Center with good directivity for Revel M126Be?

[napilopez]

This was my issue as well. I tried so hard to just stick with stereo, but many shows and movies the dialogue just gets lost in the mix. I try to avoid the dialogue boost function whenever possible considering it just boosts everything in those frequencies and can get boomy/muddy. Being able to independently increase center channel content volume is the key benefit to adding that speaker.

It sounds like the M105 would be your best bet.

I did just receive the M105 a few hours ago and thankfully it's immediately an obvious improvement! I try to avoid making immediate claims about SQ but in this case it was so obvious. Yay me!

Definitely agree about being able to adjust center volume independently.

Never move the cat. They hold grudges and attempts to appease them with wet meat will be seen, rightly, as signs of weakness.

Interesting point about the dialogue mix, but I think this is a common misconception about the centre channel. It is not a dialogue channel! When you have a centre speaker it becomes the main speaker for the front of your system, and the front L+R become effects channels. This makes sense if you think about a movie theatre, where the centre channel covers the width of the screen. See this post and thread for interesting discussion on the topic.

So adding a centre won't for most movies solve the dialogue intelligibility problem. There are movies where the centre is mostly dialogue but they are rare. In fact unless it has ~double the SPL capability of either of your L+R speakers it will actually reduce power handling and presumably increase distortion as well.

Thank you for sharing this. I was aware of this to some extent, but this thread does help me see the degree of how much movie content is sent to the center.

That said, there are some other benefits to using the center channel for movies and music. I think one part of it is it avoids the interaural crosstalk dip around 2kHz. From Toole's book

The other part of it is being able to adjust dialogue separately by adjusting the center channel volume. I know more than just dialogue goes here, but subjectively it seems to be significantly more effective than raising the dialogue level with my reciever (at least on my Yamaha. Maybe it's better with others). I like being able to clearly hear the score from the non-center speakers while the dialogue comes from the center.

I think there's some additional clarity gained with having various sounds coming from different sources/directions when the scene gets busy.

One extra reason of the other reasons I wanted a better matching center was that the Center Spread function on my reciever should be better. So far that's true -- the center snaps very nicely between thee speakers even if I move somewhat away from the center. Just did some A/Bing and the center is significantly fuzzier with the Def Tech.

Raise the TV a bit? IMO the answer is rarely anything but identical LCR.

Going from 708i L/R and SCL-3 center in the old house to 708i LCR is probably the biggest improvement in this system compared to that one.

Thankfully the M105 seems quite close enough for now. Certainly can't immediately notice an immediate drop in quality/tonality compared to the M126Be. Maybe a little brighter but I think I can EQ that away. I suspect it's the more linear(less sloping) response at 4-6khz off axis.

 

[rynberg]

Interesting point about the dialogue mix, but I think this is a common misconception about the centre channel. It is not a dialogue channel! When you have a centre speaker it becomes the main speaker for the front of your system, and the front L+R become effects channels. This makes sense if you think about a movie theatre, where the centre channel covers the width of the screen. See this post and thread for interesting discussion on the topic.

So adding a centre won't for most movies solve the dialogue intelligibility problem. There are movies where the centre is mostly dialogue but they are rare. In fact unless it has ~double the SPL capability of either of your L+R speakers it will actually reduce power handling and presumably increase distortion as well.

You would be mostly correct if surround channels were being used. But when 5.1 content is downmixed to stereo, now the L+R contain all of the center channel content (which you are correct, is not just dialog) and the surround channel content. The clarity of listening to a native 5.1 mix in true 5.1 (or in this case, at least 3.1) over stereo is undeniable.

EDIT: one thing I always find fascinating is how many people forget or never knew that "stereo" content was never meant to only be two channels, the original goal was three with a "center" channel. We are all stuck with compromised stereo today thanks to media limitations (and cost) in the 1960s.

 

[napilopez]

I totally forgot that I'd put in an offer for a Revel C12 and the seller accepted so I guess I'll have a chance to see how such an old relatively affordable center holds up.

Wondering if it's worth the trouble of making a full spinorama for it myself. Sure would be nice if revel would just, you know, send me the spin they have. They haven't responded to my email about it, and when I asked JBL for a spin of the L52s, they told me they don't provide spins to customers anymore. That's really unfortunate.

 

[napilopez]

Well, as mentioned on another thread, I ended up selling the M126Be's because I found I was plenty happy enough with the JBL L52s plus a little EQ. Very similar tonality out of the box other than bass below 200Hz, and extremely similar tonality after some EQ. Directivity is very close too. So the center channel quest now applies to the JBLs instead.

I received the Revel C12's yesterday. My impressions compared to using a single M105 as a center are mixed.

Out of the box, the first thing I noticed is that they seem a good deal more rolled off than either the M105's or L52s. This was quite apparent when playing white noise. I think EQ fixes it, but I have to run more tests.

I can't see much to indicate this roll off in measurements. Being such an old model, its possible something is wrong with the tweeter, even though I bought it "new". Age aside, the box had clearly been opened before and the grille arrived broken,, though the speakers themselves seemed almost mint.

I also wonder if this is one of the few cases where break in actually matters a bit, given the speakers appear to have been sitting in a box for years.

On the other hand, they do seem very dynamic, they look much nicer in my space and next to the L52s, and they have a large soundstage with an image that's a little less localized than the M105.

I was hoping to avoid getting my Umik out. Alas...

 

[napilopez]

Well, as mentioned on another thread, I ended up selling the M126Be's because I found I was plenty happy enough with the JBL L52s plus a little EQ. Very similar tonality out of the box other than bass below 200Hz, and extremely similar tonality after some EQ. Directivity is very close too. So the center channel quest now applies to the JBLs instead.

I received the Revel C12's yesterday. My impressions compared to using a single M105 as a center are mixed.

Out of the box, the first thing I noticed is that they seem a good deal more rolled off than either the M105's or L52s. This was quite apparent when playing white noise. I think EQ fixes it, but I have to run more tests.

I can't see much to indicate this roll off in measurements. Being such an old model, its possible something is wrong with the tweeter, even though I bought it "new". Age aside, the box had clearly been opened before and the grille arrived broken,, though the speakers themselves seemed almost mint.

I also wonder if this is one of the few cases where break in actually matters a bit, given the speakers appear to have been sitting in a box for years.

On the other hand, they do seem very dynamic, they look much nicer in my space and next to the L52s, and they have a large soundstage with an image that's a little less localized than the M105.

I was hoping to avoid getting my Umik out. Alas...

Update: Well, lesson learned. Measure when you buy used... and don't get rid of the box.

I was a little suspicious the speaker wasn't actually new as promised by the selelr given the grille arrived broken and there seemed to be a few scuffs. But given it was in the original box I was willing to give them the benefit of the doubt.

After some switching back and fourth with the M105, it was clear the M105 was much better out of the box, but I wanted to give the C12's a try. A week later I kept feeling like something was really off with the speakers. I tried EQing problem areas away but no luck.

So I dug out the old Umik and, it's much worse than I imagined:

Fun!

That's an on-axis gated measurement ~4.5ms at 1m. A 2m in-room measurement is not quite so bad, but still there's obviously something wrong.

As a sanity check, I did a quick quasi-anechoic measurement of the M105, which matches harman's on-axis measurement nicely, even without taking much care on microphone placement.

So, M105 it is. In the meantime, I guess, I'll try to see what exactly is wrong with the C12. The low woofers seem fine, at least.

 

[thewas]

That narrow dip at 3 kHz could be due to wrong polarity of one of the drivers, if I were you I would just reverse it and measure again, maybe you are lucky and it was just that.

 

[napilopez]

That narrow dip at 3 kHz could be due to wrong polarity of one of the drivers, if I were you I would just reverse it and measure again, maybe you are lucky and it was just that.

I was going to make a separate thread asking, thanks for that!

I don't know enough about speaker design to be sure, but compared to other nearfield measurements I've made, the individual driver responess seem pretty normal (the tweeter is always wonky due to the waveguide and phase shield)

Measuring the drivers individually the responses all seem pretty normal without baffle step compensation, although I would've thought the tweeter should extend a little lower:

The bass and mid woofers sum nicely as expected with baffle step correction (very rough sum, by eyeballing it, I didn't properly measure driver sizes and levels):

But can't quite get the tweeter to integrate smoothly, even when inverting polarity and messing around with different timings in REW. Still, I'll see if physically reversing the polarity helps.

 

[ernestcarl]

can't quite get the tweeter to integrate smoothly, even when inverting polarity and messing around with different timings in REW. Still, I'll see if physically reversing the polarity helps.

Yes, you really have to measure each driver individually at the same reference distance. You probably already know this, but REW's alignment tool feature can be a great visual aid here:


*polarity of LF MW inverted in second plot

----

Assuming that none of the drivers are broken, perhaps a bit of FIR DSP can help improve the xo summation? Unfortunately, this would be an additional expense.

 

[napilopez]

Yes, you really have to measure each driver individually at the same reference distance. You probably already know this, but REW's alignment tool feature can be a great visual aid here:

View attachment 359506 View attachment 359507
*polarity of LF MW inverted in second plot

----

Assuming that none of the drivers are broken, perhaps a bit of FIR DSP can help improve the xo summation? Unfortunately, this would be an additional expense.

Thank you! But indeed, the alignment tool was what I was using when I said I messed around with different timings and polarities. I ended up physically reversing the polarity on the tweeter and while it gets rid of much of the dip, the rest of the problems remain.

If I'm lucky, revel may have a replacement tweeter that isn't too expensive, or maybe it's becomes a DIY project to dabble in speaker design. The seller agreed to refund me half the amount so if I can find a ncie replacement tweeter it may not be a total loss. Maybe one of the DIY guys can help me find something third party that'll work