Why is your audio set up the way it is?

I’ve been recording myself playing games for a very long time.  Audio has been a constant struggle for me because my situation isn’t common.  The short answer is, there’s a lot of caveats with what my live show is and there’s certain restrictions that come with them. I’ll break this up into sections:

I want a simple, unobstructed camera scene

I don’t want any microphones, pop filters, wires, mounts, headphones or any other equipment visible onscreen when I’m streaming.  This is both a practical and an artistic requirement of mine.  I like the idea of my stream simulating that you guys have been invited into my living room and we’re having a conversation together while you watch me play games. If there is a mic in front of my face, that just feels wrong.

Microphones definitely make a person’s voice sound better when they’re close to the speaker’s mouth.  When I do scripted videos I use a mic close to my mouth and it sounds crisp like you are expecting it to sound. Unfortunately, I haven’t found a way to accomplish that without having a microphone visible in the shot or causing other complications (I will get into that in a bit).

I feel a lot of what I’m trying to do depends a lot on the onscreen people’s facial expressions.  On stream, we play games and yell at them, we laugh, we get silly, and I feel like a lot of that is ruined by having a person’s mouth and most of their face hiding behind a big circular pop filter and a microphone.  I know my audio quality will suffer for it but it’s a tradeoff I’m willing to take.  Not everyone is going to agree, but I hope you can understand my reasoning.

I also have tons of consoles set up with tons of monitors, sometimes I want to use game accessories like light guns, I want to have virtually unrestricted movement… there’s lots of reasons I need the space and the camera pretty much has to be positioned where it is. And speaking of light guns, they make a particular “cling” sound when you press the trigger. I want people to hear those sounds and my button presses. My chat has said to me before they like hearing my SNES controller creaking. 

The other problem with visible equipment is obstructing movement.  I want to be as animated and rowdy as I want to be without worrying about bumping into stuff.   Having a microphone in front of you sort of forces you to sit in one single position and you’re always worried about hitting it with your knee or your arm and making a loud noise go out into the stream.  That setup works great for podcasts and radio but I’m doing a different kind of show.

The other big factor in my audio setup:

I scream.

I honestly don’t get accused of hamming up my reactions to games very often, but I promise you, hand to God, I don’t.  I am supremely lucky that people find my rage funny because I would do it even if people hated it.  This is the way I am.  I will play a game I absolutely love in every way and I will still raise holy fucking hell when I fall off a cliff or whatever.

The yelling, as it turns out, is a technical nightmare.  Microphones clip when they’re set too high and someone screams into them.  Turning down the microphone makes the audio way too quiet for everything else that isn’t one of my trademark screams.  Not everyone listens to a Twitch stream with studio-quality headphones, a lot of people are just listening with their laptop speakers.

The laptop speakers folks and other people who aren’t in a situation where they can’t boost their audio very loud are a group of people I want to make sure aren’t left out.  The overall audio has to be loud enough, there’s no way around it.  I can turn the gain on the mics way down and blow an air horn into them without clipping the audio, but everything else won’t be audible.

My situation means that my audio will run the gamut from quiet to extremely loud, and that severely restricts my options.  I tried using a lapel mic for awhile and the screaming made it impossible to compress the screams without a ton of distortion.  The lapel mic also restricted my movement and there was a lot of rustling noises that I felt was ruining my gameplay footage.  It was worth trying for an experiment but I came to the conclusion that a lav mic was not the magic solution that I was hoping it’d be.

I have recently purchased a good compressor that actually works unlike my old one, and the screaming has been managed to a good level.  But, the fact of the matter is the screaming is always going to limit my options for audio, and not screaming just isn’t an option for me.

Why don’t you use direct audio capture from the games while you stream?

In order to direct capture audio from a game, the game audio pretty much cannot be heard by the mic.  You can try your best to ride the line but there’s pretty much always going to be an echo of some kind if the streaming software is piping game audio to the stream and the microphone is also picking up the same game audio.  There’s a lot of latency issues and it’s infeasible to fine tune the latency perfectly day to day with every single one of the many game consoles I’m playing on.  Even the slightest offset is going to create undesirable audio.  Dealing with old game consoles involves a lot of converters and cables going into different things and coming out of other things, and every thing added to the equation changes the latency.  It’s a nightmare.  If I was just streaming myself playing PC games it would be easier but oftentimes I’m working with bizarre old consoles that need extra work to interface with modern streaming equipment.

Not to mention, the sound coming from the TV speaker is going to reverberate in the room and arrive at the mic at different times than the direct game capture.  Having these two audio streams overlapping perfectly is supremely difficult not only to line up perfectly, it’s even more difficult to maintain day to day.  It’s a difficult moving target.

The solution most people use is to monitor the game audio through headphones.  I explained in the previous question why headphones are not an option for me.  In short, I don’t want headphones visible on camera and I honestly cannot stand wearing them.  We would also have to pipe in audio from the speakers into the headphones so people on the couch can hear each other speaking, which feels unnatural and complicates things.  Unfortunately, headphones are off the table for me.  Obviously, I need to be able to hear games in order to play them, so turning off the sound for myself and having the direct capture going out isn’t an option either.

Another issue with capturing game audio is that creates an extra problem of audio balancing.  I routinely edit footage from my streams into video and game audio being too loud can make footage completely unusable.  Having the audio going out on the TV speaker allows me to figure out in real time what the volume level should be at while I’m streaming instead of having to play guess and check every time I switch games.

Another thing I want to mention is, and I’m fine if you disagree, but I don’t think perfectly captured game audio overlayed on top of people speaking is necessarily the only good way to go about things.  Direct captured game audio has to be lowered in order to keep the speakers’ voices audible anyways, the game cannot be louder than the person speaking.  The people on the couch are closer to the mic than the TV is, so the peoples’ voices will be dominant over the game, which is what I want.

My solution to the problem is to just have the microphone be the one and only source of all audio on the stream.  It’s a very useful simplification for all the problems I’ve listed on this page.  If I can hear the person sitting next to me talking over the game, then the audio going out to the stream is going to reflect that.  The room environment “softens” the game audio somewhat, almost like a filter, to make sure the voices are coming through clear over it.

That all being said, at the time of this writing, my current audio equipment seems to be capturing the game audio in the room quite clearly.

After all that explanation, there will still be those out there that disagree.  I still see my audio as a work in progress and I’m happy to listen to any suggestions.  But, even if you think my audio isn’t perfect, please at least realize that the audio is the way it is based on a lot of uncommon circumstances based on my specific scenario and these decisions were arrived at after a lot of thought, experimentation, and many different people weighing in on the issue.

There’s tons of articles written by real audio engineers online that say that audio is an open-ended problem.  Audio setups have to change based on the type of sound you want, the type of show you’re doing, the specific needs of the performers, and lots of other reasons.  I’ve spent a lot of time trying to find a compromise between all these various elements, and this is the best I’ve come up with so far.

Here and there I will get a comment like “Get a better mic”. I can assure you, my microphone is an extremely high quality mic. I use the same Sennheiser microphone on my streams that game grumps use, I have talked to Arin before about this. It is an expensive high quality microphone. That is not the issue. It’s that I prefer to use room sound and most listeners aren’t used to that because most other people wear headphones and have a mic up near their mouth. I don’t want it that way.