Variations in LFE volume

[jimfisheye]

Is automatic plugin delay compensation from latency the same as delay caused by strong filtering?...

No. Latency comes from computerized shuttling of data by clock cycles. Rather than an analog circuit with electrons running at near light speed.

EQ is phase manipulation and targeted comb filtering leveraging that.
You could only be off by part of a cycle from EQ. There are linear phase eq techniques to address this, FYI. Pros and cons from either direction.

The offset errors that have made it to consumer delivery are truly eyebrow raising mistakes.

 

[humprof]

Yeah, so the weird Lfe offset mistakes!
Can anyone offer a clue on how some of these happened? What was the screwup? I'm just really curious to know if this was a software bug led blunder or if this is genuinely a number of different people all with an outboard fx path on their Lfe track and all screwing up their DAW system offsets with that. Just typing that out makes it feel absurd to suggest... but here we are!

IIRC, the answer is somewhere in these ten pages:
https://www.quadraphonicquad.com/fo...ave-a-lfe-issue-info-list-fixes-inside.30048/

 

[LTW]

IIRC, the answer is somewhere in these ten pages:
https://www.quadraphonicquad.com/fo...ave-a-lfe-issue-info-list-fixes-inside.30048/

Interesting! The waveforms posted in that thread look exactly like what I was seeing from the LFE low pass filtering I was doing (before I used a linear EQ).
Then later in the thread this was quoted by @Eclectic:

However, this does raise the spectre of potential latency issues, since the main channels remain unfiltered. If the LFE channel is delayed even slightly, it can completely cancel redirected bass content from the main channels when monitored on a bass-managed system. Therefore, the surround mastering engineer must listen on a bass-managed system to check the results of combining the low frequency signals. In some cases, it may be necessary to time-shift the main channels in order to ensure that all channels remain in absolute perfect phase after the filtering process.

I only discovered the delay issue when I listened to the mix through my bass managed AVR. The L+R signal was interfering with .1 LFE channel due to an approx 180 phase shift.

These delays must be from incorrectly implemented LPF and not checking on bass managed systems.

 

[jimfisheye]

I don't think there were any conclusions as to root cause.
A lot of very well and complete documentation and presentation of the evidence! I know it's academic but I'm just interested in what software bug or whatever it was that led to this.

A random consumer might not just know how to calibrate their system... A studio sure as heck does! It's not a guess or anything cryptic. Literally run a calibration loopback test with test pings and set the offsets if there are any. It's a matter of fact setup step you do on day 1 with any DAW.

If you have a complex system with analog outboard gear interfaced and so on... It's not a mystery to you. You run tests through the different audio paths and dial it in.

Not to diss on anyone either because this IS an insidious thing that can easily catch someone off guard! My thought here is more to look for a gremlin in some software to point out and say "Watch out for that one!"

 

[DuncanS]

No. Latency comes from computerized shuttling of data by clock cycles. Rather than an analog circuit with electrons running at near light speed.

EQ is phase manipulation and targeted comb filtering leveraging that.
You could only be off by part of a cycle from EQ. There are linear phase eq techniques to address this, FYI. Pros and cons from either direction.

The offset errors that have made it to consumer delivery are truly eyebrow raising mistakes.

All digital filters will also add delay, its inherent in filtering. So the DAW tools should compensate for processing delay i.e. the latency caused by running an algorithm etc., and the delay introduced by the actual filtering process.

 

[jimfisheye]

All digital filters will also add delay, its inherent in filtering. So the DAW tools should compensate for processing delay i.e. the latency caused by running an algorithm etc., and the delay introduced by the actual filtering process.

There the plugin delay itself. The processing time in samples.
That should be normalized by PDC or someone can take over manually.

Then the actual EQ function. This normally would be measurable as phase offset. If it's to the point of significant offset then something out of the ordinary is going on.

It's like using any other plugin anywhere else on the mixing board. We also do parallel processing tricks often enough where perfect phase locked operation across tracks is required. Seeing this happen with a commercial release from a big studio is still a big surprise is all I'm saying.

 

[LTW]

For anyone curious, this is the waveform for a 5.1 mix of mine where the peaks are delayed approx. 11ms in the LFE due to a 24db 120hz LP and 24db 20hz HP. This was a Fabfilter Pro-Q3 EQ set to zero latency. Sucked all the low end life out in my home theater system which I couldn't hear in my mixing room.

And this is the unfiltered, corrected, version.