The wonderful world of LFE

[ssully]

So then to your point and mine --

setting the LPF to 120hz as recommended - keeps the low frequencies in the .1 channel up to 120hz going to the sub and then setting your crossovers to 40 or 50 or 60 or 80 or 100 or 120 or whatever for your speakers will keep those frequencies either going to your sub (the lower) or to your speakers (the higher). So setting crossover at 80 means 80 and lower goes to Sub and 81 and higher goes to speakers, etc, etc. Setting the LPF to anything lower than 120 means losing information that could be present in the LFE channel. If your equipment has capability of setting LPF higher than 120hz then the risk is going to be the ability to locate the sub.

Content with full frequencies in the .1 will be bass managed.

Problem solved 100%.

So, first you seemed to be questioning whether >120Hz content existed in consumer product LFE. When I pointed you to evidence, instead of acknowledging it, you pivoted to this recap of things already said in this thread. Including by me, who said that the >120Hz content was unlikely to affect a typical installation, where multiple low pass filter steps are in place for subwoofer content, but that there are use cases where it could.

Thanks, I guess?

 

[EricKalet]

So, first you seemed to be questioning whether >120Hz content existed in consumer product LFE. When I pointed you to evidence, instead of acknowledging it, you pivoted to this recap of things already said in this thread. Including by me, who said that the >120Hz content was unlikely to affect a typical installation, where multiple low pass filter steps are in place for subwoofer content, but that there are use cases where it could.

Thanks, I guess?

Sorry I didn’t acknowledge your hard work. Agreed you did show there is some content recorded to the LFE that is above 120hz. 99% of us won’t hear it. I don’t have the equipment or the wherewithal to do the experiments you did so thank you sincerely for that.

I don’t know how this evolved from my original statement of setting LPF to 120hz is the right setting. Someone argued that maybe 120hz isn’t a one size fits all. I think after all this discussion - 99% of the time it is a one size fits all. Maybe we could all agree on that, but probably not. I think I’ll stick with what has been working for me for 30 or so years.

 

[jhw59]

So my AVR equalizer was on so I turned it off. I set the LPF to 120 as recommended. Bass is more to my liking but the tweaking continues. I really don't want to go to optical to get sound from the TCL TV apps through my AVR but so far that seems to be the only solution.

 

[EricKalet]

So my AVR equalizer was on so I turned it off. I set the LPF to 120 as recommended. Bass is more to my liking but the tweaking continues. I really don't want to go to optical to get sound from the TCL TV apps through my AVR but so far that seems to be the only solution.

Always tweaking! Seems like that never ends.

IMO never use the equalizer and change the bass volume when needed from the AVR settings (mine has a quick menu that lets me adjust bass and treble on the fly) and not from the woofer.

 

[jimfisheye]

You need to be able to play sound from the computer and know what a DAW app like Audacity or Reaper is to do this.

Still here? OK.
Pull up a sound/tone generator plugin. (Reaper has a JS tone generator plugin. Audacity probably has something.)
You can play different tones and frequency sweeps.
If you have a speaker/bass managed system and want to hear how your crossover point choice is working out, do a frequency sweep through the bass end. If you have small tops + sub, send that sweep to any main channel. You should hear a smooth transition between the speaker and the sub. In other words, you shouldn't hear any artifact at the crossover point. Listen for a dropout, a volume lurch, or a volume change. This gives you the ability to test as you adjust and dial in the best crossover point.

You'll also discover if you have anything weird going on. A level off by 10db somewhere or a speaker management setting doing something odd. You have control over the channels and test tones so you're not guessing by listening to random mixes.

There's a lot of sound coming out of a lot of speakers even if some settings are wrong. I don't want to suggest anyone around here might be easy to please and having that experience but there's a matter of fact way to find out.

I guess a test DVD or bluray could be authored or maybe there are images to download?

You don't have to pull out calculators and oscilloscopes for any of this either. Just basics:
Channels are intentional and balanced.
No weirdness with a crossover point in any bass management scenarios.

 

[Owen Smith]

I have commercial test DVDs and Blu Rays with exactly these sorts of frequency sweeps, in each channel one by one.

 

[jhw59]

What's the best overall test DVD/Blu Ray?

 

[Owen Smith]

What's the best overall test DVD/Blu Ray?

I only have one Blu Ray and two DVD ones, and they're all long since out of print and hard to get. And I wouldn't say any are better, I used different ones for different things.

 

[ssully]

let us return to the wonderful world of LFE....

What happens when a release is initially DTS 96/24, but later rereleased as SACD? The Genesis catalog, for example.

Here's what happens with the LFE (for 'Squonk'...lots of bass frequencies on that one*). This is a linear plot of the averaged level (Y-axis) at each frequency (X-axis), displaying the part up to 400 Hz, at which point the SACD LFE average level is -100dB (let's call that inaudible). The lines start to diverge significantly at 120Hz, the SACD having higher level content from there on. A steep rolloff of LFE in DTS and Dolby Digital after 120 Hz is standard for those formats (this plot shows an approximate 50dB/octave drop after 120 Hz in the DTS signal vs 30db in the SACD).

Here is the Center channel, for comparison (displaying the whole audible spectrum). The two lines are essentially the same. The ~+4dB in DTS is consistent over the range, I can't explain where it comes from, but it amounts to a simple level mismatch (which in a blind but nonlevel matched comparison could well lead to the DTS being *preferred* over the SACD ) I *predict* that the same holds for the L/R and surround channels too.

What does this mean in practice? Short answer is, often nothing, but it *could* make an audible difference if you use a subwoofer...depending. It could make even more of a difference if you don't use a subwoofer....depending.

A. Consider the 'choke points' for LFE output., in a system with a subwoofer:
(1) the 'LPF for LFE' of the AVR -- the low pass filter applied to LFE content alone. Here's what one of my Denons offers:
LPF for LFE: Set LFE signal playback range. Set this when you want to change the playback frequency (low pass filter point) of the subwoofer.
80 Hz / 90 Hz / 100 Hz / 110 Hz / 120 Hz / 150 Hz /200 Hz / 250 Hz (Default : 120 Hz).

(2) the LPF ('crossover') of the subwoofer itself. This is both the adjustable electronic one built into the sub, and the actual usable FR of the 8/10/12/15-inch woofer + enclosure itself. Users of .1 systems are typically advised to either run the sub crossover as high as possible, or better yet turn it off if possible, so it doesn't pile on top of the upstream filters.

The upshot: for any settings or aggregate of these effects that allows >120Hz content to be output, there might be audible subwoofer difference between these two releases. If there's a 120 Hz LPF choke at any point, the LFE difference should be inconsequential.

B. What about in a system WITHOUT a subwoofer, where LFE is routed to 'full range' mains?
Things could get interesting... depending on if the AVR's LPF for LFE is set higher than 120 Hz. Look up there again: Denon permits a LPF setting up to 250 Hz... and notice there's a +30dB higher average level in the SACD vs DTS at that frequency. Admittedly it's still -80dB , quite low, but everything between 120 and 250Hz is louder than that. And this is all coming out of loudspeakers, not a subwoofer. Unlike subs, speakers do not have a large intrinsic rolloff of their own at those frequencies.
The implications of this can be heard if you audition both LFEs 'raw', unfiltered, through headphones (standing in for 'full range' speakers). In this brief (fair use) clip from partway through the track, the first instance is the DTS LFE, the repeat is the SACD LFE. No level matching done (though even after level matched, they are distinct).
https://www.dropbox.com/scl/fi/pvc6...mple.wav?rlkey=nqyx1lcli710u5lxxev8ecu2t&dl=0
You'll hear somewhat more attack on the bass and kick in the SACD LFE, as well as a few bass notes high up the neck that are barely heard at all in the DTS LFE. This is because of those stronger SACD LFE frequencies above 120Hz

For the record, I don't think those >120 Hz frequencies belong in an LFE at all, and certainly not any content above 250Hz! I think mixers lazily expect that the consumer will have a Dolby-compatible LPF in their system somewhere. But if you don't, be aware that you could hear a difference, depending on how your system is set up. 'System' means gear + room, btw.

And please NB that the LFE rolloff in DD and DTS is an industry *choice*, it has nothing to do with them being lossy formats! You can see from the Center comparison that there's no need to be hysterical about 'lossy' (and yes, DTS 96/24 is still 'lossy') .



* Method notes: both versions were ripped from discs; the SACD rip was converted to PCM at 96000 kHz/24 bits to match the format of the decoded DTS 96/24. Frequency analysis was in Audition, using a Blackman-Harris FFT of ~65,000 bins. FA data was plotted in Excel.