Print

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

[flac.rules]

BTW, subtracting 10 dB BEFORE the normal channels are added to LFE is important, if not relative bass levels will be wrong between the existing LFE--track and the new data added from the normal channels.

[mojave]

I just received a response back from Dr. Mark Waldrup, the president of AIX Records. He said their LFE track is attenuated by 10dB during the mix.

By the way, iTrax/AIX Records is to high resolution music as MC is to audio players.

[Matt]

I just received a response back from Dr. Mark Waldrup, the president of AIX Records. He said their LFE track is attenuated by 10dB during the mix.

Thanks.

Are there _any_ mainstream sources that have an LFE track that aren't assuming a +10 dB boost by the amp / subwoofer?

It won't give you those "flat curves" mojave, but I think maybe we could just drop the option and always attenuate redirected bass by 10 dB? (the curves would be offset by 10 dB)

[flac.rules]

Thanks.

Are there _any_ mainstream sources that have an LFE track that aren't assuming a +10 dB boost by the amp / subwoofer?

Not as far as i know, there are some examples of SACD and the like not doing it correctly, but I think as good as everything assumes a 10db boost. I think its wise to just assume 10 dB boost, and eventually create a DSP plugin that lets you control volume on all channels independantly, for people with special needs.

[Matt]

and eventually create a DSP plugin that lets you control volume on all channels independantly, for people with special needs.

Just a little note to say the 'Room Correction' DSP already allows per-channel volume adjustments.

[mojave]

Are there _any_ mainstream sources that have an LFE track that aren't assuming a +10 dB boost by the amp / subwoofer?

To be correct, I think most mainstream sources are assuming bass management will be used by the consumer in a DVD/Blu-ray player that is using analog outputs, receiver, or preamp/processor and that the LFE track will be boosted +10 dB.

Please read Alex's link from post 11. It is excellent for understanding the LFE channel. Here is the link again:

Feature Article - "The Misunderstood 0.1 LFE Channel in 5.1 Digital Surround Sound"

[flac.rules]

Just a little note to say the 'Room Correction' DSP already allows per-channel volume adjustments.

Indeed it does, then I say, ditch all the other unusual settings, with the possible exception of not attenuating bass data from channels 10 dB before adding them to LFE, and if people have very special needs, they can use the per-channel volume adjustments. That should take care of most problems(?)

[Alex B]

Are there _any_ mainstream sources that have an LFE track that aren't assuming a +10 dB boost by the amp / subwoofer?

The quote I posted earlier mentioned the DTS Music CDs. They were never mainstream, but there are lots of them in use. I have a few.

You can still buy DTS Music CDs, but I don't know how the LFE channel is calibrated in the available discs:
http://www.5point1.com/Products-5_1_DTS_Music_Disc_2.html

some other links:
http://en.wikipedia.org/wiki/5.1_Music_Disc
http://www.dts.com/Pro_Audio_Products/Products_by_Media/DTS_Music_Disc_CD/Overview.aspx
http://www.avrev.com/equipment-reviews/music+disc+reviews/music+disc+reviews/dts+5.1+cd.html


I'd suggest to preserve the option so that the calibration problem can be fixed if needed.

[Matt]

I'd suggest to preserve the option so that the calibration problem can be fixed if needed.

I think you had it right that there are really two questions:

[ x ] Something outside of Media Center will make the subwoofer channel 10 dB louder
[ x ] The data already in the LFE channel expects something to make it 10 dB louder

We added an option "Calibrated to +10 dB (Dolby standard)" under Subwoofer in DSP Studio > Output Format for the first question, but I think you're talking about the second question.

Any suggestions as to option names if we do want it to be optional?

Subwoofer
[ x ] Calibrated to +10 dB (Dolby standard)
[ x ] LFE soundtrack expects +10 dB

[flac.rules]

I think you had it right that there are really two questions:

[ x ] Something outside of Media Center will make the subwoofer channel 10 dB louder
[ x ] The data already in the LFE channel expects something to make it 10 dB louder

We added an option "Calibrated to +10 dB (Dolby standard)" under Subwoofer in DSP Studio > Output Format for the first question, but I think you're talking about the second question.

Any suggestions as to option names if we do want it to be optional?

Subwoofer
[ x ] Calibrated to +10 dB (Dolby standard)
[ x ] LFE soundtrack expects +10 dB

I dont really see the difference the way you have worded it, but I might just misunderstand you. The important question is this, is the existing LFE in the material 10 db lower than the rest of the channels? This decides if MC should mix in added content to LFE-channel 10db lower or not. The handling outside of MC is easily fixed by just dropping the voulme on the subwoofer channel, the important is that LFE-content and "the rest of the channels"-content mixed into LFE-channel are at the right level relative to each other.

[Matt]

The handling outside of MC is easily fixed by just dropping the voulme on the subwoofer channel

For some users, this is true.

But for others, there's no such thing as "handling outside".  I only have a power amp.

So to play 5.1 material mastered different ways, the program needs to know how the material was mastered and how the stereo is configured so that it can deliver consistent results.

[jmone]

For some users, this is true.

But for others, there's no such thing as "handling outside".  I only have a power amp.

So to play 5.1 material mastered different ways, the program needs to know how the material was mastered and how the stereo is configured so that it can deliver consistent results.

Looks like there will be three options to Calibrate the LFE:
1) MC
2) Receiver (which Matt does not have)
3) Vol Control on the Sub itself (which Matt is going to make!)

I'm looking forward to hearing these changes (as I thought the bass overemphasised in the 5.1 mix) once my HT is back together.   As usual I plan to calibrate my 5.0 speakers (using a Sound Pressure Meter and test tones) via the Receiver and then bring the Subs levels up to match using the Volume Control on the Sub itself. 

Thanks
Nathan

PS - I'm also hoping that these changes don't make MC's output levels different with other HTPC apps, eg TMT which I use for my blu-ray playback, or other sources such as the STB as then there becomes nother layer to calibrate to bring the entire HT into balance!

[Alex B]

The questions:

[ x ] Something outside of Media Center will make the subwoofer channel 10 dB louder

[ x ] The data already in the LFE channel doesn't expect something to make it 10 dB louder
(in other words: the poor guys in the studio didn't have their subwoofer correctly set to +10 dB and the recording sounds good only if also the playback system doesn't boost LFE)


The options:

Subwoofer
[ x ] Already calibrated to follow the Dolby/DTS +10 dB LFE standard (the internal LFE gain will be disabled)

LFE input
[ x ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)


EDIT: changed Dolby/DTS to +10 dB LFE

[mojave]

I agree with Elvis133's entire post.

Here is what I think you might be getting at, but I'm probably misunderstanding, too:

1.  You are mixing the redirected bass at -10 dB to match the LFE. This is then being sent out and the subwoofer is set at a higher level to match the rest of the channels. This allows more headroom for the other channels.

2.  You are mixing the redirected bass at -10 dB to match the LFE channel. This is then increased by 10 dB (or other channels attenuated) so that all channels including are output at the same level. This is how all receivers do it.

I'm not sure why you are referencing the Dolby standard because it isn't just Dolby that does this. Whether a Blu-ray outputs in Dolby TrueHD, DTS-HD, or PCM, the LFE track has still been attenuated by 10dB in the studio. This has become the standard for all multi-channel mixing except for the rare case of some DTS audio and some SACD's. Probably these are even mixed to this standard now.

[mojave]

The issue isn't about how the subwoofer is calibrated, it is whether or not the LFE track has been attenuated in the studio by 10 dB. The two options should be:

LFE & Redirected Bass Mixing
1.  LFE track is -10dB per normal mixing standards (DVD, DVD-Audio, Blu-ray)
2.  LFE track is the same as other channels (some DTS Audio and SACD)

[Alex B]

The issue isn't about how the subwoofer is calibrated, it is whether or not the LFE track has been attenuated in the studio by 10 dB. The two options should be:

LFE & Redirected Bass Mixing
1.  LFE track is -10dB per normal mixing standards (DVD, DVD-Audio, Blu-ray)
2.  LFE track is the same as other channels (some DTS Audio and SACD)

The two options can be set by the latter tickbox:
LFE input
[ x ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)


The subwoofer calibration is a different issue. Apparently Matt would like to provide an easy setting for it. The first tickbox would do the trick:
Subwoofer
[ x ] Already calibrated to follow the +10 dB LFE standard (the internal LFE gain will be disabled)


These two tick boxes will actually provide the funtionality of my # 1, 2, 3 and 4.

[mojave]

I posted the above so it could be read while typing some more.

There are two ways #1 could be handled which was what I posted earlier:

1.  You mix the redirected bass at -10 dB to match the LFE. This is then being sent out and the subwoofer is set at a higher level to match the rest of the channels. This allows more headroom for the other channels.

2.  You are mixing the redirected bass at -10 dB to match the LFE channel. This is then increased by 10 dB (or other channels attenuated) so that all channels including are output at the same level.


If you have a subwoofer with a pro amp or no volume control then #2 might be necessary. If you have volume control in your amp then #1 is preferred. This brings us to the following options:

LFE & Redirected Bass Mixing
1.  LFE track is -10dB per normal mixing standards (DVD, DVD-Audio, Blu-ray)
     a. Summed bass is -10 dB for maximum channel headroom
     b. Summed bass is the same level as other channels

2.  LFE track is the same level as other channels (some DTS Audio and SACD)

However, 1b can be accomplished by just using the subwoofer gain control in Room Correction. You would never just output directly to an amplifier without some volume compensation because of varying driver efficiencies. I don't think the equal levels is necessary and 1a would be fine as the option. Now we are back to:

LFE & Redirected Bass Mixing
1.  LFE track is -10dB per normal mixing standards (DVD, DVD-Audio, Blu-ray)
   
2.  LFE track is the same level as other channels (some DTS Audio and SACD)

[flac.rules]

For some users, this is true.

But for others, there's no such thing as "handling outside".  I only have a power amp.

So to play 5.1 material mastered different ways, the program needs to know how the material was mastered and how the stereo is configured so that it can deliver consistent results.

Then you can just adjust the LFE-level in the room correction DSP as long as the relative levels of material added to LFE, and the original LFE-channel is the same.

[flac.rules]

However, 1b can be accomplished by just using the subwoofer gain control in Room Correction. You would never just output directly to an amplifier without some volume compensation because of varying driver efficiencies. I don't think the equal levels is necessary and 1a would be fine as the option. Now we are back to:

LFE & Redirected Bass Mixing
1.  LFE track is -10dB per normal mixing standards (DVD, DVD-Audio, Blu-ray)
   
2.  LFE track is the same level as other channels (some DTS Audio and SACD)

I agree 100%, those are the only two options we need, the LFE-bass material is either 10 db lower or the same as the bass-material in the 5 other channels, as long as this material is added togheter correctly, the absolute level of the LFE-channel is much easier to fix/modify later. If the bass is not mixed together properly, it cannot be fixed by a simple gain adjustment in MC, receiver, sub or other part of the chain.

[mojave]

Alex, I guess I am breaking down your LFE option into two options instead of an on/off. This is just to help make it more clear to the user.

I see how the Subwoofer option might be necessary, but anyone that is using other sources and is calibrated to those will be astute enough to make sure the sub is calibrated in MC, too, and just use the levels control in Room Correction.

[Alex B]

Here is how the four options can be set:
1.

LFE    redirected
+10          0         could be the default. Useful when you have a calibrated HT receiver, but the digital calibration is bypassed when the analog 5.1 inputs are used.

Subwoofer
[   ] Already calibrated to follow the +10 dB LFE standard (the internal LFE gain will be disabled)

LFE input
[   ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)

2.

LFE     redirected
0            0        the content is mixed without taking the +10 dB boost into account

Subwoofer
[   ] Already calibrated to follow the +10 dB LFE standard (the internal LFE gain will be disabled)

LFE input
[ x ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)

3.

LFE    redirected
0        -10        the boost happens after MC in some other component. Provides louder output and more dynamic range (to all channels) if you can set the subwoofer permanently to +10 dB outside MC

Subwoofer
[ x ] Already calibrated to follow the +10 dB LFE standard (the internal LFE gain will be disabled)

LFE input
[   ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)

4.

LFE     redirected
-10        -10        the content is mixed without taking the +10 dB boost into account & the boost happens after MC in some other component

Subwoofer
[ x ] Already calibrated to follow the +10 dB LFE standard (the internal LFE gain will be disabled)

LFE input
[ x ] LFE soundtrack doesn't expect +10 dB (use only if the recording does not follow the standard)

[Matt]

It sounds to me like everyone is in agreement, and only haggling over naming.  That's good.

I was going to ask if anyone would ever not check the option "Something outside of Media Center will make the subwoofer +10 dB".  Then AlexB recommended not checking it by default and gave a good reason -- analog paths may bypass the boost of a calibrated receiver.

Thanks everyone.

[Matt]

I have a proposal for how to do the user interface.  In DSP Studio > Output Format, add a root level section 'Bass Management'.  Remove 'Subwoofer' settings from 'Channels'.

The new section would look like this:


=== Bass management ===

[ x ] Something outside Media Center will make the subwoofer +10 dB (standard calibration)

* When playing sources with no LFE (CD audio, etc.)
    [ Crossover combobox (Silence, No crossover, 20 Hz, 25 Hz, etc.) ]
    [ x ] Subclarity (tm) for cleaner, tighter subwoofer output

* When playing sources with an LFE (DVD, HDTV, etc.)
    [ x ] LFE is correctly mixed to +10 dB (only uncheck if the recording does not follow the standard)

Bass management is used to control how the subwoofer is handled.  The settings only apply when the output format has a subwoofer channel.

You can redirect bass to the subwoofer using the 'Room Correction' DSP.


By default, all checkboxes would be checked.

This way, by default, no audio changes would be needed when playing a 5.1 or 7.1 source.  Both the system and the LFE would be at the same level.

[Matt]

One final question: how technical should we be with regards to the terminology "subwoofer" vs "LFE"?

My inclination is to just always say "subwoofer" because I think it's easy to understand.  My wife does not know what an LFE is, but she knows that a subwoofer is that annoying thing behind the couch.

But I'm afraid this less-than-technically-pure usage may offend some people.  Advice?

[mojave]

Well, I finally understand why the electronic measurement should not be flat because of preserving energy in the satellite channels.

I really like the idea of having a bass management DSP. However, I still don't like the idea of the crossover combo box (silent, no crossover, 20 Hz, etc.). It only confuses things with the crossover settings in Room Correction. You can still do all this with the settings in Room Correction without someone creating double crossovers and filtering out part of their music. You (Matt) mentioned before that you like a different setting for music.You still need to turn off the crossovers in room correction before enabling them in the bass management DSP (or Output Format). It really isn't saving anything. You could have two zones setup:  one for music and the other for movies. You could even make an option in the Video settings to use the Video Zone.

I think the Bass Management DSP should have some default settings that are used by room correction (and it should be enabled whenever Room Correction is enabled). These settings would be the crossover frequency for each speaker (same for all speakers); whether the redirected bass is moved, copied, or removed; and the High and Low Pass Crossover slopes. It gets very tedious to change things all the time in Room Correction for each channel - at least when trying different stuff and calibrating the system.

In Room Correction you just add "Default" as an option to each of the above mentioned items. This allows for global changes in Bass DSP and variations in Room Correction.

Regarding the use of LFE and Subwoofer, I think you are using them correctly in the above bass management rough draft. I would actually add the word "channel" after LFE and call it the LFE channel. I thnk it is best to be technical. It makes it better in the long run when consistent terminology is used among varous manufacturers, magaines, software, etc.

Good Night!

[Matt]

However, I still don't like the idea of the crossover combo box (silent, no crossover, 20 Hz, etc.). It only confuses things with the crossover settings in Room Correction.

A 5.1 source has LFE and redirected bass, so why wouldn't 2.0 upmixed to 5.1 (or 2.1) have an LFE and redirected bass?  It seems like consistently structured audio coming out of 'Output Format' (always the first DSP) is desirable and makes getting consistent results from Room Correction easier.  The only counter-argument I can think is that the amount of redirected bass might be less for a 5.1 source depending on how they filtered it during mastering.

[mojave]

A 5.1 or 7.1 source does not have redirected bass. It only has the LFE channel and you are handling that in Room Correction and the Bass Management DSP. Dolby Pro Logic II or IIx only creates a 5.0 or 7.0 mix from a 2.0 source (and DTS:NEO 6 creates 5.0 or 6.0). It does not make an LFE channel. The LFE channel is created from redirected bass by the bass management used by the receiver. You used to not have bass management so JRSS had to create the .1 channel. JRSS should now do the same as Dolby Pro Logic to prevent problems with multiple crossovers. Have JRSS create 4.0, 5.0, or 7.0 mixes from 2.0 content. For 5.1 content, have it just create the surrounds and leave the LFE alone (like it does now). Then let Room Correction/Bass Management determine the crossover, etc.

Like I said before, you can't use both settings at the same time and Bass Management accomplishes the goal in all circumstances.

Dolby Surround Pro Logic II Decoder Principles of Operation

[flac.rules]

A 5.1 or 7.1 source does not have redirected bass. It only has the LFE channel and you are handling that in Room Correction and the Bass Management DSP. Dolby Pro Logic II or IIx only creates a 5.0 or 7.0 mix from a 2.0 source (and DTS:NEO 6 creates 5.0 or 6.0). It does not make an LFE channel. The LFE channel is created from redirected bass by the bass management used by the receiver.

AFAIK that is not entirely precise, the receiver takes all material below crossover and outputs it to the sub, no matter if its in the LFE-channel or any other channels, I don't think it creates a LFE-track per se if it gets 5.0 input.

[jmone]

I'm looking forward to testing all this on the WE, I plan to:
1) Recalibrate the 5.1 speakers from the reciever's test tones using a SPL meter
2) Check calibration output from MC using it's test tones

Thanks
Nathan

[Matt]

A 5.1 or 7.1 source does not have redirected bass.

Are you saying you want bass redirection (and speaker crossovers?) disabled when playing 5.1 or 7.1 content?  I think this is the first I've heard of this, but maybe it could be an option.

[Alex B]

I think his point is that you should first create the full range channels from the source signal and only after you have the finished new audio signal ready apply the bass management.

If a separate LFE channel exist in the original unmodifed source it should be kept out of all processing except the +10 dB adjustment that will be applied only to the LFE signal (if enabled).

Naturally in the case of the separate LFE source you must maintain the correct balance between the volume levels of the other channels and the LFE channel (in this equation the LFE channel is normally set to be +10 dB louder than it was in the original source data.)

So, for instance, when you create surround from a stereo source create only the five (or seven) full range channels and only after that let the bass management determine if the lowest frequencies should be redirected to the subwoofer. (Isn't this already how it works?)

Another example could be 5.1-> 7.1:

1. Process the five full range channels and create seven channels.
2. Manage bass, i.e. create the redirected bass mix and the seven high passed channels
3. Mix LFE (+10dB) and the redirected bass mix (0 dB) and create the subwoofer channel

Which is exactly the same as this:



- except that the subwoofer mixer is not implemented in the UI as a mixer with adjustable sliders. The only available adjustment is the on/off switch for the +10 dB gain so that incorrect recordings can be fixed.

Here's one more step (not pictured):
4. A setting that attenuates the final subwoofer channel by -10 db if:
    "Something outside Media Center will make the subwoofer +10 dB"


This is the logically correct order of the steps. Naturally you must decide how the processing is actually done under the hood. You may need to change the order, combine some steps and adjust the gain in several places in order to maintain the correct balance and prevent the final output from clipping. As long as the end result is the same that doesn't matter.

[mojave]

For clarity, I'm going to start calling the output that contains the LFE and redirected bass "Channel 4." This is the actual channel number that is being used. On Channel 4, you might have an LFE track which is the Low Frequency Effects (LFE) created in the studio for movies, you might have a bass track on multichannel audio, and it will be blank with a stereo source. Channel 4 is also used for the redirected bass from other channels.

Are you saying you want bass redirection (and speaker crossovers?) disabled when playing 5.1 or 7.1 content?  I think this is the first I've heard of this, but maybe it could be an option.

No, I don't want that at all. I was responding to your statement that "A 5.1 source has LFE and redirected bass." I was trying to say that a 5.1 source has all discrete channels and the redirected bass is the function of bass management. I was also pointing out that Dolby and DTS matrixing leave Channel 4 blank when matrixing. It is the bass management that either copies or moves the redirected bass to Channel 4.

You aren't really "creating" anything on the channel 4 with JRSS 5.1 or 7.1. You are leaving the other channels full range, and copying the bass below the crossover to channel 4. This is the same as using Bass Management and setting the routing in Room Correction to "Copy bass to Subwoofer (LFE)." By having it done in JRSS, you are removing the ability to set High and Low Pass Filter slopes. You are also causing confusion because some will set the crossover both in JRSS and in Bass Management. This will result in problems with the frequency response and phase.

Try this:  Listen to a stereo audio track with JRSS on with 5.1 output and Subwoofer to 100 Hz and disable the bass routing for the L & R channels in Room Correction.

Now set the JRSS's Subwoofer setting to silent and in Room Correction set the routing for the L & R channel to a crossover of 100 Hz with "Copy bass to Subwoofer" selected. Set the slopes to match whatever you use when the crossover is active in JRSS. Now listen again. These should both sound exactly the same. Below are what they look like.

[mojave]

The High and Low Pass Crossovers in Room Correction need to be labeled. I know what they are, but most people wouldn't be able to tell them apart. They also shouldn't be selectable in Room Correction:  Subwoofer.

[Alex B]

To continue my previous reply:

A different and very common case would be

5.1 -> 2.0

According to the links in this thread most HW players simply drop the LFE channel from the mix when the stereo analog outputs are used. This may be good if the speakers are not up to the task of reproducing the low frequency effects.

However, the user may have an standard analog stereo system with a good subwoofer (or perhaps just a pair of huge full range speakers). It might be good to provide an option to include LFE in the mix. (The channel mixers in the AC3filter and FFDSHow decoders have this option, but not all 5.1 formats are decoded through DirectShow.)

[Matt]

I think it's correct to allow the option, like we do now, to fill the subwoofer on a 2.0 to 7.1 upmix using JRSS or to leave it silent so a user can use only redirected bass.

A native 7.1 mix would have information in the LFE channel, so it seems reasonable for JRSS to do the same when simulating 7.1.

This option also makes it straight-forward for a user to select 7.1 output and get something good on all the channels.  They could later explore the more complicated Room Correction with bass redirection.

I also believe that bass management, using redirected bass, is not going to be able to build as good of a subwoofer output from a 2.0 to 7.0 JRSS mix as JRSS itself could have built.  If bass redirecting did not redirect simulated channels (so 'Move bass' would become 'Remove bass' for simulated channels) it might be better.  But then it would be acting differently than configured, which could be a problem.

[mojave]

I think it's correct to allow the option, like we do now, to fill the subwoofer on a 2.0 to 7.1 upmix using JRSS or to leave it silent so a user can use only redirected bass.

If you are going to allow JRSS to create a bass channel when upmixing, then you should, at the very least, have this crossover disabled when someone uses bass management.

A native 7.1 mix would have information in the LFE channel, so it seems reasonable for JRSS to do the same when simulating 7.1.

No other matrixing method makes a .1 channel because of the problems it causes when Bass Management is used.

I also believe that bass management, using redirected bass, is not going to be able to build as good of a subwoofer output from a 2.0 to 7.0 JRSS mix as JRSS itself could have built.

I disagree for several reasons. First, JRSS assumes that your speakers are full range so it doesn't move bass, but just copies it. Now one has not just two, but 5 or 7 channels of full range audio. 30%- 40% of the power of the amplifier is used for the bass frequencies. This is putting an extra load on the amplifier and it will clip a lot sooner. Also, most receiver's maximum output is rated for two channels. When the surround channels are in use, the available watts can drop quite a bit. Having all the frequencies playing will cause the drivers in the speaker to distort at lower volumes.

Second, JRSS only uses a 12 dB/octave slope on the low pass filter (at least that is what I measure). The standard for all receivers is a 24 dB/octave slope. Have such a shallow slope means that you will be playing out of the linear range of the of subwoofer reducing sound quality. It also makes the subwoofer more directional. In other words, it makes it easier to hear where the subwoofer is located.

Third, having the bass in 6-8 sources makes it very difficult to calibrate and eliminate room modes and phasing problems. This is why the default for a receiver's room calibration software such as Audyssey or YPAO is to redirect all bass to the subwoofer using Bass Management. It is much easier to EQ one source that multiple sources.

Fourth, have the bass in 6-8 sources that might vary in quality will mean that your bass is only as good as the worst speaker. For example, someone might have nice mains, center, and even subwoofer, but have crappy surrounds. The muddy bass coming out of the surrounds will mask the other bass and reduce its quality.

There are good reasons one might want full range bass in all speakers, but for JRSS to do this and be considered the beginner's option (with Bass Management being more advanced) is a little backward. Routing all bass to Channel 4 is the beginner's option and will produce the best quality in the most systems. Choosing full range audio in all channels or a 12/db slope on the low pass filter are more advanced settings/options that require better amplification, better drivers, and a better understanding of calibration in order for it to work well in someone's room. 

[Matt]

I disagree for several reasons.

You've made some very good arguments.  Thank you.

One little note is that JRSS is using a 48 dB/octave slope for the subwoofer since build 15.0.74.  I just measured it and it seems to be working correctly.

[Alex B]

I also believe that bass management, using redirected bass, is not going to be able to build as good of a subwoofer output from a 2.0 to 7.0 JRSS mix as JRSS itself could have built.

I can understand this because the signal in the surround channels probably isn't in the original phase. If you combine bass after creating the surround channels the signals in the opposite phase will cancel each other out (more or less).

I suppose you could filter and combine the bass content according to the bass management settings from the stereo signal and only after that create the other channels and apply the high pass filter (if bass management is set to do so).

In any case, the fake surround mode is not important to me. I normally prefer to have the original channels.

Correctly implemented bass management is a lot more important thing to have. I have not even hooked up the analog 5.1 outputs yet (except some experiments years ago), but I will need the bass management features when I do it. I have a growing number of multi-channel media files that don't contain Dolby or DTS audio and thus cannot be played through SPDIF (without re-encoding to AC-3). My receiver can apply bass management only to digital signals.

[Matt]

I did some real world testing tonight.  I'll start by saying I'm only trying to achieve sound I like, not sound that is correct according to some text-book definition.  I appreciate both neutral bass and exaggerated bass.  Below are my personal opinions with regards to the results:

I was testing with stereo files played on a 5.1 system, using all the speakers.

First, I found the highest crossover points for my speakers at which I couldn't hear any audible effect from the crossover.  This seems like the logical place to crossover the speaker since any energy below that frequency would be wasted (and possibly distort the other sound).

It worked out to:
L / R: 40 Hz / 24 dB/octave
C: 80 Hz / 24 dB/octave
SL / SR: 100 Hz / 24 dB/octave

Next, I experimented with how to get the best sounding subwoofer output.

I used the two main approaches we've been discussing:

1) JRSS (using 'Remove bass' in bass management)

I simply love the way this sounds.  Media Center + JRSS + Room Correction delivers the best sound I've ever gotten from any stereo.

I'm able to dial in a very clean subwoofer output in JRSS.  This is especially good with rock music, where it does a nice job of keeping clean bass drum hits on the subwoofer, while keeping drone from bass guitars off the subwoofer.

2)  Redirected bass only (using 'Move bass' in bass management)

This approach is energy neutral, so there's a mathematical purity to it.  But I just couldn't get a bass output that sounded as good as JRSS.  Using the crossovers established above, it put too much high bass on the subwoofer.  Instead of tight punches, it droned and made bass guitars slide around the sound stage.

Moving sound below the 100 Hz crossover from the SL and SR speakers puts frequencies I don't think sound good on the subwoofer because of directionality, inability to exaggerate bass without exaggerating unwanted drone, etc.

Allowing a final (optional) subwoofer crossover might solve some of my problems, although then it's not clear the advantage of using #2 instead of #1.


This is all relevant because we're trying to figure out what to make most prominent in the user interface and what to recommend.  From my testing, the best sound was with letting JRSS fill the subwoofer and using bass management to remove unsupported frequencies (#1 above).

[flac.rules]

I disagree for several reasons. First, JRSS assumes that your speakers are full range so it doesn't move bass, but just copies it. Now one has not just two, but 5 or 7 channels of full range audio. 30%- 40% of the power of the amplifier is used for the bass frequencies. This is putting an extra load on the amplifier and it will clip a lot sooner. Also, most receiver's maximum output is rated for two channels. When the surround channels are in use, the available watts can drop quite a bit. Having all the frequencies playing will cause the drivers in the speaker to distort at lower volumes.

Ok, here I am getting confused, JRSS copies bass instead of moving it? Why would it do that? That woud create a lot more energy at the frequencies copied, and make the mix much more bass-heavy (wich have been a problem with JRSS, could this be the explanation?)

[jmone]

Ahhh what fun!  I think most of us will find the "correct" settings will be a mix of personal preference with the equipment we have (and the effort needed to go to smooth out any issues with the speakers).  Looking at the frequency curves of my speakers I see that may front speakers start to roll off at 80Hz, with my sub flat from 18-90hz.  Looks like the THX recommended cross over of 80hz suits my setup well.

That said I personally find the following listening scenarios:

1) "Sweet Spot" - Occasionally (very occasionally) I will sit in the sweet spot to listen to 2.0ch audio and much prefer to sound stage of a "pure direct" 2.0 output.  I find any of the DSP (either the receiver or JRSS muddies the sound stage when upmixing to 5.1 or even (though to a lesser extent) 2.1 with bass redirection).

2) Casual Listening - Much more common is just "cranking it up" and listening while moving around.  I really like the 5.1 upmix for this scenario and think the JRSS provides a great option to do this to ONLY my 2.0 stuff unlike the receiver that requires you to manually change the listening mode.  That said JRSS to my ears was bass heavy in it's operation.

While I like this option for 2.0 material, I'll never use it for original 5.1 encoded material - I'm just happy with the mix done by the studio and don't feel the need to play with it.

Thanks
Nathan

PS - Matt, I did not think you had a sub?

[flac.rules]

Yeah, jmone, i have it more or less the same, i use 5.1 source as it is, and upmix 2.0, however I have also had problems with JRSS being to bass-heavy. The upmixer should keep the bass energy the same as in the source, do it correct, and le people play with an EQ or similar if they want a more bass-heavy mix than the source.

[mojave]

While I like this option for 2.0 material, I'll never use it for original 5.1 encoded material - I'm just happy with the mix done by the studio and don't feel the need to play with it.

You may know this, but if your output is set to 5.1 and your source is 5.1 and you have JRSS on, it doesn't do anything to the original mix. However, if output is set to 7.1 and the source is 5.1, it will matrix the rear surrounds. The only time JRSS does anything to Channel 4 (LRE/Subwoofer) is when the source has no info on that channel.

I have 7.1 output, JRSS on, "For stereo sources, only mix to 2.1" checked, and the crossover set to silent. This setting doesn't touch 2.0 or 7.1, but it matrixes 5.1 to 7.1 by adding the rear surrounds. Using Bass Management, I listen to 2.1 for music (if it is a stereo source) and 7.1 for movies.

[mojave]

I did some real world testing tonight.

Somehow I missed this post. I see how JRSS could be producing better bass because it is probably working with just the original channels to create the subwoofer output. I just did some listening myself using JRSS and "Remove Bass" in Room Correction. It is very nice. However, when I want to watch a movie I have to change all the setting for each channel back to "move bass" or create two different zones with different settings and then instruct the family on how to change zones, etc.

Is there a difference between have JRSS create the subwoofer output or having Bass Management create the subwoofer output when I am only outputting to 2.1? If they are different then you could have the bass management work differently when a 2.0 source is expanded to 2.1, 5.1 or 7.1 by JRSS. Otherwise, you would only need to make a change when JRSS is upmixing to 5.1 or 7.1.

Maybe this would work. If the user is going to use bass management for movies, they will want to always leave their settings the same. However, like you pointed out, it might be best to use JRSS in some instances. The following then needs to happen:

1.  The only place to setup crossovers and bass management is in Room Correction.
2.  When JRSS is enabled and there is no subwoofer channel, then it looks at the crossover setting for the L & R (it would be nice if these were linked so changing one would change the other) and uses this crossover setting to send bass to the subwoofer channel using the Low Pass filter slope selected by the user. Contrary to regular bass management, it ignors the other channels when mixing to 5.1 or 7.1.
3.  If Routing in Room Correction is set to "Remove" or "Move" JRSS automatically uses "Remove" since the bass below the crossover is already being sent to the subwoofer. It uses the user selected High Pass slope. If it is set to "Copy" then JRSS still sends the fullrange signal to the speakers.

This integrates JRSS with the Bass Management and provides the best of both without any user intervention required for differences between movies and music. It allows users to use the high and low pass filters of their choice. It only requires the crossover be set in one place. It still follows the JRSS mixing method for going from 2.0 to 2.1, 5.1, and 7.1. It allows the user to set everything once and not have to make changes depending on the source.

The only thing it actually changes in JRSS is allowing the Low Pass filter slope to be varied rather than set at 48dB.  

[jmone]

Matt's done a good job and I think I've got my Bass where it should be - and it could have been multiple devices applying DSP / config issue on my part (and/or JRSS overemphasis?) - http://yabb.jriver.com/interact/index.php?topic=58771.0  Who knows but it is sounding pretty good to my ears.

[Matt]

This integrates JRSS with the Bass Management and provides the best of both without any user intervention required for differences between movies and music. It allows users to use the high and low pass filters of their choice. It only requires the crossover be set in one place. It still follows the JRSS mixing method for going from 2.0 to 2.1, 5.1, and 7.1. It allows the user to set everything once and not have to make changes depending on the source.

How about a checkbox in Output Format > Subwoofer > When source has no subwoofer >

• Disable redirected bass from Room Correction

[jmone]

How about a checkbox in Output Format > Subwoofer > When source has no subwoofer >

• Disable redirected bass from Room Correction

I would have thought that "Room Correction" would be the only place you needed to config all the Speaker Settings including for the Subwoofer for your particular environment.  Output format would then just determin what channel mixing, sample rates and bitdepth you would use etc.

[mojave]

How about a checkbox in Output Format > Subwoofer > When source has no subwoofer >

• Disable redirected bass from Room Correction

That still doesn't address the issue of the user needing to change from "remove bass" to "move bass" in Room Correction depending on whether or not the "source has no subwoofer."

You could do this behind the scenes when the source has no subwoofer. I don't even see the need for a checkbox.:

1.  Disable redirected bass from room correction.
2.  Use the crossover setting from front left speaker in Room Correction for JRSS.
3.  Get rid of the "remove bass" option. It is only needed in this scenario with JRSS. When "move bass" is selected and JRSS is creating the subwoofer output, then silently "remove bass." Otherwise, copy bass if the user has selected "copy bass."
4.  This might be more difficult to implement, but you could have the JRSS low pass filter use the low pass filter setting in Room Correction. If it is too difficult, just leave it at 48 dB/octave regardless of what the user wants.

This makes Room Correction the only place to set a crossover, etc. It is just a few extra lines of code, right.  ;D

[mojave]

Last night I watched Valkyrie in a home theater with extensive room treatments. The treatments make a great difference (I heard the room before the treatments, too). This weekend I plan to build some bass traps in the front corners of my living room. I also want to do some more calibration after they are installed. I was reading about the DIY Audio Test DVD at AVSForum and came upon the post I quoted below (post #291). Does MC's energy preservation calculations of moved bass arrive at similar numbers to the incoherent sound summing figures below?

   
OK, I keep getting asked so here it goes.

To set your system up for the recommended DD theatrical range you just follow the prompts on the DVD. Forget anything you read anywhere else about compensation, 75dBSPL, and anything else.

This will give you a max of 115dB from the LFE at 0dBFS, and 105dB from all the other speakers individually.

DTS is not recorded "HOT," neither are any DVDs. Most DVDs are dropped by -4dB. The ones that seem "Hotter" are just not as limited. There is no limiting on DTS (unless your processor does something funny.) Either way it is not "Hot," it just isn't attenuated by default.

Yes, some AVRs say +2 or -3 for DialNorm. These are setup where 0 on the screen(or no DialNorm screen in some cases) is actually -4dB below 0dBFS. In this case +4 is actually 0dBFS, and full loudness(DNR.) If you notice my DVD will pull up as +4. This is because there is no signal attenuation with my disk.

If you turn on Bass Management of any kind, and set at least one speaker to small you could get peaks above 115dB from your subwoofer. This is where it gets complicated.

If the sounds are not exactly the same this is called incoherent sound. There is a formula to figure it out but lets skip it. Lets assume this are 0dBFS level sounds;
LFE alone=115dBSPL
LFE+1Sp=115.41
LFE+2Sp=115.78
LFE+3Sp=116.12
LFE+4Sp=116.44
LFE+5Sp=116.74
LFE+6Sp=117.02
LFE+7Sp=117.28

If they are exactly the same they are coherent, and are a pain to sum, but here it is.
LFE alone=115dBSPL
LFE+1Sp=117.32
LFE+2Sp=119.25
LFE+3Sp=120.78
LFE+4Sp=122.09
LFE+5Sp=123.23
LFE+6Sp=124.23
LFE+7Sp=125.13

Now before you get too worried this is unlikely to ever happen. Why would you have full signal bass on every channel at the same time? There is also one other hurdle to clear, that never seems to get talked about.

Most people seem to know that the LFE channel is boosted +10dB. This is why 0dBFS is set at 115dBSPL on the LFE channel. What is mostly unknown is that before the crossover section in the process the LFE is digitally lowered -5dBFS, and is then boosted up to +15dB in an analogue output section to the RCA. All Bass Managed sound from the mains/surrounds are added in after the -5dBFS LFE step, and they are lowered by -15dBSF before being summed in with the LFE signal. (It is supposed to be boosted ~+15dB, but it isn't always.)

What this means is that if a sound on the Front Right speaker was supposed to play back at 105dB, it will still be output at 105dB. The thing that is missed is that there is only enough digital headroom left on the LFE channel for +5dB. Anything over that will clip the DSP. This means the max signal from the LFE should be 120dBSPL. Anything else is clipped off. The mastering process should have fixed any gross offenders, so around +5dB plus an extra dB should be it. See below for the extra dB.

There is the problem of Digital to Analogue conversion where the DAC can create a signal higher in dBV than the dBFS. This is known as 0dBFS+. If this perfect storm of chance were to happen you could end up with a peak higher than 120dBSPL. This shouldn't happen with commercially produced products, but it could happen. (Sound guys are human too, and forget the lessons from the analogue world still carry over.) It would be an abnormal situation though.

For more info:
Dolby Section 3.5:http://www.dolby.com/uploadedFiles/z...Guidelines.pdf
0dBFS+:http://www.tcelectronic.com/media/Le...per_AES109.pdf

[Matt]

Last night I watched Valkyrie in a home theater with extensive room treatments. The treatments make a great difference (I heard the room before the treatments, too). This weekend I plan to build some bass traps in the front corners of my living room. I also want to do some more calibration after they are installed. I was reading about the DIY Audio Test DVD at AVSForum and came upon the post I quoted below (post #291). Does MC's energy preservation calculations of moved bass arrive at similar numbers to the incoherent sound summing figures below?

I'm not exactly sure the question, but soho54's math makes perfect sense.

This was actually a helpful post because it made me realize it would be best to work the math so that you calculate the correct decibel level first, and work backwards to channel levels.  We were working in the other direction.

So there will be some refinements to redirection energy preservation in a coming build.

Thanks.