Happy almost New Year!!
Thanks for this. I'm really glad we're talking about this.
I may have implied this is a Steven Wilson issue, since it is in the thread about him,
but I don't mean to imply that. I was just hoping he was still looking at this site and might be able to provide some insight as an obvious expert in surround sound engineering.
The best example of this making a huge difference is with American Beauty by the Grateful Dead. If you listen to Box of Rain, from the opening notes,
there is this weird void in the sound when .1 is sent directly to a subwoofer. I've heard it from the day I bought the disc way back at the beginning of all this DVDA stuff and for that reason, I never listened to that disc much. I just didn't like it. It wasn't until I started messing around with authoring hi-res discs of my own mixing SACD and MLP surround tracks on the same disc that I discovered this extra content. (And before anyone objects to that, I converted the DSD to FLAC to MLP. I wasn't suggesting that I was mixing formats).
Even that was by accident when I accidentally assigned the .1 to center channel (SUPER easy to do BTW) when putting together the disc that I discovered the higher frequency stuff in the mis assigned .1/LFE. I heard the full range kick drum and thought WTF is that in center? I didn't remember anything with kick drum in center so I checked and realized that I assigned channels incorrectly but, wait a sec, why is that in the .1? I'd never hear any of that stuff on my 80hz low pass sub. So I started fooling around and voila, most of the discs I had were the same, hence my original posting.
And, when I change over my system config to 5.0 with .1 distributed across the front two channels, with low pass sending only 80hz and below to the subwoofer and everything else to the L and R, it has the same sort of frequency distribution, albeit in surround, as what the vinyl sounded like. In other words lump it all together and let the hardware decide what goes where instead of just assuming there's nothing above your sub's crossover frequency in the .1/LFE channel.
So, I am starting to draw the conclusion that the weird void I hear in lots of DVDA 5.1 discs is this loss of anything in the .1/LFE channel that is above what frequency the sub can reproduce. Before I put this to rest, I thought I'd ask Mr. Wilson. If I ever get a chance to talk to any surround engineer, I'll ask them as well.
My apologies to Mr. Wilson for implying that his recordings have this issue. I can't say either way, as I don't have an example of one of his mixes that I can cite off the top of my head. That was not my intention.
My guess is that most engineers just assigned the kick drum off the 24 or 16 track tape into the LFE channel, set levels and were done with it. I'm also guessing that most didn't have a really good surround monitor system so that they'd notice. What choice did they really have?, if you were going to send signal to .1 by frequency, what authority would decide what that frequency is for all recordings? Is it 80hz or 40hz? Or maybe 120hz so people can use less wide ranging (read cheaper) L and R speaker systems. In any case, there would be absolutely no practical way to ensure that the engineer's intention would be reproduced on everyone's system because of the variations in subwoofer crossover frequencies even in properly configured surround systems. I know, I know, there are SO MANY other variables that prevent that, but this one is DOA at the source. And in fact, that may be why the engineers didn't care that much to begin with because they know there are bigger fish to fry to get what they want to be heard. I suppose you could print the .1/LFE frequency on the cover of the disc, then the owner could go set their frequency on their sub for each disc. ROTFL
I'm REALLY thankful that people are contributing to this topic because it's an interesting bit about our hobby.
BTW...I'm going to open another thread about why sample rate really does matter (aka why Nyquist quoters are wrong), why there's more to sample size than just dynamic range(ie. 16 bit versus 24 or even 32 bit), and why digital is inherently, by definition, and permanently, lossy if you compare to the original sound. But, that may not belong on this site.
I hope that thread is as lively as this one.
Thanks again and Happy New Year!!!
Hi, thanks for your reply.
I didn't think of coming back here but I did make some other findings in the mean time. I
did have a channel assignment issue, but this issue is limited to the PCM tracks - because PCM on blu-ray uses a slightly different channel order (L R C SL SR LFE as I suspected) compared to Dolby or DTS tracks (L R C LFE SL SR). Depending on which application I used to rip or playback those PCM tracks, I would or would not get the correct channel order. So that's where my above post comes from - but now I know better and I understand this has nothing to do with the sound engineers. This might be happening to some other users though, especially those ripping all their discs to FLAC like I do (I do not have the issue when playing straight from the discs).
Ok, now that's out of the way, let's get back to the subject at hand. I did find that
some tracks on Steven Wilson mixes seem to have a bit of mid/high frequencies in the LFE as well. But it's rare and on most tracks I found that it doesn't sound different - even if there's vocals and other things they are so weak relative to the other channels that they are perfectly inaudible unless you listen to nothing but the LFE (or your LFE is super loud). I did my testing like you said - by downmixing the LFE to all channels. I then toggled a lowpass filter (120hz) on and off on the LFE channel alone,
before the downmixing, to see if I could actually hear any difference. The vast majority of the time, no audible difference.
In a few cases, I can definitely hear a difference. But it seems to me as if it just makes some content present in the other channels louder, and so I'm thinking it's (maybe) not intentional. Some films on blu-ray, for example, have hi/mid frequency noise (and I do mean
noise) in the LFE channel because they (the engineers) forgot to lowpass it (and they played it back on a system with a subwoofer so they didn't notice), more about that over
there if anyone is interested.
Do you have some examples in mind (tracks or even albums) where the difference is really noticeable to you and you feel it is really the right way of listening to the content?
