Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Andrew Broughton]

Remember that all FM systems need to have a limiter to prevent (illegal) overmodulation. The Shure transmitter compander looks something like this:

Right - Karl was speaking about some limiter settings. The compander is non-adjustable.

[Jason Glass]

The other strange thing we noticed was an issue with inputs that were what I call "Super Stereo."  The show utilizes a good deal of playback along with a large live band, so there are some verrry produced tracks in there that are in super duper stereo.  One particular track made us stop everything, as an acoustic guitar part (in super stereo) was either swishing around in our ears or cutting in and out.

Once upon a time, all professional audio engineers were expected to know about this effect, and avoid it in their mixes by monitoring the mix bus with a vectorscope.  This is because all stereo multiplexed transmission systems, from Shure, to Sennheiser, to Optimod for commercial FM, to NTSC television, have maximum L to R separation specifications given in dB.  As many here have noted, the Senny gear seems to have a higher tolerance for separation issues than the Shure stuff, but they all have their limits.

It is worth noting that these specs were also relevant to LP mastering because excessive separation would cause a record player stylus to jump out of the groove.

The problem is that most upcoming engineers in the studio and broadcast world are mixing for digital media, which are generally immune to artifacts from excessive stereo separation.  In many cases they have never experienced the "old school" phenomenon of subcarrier overmodulation and dropout (your "swishing and cutting in and out") that occurs when the separation spec is exceeded.  So, for example, your band is mixing tracks for the tour in a bedroom studio and decides to pan the acoustic guitar hard L&R and flip the polarity on one side.  They think it sounds great on the studio monitors, but have no idea that they are sabotaging their sound.  Anytime that this type of content is transmitted via any analog FM stereo multiplex, artifacts will occur to some degree.

Your description of "super stereo" is something that I run into frequently in live background tracks.  It is a result of what I described in the above acoustic guitar example.  If you have those channels at maximum levels in a mix that has 96dB dynamic range, the separation difference that the transmitter sees can be as high as 192dB!  Most of our in-ears have separation limits in the neighborhood of 60dB, and I have never seen the spec for analog broadcast gear exceed 96dB.  This is why old school engineers would judge such a mix as utterly unacceptable and the mark of an amateur.

The bottom line here is that folks are blaming equipment for bad sound when the real issue is poor mix engineering that violates ages-old standards and practices for multiplexed stereo transmission platforms.  The deeper you understand the media on which your mix will be played, the better you will be able to compensate for its limits and improve the listener's experience.  BTW, I don't completely blame the monitor guy for this, but rather the fault lies with whoever has mixed any portion of the final mix (submixes, stems, and/or tracks) out of spec.

[Adam Robinson]

Once upon a time, all professional audio engineers were expected to know about this effect, and avoid it in their mixes by monitoring the mix bus with a vectorscope.  This is because all stereo multiplexed transmission systems, from Shure, to Sennheiser, to Optimod for commercial FM, to NTSC television, have maximum L to R separation specifications given in dB.  As many here have noted, the Senny gear seems to have a higher tolerance for separation issues than the Shure stuff, but they all have their limits.

It is worth noting that these specs were also relevant to LP mastering because excessive separation would cause a record player stylus to jump out of the groove.

The problem is that most upcoming engineers in the studio and broadcast world are mixing for digital media, which are generally immune to artifacts from excessive stereo separation.  In many cases they have never experienced the "old school" phenomenon of subcarrier overmodulation and dropout (your "swishing and cutting in and out") that occurs when the separation spec is exceeded.  So, for example, your band is mixing tracks for the tour in a bedroom studio and decides to pan the acoustic guitar hard L&R and flip the polarity on one side.  They think it sounds great on the studio monitors, but have no idea that they are sabotaging their sound.  Anytime that content is transmitted via any analog FM stereo multiplex, artifacts will occur to some degree.

Your description of "super stereo" is something that I run into frequently in live background tracks.  It is a result of what I described in the above acoustic guitar example.  If you have those channels at maximum levels in a mix that has 96dB dynamic range, the separation difference that the transmitter sees can be as high as 192dB!  Most of our in-ears have separation limits in the neighborhood of 60dB, and I have never seen the spec for analog broadcast gear exceed 96dB.  This is why old school engineers would judge such a mix as utterly unacceptable and the mark of an amateur.

The bottom line here is that folks are blaming equipment for bad sound when the real issue is poor mix engineering that violates ages-old standards and practices for multiplexed stereo transmission platforms.  The deeper you understand the media on which your mix will be played, the better you will be able to compensate for its limits and improve the listener's experience.  BTW, I don't completely blame the monitor guy for this, but rather the fault lies with whoever has mixed any portion of the final mix (submixes, stems, and/or tracks) out of spec.

Thanks for the info, Jason.  Your posts on RF are always incredibly insightful.  I'm thankful for your contributions to the community.

[Josh Stevens]

Thanks for that info Jason... greatly appreciated.

So if one shows up to a show where tracks of this nature of being used, would not panning hard left and right, (reducing to 10 and 2) help?  Or at this point are they doomed due to the mixing?

[Jens Palm Bacher]

Thanks for that info Jason... greatly appreciated.

So if one shows up to a show where tracks of this nature of being used, would not panning hard left and right, (reducing to 10 and 2) help?  Or at this point are they doomed due to the mixing?

If the content is really out of polarity your are pretty much out of luck, as panning it to centre Will just make it disappear. One could try to only use one side of the track... (Or use a iem system made in germany..)

[Jason Glass]

If the content is really out of polarity your are pretty much out of luck, as panning it to centre Will just make it disappear. One could try to only use one side of the track... (Or use a iem system made in germany..)

Jens has pretty much nailed it.  Although you might reap some benefit if you could insert a mastering mid-side processing plugin on the offending tracks and tweak the available parameters to narrow the width.

[Chris Johnson [UK]]

The bottom line here is that folks are blaming equipment for bad sound when the real issue is poor mix engineering that violates ages-old standards and practices for multiplexed stereo transmission platforms.

You are absolutely correct.

However, I believe the issue that the OP was referring to is not specifically related to this, although could well be a component of the overall problem.

The Shure systems do include some kind of audio compression (seperate, it seems from the necessary pre-transmission companding) that is in-circuit and invisible to the operator. This effect is clearly audible, especially when you drive the Tx hard, as you typically would with a G3 system. Part of the problem with the PSM1000, which i believe was the primary reason a lot of the units sent down for the Olympic opening ceremony were replaced with Sennheiser at the last minute, is the position of the metering on the Tx in the audio signal chain. It appears that this compression happens post-metering, and is therefore not visible to the operator.
Luckily, the PSM packs have much more powerful headphone amps than their sennheiser counterparts, so you can get away with less audio signal into the Tx, and make up for it with clean headphone amplification on the Rx.

[Jens Palm Bacher]

I'm testing an italian IEM system at the moment... a bit wobbbly at times: https://dl.dropboxusercontent.com/u/18154497/IMG_1733.MOV (look at the RF lamp as i turn it on/off)

[Jason Glass]

You are absolutely correct.

However, I believe the issue that the OP was referring to is not specifically related to this, although could well be a component of the overall problem.

The Shure systems do include some kind of audio compression (seperate, it seems from the necessary pre-transmission companding) that is in-circuit and invisible to the operator. This effect is clearly audible, especially when you drive the Tx hard, as you typically would with a G3 system. Part of the problem with the PSM1000, which i believe was the primary reason a lot of the units sent down for the Olympic opening ceremony were replaced with Sennheiser at the last minute, is the position of the metering on the Tx in the audio signal chain. It appears that this compression happens post-metering, and is therefore not visible to the operator.
Luckily, the PSM packs have much more powerful headphone amps than their sennheiser counterparts, so you can get away with less audio signal into the Tx, and make up for it with clean headphone amplification on the Rx.

Hi Chris,

Yes, of course, this is only one of many things that get overlooked in IEM mixing.  Here's another:

Multiplexing is also the reason that certain systems are sensitive to excessive 19KHz content, even when panned to center.  The multiplexed signal contains a 38KHz subcarrier that gets modulated with L minus R audio, and also contains a straight or coded 19KHz pilot tone that the receiver uses as a reference to unmodulate the subcarrier.  When we add too much 19KHz to the content (which is easy to do because our cue system and earpieces often can't reproduce it, and we sadly may have anatomical hearing loss up there) the signal may sum with constructive interference and overmodulate, or destructively interfere and cancel, the pilot tone.  Either scenario destroys the signal by defeating the demultiplexer.  Many systems also depend on the pilot tone to open the receiver squelch.  If the pilot gets cancelled by content, the squelch doesn't function correctly.

[Jason Glass]

I'm testing an italian IEM system at the moment... a bit wobbbly at times: https://dl.dropboxusercontent.com/u/18154497/IMG_1733.MOV (look at the RF lamp as i turn it on/off)

Hmmm...  It looks like it's not seeing the pilot.  Nice metering, though!

[ProSoundWeb Community]