Phase Cancellation Coffee

Music Production, Audio Engineering and Technology Music Blog

Social Media

Saturday, 11 July 2020

Q&A - Volume Two


Hi Everyone!

I apologise for being away for so long! As I am awesome human, I have decided to post up a few of the Q&A emails I have received over the last couple of months. Hopefully these questions are some that you have had/having at the moment while being locked up in Quarantine. If you have any questions, please drop a message over so I can help with any recording/technology issues you’re having!

Catch you on the flipside!


“Which omni mics are best for small ensemble recordings?” – Carl, age 33.
My personal preference would be for the Microtech Gefell M221s. The Josephson C617SET uses the same capsule, of course, and their electronics are fractionally quieter, but the Acoustic Pressure Equalising spheres which are supplied with the Gefell mics give them a significant edge in versatility to my mind.

Earthworks make some very nice, neutral-sounding mics, but they tend to be noisy in comparison with the Gefell (22dBA versus 15dBA) because of the very small capsule size. That’s something that’s necessary to achieve the extended high-frequency bandwidth which Earthworks prioritise, but I didn’t feel that the Gefell lacked anything in the upper regions.

The DPA 4060 microphones are astonishingly good for their size and price but are inherently slightly compromised on the self-noise front, again, and have a tendency towards brightness that I don’t think you would appreciate. DPA’s dedicate range, reviewed in last month’s issue, now includes the MMC2006 omni capsule, which essentially contains a back-to-back pair of 4060s internally (with a self-noise advantage). This ‘twin-diaphragm’ technology is presented as a lower cost alternative to the classic MMC4006 capsule, but the MMC2006 is not compatible with the company’s range of APE spheres.

As for other alternatives, I remain a big fan of Sennheiser’s MKH20s, which I think still sound slightly better than the newer MKH8020. I like the ability to switch them from nearfield to diffuse-field equalisation, to suit different applications, and I relish their amazingly low harmonic distortion, ruler-flat frequency response, and very low self-noise.
Does pan placement change if I place my speakers further apart? – Micheal, age 19.
The maximum image width is obviously determined by the physical separation of the speakers, so switching to the 55-inch set moves the outer edges further out, as you’ve noticed. The whole stereo image has been stretched from the centre outwards in both directions. Imagine an elastic band, with the centre pinned in the middle of your sound stage, and the outer edges fixed to the monitors. If you mark the positions of different sound sources on the band and move the monitors outwards, the elastic band stretches and so too does the spacing between your marked sound sources. So, if the saxophone is panned 30 percent left in the image, then that’s where it will always be. When you switch to the wider speakers ‘30 percent left’ is actually going to be physically further left than it was with the closer speakers.


Don’t get your percentages confused with absolute measurements! When setting your speakers further apart, the placement of a panned source will inevitably change in degrees/distance, but not in terms of the relative distance from the centre to the extreme of the stereo panorama.
I’ll assume you’re listening position is at the apex of an equilateral triangle, with the other two points being at your 40-inch spaced speakers. Rough trigonometry calculations suggest that with the closer speakers the sax will appear roughly 10 degrees left of centre. Switch to the second set and this perceived angle increases to about 14 degrees. But it is still panned 30 percent left within this wider overall image!
Can I use an effects pedal for vocals? – Sam, age 21.
There will be people who tell you to track clean and add this sort of effect to a vocal only while mixing — but it can be both fun and inspirational to try mangling things while tracking can’t it? Still, they have a point: the sort of fuzzy distortion a Big Muff Pi can be responsible for is not something you can undo. For that reason, it makes sense to track a clean part alongside your distorted one, and there are a few ways of doing this. If you have a small mixer (or even a large one!) you could simply multi the clean mic signal out to another track and process that. You could try patching the pedal in as an insert effect on that channel, and this will work to some extent, but there’s likely to be both a level and impedance mismatch, which means the pedal probably won’t operate quite as it would on the instrument signal for which it’s intended. Whether it’s working, though, is a subjective matter — use your ears, and if you like what you hear then great! If not, then you need some way of overcoming those problems.

Using the Big Muff as a send effect might improve things, as you can change the level going into the pedal using the mic channel’s aux send control. If that doesn’t work, what you’re looking for is a DI/re-amp box. The re-amp signal goes to the Big Muff’s input, and its output goes via the DI to a line input on your mixer or recording device. If you have no mixer, you could do pretty much the same thing, but beware of latency. You’ll need an interface with zero-latency monitoring with which the input signal can be routed straight to an output without passing through the A-D/D-A converters. The incoming mic signal is routed both to your DAW and to a physical output. That physical out goes into the Big Muff (the same level/impedance considerations apply) and the output comes back, either via a DI box to a mic input, or straight into an instrument input if your interface has one. Alternatively, you could just use the processed part in your monitor mix and ‘re-amp’ the clean signal through your pedal later, when you might have a little more control over the tone.


Why would I want to bounce out mixes for referencing? – Mike, age 25.
This is an interesting question that I’ve been asked on a number of occasions, but I’m not sure I’ve ever written down my answer to it before! I realise that it’s perfectly possible to compare a mix in progress with commercial releases using something like Magic AB, Melda MCompare, or Meterplugs Perception — or indeed just using a multi-channel switcher plug-in within Reaper, which is my own normal method. However, I do still prefer to bounce out my mix as a WAV for referencing purposes most of the time, for several reasons — although not, funnily enough, for the reason you suggested!

On a practical level, I like the flexibility the DAW offers in terms of editing out and looping the most relevant pieces of each reference track, and the way it lets me easily adjust the time offset between my mix file and each reference track, something that I’ve not found as straightforward in the referencing plug-ins. I also often experiment during referencing to see what impact loudness processing might have on my mix, but mastering-style processors can cause CPU or latency-compensation problems when applied to an already heavily loaded mix project, and I can do without glitches or crashes while mixing. Besides, anything that encourages people to apply mastering processing to their mix project is a bit hazardous in my view, because I’ve seen a lot of people come unstuck that way, effectively trying to master a quick fix to complex mix problems.

However, the main reason I like to bounce out the mix is purely psychological. You see, when I reference using a bounce-down in a separate project, I can’t change the mix while I’m listening, so it encourages me to take decisions much more rigorously before acting on them. In other words, I’m reminded to cross-check each decision across several different references and several different listening systems before actually tweaking any mix settings. It’s enormously tempting when referencing within your mix project to hear, say, that the hi-hat’s too loud in comparison with one of your references over your main monitors and then to immediately charge off and change it, without checking whether that hi-hat’s also too loud compared with another of your references, or on a different listening system. Referencing within the mix project is therefore all too often a recipe for tail-chasing, in my experience, and I prefer to remove that temptation from my workflow.

The other psychological advantage of the ‘separate reference project’ approach for me is that it makes me more confident of when the mix is finished. At each referencing iteration, I’ll build up a properly cross-checked list of tweaks I want to do, and then check the effectiveness of those tweaks at the next iteration. Once everything’s crossed off the list, I can feel pretty confident of signing off the mix. If you reference in a less structured ‘hunt and peck’ kind of way, I find it’s a lot trickier to know when you’re actually done.

The last thing to say is that while referencing I prefer to step back mentally from the technical details of a mix and listen more like a typical punter, which is far easier to do when I’m listening to a bounce-out. Because I can’t change anything, my whole mindset changes. Thanks to pure paranoia, I actually do most of my bounce-outs in real time, and I’m constantly amazed at how often I’ll spot some glaring oversight even during the bounce-down itself that I haven’t noticed for the last five hours of mixing, simply because of the change in mental perspective that occurs once I think “now I’m bouncing down the mix”. Also I’m more likely to transport the bounce-down to the car, the office PC, iPod or wherever.

Sure, you could work around all of these issues when using a referencing plug-in on the mix project, but you’ll need a whole lot more self-discipline than I have, frankly! And besides, I think the little breaks you’re forced to have while bouncing things out and switching projects are good for perspective in their own right, but that might be the Luddite in me speaking.




Share:

No comments

Post a comment

Blog Design Created by pipdig