Thanks for the very clear explanation. 
The basic idea behind the shuffler is to eliminate the comb filtering by not allowing left and right to have identical phase above 900Hz. Shuffling kills the comb filtering with minimal audible affect on the left and right signals. I.E., if you are to listen to the left or right channel independently with or without the shuffle, you would hear little or no difference.
The advantage of the shuffle over EQ is that the shuffle gets to the root of problem, and will eliminate the comb filtering dips no matter where you are sitting, or what size your head is. Any fixed EQ would be good only for certain listening postilions.
Both the shuffle and EQ have their advantages and disadvantages.
Of course there are non-electronic ways to fix the imbalance. In fact most rooms, and most system, do it automatically.
In most systems, diffraction, early reflections and other acoustic messes will "shuffle" the phase enough to eliminate any stereo comb filtering. So of course one could introduce these mechanical shuffles into the system to kill the problem.
The other to fix the gaps could be electro-mechanical, such as Wesayso Hass kickers. Or perhaps other drivers near the front to splash some mid-high around and break up the phase. Remember, as noted, the most important dips are circa 1800Hz and 6KHz. That's where you'd want to concentrate your efforts.
The basic idea behind the shuffler is to eliminate the comb filtering by not allowing left and right to have identical phase above 900Hz. Shuffling kills the comb filtering with minimal audible affect on the left and right signals. I.E., if you are to listen to the left or right channel independently with or without the shuffle, you would hear little or no difference.
The advantage of the shuffle over EQ is that the shuffle gets to the root of problem, and will eliminate the comb filtering dips no matter where you are sitting, or what size your head is. Any fixed EQ would be good only for certain listening postilions.
Both the shuffle and EQ have their advantages and disadvantages.
Of course there are non-electronic ways to fix the imbalance. In fact most rooms, and most system, do it automatically.
In most systems, diffraction, early reflections and other acoustic messes will "shuffle" the phase enough to eliminate any stereo comb filtering. So of course one could introduce these mechanical shuffles into the system to kill the problem.The other to fix the gaps could be electro-mechanical, such as Wesayso Hass kickers. Or perhaps other drivers near the front to splash some mid-high around and break up the phase. Remember, as noted, the most important dips are circa 1800Hz and 6KHz. That's where you'd want to concentrate your efforts.
Pano: I have recently completed some mods on my Behringer DEQ 2496. As I was doing this I had a more detailed look at the owner's manual and discovered that the DEQ does have a shuffling option of some sort. Have you had a chanced to use the DEQ version? Do you consider it a good implementation of a shuffling circuit?
I am not surprised that folks don't know about the Shuffle feature since it is buried in the menus and the in the manual. It is described on page 11 of the owner's manual and can be found under the "Width Menu". I confess that I have not yet played with it (I am still fine tuning the output section modifications)
I understand that is what the "Width" menu performs. However, the manual implied the "Shuffle" menu was an additional circuit/function that did other things beyond what the "Width" circuit did.
I really need to get my unit re-assemebled so I can see if there is a hidden treasure in there. I am only guessing at this point ....
Thanks WT. I looked up and the DEQ Shuffle and found it's not a new dance. Seeing it again, I remembered that it's part of the Width function. It's not going to work for the center image, as it's a bass shuffle meant to widen the bass. You can adjust its low pass filter up to 1500Hz, but that unfortunate the opposite of what we want to do. We want to high pass the shuffle above 900Hz.
As Wesayso mentions, this is a different effect. I probably tried it, as I did play around with the Width function, but it never stuck with me. Guess it didn't do anything useful on my system.
Anyone else ever try it?
Seeing it again, I remembered that it's part of the Width function. It's not going to work for the center image, as it's a bass shuffle meant to widen the bass. You can adjust its low pass filter up to 1500Hz, but that unfortunate the opposite of what we want to do. We want to high pass the shuffle above 900Hz.As Wesayso mentions, this is a different effect. I probably tried it, as I did play around with the Width function, but it never stuck with me. Guess it didn't do anything useful on my system.
Anyone else ever try it?
I dug around all over their website and a lot of other places looking for a demo. Could not find nothing.
You'd think they would have some sort of demo file.I have heard the DTX virtual headphone thing and it was good.
https://www.manualslib.com/manual/677558/Behringer-Ultracurve-Pro-Deq2496.html?page=9#manual
It's based on the Blumlein shuffler, more info here about the principles: http://www.audiosignal.co.uk/Resources/Stereo_shuffling_A4.pdf
What it does is mid/side EQ but different for low frequencies than higher ones.
This is similar to the S-curve mentioned in this thread a couple of times. It's related (and probably fun to experiment with) but does not solve the combing problem.
I never played with it. My behringer DEQ is still new in its box since I bought it in 2011 [emoji15].
Last edited:
Hi,
I have a Pi seven setup whose loudspeakers are positioned in corners and fire into the room at 45 degree angle. If I sit in the sweet spot where two speakers firing lines intersect, the center tonal shift is really negligible. I guess this position get more mid-hi sound from the speakers to the ears (45 degree angle). So maybe besides the comb filtering effect, there is also a problem of the insufficient amount of directional sound waves reaching to ears.
Could you show an IR measurement at the listening position? That will show if there are reflections (and at what level) etc. which can hide the phenomenon we are talking about in this thread. It would help to get a grip on what you are hearing at that spot. An APL_TDA plot would help too.
An IR at my listening spot has been posted in this thread on several occasions.
As well as the separate left and right and it's sum at the listening location.
Reflections will hide the tonal defects caused by the combing problem. I made sure to make my listening spot as "reflection free" as I could (*). Only after that step did this comb filter "problem" show up.
It's always better to measure than to assume. Even with corner speakers. I'm assisting someone with a corner horn setup to improve his results at the listening position which does show it has some problems. A system like that isn't always a 100% guaranteed ticket to a reflection free listening spot. Measure to be sure.
If any of you don't have this problem? Good for you and I can accept that as an answer.
If you're happy with your system: even better! (*) = Personally, I had very specific goals and those goals have made me more aware of the existence of this problem. One of the goals was get more direct sound compared to indirect sound, a clean first wave front.
Even before the launch of this thread I was already working on a solution for myself. I found a solution that works for me and shared it here.
I don't know where the nulls would be with speakers at a 90 deg spread. Different from 60 deg. Might have to figure that out. My speakers were at less than 60 deg when I noticed the problem.
But as stated above, early reflections will kill the comb filtering, and you won't hear the tonal imbalance.
But as stated above, early reflections will kill the comb filtering, and you won't hear the tonal imbalance.
I will see what I can do about the IR measurement. I have never been into this before, relying rather on other people's expertise so far.
I hope you do. It will help us see what you are listening to at the listening position. Even though a corner horn should avoid a lot of reflective surfaces we should still check to see what it does. We won't get ahead by assuming things.
I'm in here for the lessons. Together we can educate ourselves way more than each of us on his own.
Do you own or have access to a proper microphone for these type of things?
I'm in here for the lessons
. Together we can educate ourselves way more than each of us on his own.Do you own or have access to a proper microphone for these type of things?
I have a little concern: when I listen facing the front center, my ears are facing the side walls. Since mid-hi frequency sounds are directional, the angle of ear in relation to speaker is important. This angle decides how much spl of mid-hi sounds your ear gets from the speakers. I think the general shape of the ear also come into play here affecting something (but I have no clue). How would position the mic towards the speaker is comparable to the ear facing the side wall when we listen?
Best would be to measure it with both orientations to see if it makes a difference .
Just one more measurement. REW would be an easy tool to do this.
Before I got a calibrated microphone I did my share of measurements in my car with a RadioShack SPL meter.
If we get an IR measurement, zoom in to the first 20 milliseconds to see what happens there. It should show a clear first peak. Depending on what happens in the room it could be followed with a few other peaks (reflections). That's what we need to see first.
.Just one more measurement. REW would be an easy tool to do this.
Before I got a calibrated microphone I did my share of measurements in my car with a RadioShack SPL meter.
If we get an IR measurement, zoom in to the first 20 milliseconds to see what happens there. It should show a clear first peak. Depending on what happens in the room it could be followed with a few other peaks (reflections). That's what we need to see first.