Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[mwillems]

So I'm at the tail end of a long speaker design project, and now have a pair of biamped speakers and a sub that I'm driving through three separate amps.  I use JRiver (in connection with an 8-channel Asus Xonar DX) as a digital crossover/EQ/signal processor.  Critically, because my computer is my DAC, and all my elements are driven by different amps, JRiver is also my primary volume control.  And it works really, really well for almost all my applications (music sounds better than I've ever heard in software), I just have some trouble with multichannel sound in movies.

I currently have output format set to "2 channel in 5.1 channel container" for bi-amping (I need at least two additional dead channels, and this setting works great for that).  I also have JRSS mixing enabled. 

I have a question in two parts: 

1) Whenever I watch a movie the dialogue is significantly lower in volume than the rest of the movie in such a way that there's no single volume setting that "works" where all dialogue is intelligible, but I'm not getting blasted out of my seat.  I recognize that movies often have wide dynamic range by design, and that's great with non-vocal audio output (I appreciate the wide dynamic shifts in music, sound effects, etc.). 

My question is, is there an easy way to boost the center channel (where almost all the dialogue is) a customized amount?  I've seen a few other threads on this topic and the suggestion there was to use adaptive volume; however I can't really hear much difference for dialogue with "light" adaptive volume enabled, and while medium makes all the dialogue intelligible, it seems to really squash the dynamic range, and the dialogue seems much too loud compared to the rest of the soundtrack.  I'd really like to be able to just boost the center channel a few dB

Are there settings (whether in DSP studio or otherwise) that I could use to boost center channel output in a controlled way?  My understanding is that JRSS mixing must happen first in the signal chain, or I would just use the PEQ to do it myself before the downmix.

2) Assuming that there's no way to do what I'm asking about in 1), I've got some extra drivers and plenty of left over wood. I could pretty easily build a center channel for the express purpose of being able to adjust the dialogue volume, but a problem with that approach occurs to me:  How would I set up my output format in JRiver?  There is no 3.1 setting (much less a "3.1 in a 5.1 container" setting, for my bi-amp channels).  So I won't be able to use JRSS for my mixing in that scenario.  If I set the output to 5.1, how do I mix the surround channels into the mains? Do I just add them together in PEQ?  Is there a better way to do that?  Will it sound terrible?  I only ask because I don't want to build a speaker and wind up in a worse spot than I started ;-)

I'd appreciate any help on this point as my WAF is stretched pretty thin at this point (what with the gigantic bi-amped speakers I just installed in the living room), and movies are pretty important to the W.

Thanks.

[Matt]

Sounds like a neat setup.  Tell your wife huge gigantic bi-amped speakers are some of the nicest furniture you can have!

1) Whenever I watch a movie the dialogue is significantly lower in volume than the rest of the movie in such a way that there's no single volume setting that "works" where all dialogue is intelligible, but I'm not getting blasted out of my seat.  I recognize that movies often have wide dynamic range by design, and that's great with non-vocal audio output (I appreciate the wide dynamic shifts in music, sound effects, etc.).

 

I've seen a few other threads on this topic and the suggestion there was to use adaptive volume; however I can't really hear much difference for dialogue with "light" adaptive volume enabled, and while medium makes all the dialogue intelligible, it seems to really squash the dynamic range, and the dialogue seems much too loud compared to the rest of the soundtrack.

You answered yourself with the advice I was going to give.

Light Adaptive Volume preserves the original mix as you would hear with a calibrated surround system and preserves the dynamic range (although the first loud passage could turn the volume down one time if you don't use Internal Volume).

Medium Adaptive Volume adds 3dB to the center and does conservative dynamic range compression.

I think you're sort of asking for a 'Light (with dialog boost)' mode.  Or maybe a discrete setting for center downmixing (although I'm not eager to complicate the Output Format dialog further).

Thoughts?

[mwillems]

Light Adaptive Volume preserves the original mix as you would hear with a calibrated surround system and preserves the dynamic range (although the first loud passage could turn the volume down one time if you don't use Internal Volume).

Medium Adaptive Volume adds 3dB to the center and does conservative dynamic range compression.

I think you're sort of asking for a 'Light (with dialog boost)' mode.  Or maybe a discrete setting for center downmixing (although I'm not eager to complicate the Output Format dialog further).

I think you hit on exactly what I'm looking for.  About a 3dB boost for dialogue/center channel without otherwise altering the dynamic range would probably solve this specific problem completely.  I understand your reluctance to complicate output format further, and I wouldn't advocate for that. 

What I was originally thinking about was allowing for JRSS to be applied later in the signal chain (i.e. after PEQ), or even just offering JRSS as a PEQ filter so people can put it wherever they want (in addition to having it as an option in output format).  My sense is that JRSS does volume leveling in addition to mixing which is what separates it from the "mix channel" PEQ filter? 

Regardless, you've got an incredibly powerful and complex set of options already, so I understand not wanting more "clutter" for unusual use cases.  A light plus dialogue boost would almost certainly resolve my issue, and if I ever build a center channel, I'll cross my 3.1 bridge when I come to it ;-) (or just build surrounds).

Sounds like a neat setup.  Tell your wife huge gigantic bi-amped speakers are some of the nicest furniture you can have!

That's what I keep telling her ;-)  She's been surprisingly supportive so far, even though the speakers are kind of at that Death-Star-in-the-Return-of-the-Jedi stage: i.e. the battlestation is fully operational, but they look like hell.  A big part of "making the sale" has been how good movies sound (thanks in no small part to your software).  JRiver has functionality that you just can't get in hardware, for a tenth of the price of the hardware that even comes close. 

Thanks for the quick response!

[mojave]

Since I use a phantom center, there are times when I would also like Light (with a dialogue boost). I think it should exclude media subtypes of music or music video. That probably makes it too complicated and ZoneSwitch should be used instead.

[mwillems]

Since I use a phantom center, there are times when I would also like Light (with a dialogue boost). I think it should exclude media subtypes of music or music video. That probably makes it too complicated and ZoneSwitch should be used instead.

Yeah, I would only use it in my movie zoneswitch "zone"

[Trumpetguy]

What I was originally thinking about was allowing for JRSS to be applied later in the signal chain (i.e. after PEQ), or even just offering JRSS as a PEQ filter so people can put it wherever they want (in addition to having it as an option in output format).   

This solution would maybe solve one problem, but would certainly create some challenges for me: I use 8 channel convolution, and upmix all 5.1 contents first using JRSS. Would not be the same if I was left with 6 channel convolution and then upmixing to 8 channels.

[mwillems]

This solution would maybe solve one problem, but would certainly create some challenges for me: I use 8 channel convolution, and upmix all 5.1 contents first using JRSS. Would not be the same if I was left with 6 channel convolution and then upmixing to 8 channels.

To be clear, I wasn't  suggesting that JRSS be changed to always happen after PEQ, I was suggesting making its position in the signal chain customizable (like most of the other filters in JRiver).  Right now it can only happen first as part of output format; I was just suggesting allowing it to happen later on in the chain as well, not only later in the chain.  

I definitely wasn't advocating for permanently/irrevocably relocating JRSS mixing to the end of the chain :-)

[Trumpetguy]

To be clear, I wasn't  suggesting that JRSS be changed to always happen after PEQ, I was suggesting making its position in the signal chain customizable (like most of the other filters in JRiver).  Right now it can only happen first as part of output format; I was just suggesting allowing it to happen later on in the chain as well, not only later in the chain.  

I definitely wasn't advocating for permanently/irrevocably relocating JRSS mixing to the end of the chain :-)

I got that, was just pointing out the obvious 

What is the reason for not using room correction and reduce all channels except the center by 3dB? Would that compromise dynamic range significantly?

[Matt]

I got that, was just pointing out the obvious 

What is the reason for not using room correction and reduce all channels except the center by 3dB? Would that compromise dynamic range significantly?

You can only change channels that are in the mix.  But the center is already gone since JRSS mixed it into the mains.

[Trumpetguy]

You can only change channels that are in the mix.  But the center is already gone since JRSS mixed it into the mains.

Sorry, didn't get the point about phantom center.