based on my own experiences thus far while playing with Subs in my leisure as a hobby. I'm quite sure an expert in the field could spot issues in my new configuration and graphs and fix them no sweat but I chose to do this on my own using info found around the web from the various Forums, posted Articles/Papers and pure trial and error.

 

Back when I first added Subs to my system the whole process was rather daunting and the learning curve was quite steep. Now that I have an additional year under my belt doing all this and more experience on what I'm looking at and what to look for during the process I've come to find out that even though the work I put in previously that resulted in a rather pretty REW Freq Response Graph wasn't really telling the whole story and had several issues lurking in the background that I was unaware of all along.

 

This post is intended to show some of those issues I was unaware of and to maybe help others keep an eye out for them should they decide to jump into this process with blinders on like I did a few years back. Now onto the findings....

 

All graphs showing the "Original" configuration was while using 2 X Rythmik E15HP2 SE Subs located at the back of my listening room. One Sub was in the far Left corner and the other was in the far Right corner of the room. Both Subs were firing straight ahead towards the Front Wall where my main towers are located. Below is a rough diagram of the Original Sub placement in the room along with the approx Volume of the space. This was generated with the REW Simulation feature

 

Config # 1 - Original 2 Sub Placement In Room (S1 & S2 are the Rythmik E15HP2 SE's)

 

Config #2 - New 3 Sub Placement In Room (S1 & S3 are the Rythmik E15HP2 SE's) (S2 is the Rythmik F18SE)

 

Lets start off with a comparison Graph of the Freq Response between the two room/sub layouts shown above
 

• 3 measurements at the Top of the graph are the "Original" 2 Subs Config #1). 1/12th Smoothing in effect
• 3 measurements at the Bottom of the graph are the "New" 3 Sub Config #2) 1/12th Smoothing in effect

 

As you can see, the 2 Sub layout in Config #1 measured quite nicely from a in room Freq Response standpoint. I can tell you I spent over 100hrs trying to duplicate or better that Top measurement trio while re-tuning the room with 3 Subs. In short, it never happened and I was quite disturbed, annoyed and many curse words where thrown around along the way.

 

So much to my dismay I added a 3rd quite massive Sub to the room and for some reason I couldn't understand why, no matter what I did or where I placed all of the Subs, I couldn't better or even match that "Original" in room measurement with 2 Subs. Despite this though, the new 3 Sub configuration sounded significantly better in all areas across the board and especially in the tactile feel and scale categories. It wasn't even close in how the two layouts sounded. Config #2 was a clear winner.

 

Now I will introduce some of the very recently discovered demons that were lurking under the covers of that Config #1 layout with 2 Subs that I didn't realize at the time. After banging my head against the wall for a few weeks I decided to circle back and take a very hard/close look at REW again to compare the two configurations side by side.

 

Lets start this segment off with the ugliest of my findings when comparing Config #1 vs Config #2

 

Spoiler...DISTORTION For Days!!

 

All Channels Distortion Overlay (Left/Right & All Subs)

 

• Config #1 w / 2 Subs = Blue Line, Crosshair located at highest THD Peak 45.8% @ 17hz
• Config #2 w / 3 Subs = Pink Line, Crosshair located @ 17hz, THD 5.5%
• But wait...it gets MUCH Worse

 

Right Channel Distortion Overlay (Right CH Only & All Subs)

 

• Config #1 w / 2 Subs = Green Line, Crosshair located at highest THD Peak 417% @ 19hz  🤯
• Config #2 w / 3 Subs = Red Line, Crosshair located @ 19hz, THD 2.74%

 

 

So it appears that I was pushing the original Subs a bit too hard and got a bit Ham fisted with the output/loudness of the sweep settings in general. Lets take a look at a few other signs of this

 

Spectrogram Graph Comparison - All Channels (Left/Right & All Subs)
 

• Config #1 w / 2 Subs - Looking pretty HOT in here across the board but especially below 75hz with lots of far reaching overhang trails in the bass region

 

 

• Config #2 w / 3 Subs - Waaay cooler across the board but with plenty of energy still down to 10hz and below. Long overhand trials in the bass region look much better

 

So what did I do differently between the first few Graphs shown so far from the Original vs the New Configuration.

 

1. In Config #2 w / 3 Subs I Level matched the Subs with the Main Tower speakers using Random Pink Noise in the bass region playing between 30-80hz to a Ref Output Level of about 75db SPL

2. In Config #1 w / 2 Subs I Level matched the Subs at the X-Over point only with the Main Tower speakers at the loudest level I would typically listen at (around 95db). So the whole system was level matched based solely on one point of the Freq scale. In hind sight, this was a bad idea 🤕

 

Now lets look at a few other problems I discovered with my original Config #1

 

PHASE

 

Right Channel Phase Overlay (Right CH Only & All Subs)

• Config #1 w / 2 Subs = Green Line
• Config #2 w / 3 Subs = Red Line

 

 

Left Channel Phase Overlay (Left CH Only & All Subs)

• Config #1 w / 2 Subs = Orange Line
• Config #2 w / 3 Subs = Green Line
   

 

SPL/Phase Graphs

 

Config #1 w / 2 Subs, All Channels (Right/Left CH & All Subs)

• Blue Graph Line = Freq Response
• Purple'ish Graph Line = Phase ---What a mess

 

Config #2 w / 3 Subs, All Channels (Right/Left CH & All Subs)

• Pink Graph Line = Freq Response
• Brown/Green'ish Graph Line = Phase ---Way Cleaner

 

 

So it appears that there were some serious Phase anomalies going on with Config #1 in addition to it being way too hot in terms of Output Level.

 

What else did I find? Let's look at the Step Response next.

 

STEP RESPONSE
 

Config #1 w / 2 Subs, All Channels (Right/Left CH & All Subs)

 

Now with this one I'll admit that its not clear there was an issue until I took all the other findings into consideration and compared it to the new Confg #2. I can see that there is an initial Spike towards positive polarity which then goes immediately negative with a less than great matching of All the channels at the initial spike. The rest of the plot then starts from the bottom and works its way back up towards 0. Again, I'm not clear on what this really means but looking at the next Graph from Conig #2 tells me something probably wasn't right with this one

 

 

Config #2 w / 3 Subs, All Channels (Right/Left CH & All Subs)

 

We can see that this Plot is the opposite of the Original one in Config #1 and to me looks more correct based on other Step Plots I've seen around the web

 

 

Overlay of STEP RESPONSE comparing Config #1 with Config #2

 

Here in this Plot it becomes clear of the differences between the two Configs #1 vs #2

 

 

Finally to bring an end to this long original post let's look at the Impulse Response comparisons

 

Right Channel Impulse Response Overlay (Right CH Only & All Subs)

• Config #1 w / 2 Subs = Green Line - Here Config #1 appears to be showing signs of Pre-Ringing Like A Bell
• Config #2 w / 3 Subs = Red Line - Config #2 looks pretty clean by comparison
   

 

All Channel Impulse Response Overlay (Right/Left CH & All Subs)

• Config #1 w / 2 Subs = Blue Line - Here Config #1 appears to again be showing signs of Pre-Ringing Like A Bell
• Config #2 w / 3 Subs = Pink Line - Config #2 looks again pretty clean by comparison
   

 

So getting back to my original point of the post that a pretty Freq Response Graph doesn't tell the whole story of whats going on. The scary part of all this is that at no time while I was listening to my system when Config #1 was in use did I get a sense of something being wrong. The system sounded great compared to how it sounded before any subs or room correction was thrown into the mix so I was living in what is known as ignorant bliss. The Freq Response Chart looked great, things sounded great and everything sounded better than it ever had before I started this journey with room correction and subs.

 

Now that I have had a chance to directly compared the two configurations/layouts with listening and somewhat scientific measurements its become clear to me that if All the measurement data is not taken into careful consideration and your initial setup process is not don't correctly you may be wasting valuable time listening to your system with some ugliness hiding under the covers.  Even though all may seem to sound just fine.

 

I'm open to any additional dialog on this subject/info posted here including dialog that may oppose what I think seems correct now about Config #2 w/ 3 Subs compared to Config #1 w / 2 Subs

 

Thanks for reading and hopefully this helps others in their own journey of pit falls to look out for

[cjf]

Thanks for your reply.

 

I'm not sure I quite understand what those two folks are saying but I do agree that the plots posted are certainly not perfect. I'm not sure I would agree that they are bad either though which it sounds like they may be suggesting.

 

I did fail to mention that the subs and system was configured using Audiolense which takes many variables into consideration when the filters are created. I agree again that it won't work miracles though if physical placement is wrong or components are malfunctioning or not working as designed.

 

Are they saying my system is broke? 🤣

 

If so, I guess I have other problems to deal with but I suspect that is not the case. I would more than welcome any hints or tips they can provide showing me where I have gone so wrong. Maybe they can forward over their own room plots/graphs so I know what a good one should look like?

 

Thanks again for your input.

[cjf]

Below is a graph showing the Group Delay of the system in its current state (ie..3 Subs Config#2).

 

All Channels appear to line up pretty nicely for the most part. The delays of all channels to the Main Listening position are taken into consideration when the correction filters are created.

 

 

Below measurement sweep settings from Audiolense shows the actual Delay between each speakers arrival time to the Mic at the MLP. My understanding is that all channels are Delayed to account for the longest Delayed speaker to the MLP found when the measurement sweep is taken and the correction filters is generated.

 

You can see in this case one of the Subs sweep is taking 9.4x /ms to arrive to the MIC

 

 

 

[cjf]

In terms of the zoom level of the first Freq Resp chart I'm not sure if adding additional hash marks is possible or not but I can say the largest dip/spike at the worst point is about 10db.

[cjf]

Hello,

 

I'm getting the sense that there is a misunderstanding occurring. The original intent of my post was to show that if one only considers the Feq Resp curve in a room measurement and not all the other measurements they may be overlooking other issues that could still be there even though the Freq Resp curve looks OK, good or even great.

 

I do not think that anything sounds wrong or strange with the system as it sits today which is represented as Config #2 in all the graphs previously posted.

 

If one looks at basically any in room speaker measurement found on "pick your favorite audio rag" you see all sorts of uneven spikes/dips and drastic drop offs at various points in the curve. Even with the uber priced super speaker going for hundreds of thousands of dollars. In most cases you see a large bump at 20hz or higher that drops off above and below that +10 to +20db mountain which in my opinion may be fine for those who never actually crank up their system to elevated levels. I suspect the speaker manufacturers do this on purpose so that when listening at low levels it gives a greater sense of bass output. But for those who do crank up the volume of their system the same Freq Resp graph sounds like a 1980's boom box.

 

Its been my understanding that the ultimate goal of any speaker manufacturer would be to have their speaker's Freq Resp look like a perfectly straight line. This goal has not been achieved in any measurement I've ever seen. Such a speaker would be able to reproduce the source material perfectly without adding its own flavor to the original sound. Then you add such a speaker to a real room. At that point all bets are off on what the Freq Resp graph will look like.

 

The fact that my measurements show a roughly 10-12db roll off at each extreme doesn't seem all that bad and in fact matches pretty closely with just my tower speakers alone and no room correction or subs. The biggest difference between Config #2 and a measurement of just the Main towers alone in this same room is that the current measurement has usable bass extension and output down into the single digits.

 

The room this system is located in is quite large (at least 4500cu/ft) for just the listening area alone and very open to other areas of the house as well. In order to fill this space with sound to make the Freq Resp chart have the very typical 10-20+ dB mountain seen in many in room measurements previously mentioned it would take far more cubic inches of Sub and power to go with it. In any case, a Freq Resp chart that looks like that is not what I prefer to hear. I'm not interested in artificial bloom or bass texture that one note sounding due to excessive overhang as a result of again that typical chart seen in the audio rags.

 

With all that said though, I do have the ability to alter the shape of the Freq Resp curve until I'm blue in the face using Audiolense. But this circles back to one of the issues previously shown in my original Config #1 layout with the better looking Freq Resp chart. The end result of trying to reproduce that better looking Freq Resp chart being way higher levels of distortion requiring me to push the Subs outside of their limitations in terms of being able to maintain good SQ at louder listening levels.

 

I can say for sure there is no lack of bass or detail in the current configuration. I don't think it even makes sense to suggest a -10db roll off from the highest peak at around 100hz or so down to 14hz and below while still outputting over 80db+ down there would have a lack of bass output. Even most of those uber mega buck speakers can be down 10-30db or more after 20hz if your lucky.

 

I guess it all comes down to personal preference. I like my sound to be as unflavored as possible and as loud as possible whenever possible.

 

 

[cjf]

18 hours ago, bibo01 said:

OK, fine. I'll stop here.

So you were very happy with your #1, but when #2 arrived you realized that you were "living in an ignorant bliss".

It seems to me that now you are happy "living an ignorant bliss #2". As you are happy, however, it's fine by me and I will not waste my time any further. 

Perhaps my explanation wasn't as clear as it could have been but my point was that the system sounded find with Config #1 and the Freq Resp looked good as well (IMO)....but despite those two things I failed to know or realize at the time the other issues shown were also present (ie..Pre-Ringing, High Distortion in the Bass Region..etc..etc)

 

The current configuration doesn't show the same signs of high distortion, pre-ringing..etc.etc but its' Freq Resp plot is not as nice (IMO) but sounds even better.

 

These are the points I was making. Could I crank up the vol knob of the Subs and make a nice big hump from 15-30hz to make the current Config #2 look like Config #1....yes indeed I can but knowing what I know now about the end result of doing that I will no longer be attempting to duplicate Config #1's Freq Resp graph unless I decide to purchase more Subs in order to achieve the same type of chart but without all the added distortion that went along with it.

 

All the x-overs I am using in Audiolense are of the Linear Phase type. Honestly I've not tried to create any Minimum Phase x-overs as of yet but certainly could if I wanted to.

[cjf]

15 hours ago, Bob Stern said:

REW graphs & axis:  

 

In REW, if you hover the mouse near the upper left corner of a graph, + and – buttons appear allowing you to expand/compress the scale.  Above the upper right corner of the graph are 4 icons, one labelled Limits.  Click that and then type your desired numerical upper and lower limits for the X and Y axes.

 

Your graphs would be much more helpful if you expand the vertical axis to show the full range.  In the horizontal axis, if you're focussing on the bass response, you should set the upper limit of the graph to the upper bass frequency of interest.

Hello,

 

I can post any additional info/graphs folks may want to look at in whatever view/format. In the case of the question above which graph would you be referring to when you mention about expanding its Vertical axis?

 

Posting these types of graphs in the format that all can agree on can be a challenge. Some folks want it zoomed out while others want it zoomed in.

[cjf]

1 hour ago, Bob Stern said:

 

I'm not making a request; I'm merely offering a suggestion on how to make the REW graphs more useful to you.  A graph in which there's lots of white space above and below the curves would be more useful if you zoom in so that the curve occupies most of the displayed Y-axis.  Likewise, if you're examining the bass, you can see more detail by reducing the upper frequency limit on the X-axis.

 

Also, when running the algorithm to determine bass EQ, the REW developer recommends setting smoothing to Variable instead of 1/12 octave.  "Variable" applies almost no smoothing in the bass and increasing smoothing in the treble.  See the help topic "EQ Window":

http://www.roomeqwizard.com/help/help_en-GB/html/eqwindow.html

Thanks and I appreciate the tip.

 

As another side note, One thing that I learned the hard way during this process was that for whatever reason or another I could not rely on the combined results seen in REW of more than one sub being measured at the same time and have those results appear the same way within Audiolense once the correction filters were generated.

 

Part of my process while determining where to place the various Subs in the room was to use REW first to get an idea if a movement here or there of a particular Sub/Subs would result in the reduction of a dip or spike across the measured Freq range I would use for the Subs (10-500hz). Once I achieved a decent measurement of each Sub on its own I would then start to do sweeps with combination of Subs together. I spent a few weeks doing this and thought I found the best spots once the combined measurements where taken into consideration only to find out those combined results never seemed to result in the same outcome after corrections were done in Audiolense. On a whim one day after continued failure I decided to just worry about achieving the best measurement for each Sub individually in REW. As soon as I did that everything started to fall into place and make sense and the correction filter Simulations within AL started to look like what I saw while using REW (for the most part). As a side note, I wish REW would allow sweeps of more than two channels at a time but unfortunately it does not.

 

I guess my point being that I'm not sure if using some of the Simulation features in REW would actually translate into the same result after corrections are generated/applied with AL. It appears the combined Sub results do not but maybe its just me and something I was doing incorrectly.  I can say that the Room Sim feature in REW doesn't translate at all in my room with the actual measurements. I assume this is mostly due to the irregular shapes of the walls, the very open layout to other rooms and the 12ft vaulted ceiling that runs down the center of the room none of which are variables that can be plugged into the Room Sim feature of REW.

 

Unfortunately my main Right channel sits directly under this peak in the ceiling and it results in a very nasty room mode to fight with. With no correction in the picture that channel is 8db louder than the LCH is at various lower Freq. There is also a nasty Null in the Center of the room a few feet in front of the MLP. Standing in that spot and almost all bass literary disappears. Its my understanding that subs on each side of a Null are out of Phase with each other so you have to get creative with "Aiming" the Sub in different directions I order to end up with all speakers being in phase with each other once all corrections are done.

 

 

[cjf]

9 hours ago, ecwl said:

5 channels (L+R + 3 subs) and audiolense are really complex.

My gut feeling about the REW frequency responses of Config 1 (2 subs) vs Config 2 (3 subs), assuming Audiolense was run and setup correctly is that Audiolense uses psychoacoustic correction so the frequency response filtering would be different than REW's measurements. That's probably why the frequency response of the bass sounds the same in Config 1 vs 2 but the REW measurements look different.

 

As for whether there is further room for optimization from Config 2, there probably is. I'm still unclear whether the speakers are running full range or not. It seems the subs are running 10-200Hz. I mean, with good speakers, you always wonder if you can run the speakers 50-20kHz and just run the subwoofers 10-50Hz or set the crossover frequency elsewhere, e.g. 40Hz, 80Hz, 100Hz, etc. Obviously, if you're running the speakers full range, it is possible that with Config 1 your speakers + subwoofers are enhancing each other in the bass region but in Config 2 your speakers and subwoofers are cancelling each other out, leading to a lower bass frequency response measured in REW.

 

The other issue is the amount of pre-ringing from the speakers. The True Domain Subwindow Cycles before and after peak is set at 5 for 10Hz and 3 for 96kHz. This is extremely system dependent and sometimes you can push the cycles to 6 for 10Hz but to prevent excessive preringing, you need to set the cycles to 1 for 96kHz. If it sounds pretty awesome right now, I would suggest trying 5 for 10Hz and 2.5 for 96kHz and see if you like it better. And then try 5 and 2. I personally use Acourate instead of Audiolense and I have my own "optimization" preferences where I tweak these numbers to 0.1 accuracy for months. I have a very asymmetric setup so I think my system is like 4 & 2 on the left and 6.2 and 2.7 on the right or something like that. But bottomline is that I think 3 for 96kHz maybe too high for your system.

 

Ultimately, I think what most people are saying is that:

1) You can't completely correct for room issues and speaker/subwoofer issues. I think that's true but all DSPs (or stereo systems) are about compromises and tradeoffs (sadly, just like life)

2) There are these ideal frequency responses we should target except from my experience, different people have different preferences for frequency responses. So if you're happy with what you're hearing, you're probably good.

Hello,

 

In both Config states I am using a cross over for the Main towers. In Config #1 the x-over was set to 62hz I believe and in the current Config #3 I am using an x-over of 60hz. In the past I tried running the Main Towers full range and just using an x-over on the Subs but the results were not very good. I might give it try again now that my system is a bit different than it was at the time when running the Main Towers full range to see what the results are just for the heck of it.

 

I have all the controls on the Subs defeated and am using their LFE Inputs which is supposed to kill pretty much everything except the Vol control level on them. In the Audiolense Speaker Setup window I have the LFE cutoff set to 250hz and am Routing all Bass to all Subs.

 

For the settings being used in the Correction Procedure Designer for my current setup (ie..Config #2) I dont think I've tried to set the 96K Window to "1" yet. I did on the other hand play quite a bit with the 10hz Window but,at least according to the Simulation results, when I did that the graph started to look like a mess and at least one of the Subs would go out of Phase with all the others while viewing the Step Response Simulation results.  I'll give a few more combinations a shot though including some of your suggestions to see how it goes.

 

The TTD per Driver setting in CPD would also make a mess of things in terms of the Simulations in AL. I hear this setting though can go either way and is very system dependent. The Simulations are showing it to not be a good choice for me for whatever reason.

 

There have been several times though that the AL Simulations would look like a mess but the REW results would look better by a decent margin. Perhaps I'm putting too much weight into the AL Simulation and need to try a few more CPD combinations that look not great anyway just to see what the outcome is.

 

Thanks for your thoughts on this post

 

[cjf]

On 4/29/2021 at 10:24 AM, Bill Brown said:

Re. your subs, I would encourage you to look at the work of Geddes and Welti re. the multiple sub approach.  I now have four and saw stepwise improvement while adding (3 was great, the 4th didn't add a huge amount).  Many of the longstanding audiophile beliefs are turned on their heads by this research/approach.  Acoustics and measurements below the room's Schroeder frequency are very different from those above.  Some of the reasons I haven't delved deeply into DRC is that I would like to do as much acoustically first and that it isn't clear to me how to incorporate measurements --> filters with all the subs playing.  It doesn't matter what each one is doing individually, but what they are all doing together.

 

https://www.harman.com/documents/multsubs_0.pdf

 

https://mehlau.net/audio/multisub_geddes/

 

https://www.diyaudio.com/forums/subwoofers/134568-multiple-subs-geddes-approach.html

 

Bill

Hello,

 

I have read those docs a few times during this journey and they are indeed very interesting. In terms of the Harman paper unfortunately the Sub placement options they show as being ideal do not work exactly in my case. I guess my placement might be considered a bit of a Hybrid of some of theirs. My placement limitations are fully dictated by available AC outlet locations in the room, the length of the power cord that comes with the subs and the fact that my Breaker Box is basically maxed out so adding additional dedicated lines would get quite expensive requiring a new larger box or a Sub box branched off of the current one.

 

During this latest reconfiguration I pretty much tried every spot in the room for the subs even going so far as breaking down my main equipment rack and placing subs in between both main tower speakers. The results in that location weren't bad but aesthetically speaking it just wasn't going to work for me having Subs there.

 

The only thing I didn't try and do was give up my couch location sitting at the Mid point on the left side of the room but there isn't a plug to accommodate a sub there anyway. It would probably be a very good place to put one though if I could.

 

Thanks

[cjf]

3 hours ago, ecwl said:

After thinking about this some more, I realized you can do this instead:

1. Turn on TTD per driver

2. Keep the subwoofer that goes out of phase at 5 & 3 for TTD cycles correction

3. Push the other subwoofers’ correction higher to 6 & 3 if the simulation doesn’t go wonky (or change it to 6 & 2.5 or 2)

4. Try lowering your speakers TTD correction to 5 & 2.5 or 2, or if you really can’t hear or mind the pre-ringing go to 6 & 2.5 or 2

 

That way, at least that single subwoofer won’t go out of phase. And you can get even more aggressive phase correction for the rest of the system in the bass.

 

Personally, for simplicity sake, I would try not turning on TTD per driver first.

Hello,

 

I may just not be seeing how this is done in AL but it doesn't appear that you can adjust those settings on a per driver/speaker/channel basis. Again, it may be there but I have yet to discover how. This could be because I am dong an "All in One" type measurement but I'm not sure. The settings I am seeing appear "Global" in nature only.

 

I did a very quick and dirty target and correction procedure using Minimum Phase settings last night just to see how it would behave and saw a few improvements in certain areas (like below 14hz where a dip there leveled off nicely) but I need to dive into it more. Interestingly enough, with Minimum Phase targets procedures in the picture I could then enable TTD Per Driver without the previously mentioned Phase issue on the one Sub. I also tried your suggestion of setting 96K TTD Window to "1". I didn't notice anything positive or negative in doing so but will need to play around this some more.

 

One side note, in the previously posted graphs showing Impulse Resp zoomed in real close comparing Config #1 to Config #2 as far as I could tell the Pre-Ringing was no longer happening with my current Config #2. The Saw-tooth graphs seen are from the original setup that I am no longer using (Config #1 w/ 2 Subs). This of course assumes I'm interpreting them correctly :)

 

 

[cjf]

Ahh no problem at all. I appreciate your thoughts on all this.

 

Yeah Flickr made a mess of the first splash page during my last batch of uploads. I need to go in there and move some of those measurement charts around a bit. The latest charts of Config #2 are closest to the bottom after the system diagram/equipment list stuff.

 

I'll play around this weekend some more and report back on findings..Thanks again

[cjf]

8 hours ago, Miska said:

As a generic rule I'd advise to not do high boost EQ's, since those will increase distortion without usually helping much. Also gentler EQ adjustments based in 1/9th octave smoothing usually sounds better, very steep narrow EQs tend to cover very narrow listening position and make the sound just weird.

 

So using EQ to force flat frequency response usually sounds bad...

 

@Miska

 

Hello can you provide an example of what you mean above when you mention using steep narrow eq at 1/9th octave in the context of an Audiolense user like myself? I'm not sure I understand what you mean.

 

Thanks

[cjf]

10 hours ago, Bill Brown said:

@cjf Certainly the difficulties you describe are real.  I was limited with outlets for the rear subs.  Thankfully I could use some old power boxes with captive cords to reach and they are hidden by the couch.

 

One of the notions that I had to overcome and that then provided a real jump was re. crossovers on mains/ sub XO frequency.  Based on my experiments, I think the traditional thinking re. localization isn't completely true.  Raising the XO to the subs adjacent to the fronts and trying to match the phase of the frequencies coming from the mains' ports and woofers was fruitful.  The subs ended up 180 degrees out of phase to the mains with the XO at 150 Hz.  Sounds crazy, I know.  My thinking now is that the sub is in phase with the port with its 24db per octave roll-off and also with the woofer at higher frequencies (low frequencies now more extended and 80-200 Hz or so "filling in" the traditional floor bounce of mains).

 

I then did a gazillion measurements with the rears and they ended up at reversed phase as well.  There XO is at 150 Hz as well.......I don't hear abnormal localization at bass frequencies.

 

Finally, I have read a fair amount re. minimum v linear phase EQ in the low frequencies and think that if the pre-ringing of LP EQ/XOs is ever audible it is in low frequencies.

 

Bil

With the current 60hz x-over I'm using there are times and certain Freqs where the big 18" Sub sitting right next to my listening chair makes its location a bit noticeable but that is one downfall of having it sit right next to me I suppose. The benefit of having it there is that it seems to help fix other MLP related Nulls that were present before putting it there. Its a Damned if I do and Damned if I don't situation 😀

 

If I move it further away along the same rear wall it ends up being very close to a corner resulting in a -10-12db Null at 19hz due to a half wall I have on the right side of the room. I've thought about ripping the wall out  😈 but then the Sub would be sitting out in the open and look less than desirable.

 

I do have one additional location along that half wall I can try if I get bored which isn't a corner and may eliminate the localization but at the moment its still livable. I think if I were to move up the x-over it would be problematic in my case though.

 

 

[cjf]

14 hours ago, Miska said:

 

With REW or other tools it is good to perform 1/9th octave smoothing to the measured response before starting to generate correction filters for it, to avoid filters that have steep (high Q) changes in the correction response. Narrow dips and bumps in the measured response are usually applicable only to a very small area in listening space and overall trying to correct such will make sound weird and dull.

 

You can check the correction filter response for example in HQPlayer to see how smooth the correction filter response is. For linear-phase convolution filters, keeping the correction response smooth also keeps the amount of correction filter pre-ringing at minimum. Steeper the variations in the filter, longer the filter gets with more pre-ringing too.

 

Ok thanks for the info @Miska

[cjf]

12 hours ago, ecwl said:

In Audiolense, you "control" how aggressive you EQ/DSP by controlling the frequency dependent windowing (FDW) which is called cycles before and after the peak in the frequency domain and in the time domain (TTD). Your frequency domain setting is 8 cycles at 10Hz and 5 cycles at 96kHz. I highly doubt that you'll have any steep narrow dips or peaks with that setting. So I wouldn't worry about it. You'll have to set the FDW to much higher numbers where it'll pick up the narrow dips or peaks. In reality, it's a bit more complicated than that but I think my simplification still holds.

Ok makes sense now. Sounds like the larger the numbers are the more granular the corrections attempt to be.

[cjf]

12 hours ago, ecwl said:

It is hard to know whether you're hearing the 18" subs because you're localizing 20-60Hz from the Sub or more likely, whether you're hearing the 60-200Hz from the Sub even with the crossover. I believe the solutions are either a steeper crossover which can create other problems, or a lower crossover frequency, e.g. 50Hz or 40Hz to see if you run into the same problem.

Yes I suspect its the later as well, meaning the higher Freqs after the x-over point before it falls off into oblivion.

 

I ended up going down another rabbit hole this weekend and moved the F18 to another location and then moved one of the E15's to a spot I've had in mind for quite awhile but never tried. Turns out I should have done this sooner. The spot I've been eyeballing for a few months now ended up being better by a noticeable margin when compared to the spot next to the MLP. My slight localization complaint has now been rectified fully.

 

I'll be doing some more shuffling around this week trying out one or two more remaining combinations before I throw in the towel and call it "good enough".

 

A few things I did notice while playing with Minimum Phase x-overs and targets with AL this weekend is that they "Simulate" quite nicely compared to the Linear Phase x-overs and targets within the AL interface but despite this my ears prefer the Linear Phase flavors instead. Plus it appears the Min Phase stuff doesn't do Time Alignment based on what I am seeing when looking at the Impulse Response measurements in REW. I can clearly see that the L/R Channels no longer converge at the same point just before the initial main spike. It also appears that the Min Phase settings do not control the curve of the top end of the Freq Resp measurement as nicely once the REW measurements are done. I'll see my initial -9db roll off at around 22,050 Khz but then the line will start to roll back up north again for a bit until 32Khz or so resulting in an Upward graph at the high Freq extreme. The Linear Phase settings don't do that even with High Freq boost enabled.

 

One final observation which is that it seems like I'm at a point where no matter where I put these Subs I the room short of being negligent the Freq Resp graph really hasn't changed all that much from Config #2. Overlays are basically mirror images of each other short of a few wiggles and dips here and there. At this point I'm in a battle of recovering a few DB's here and there but mostly these changes are to try and dial in the other stuff like Step, Impulse response, Group Delay and a as smooth as possible Spectrogram with  as few discontinuities as possible.

 

Once I get done shuffling things around again I'll add a new series of graphs and comparisons here. One things for sure, this is beginning to feel allot like work. Moving and dialing in Subs a hundred times or so can begin to wear a man down both physically and mentally 🙂

[cjf]

@Bill Brown

 

Hello,

 

I've not seen the ability to combine both in AL but I'll admit to not knowing how several features and capabilities work in there. I wish there was a deep dive available on all features available but most explanations of how to use the application are extremely high level in nature within the help file.

 

I've been in a listening phase of my new configuration (will call it Config #3) and the results have been very positive thus far and noticeably better than previously reported for Config #2.

 

It will take me a few days to gather all the graphs and comparison charts to post here but will certainly do probably over the coming weekend.

 

I've managed to smooth out the Freq graphs below the 60hz x-over point with the new Sub placement locations and will be focusing on trying to do the same for the upper Freq range next in terms of Group Delay..etc while attempting to keep Distortion figures as low as possible across the board.

 

At one point I considered adding a 4th Sub but honestly I don't think there is anywhere I could place it in the room that would add value based on all the measurements that have been taken thus far and seeing the new results compared to the previous ones posted.

 

One thing is for sure, the Rythmik E15HP2-SE Subs are far more easy to find spots for than the F18 is and the measured differences between them is pretty much unnoticeable; at least at the lower Freq limit I've been measuring to (10hz). If I try and sweep lower the protection circuit on the Sub kicks in due to a power sag in the electrical system in the house. Guess I'll need to address that next. There is no end to the rabbit hole it seems 🙂

 

 

[cjf]

On 5/5/2021 at 8:29 AM, MarkusBarkus said:

@cjfEarlier in the thread you had mentioned your panel-box was full. I intended to mention then that there are tandem breakers available that might help, FYI.
 

I have used them in scenarios where I can identify that circuits have mainly lighting load, or little used bedrooms, etc. to put on a tandem breaker, freeing up a spot on the buss for a new line.

 

If possible, you may be able to rearrange breakers, or select a spot that is on a more favorable phase, re: fridges, pumps, etc.

 

That said, if you're finding your aggregate load is nearing your service max, it wouldn't help you there. And as always: be safe and follow NEC (National Electrical Code, for non-US readers) guidance or use an electrician! 

@MarkusBarkus

 

Thanks for the tip on the breakers. I have two slots I believe I could steal if I decide to start messing with the power. They are currently doing nothing and are used as a 240v feed for an old hot tub which is now gone. I was thinking of taking that and putting a single width 50v breaker in is place which would then feed this Sub panel I mentioned. From there I would add several 15a receptacle's for the audio components but this is just theory on my end at this point as I've not brought an electrician in to validate if it feasible, safe or legal from a code standpoint.

 

I'm hesitant to go down path even though I'm sure it would bring about noticeable sonic benefits. I went a bit wild in my old house with this stuff and lost quite a bit of cash when I sold the house because most people don't care about such things or what it cost to install it thus will low ball you anyway.

 

I haven't ruled it out yet by any means though so thanks again for the alternative option to consider

[cjf]

So I've been messing around with various Correction Procedure settings within Audiolense to see if I can tighten up the Step Response graph a bit more but there are difficulties trying to do it all within AL only via the simulations. Probably one of few complaints with AL I have is that it doesn't offer an Overlay feature between graphs. My current approach has been pretty caveman like in nature by making a few changes in CPD then generating a filter. I would then move over to the Step simulations and look at them. After Im done I take a screenshot of the plot so I can use it to compare another round of CPD changes. Rinse and repeat.

 

For anyone familiar with that you can imagine that its not very easy to see how the two plots compare with another, especially when viewing Step Response plots and making only .25-1.0 changes in values. The only way to really tell is to do another REW measurement and overlay them in there. Depending on your setup this may be easy enough but in my case I need to unhook all the connections from my Lynx Hlo and my Benchmark DAC3 then reconnect them to my Hapi DAC in order to get a real picture of how it will all look using my full-time DAC.

 

The Hilo and the DAC3 combo are used in order to have enough XLR output channels for my purposes. I use all of the TRS and XLR Analog Outputs on the Hilo then feed two more channels out thru the AES Output to the AES Input on the DAC3. The Hapi has enough channels but I don't have an AD Card in it so use the other two DAC's together for measurements only.

 

These are all just first world problems of course but they add quite a bit of time to the evaluation process.

[cjf]

1 hour ago, Jean Paul D said:

you can x your measurement by your filter in REW and that simulation is quite accurate 

Hello,

 

Would you mind expanding on how this is done? I'm not sure I fully understand what you mean.

 

Thanks

[cjf]

1 hour ago, ecwl said:

Sounds like you’ve pushed your system very far.

 

If you still have suspicion that you can push it further, you can always ask @mitchco or just hire him through his new service to setup the convolution filter for you. You already have all the latest measurements. You already know what your preferred frequency response is. So it’s just a matter of the optimal way to build the filter so I presume the cost would not be exorbitant.

 

That said, sometimes, part of the fun is learning and understanding more about convolution filters. On the other hand, sometimes, it’s just more fun to enjoy all kinds of music without the need to fiddle around with filters and doing repeated A/B testing.

Hello,

 

Yes I've enjoyed reading about the information Mitchco has been generous enough to share around the web and agree, at least in my case, part of the fun is forcing myself to understand how all this works. At some point I may consider contacting him to see how much further I can go once a hack like me is done playing with it all.

 

Thus far though I've been quite pleased with the new Sub layout and filters I've managed to cobble together with this last round of changes I will be posting up. I'm sure they won't check all the boxes for what some might want to see in an ideal set of measurements but for my listening habits and music preferences it seems to work quite well IMO.

 

It seems to come together well at higher SPL levels and the Subs are blending in as if they aren't there and it sounds like you are just listening to much larger pair of main speakers with a bunch more power behind them. When the bass is there in the source material the Subs make themselves known but otherwise despite having a bunch more additional cone space pushing air around the room they are basically invisible and blend in nicely with the mains which was my ultimate goal.

 

When doing demos for buddies I know they are probably expecting to be vibrated out of the listening chair based on the number and size of the subs involved but it just doesn't sound that way at all unless the source has such low Freq content in it...which almost none of it does. What is quite obvious though is the scale and size of the sound of the venue which is hard to explain to those who are not exactly Audiophiles and don't know what to listen for. They know know it sounds good but don't necessarily understand why which is fine by me as long as it works for me, the primary listener 🙂

[cjf]

7 minutes ago, Jean Paul D said:

 

@Jean Paul D

 

Thanks for the info on that. How would you describe the level of detail in the documentation for REW for how to use this feature? I've never tried to use it or noticed it before so do you think the documentation is sufficiently detailed in describing how one can use this feature?

 

Thanks again

[cjf]

Hello,

 

This post is a follow-up to show how things are now measuring after moving the Subs around one last time and after dialing a bit more some of the Audiolense Filters/Corrections. Will call this last set of measurements "Config #3" and it will be compared to the previous "Config #2" which as shown before did not have as nice a Freq Response graph as the original "Config #1" but despite this, Config #2 did beat Config #1 in all other measured categories minus the Freq Response chart...IMO.

 

I'll dive right in with the comparisons but first I will show where the new Sub placements for Config #3 ended up and why I decided on this final placement.

 

S1 = Rythmik F18-SE

S2 & S3 = Rythmik E15HP2-SE's

 

 

Its not shown in the diagram above but there is a half wall that runs in between the placement of S2 & S3. This half wall is basically just providing a separation of space between the main hallway and the listening room. I've suspected for quite awhile that this wall was causing me all sorts of headaches in terms of room acoustics (in addition to the 12ft peaked ceiling above it). This time around I decided to place one of the Subs on each side of that half wall which I hadn't tried to do before. I hadn't tried it mostly because I felt it would have a negative effect on aesthetics and the livability of S2's location which is near the front door. But as it turns out, its not as bad as I expected it to be. I will need to relocate an AC Outlet because of S2's new location but that's not too horrible of a compromise IMO and I've already found a nice table that covers S2's location pretty nicely.

 

For those interested in a pretty cool looking table that can cover up a Sub the size of a Rythmik E15HP2-SE (and smaller of course) below is a link to the one I bought. I did perform before and after measurements with the table in place or not and it measured almost exactly the same except for an inconsequential difference way up in the Freq Range of the Sub long after its usefulness had already started to die off (250hz ++).

 

https://www.wayfair.com/furniture/pdp/gracie-oaks-tannenbaum-end-table-w000573359.html

 

 

Ok now onto the measurements:

 

Frequency Response - ALL CH Combined

 

Config # 2 = Blue Line (1/12th smoothing in effect)

Config # 3 = Pink Line (1/12th smoothing in effect)

 

 

Frequency Response - Left/Right Channels (ie..Ch Matching)

 

Config # 2 = Orange/Green Lines (1/12th smoothing in effect, Line values 1&2)

Config # 3 = Red/Purple Lines (1/12th smoothing in effect, Line values 4&5)

 

 

So comparing these first two charts the main difference is seen from the X-Over Point (60hz) and below where I was able to get Config #3 to measure a decent bit smoother than Config #2. Channel matching is pretty much a wash and both look OK to these eyes.

 

 

Distortion - ALL CH Combined (Cross-hair at largest peak @70hz)

 

Config # 2 = Blue Line (8.4% THD)

Config # 3 = Pink Line (1.4% THD) ..picked up a few percentage points reduction here with the new configuration and is pretty close everywhere else across the board to Config #2

 

 

 

Impulse Response - Left/Right CH

 

Config # 2 = Bottom Red/Green Lines (Line values 1&2)

Config # 3 = Top Orange/Green Lines (Line values 4&5)

 

 

For this comparison it appears both Config #2 & Config #3 are in Time Alignment with their L/R Channels. I'm not sure if one could be considered better than the other or not. The obvious signs of Pre-Ringing as seen in Config #1 are not seen here but there are a few more wiggles before the initial spike with Config #3 than with Config #2. My graph interpretation knowledge of Impulse Response measurements is a bit lacking here to say which one could be considered better.

 

Step Response - Left/Right CH's - Config #2

 

 

Step Response - Left/Right CH's - Config #3

 

 

For the Step Response comparison I was able to tighten up the Config #3 chart compared to Config #2 across the board. The channels appear to be better matched with Config #3 and as you can see the lines converge with each other at the far end of the graph unlike in Config #2 where they follow their own path separate from each other after the mid point

 

Group Delay - Left/Right & Combined CH's - Config #2

 

 

Group Delay - Left/Right & Combined CH's - Config #3

 

 

For the Group Delay comparison Config #3 shows a smoother series of lines compared to Config #2, especially from the X-Over point (60hz) and below.

 

So for these last few sets of measurement comparisons (Impulse/Step/Group Delay) the changes were dialed in by tweaking the Correction Procedure Designer settings in Audiolense in addition of course due to the new Sub placements.

 

The only difference between Config #2 & Config #3 CPD settings are that in Config #3 I have Disabled "Prevent Treble Boost" and increased the TTD Window at 10hz from "5.00" to "5.20".

 

 

Waterfall - All CH Combined - Config #2

 

 

Waterfall - All CH Combined - Config #3

 

 

Spectrogram- All CH Combined - Config #2

 

 

Spectrogram- All CH Combined - Config #3

 

 

So comparing the smoothness, coverage and discontinuities of these last series of graphs between Config #2 & Config #3 we can see a pretty big difference in the Waterfalls plots between the two. We can see far fewer peaks/valleys in the lower Freqs and a less overhang as well in those same lower Freqs.

 

The Spectrogram plots show another way to see the same smoothing effect between Config #2 & Config #3 but the Waterfall plot makes the differences much more obvious to see.

 

To bring a close to this post I want to circle back to where it all started where I made the statement that the Freq Response graph alone doesn't really tell the whole story. When looking at all the comparisons thus far Config #2 measured better than Config #1 in all categories besides Freq Response. And now, at least as far as my interpretations go of these last set of graphs between Config #2 & Config #3 it appears that the new Config #3 now betters Config #2 in almost all of those same measured areas.....but yet again, it does not appear to better Config #1 in terms of Freq Response as shown below:

 

Frequency Response - Left/Right Channels & Combined

 

Config # 1 = Top Brown/Teal/Purple Lines (1/12th smoothing in effect, Line values 4,5&6)

Config # 3 = Bottom Orange/Green/Blue Lines (1/12th smoothing in effect, Line values 1,2&3)

 

 

I guess its debatable depending on who you ask which Freq Response graph looks better but at first glance I suspect most people would choose Config #1 over Config #2 & Config #3 if they didn't look at any of the other measurements involved but the way I see it now, after going thru this whole exercise and analyzing all the data, a better way to chose between two or more sets of Convolution filters/corrections would be to pick the one that measures better in the most number of areas being analyzed with REW.

 

In this case we looked at 7 different sets of measurements. Config #1 wins in one out of seven categories while Config #3 wins in six out of seven categories. Based on my listening tests the configuration that has more "Wins" (Config #3) sounds the best in a very noticeable way when compared to the configuration that looks best in just one area.

 

I want to say Thanks again to all those who have contributed to this thread and I appreciated hearing the different insights and opinions on the information I have posted here.

[cjf]

4 hours ago, pompon said:

Scaling bad ...

 

try 60 to 110dB unless your graph is not very precise.

I can post a forensic view of the Freq Response plots tomorrow if that's what you mean?

 

I personally don't find them useful once they start looking like an upside down "U" just to squeeze the whole range into the zoomed view but understand others may feel differently.