"Sparkle in the Rain" Separating Fact from Fiction

[rtbluray]

OK so after people have been complaining about the surround mix too, I decided to crack open my newly acquired disc and actually dissect these mixes rather than just send it back.

So the first question on everyone's minds is (particularly those who don't have it yet) is this mono or stereo?

Answer: It's mono.

How can we know this for sure?
Here's what I did:

I extracted the 'stereo' mix from my disc using DVD Audio Extractor (which can extract Blu-Rays too) and I imported the audio file into Pro Tools (though any Digital Audio Workstation can do what I did)
I took the file and split them into L & R files. Both were automatically panned far left and far right, but I panned both of them to the center, and I inverted the polarity (or phase) on one of them. That way if it was a stereo file I should hear the difference between the left and the right channels, also known as the 'sides'.
But you know what I heard instead? NOTHING! When the polarity was inverted on one of the sides, I got complete cancellation. This would only happen if both sides were completely correlated (the same thing), hence we get everything clumped together in the middle, which means 'mono'!

This is something that is so blantly obvious when listening on headphones. It's less blatant when listening through loudspeakers, but if you know this album well, you know it's a big and wide stereo mix with gigantic drums, big reverbs, and lots of keys. Something that doesn't work well at all in mono!

 

[rtbluray]

First important thing to mention about my analysis of the surround mix is that I am listening on a calibrated system. This is extremely important as it means that all my speakers should be outputting the same level so proper judgment can be made on whether particular elements or whole channels are too low or too loud.

First point to address:

I'm finding the vocals on the 5.1 PCM on BluRay to be a little "distant" as in not quite high enough and very wet with reverb. Anyone else finding this?

Yep, I'm finding the same thing, but it's not something that is unique to this Steven Wilson mix. The fact is this has been an issue I have been noticing with Steven Wilson's surround mixes for at least a year or more.
Why is that? Well here's an important point I have picked up from my Master's Project on Surround Mixing. DAWs (Digital Audio Workstations) do not compensate for pan law when going from phantom center to the center channel. There's a dip that's usually built into all stereo pan pots at the center position, which is usually -3 dB down from far left and far right. This is so that the pan stays at uniform level when moving an element from far left to center, and then the level increases again when going from center to far right. After all, when something's in the center, it's now being output of two speakers equally, so double the power.
Steven seems to think that there is no level difference between phantom center and the center channel, but there is. When I was working on my surround mixes for my Master's Project (which everyone will get to hear very soon!), on average, I had to add around 3 dB to the lead vocal when I moved it from phantom center to the center channel, so that way when I switched back and forth between my stereo and surround mixes, the vocal level would remain consistent.
Steven does not compensate for that, so if you have your system set up correctly, the vocal will appear to be more distant and lower in surround than it was in stereo.
Some people compensate for that by turning up their center channel, but I do not as it could affect the level of other mixes that are utilizing the center channel correctly.

This problem seems to be exacerbated on the Simple Minds and Tears for Fears mixes as the vocals are not dry like on the 70s mixes. They have lots of reverb on them, so the reverb increases the spaciousness and depth, which is not really something you want for clear and articulate vocals.

Next point coming up, so stay tuned!

 

[wavelength]

First important thing to mention about my analysis of the surround mix is that I am listening on a calibrated system. This is extremely important as it means that all my speakers should be outputting the same level so proper judgment can be made on whether particular elements or whole channels are too low or too loud.

First point to address:


Yep, I'm finding the same thing, but it's not something that is unique to this Steven Wilson mix. The fact is this has been an issue I have been noticing with Steven Wilson's surround mixes for at least a year or more.
Why is that? Well here's an important point I have picked up from my Master's Project on Surround Mixing. DAWs (Digital Audio Workstations) do not compensate for pan law when going from phantom center to the center channel. There's a dip that's usually built into all stereo pan pots at the center position, which is usually -3 dB down from far left and far right. This is so that the pan stays at uniform level when moving an element from far left to center, and then the level increases again when going from center to far right. After all, when something's in the center, it's now being output of two speakers equally, so double the power.
Steven seems to think that there is no level difference between phantom center and the center channel, but there is. When I was working on my surround mixes for my Master's Project (which everyone will get to hear very soon!), on average, I had to add around 3 dB to the lead vocal when I moved it from phantom center to the center channel, so that way when I switched back and forth between my stereo and surround mixes, the vocal level would remain consistent.
Steven does not compensate for that, so if you have your system set up correctly, the vocal will appear to be more distant and lower in surround than it was in stereo.
Some people compensate for that by turning up their center channel, but I do not as it could affect the level of other mixes that are utilizing the center channel correctly.

This problem seems to be exacerbated on the Simple Minds and Tears for Fears mixes as the vocals are not dry like on the 70s mixes. They have lots of reverb on them, so the reverb increases the spaciousness and depth, which is not really something you want for clear and articulate vocals.

Next point coming up, so stay tuned!

I really hope Steven reads this. Nice work rt!

 

[grill]

... When I was working on my surround mixes for my Master's Project (which everyone will get to hear very soon!), on average, I had to add around 3 dB to the lead vocal when I moved it from phantom center to the center channel, so that way when I switched back and forth between my stereo and surround mixes, the vocal level would remain consistent...

How about to formulate a question to SW on this topic and put it on Ask Steven Wilson! forum? Or have you already done it?

 

[fredblue]

Absolutely amazing! Thank you so much! You know, it's funny because I'n not techhy like you and so many others here but I first came across something along these lines completely randomly by accident a few years back (and really thought nothing of it! ) with the Motown/Michael McDonald 5.1 (I forget now if it was DVD-A or SACD). Anyway, I had left the gear setup in 4.0 configuration from last time and immediately found when I put on this 5.1 disc that the lead vocals seemed unusually loud.. I then noticed I had knocked out the centre channel as it was silent.. and when it kicked back in the vocals suddenly became more muted yet more balanced! I'm guessing that's the same kind of phenomenon as you explain here? Blimey, we never stop learning with this surround music stuff, do we..!? Thanks mr.! Superb work!

 

[wavelength]

How about to formulate a question to SW on this topic and put it on Ask Steven Wilson! forum? Or have you already done it?

Is it really a question? Just tell him what you believe.

 

[grill]

Is it really a question? Just tell him what you believe.

Yes but imho it would also be interesting to know what Steven Wilson believes.

 

[rtbluray]

How about to formulate a question to SW on this topic and put it on Ask Steven Wilson! forum? Or have you already done it?

Maybe…maybe not…

 

[rtbluray]

Maybe…maybe not…

Ok, I was gonna wait until next week to reveal, but…

There's no doubt in my mind this is what he does, not only cause I have read several interviews about his surround mixing technique, but he told me so himself!

When asked, he allowed me the opportunity to interview him as part of my Master's project, and one of my questions was: Do you generally find that you need to tweak the processing settings when moving elements from stereo to the other surround channels?

His response (edited down):

You would think so, but I never have. Once the stereo mix is done, I move things out into the surround field, and that’s the mix. When you listen to one of my surround mixes, you are listening to exactly the same mix, by which I mean, if you take out the decision making process in terms of spatial positioning, it is the same mix. It’s the same EQ, compression, processing, rides, balances. Even the panning takes its cues from the original stereo mix panning.

So there you have it. His idea of creating the consistency between mixes is to move things into the other channels without changing the settings and balances from the stereo mix. Hope that helps!

 

[MarkN]

The observation about how Steven apparently treats the center channel is interesting. But I would think that he would listen to his 5.1 mix as he is creating it on a system that actually has a center speaker. If so, wouldn't the final result be what he intends for the levels of the various channels?

 

[rtbluray]

The observation about how Steven apparently treats the center channel is interesting. But I would think that he would listen to his 5.1 mix as he is creating it on a system that actually has a center speaker. If so, wouldn't the final result be what he intends for the levels of the various channels?

Of course he has a center speaker, and by posting about this, I'm not at all questioning his intent but simply explaining why some people feel the vocals are lower on his mixes than they think it should be.

 

[Laird_Williams]

Of course he has a center speaker, and by posting about this, I'm not at all questioning his intent but simply explaining why some people feel the vocals are lower on his mixes than they think it should be.

1) What a fantastic analysis at the start of the thread. Thanks for clearing things up in an objective manner. I managed to learn a couple of new things in the reading.
2) The phrase "...think it should be." implies a subjectivity about which I wonder. For a lot of us, the lyrics are THE most important part of a song. Certainly, even at their most abstract, they are far less abstract than the music. Others of us, however, are more tuned to the music, with the lyrics being "frosting on the cake.". A lot of times, for example, I will listen to a song and imagine the vocals replaced by an instrument - dropping the words and leaving just the melody. An awful lot of pop and rock (old and new) is total crap when you do that. A lot of good songs are "Beautiful words with incredibly-mundane accompaniment." There are also a lot of incredible musical compositions that have, well, really lousy, stupid lyrics. Perhaps SW's surround mixes indicate a slight bias towards emphasizing the music. I happen to be aware of that bias in myself, and I generally love the way he mixes vocals. But I can see where other very discriminating people from other point of view might have a mild complaint.

Maybe it is less about calibration and more about personal preference. Or it could be the "pot detente position" problem. Or both.

 

[Rickko]

First important thing to mention about my analysis of the surround mix is that I am listening on a calibrated system. This is extremely important as it means that all my speakers should be outputting the same level so proper judgment can be made on whether particular elements or whole channels are too low or too loud.

First point to address:


Yep, I'm finding the same thing, but it's not something that is unique to this Steven Wilson mix. The fact is this has been an issue I have been noticing with Steven Wilson's surround mixes for at least a year or more.
Why is that? Well here's an important point I have picked up from my Master's Project on Surround Mixing. DAWs (Digital Audio Workstations) do not compensate for pan law when going from phantom center to the center channel. There's a dip that's usually built into all stereo pan pots at the center position, which is usually -3 dB down from far left and far right. This is so that the pan stays at uniform level when moving an element from far left to center, and then the level increases again when going from center to far right. After all, when something's in the center, it's now being output of two speakers equally, so double the power.
Steven seems to think that there is no level difference between phantom center and the center channel, but there is. When I was working on my surround mixes for my Master's Project (which everyone will get to hear very soon!), on average, I had to add around 3 dB to the lead vocal when I moved it from phantom center to the center channel, so that way when I switched back and forth between my stereo and surround mixes, the vocal level would remain consistent.
Steven does not compensate for that, so if you have your system set up correctly, the vocal will appear to be more distant and lower in surround than it was in stereo.
Some people compensate for that by turning up their center channel, but I do not as it could affect the level of other mixes that are utilizing the center channel correctly.

This problem seems to be exacerbated on the Simple Minds and Tears for Fears mixes as the vocals are not dry like on the 70s mixes. They have lots of reverb on them, so the reverb increases the spaciousness and depth, which is not really something you want for clear and articulate vocals.

Next point coming up, so stay tuned!

I am very impressed by your analysis. It gave me an explanation as to the vocals on some songs for the Tears for Fears 5.1 mix of Songs From The Big Chair.

I recently posted a review on SACD.net for this 5.1 mix and referenced the vocals that seemed overpowered and pushed back in the mix. I corrected on my system by increasing the centre channel by +2.5 db (my centre always set to 0 db). I certainly did not criticize Steven Wilson's mix as it is magnificent, i.e. just make the slight adjustment to enjoy.

 

[Robert van Diggele]

Hi Ryan, thank you for your analysis, interesting to read. Going a bit off topic, but will you publish the interviews you had with the various mixers? And your own surround mix you mentioned yeah, bring it on

 

[MarkN]

Of course he has a center speaker, and by posting about this, I'm not at all questioning his intent but simply explaining why some people feel the vocals are lower on his mixes than they think it should be.

Thanks again for the analysis. I guess ultimately, it comes down to personal preference about how loud the vocals should be. I guess that Steven is happy with a less emphasized center channel than some folks here. I've personally never had an issue with how his 5.1 mixes treat the center channel level, but I can understand how some folks might like more volume on the vocals.

 

[rtbluray]

Maybe it is less about calibration and more about personal preference. Or it could be the "pot detente position" problem. Or both.

As a follow-up question to what I originally asked Steven, I'm curious to know if he does a lot of A/B comparisons between the stereo and surround mix while mixing. I know I do that a lot once I think I've got a good surround mix as it allows me to see how levels and processing are translating from mix to mix. If things are radically different, I know I've still got work to do, but if there is minimal difference, I know I'm hitting my mark pretty close to what I think it should be, and that's always a good sign!

 

[rtbluray]

Hi Ryan, thank you for your analysis, interesting to read. Going a bit off topic, but will you publish the interviews you had with the various mixers? And your own surround mix you mentioned yeah, bring it on

Maybe so. I think I should check with Steven and some of the other engineers I talked to to make sure that it's OK to post these "private" conversations in a public forum. I've transcribed almost all of my interview with Steven, and I'm really happy with it as sometimes he'll give some incredibly long and in-depth answers to what I was asking him. (A few things might be more "off-the-record" so to speak, so I should probably check with him if I can before posting).
Besides Steven, there's one other very prestigious mixing engineer that I got to speak with, so again, I'll probably have to listen back, transcribe, and check with him before posting.

But the mixes will definitely be posted as I have permission from my artist to do so, and once I do, I'll encourage all of you to ask me questions about the mixes, my mixing process, and other project details in a forum thread.

 

[Jeffrey Michael]

Besides Steven, there's one other very prestigious mixing engineer that I got to speak with...

Elliot Scheiner?

 

[rtbluray]

Elliot Scheiner?

Like I said before, "Maybe…maybe not…"

 

[rtbluray]

Next point:

There is a poster on another forum that thinks the 5.1 mix is messed up also as in front and rear channels swapped. Evidence is that there are re-enforcing lead vocals in the surrounds along with most primary instruments. Fronts have mostly bass and drums and reverb.

Given the complaints that the 5.1 mix sounds poor, this sure sounds like a possibility.

This is most definitely fiction.
Not only have I been able to listen to the surround mix several times on my surround system, but I extracted the mixes so I can check out each channel (or pair of channels) over headphones (my audio microscope).
Start with Track 1 "Up on the Catwalk". First thing that strikes me about the front channels is how direct the drums sound. Contrast with the rear channels, which contains mostly keyboards. There are some drums back there, but it's definitely less direct than the front channels (as it should be IMO)
Same goes for the next track "Book of Brilliant Things". Very direct drums and bass in the front channels, while the surrounds have less direct drums and some keyboards.
On "Speed Your Love to Me", different song but same front and surround arrangement. With "Waterfront", the drums seem to be more evenly split between the front and surround channels, but the bass that starts off the tune is clearly anchored in the front.
"East at Easter" once again has very clear direct drums in the front, while the surround channels have more drum ambiences and keyboards.
The same goes for "Street Hassle", "White Hot Day", etc.

Now the post also says there is re-enforcing lead vocals in the surrounds, but they are there in the front channels too. Depending on where you are sitting or whether your system is set up correctly you might not be hearing it that way, but the vocals are in all five channels, just diminished in the front and surround channels so they are clearly heard from the center.

So there you have it.
As always, your mileage may vary, but I hope this is helpful!