Jake’s Take: How I Stream Our RPG Actual Play, Part 1 – Overview

Hi folks! We hit episode 50 of Chronicles of Rinn Season 1 recently and I wanted to mark it in some way, so I thought I’d give you folks a look at how I produce our stream and what goes into it, in case it’s of use to any new or veteran RPG streamers out there.

I’ve spent a lot of time trying to improve our production (which I’m now quite pleased with, given our circumstances), and there was very little advice out there when I started. What was there wasn’t tremendously useful to me either – I couldn’t find any detailed case studies of the nuts and bolts of how people streamed RPGs, it was mostly people making the same suggestions of C920 webcams and Blue Yeti microphones, so most of what I’ve learned I’ve had to figure out myself. Given that I’m now in a position to give a more detailed look at how we do ours, I feel I owe it to past me to do so!

I started writing this as one big long post covering everything but that quickly got out of hand, so I’m splitting it up into separate instalments. This week we’re taking a look at a broad overview of everything, and then in later issues I’ll dig more deeply into certain aspects (probably audio and video but leave a comment below if there’s something you’re particularly interested in hearing more about) and give some alternatives and a generally more detailed look at why I use what we use, and what you might wish to use instead.

One thing to note is that I do everything for Chronicles – I’m DMing and producing the stream at the same time so there will doubtless be better ways of doing things but which require more time or focus than I’m able to give it during play. If you have a dedicated production person or are running a stream as a player, for example, there may be better choices for your use case. Hopefully this still proves useful, though! I should also note that this is all coming straight from me, there’s no affiliate links, no sponsorships, nothing like that.

We used to play in my living room but thanks to 2020 and people in our group moving away from Glasgow, we’re now playing online, so we’re going to look at our digital setup today, not our in-person one. I think that’s also the one folks will be most interested in because it seems as if you need to have a very unique circumstance to be streaming a physical actual play at the moment and even in normal times the bulk of streams I’ve seen have been remote play. If we get back to playing in-person again I’ll probably do a follow-up look at how we do that, but until then let’s take a look at the digital setup!

Video Calling

There’s a lot of rival solutions out there for video calling, but we use Discord. We just use a group chat currently rather than a full-blown server, mostly because when we started Discord didn’t play ball with my camera while the stream was running so I had to screenshare it, which you couldn’t do neatly in servers. This isn’t an issue anymore so I might look at shifting over to the Animancer server or a dedicated Chronicles server, but right now we get by fine as is.

I’m mostly using Discord because everyone in our group already uses it. There are other solutions I think suit better (we’ll talk about them in another instalment), but Discord currently gets us by well enough and it doesn’t mean another program for people to learn.

The audio and video quality is alright – it certainly doesn’t seem meaningfully worse than anything else I’ve used – and it has some nice features like individual audio mixing for people in the call. This is the standout thing I like about Discord, other software relies on people mixing their own audio and there aren’t many that let you set the volume you hear others at. I can’t overstate how useful I find this; what I hear on Discord is what the stream hears and a lot of the time people’s audio can vary from week to week or moment to moment, so being able to just touch up the overall mix as and when I need to is really helpful.

I pop out the Discord video stream and put it on one of my side monitors, where I capture it and bring it into OBS. Then I crop it to 7 different video feeds for each of the players. I pull their audio from my headphone feed.

Tabletop & Digital Game Tools

The next part of the equation is the virtual tabletop. Not everyone needs one and there’s a lot out there, but we’re presently using Fantasy Grounds. I pay for the Ultimate subscription so that the players don’t have to pay for anything, and I just get by with the SRD data in Fantasy Grounds. We use the Unity version.

We’ll talk about this more in a later instalment but Fantasy Grounds is the third tabletop solution we arrived at, and so far it’s done basically whatever I’ve needed and hasn’t caused any glaring problems. It has a specific way of doing things so until you learn the sort of ‘Fantasy Grounds mindset’, as it were, it can feel a little obtuse and confusing. It’s not perfect, I can certainly think of things I’d have tried to do differently, but it does have some handy features, mainly in automation.

The players have varying levels of automation set up (Niall has his characters quite well automated, other folks not so much), but by and large Fantasy Grounds will handle a lot of the tedious stuff for us. I think the best example is for spell saves – as long as a few things are done right, Lidda can target a bunch of enemies and cast something like a fireball and Fantasy Grounds rolls saves for all the targeted enemies, remembers who saved and didn’t, and applies the appropriate damage to each enemy. That’s a big convenience for me, and I don’t know of other tabletops that do this (although they may exist).

I store the Fantasy Grounds campaign locally on my computer and run a cloud server through the software which the players all connect to, and I’ve also got a secondary laptop next to me when we play which I use for the stream feed of the tabletop. I used to just capture my main screen for when we show Fantasy Grounds on stream but then people can see all my secrets, so I needed a different way of doing it. I’ll go into this in depth later but I get this laptop feed into my desktop via an NDI feed, if you’re curious (NDI is basically a way to send audio and video streams over a local network to another machine, it’s great).

Some of the players manage their character entirely through Fantasy Grounds, others manage some stuff on Fantasy Grounds and some stuff on D&D Beyond, and others I think manage basically everything on D&D Beyond and basically just use Fantasy Grounds for positioning. I still get the players to track health in Fantasy Grounds so I can have monster damage applied automatically and speed things up.

My Audio

This is where the ‘do it all as one post’ idea kinda fell apart – I study and work in audio so I know a lot about it (there’s still far more that I don’t know than I do but I know more than the average person at any rate). We’ll do a hardcore dive on audio later, but we’ll give a brief overview now.

In basic terms, I use an AKG P420 condenser microphone with a cardioid pickup pattern (it listens at the front but not really the sides or back). This microphone connects to my Focusrite Scarlett 2i4 audio interface via XLR cabling, and then I process that audio in REAPER, my digital audio workstation (DAW). Finally, my audio interface sends the processed audio out the interface and into the line in of my computer via a phono cable, which is what everyone hears.

A lot of that might not make sense to you, don’t worry if that’s the case – we’ll explain it better in the audio instalment and if you need to know urgently, those terms should be easily google-able. In even simpler terms, microphone goes into computer, computer processes microphone signal, sends microphone audio back out and into a different input on the computer, and everyone hears that.

Most of my audio gear I use because it’s what I have rather than choosing it specifically for the job of streaming. That said, it does a pretty good job! The biggest factor is the fact that I process my audio in Reaper before it gets to people, which makes for a huge improvement. My computer sits about six feet from the boiler in my house which can make a fair racket so I run noise reduction plugins and some other audio sweetening to get a better and clearer sound on my voice (I’ll go over that in greater detail later).

Reaper also allows me to take the music we use and send that in the same audio feed as my microphone signal, literally by running a standard aux lead from an audio output on my PC which I’m not using into one of the inputs on my audio interface. I just point the music player and my ambient sound player to output on that otherwise unused audio connection and they arrive in Reaper ready to be sent out to the players. I use MediaMonkey to play my music and I actually use a game audio middleware program called FMOD Studio to create my sound ambiences (when I remember).

In Reaper I set up my microphone track and the track for the music to output to the secondary set of phono outputs on the back of my interface, then I just run a phono to minijack cable from those phono outputs to the line in on my PC, and set my ‘microphone’ in Discord to that Line In input.

There are a lot of guides out there on using free virtual audio cables and similar stuff to do that signal routing inside your computer but those solutions are unreliable and confusing in my experience, and I like the simplicity of a single cable. Purists will tell you that you shouldn’t do what I’m doing because you’ll get noise and they aren’t wrong, but those issues are mostly theoretical – if you use a short cable of a decent quality, i.e. shielded and made of reasonable materials (don’t buy into the ‘oxygen-free’ nonsense though), any noise introduced is unlikely to be noticeable at all to the end user. Don’t let the perfect be the enemy of the good.

I know there are better ways of doing all of this out there and I do intend to move towards one of those solutions in the future. It’s just a question of having the spare time and money to make a switch that won’t improve my audio that much, given I’m already processing my audio in my DAW which makes 90% of the difference.

Reaper is my DAW of choice, other options are available, but Reaper is generously priced and extremely customizable. One of the customizations I’ve made is to set up a keyboard shortcut to switch between ‘normal Jake voice’, and a voice modulator track so that I can create voices for otherworldly or non-human entities like gods, robots, cosmic beings and all of that good stuff.

There’s a look at the Reaper project I use for my audio. You can see I have a Voice track for me and a Tunes track for our music. The Voice Mods folder is where I store various different voice processors like gods or Lidda’s ghost pal White Tongue. My hotkey simply toggles between my normal voice and whichever modulator is highest in the list (Track 4).

That’s a very quick tour and there’s a lot more detail and expansion to go over my audio with, that’ll come in the future. The players largely just use whatever microphones they have, some process them in audio programs if they know how, others don’t. They’re just gathered and brought into the streaming software as one big mass.

My Video

Ok, well, if you’re hoping for a similar sort of breakdown for my video, I’m afraid you’re probably going to be disappointed. I’m neither a photographer or a videographer and have a fairly limited understanding of the technicalities of cameras, but I’ll do my best.

My video feed comes from my Sony NEX-VG20E camcorder (now fairly elderly, think they initially released in 2008 or 9). I picked this up secondhand from eBay at a good price based purely on the reasoning that others I knew of used camcorders in that range, and that it could take interchangeable lenses. In order to get that into my computer I have it connected to an Elgato HD60 Pro capture card, which is a PCIe card that lives inside my computer. I’ve got no complaints about this capture card although I suspect cheaper ones are available. It was the only manufacturer name I recognised when I was shopping for one but as a ‘streamer brand’, I reckon that comes with a reasonable markup. If you’re looking for one, see if you can find recommendations from someone more knowledgeable than me.

I use some kind of Sigma prime lens (a prime lens is one where you can’t change the focal length, sort of how zoomed in the camera is), I think it’s a 19mm one because it needs to sit quite close to me. I just have the camera mounted on a little clamp so that it rests on top of my monitor – it’s just a cheap little thing that I got off eBay or Amazon or something some time back but I need a better one because it sags.

That’s…basically it. Camera points at me, sends video to the capture card, capture card passes it into the computer. I have a big LED video light behind me, I try to bounce that off the wall behind my computer to brighten me up – makes a big difference!

The players have various different webcams that they use, mostly the ones I used to use for our in-person stream which were various Logitech ones – the C920 that everyone recommends, a C922 I got from eBay (I actually find that one worse than the C920 despite it ostensibly being better), and two Logitech BRIO webcams.

Streaming Software (OBS Studio)

The final part of the equation is the streaming software that gets us from my computer onto Twitch. I use OBS Studio as I think basically everyone does. Previously I used Streamlabs OBS which is basically the same but is always a little ways behind OBS Studio, but I switched over for a reason I no longer remember. I think it was something to do with plugins being available in OBS but not Streamlabs.

I’ve got a lot of different scenes (partially because I keep the in-person ones around) – an intro slide with a slideshow, the title credits, a main scene with all the cameras and a version of that with Fantasy Grounds as well, plus an end screen. I use those ones every week, but I have some other ones if we need to take a break, if I’m running one-on-one for either a segment or a full session or the like.

I made the title sequence in Adobe After Effects a while ago with music I composed and the character portraits Izzy did. You can see it in action below. I’m still pretty pleased with it but it’s probably due a refresh soon.

I record everything at 1920×1080 50fps (my camera FPS), but I rescale that to 1280×720 for the stream as it seems to play a little better data wise and not everyone out there has speedy internet connections. I stream at a constant bitrate of 4000Kbps, mostly because it seems to work quite well and I’ve not had any complaints. If your streaming service reliably allows your stream to be transcoded (set to different resolutions by the viewer), you can probably go much higher than this and stream at whatever resolution you want. Whatever service you stream to will suggest stuff to you, or OBS can just set up what it feels is reasonable.

I record the stream on the CRF Rate Control with a CRF of 17, which seems to give me a reasonable file size and video quality (most of our sessions come out at around 4 to 5 GB). OBS defaults to .mp4 recording I think, but I record to .mkv instead.

My understanding is that if your recording software crashes in the middle of a recording, a .mp4 file is entirely lost but a .mkv file is just cut off, so this is mostly a peace of mind thing. MP4 and MKV files are the same in terms of data, but they have a different container, so you can actually convert from MKV to MP4 very quickly and easily because it doesn’t have to rewrite any data, just how the data is housed. If I need to do any editing to a session after we stream it I’ll convert it to MP4 for that, otherwise I just upload the MKV file to YouTube and use the built-in editor on YouTube to slice off the preamble bit (it saves me having to re-encode the whole session if I did it myself).

I have all of our sessions archived on a 2TB hard drive, going back to before Chronicles even started. I need to get a new drive soon because it’s almost full, so bear storage in mind!

And That’s…Mostly It

That’s basically what goes into putting a session together for me, at least in broad terms. Hopefully it’s useful to someone, and over the next few weeks I’ll do more detailed dives into specific elements. If there’s something you’d like to know more about, leave a comment down below and I’ll either reply in there or I’ll address it in a future instalment. Or both!

We’ll probably look at audio next as it’s most in my wheelhouse and I feel it’s where I can offer the most assistance. Until then, you can watch this all in action over on our YouTube channel or live on Twitch on Tuesdays at 19:00 UK time – see you there!

Part 2 in this series, ‘Digging into Audio’, can be found here.