How to widen stereo image

[Ping Pong]

Greetings and thanks in advance for your time! I am using stereo imaging plugins (like PAZ, and others) to view my bounced mixes. As I compare to commercial CD's and othe 001 & 002 users I see that my mixes lack the depth that others achieve.

For the instrumentation that I record, some are double miced using classic configurations (x,y - etc.) and others are mono (DI). Vocals are a single mic. The micing plays a beginning role in creating a wide image. Panning in the mix is another step that contributes towards a wide image. The use of verb and delay adds yet another enhancement to overall image. Adding multiple tracks of the same instrument and part (not copied) seems to also contribute.

All the above are performed prior to bouncing the mix to a 24 bit stereo file. Once bounced, mild compression and EQ are applied and a Limiter (and dither) across the master fader and it is bounced again to a 16 bit file. The Limiter seems to increase the stereo image (when viewed by your favorite plugin (PAZ, etc.) rather dramatically but......

I feel that I do not want to Limit very heavily so as not to destroy "Dynamic Range." And the amount of Limiting required to get my mixes up to the other guys images seems to be excessive. So........

In particular the image of the vocal in the mix seems to suffer the most and appears way to mono in it's size (image). Its a single micing. Got extra stereo vocal track lying underneath. Got verb, delay, chorus. Using a Shure KSM27 (not horrible). Some how I am missing something and I am having a hard time figuring out what it is. I want that vocal image to swirl all the way out to the edges. Any ideas?

[Rail Jon Rogut]

Are you applying dither when going from 24 bit to 16 bit? If not try using the pow-R dither.

Rail

[Cliffy_Boy]

You can try the S1 Stereo Imager from Waves. I've gotten the results.

[max cooper]

I used to swear by mult-micing and stereo micing, as well as double-tracking guitars and panning them hard right and left. I also have done a lot of duping tracks and changing the sound of the dupe a little (or a lot) in order to make things sound bigger. Then one day I re-recorded a song with one vocal, one guitar, etc. and just the room ambience instead of reverb, etc. In other words, very spare instrumentation and efx. The difference is that when I use fewer tracks, each one seems much larger. You know how some early Beatles tracks seem really fat? And early Blue Note recordings? And Motown stuff? Each of these has a very limited number of tracks mixed together. Then listen to something that's got scads of tracks on it; maybe '80's Def Leppard. Sounds really thin in comparison (to my ears). Seems like every time you add a track, the soup gets watered down a little. Try just soloing a single vocal track and listen to it dry. If it's well recorded, it usually sounds really nice to me. Once I realized this and started being really careful about everything I added, my tracks started sounding a lot bigger. (For one thing, every time you add a mic track, there's that little bit of noise. When you're mixing down fifteen or twenty, it gets to be a big thing)

Max


What Would AC/DC Do?

[MissHiss]

Try encoding the stereo tracks into an M/S matrix, then cut the highs in the middle by about 2 dB while cutting the lows on the sides by 2 db. You can also try the following:

1. apply extremely mild (practically nonexistent) compression to the middle while compressing the sides somewhat more heavily.

2. apply the same reverb to both the middle and side channels, but use a slightly longer decay for the middle and about a 15/85 wet/dry ratio, while applying a slightly shorter decay to the side with about a 25/75 wet/dry ratio.

Then, of course, it needs to be encoded back into L/R.

This will definitely take some tweaking, and occasionally you'll find some weird circumstances things work better in reverse (eg: compressing the middle heavily but applying only light compression to the sides).

It is CRITICAL to run the same plug-in chain on both channels, even though you are using different settings on each.

[max cooper]

Do you mean running the same chain on M + S channels? Is this because of latency or something else?

Do you know if there's a plug that will go L/R to M/S to L/R?

[MissHiss]

Do you mean running the same chain on M + S channels? Is this because of latency or something else?

Do you know if there's a plug that will go L/R to M/S to L/R?

I mean running the same chain on the track designated as the "M" channel as you are running on the track designated as the "S" channel.

The Waves S1 Matrix will encode L/R into M and S channels, and several Waves plugs, including the S1 shuffler, will accept an M/S input and automatically encode it back into L/R. What I usually do is bounce the session to disc with the M/S matrix on the master fader. What is then identified as the "left" channel is the middle, and what is identified as the "right" channel is the sides. You create a seperate mono track for each, but run the same plug-in chain on each channel. Running the same plug-in chain helps you avoid phasing issues. On the master fader in the M/S session, I then have an S1 set to M/S input, which encodes the signal back into L/R. This is then typically followed by an L1 or L2, unless the material is to be sent out for mastering.