Friday, May 17, 2019

Getting Your Stream Output Settings Just Right




There's a lot to think about when making a Twitch stream, but some of the most mysterious aspects are the stream output settings. These don't deal with how your stream is designed or how you as the host act on camera, but rather how the stream gets to the internet, how faithfully the image you see on your PC is conveyed to the viewer, and how reliably the audience is able to watch the stream without buffering. When doing their job right, your output settings will be completely invisible. When set up wrong however, your stream settings might cause the show to look ugly, create performance issues, or even completely crash the stream.

Choosing a resolution is like choosing the size of
a barrel. Choosing a bitrate is about how much
you can pack into that barrel.
I'm going to help you arrive at just the right settings for broadcasting your streams- not by giving you the exact numbers to punch in (everybody's games, PCs, layouts, accessories, and internet connections are too different to give one definitive configuration), but by handing you the tools to arrive there yourself.

Please be aware though, especially if you've never read The Twitch Playbook before and you skipped to this entry first: ON CAMERA EXPERIENCE is always more valuable than a perfect stream. If you haven't done at least a dozen combined hours of streaming already, put this entry down, enable all the "auto" settings on your broadcasting software and get out there, live on the internet. If you need an extra motivational boost, read the entry Start Your Twitch Channel With NO MONEY.


➢ WHAT ARE THE MAIN STREAM OUTPUT SETTINGS?


Depending on what broadcasting software you use, there may be a crazy amount of sliders and menus for you to tinker with. There are only two truly top-tier settings though, which just about every piece of streaming software will have, and these govern all other aspects of your stream. They are: RESOLUTION and BITRATE.

If you got Da Vinci to paint your portrait, it would
be totally lifelike. But this may not always be
necessary.
The interesting thing about streaming is that what you see in your game, or even on your computer's broadcasting software, is NOT what your viewers see. The nature of the internet is such that only a portion of your stream's original quality can be conveyed to viewers. There's a lot of science behind resolution and bitrate, but it all boils down to the following concept:

Before cameras were invented, if you wanted a picture of yourself, you'd have to hire an artist to paint a portrait. The resolution and bitrate settings you choose are equivalent to choosing a better or worse artist to paint the 'portrait' of your stream. Low settings may be severely lacking in detail, and high settings may be completely identical to the source, but it's important to always remember that there is an intermediary between you and the viewers.

Resolution denotes how many pixels of video are delivered on your stream. This is typically measured by two numbers, which represent the horizontal and vertical pixel counts. It's not necessary to get into more details for these purposes, but suffice it to say that 1920x1080 is the HD resolution commonly known as 1080p, and 1280x720 is the HD resolution commonly known as 720p. There is also the 'standard definition' resolution of 848x480. These are the most common output resolutions for Twitch streaming.

Bitrate is trickier for many to understand. This number, measured in Kbps, or 'Kilobits Per Second', denotes how much data is packed into each second of your stream. The lower your bitrate, the less internet speed is required on your end to broadcast. The tradeoff is, the lower your bitrate, the blurrier your video will look.


➢ THE PROBLEM WITH GOING TO 11


"Very interesting," you might be thinking. "But higher resolution and bitrate mean higher quality streams, right? I'll just crank everything up as high as it'll go and then my streams will look amazing!" This could work out for you, but in most cases, extremely high settings are a terrible idea. There are three main factors to be aware of: Processing Power, Your Internet, and The Viewer’s Internet.


PROCESSING POWER

Don't make your PC explode.
The higher your resolution and bitrate settings, the higher the processing load on your PC. As it is, your stream software is sharing your PC's processing power with the game you’re playing, as well as all other pieces of equipment connected to your computer. How much this affects you depends on what kind of computer you have, and how much strain your game puts on your PC. Playing a pixel based sidescroller won’t likely cause your computer to bat an eye, but attempting to play a next-gen visual powerhouse game on its max graphics settings, while also outputting a stream at max settings will likely crash your computer or cause major performance issues. Playing a game while streaming isn’t the same as playing it off stream either- just because you get a solid 60fps in Metro Exodus while playing on your personal time does NOT mean you’ll get the same results while streaming. If you're running into performance problems, it's likely because of these warring juggernauts: try lowering your game's graphics settings, and if that doesn't work, lowering your stream settings the next time you go live.


YOUR INTERNET

Everybody knows that you can't stream without an internet connection. What most people don't know is how much internet is required to stream, and how much of your home internet is safe to allocate for streaming. If your bitrate exceeds your amount of available internet speed, even for a half second, you'll start 'dropping frames', meaning your stream will appear frozen to everyone watching, until the internet speed raises above your attempted bitrate again. Here's a simple metric to find out how much internet you should be allowing your stream software to use, so that you won't have this problem:

Find out your internet plan's upload speed (not the more commonly advertised 'download speed'). A good bitrate to start with is roughly 30% of your upload speed.

This accounts for the huge swings in connectivity most internet companies suffer from (though they don't advertise it much), as well as anyone else who might be using the internet in your house.

Your internet provider may advertise their speeds in Mbps, or Megabits Per Second, which is simply the Kilobits Per Second number that your stream software uses, divided by 1,000. So for example if your internet plan has a 10 Mbps upload speed, this means you can use 3 Mbps to stream with. That's equivalent to 3,000 Kbps in your stream software.


THE VIEWER’S INTERNET

One other important point that many new streamers don’t consider is how the viewer RECEIVES the stream. Let’s say your PC can handle anything you throw at it, and your internet is screaming fast. If your stream outputs at 10,000 Kbps, this means your viewers need to have at least that fast of an internet connection to actually watch it. Bitrate is not only the setting that dictates whether you can stream from your end, but also the setting that dictates what whether someone can watch on their end. Have you ever watched a YouTube video that froze and started buffering, so you had to wait for it to load? This means the transmitted bitrate was higher than your internet connection could pull down at that moment. And you don’t want your viewers sitting around watching a ‘buffering’ screen all day. Otherwise they won’t likely be viewers for long. So you shouldn't just be lowering your Bitrate to the maximum that YOUR internet can handle, but one that most people can actually watch. If a lot of people complain that your stream is buffering, especially if they mention that other streams on Twitch load just fine, don't ignore these comments: lower your bitrate. You don't want to drive people away just to make your stream look better.


➢ THE MAGIC NUMBER


If you're on a pretty good PC and have decent internet, try starting with a resolution of 720p and a bitrate of 2,500 Kbps. From here, you can go higher if everything is smooth, or lower things if you notice problems. But there's a reason I waited until the end to mention these two numbers. Hopefully by teaching you what these metrics actually mean, and by demystifying the numbers themselves, you'll be able to get more performance and quality out of your streams than if you simply punched in a 'one size fits all' value.

Your ability to make an entertaining show will always be more important than how clear your stream looks. But if you've been improving your stream and are looking for another way to get better quality for no money, optimizing your output settings is a great way to do it. Always keep the 'trial and error' spirit I've mentioned in previous entries- you'll likely want to tinker until you find the sweet spot. But now, when raising or lowering these settings, you won't be shooting in the dark. Once your stream output settings are just right, you'll have one less thing to worry about when going live. So take some time behind the scenes, and make that stream shine!

Friday, May 10, 2019

Attempt Your Worst Idea for a Twitch Stream




Let's say you've been streaming for a while and your shows are going well. Your streams may not have the fanciest equipment or newest games, but you've got your audio mixed properly and your hosting persona is getting more refined. You might have also found a comfortable groove for your channel's schedule- whether that's always going live at a specific time, or just making sure you never miss a scheduled day. All is going well.

Do you love playing with action figures? Stream it!
But you probably have an idea kicking around in the back of your mind- a style of stream that's outside of your comfort zone. You haven't done it before, you're not sure if you'd be able to do it well, and you may have never seen anyone else on Twitch doing it- it lives in completely uncharted waters. For these reasons, you are terrified of attempting that stream.

This is good!

Fear of trying something on stream is the best way to know that it's the exact kind of idea you SHOULD be implementing. You should always try your worst idea for a livestream. You never know what's going to work.


➣ BAD IDEAS ARE SUBJECTIVE


It's hard to pin down what seems like a 'bad idea' to anyone out there, because anything that seems crazy to one person could be another's bread and butter. Suffice it to say that it's all relative- the kinds of streams someone typically makes are what cause a new style of show seem strange: a Fortnite streamer doing an IRL show where they repair their motorcycle, a competitive Overwatch player strapping their phone to their chest and livestreaming their morning runs in first-person, or someone known for painting miniature Warhammer figurines doing live karaoke. Think of the hobby you love most outside of livestreaming- what would it be like if you were able to incorporate that into your Twitch channel?

If you love building or customizing your PC,
don't assume others will find it boring.
Try sharing it with the world!
"But Nick!" you might complain, "My passion for LARPing would make a terrible stream! Why would I attempt something when I don't even know if it'll work?" There are many reasons. First, you can meet new kinds of viewers- if you have a passion for beatboxing, or photo editing, or sandcastle building and decide to do it on stream, there will most likely be others out there who love the same thing. You're making your channel more varied, and therefore more discoverable. Your existing fans would be able to look forward to new kinds of shows as well, having a welcome break in the typical style of stream. And most importantly, you'll be more creatively fulfilled- you'll get to do something you love and share it with everyone! Don't think of this as a permanent change- it's just a one-time thing. See how it feels to try on stream- if you like it, you might find you want to bring it back.

There's a saying by Confucius: "Better a diamond with a flaw than a pebble without." Don't let your stream be a perfectly flawless pebble- keep trying new things even if you don't know whether they'll be any good. This is how great ideas grow!


➣ MY PERSONAL QUEST: THE TRAVEL STREAM


On my streams, I've implemented many of these kinds of ideas, and they have done wonders for my channel. That's not to say that I've kept every crazy idea I've tried, but rather that I have never stopped trying my worst ideas for streams, and through this constant experimentation have been able to significantly improve my channel. The following was my personal journey in arriving at one such stream idea.

For work, I typically fly around the country about a dozen times per year. When I started on Twitch, I didn't want to let this interrupt my daily livestream schedule, so I came up with a way to still stream while away from home. Several streams ensued, some of these tried and refined multiple times, but here are the overarching concepts.

I tried to hard to stream video games away from home, before I realized
I could branch out into the other things I love!

1st Attempt: The Portable Console

When I first started, I didn't do anything except stream story-based games. And I had a hard time deviating from that rigid concept in any way. So my first thought was to bring my Nintendo Switch, laptop and a whole nest of capture equipment and cables with me. I'd plug my Switch into the computer, route it into the hotel's TV through an HDMI splitter and play Zelda while away from home. It worked (in theory) but there were too many variables: some hotel TVs weren't compatible, there was too much hardware to lug around, and most importantly: my laptop couldn't handle it.

2nd Attempt: The Less Intensive Game

On my second trip, I tried streaming Stardew Valley, a simpler pixel-art game, from my laptop. Using a camera would still cause performance issues on the stream, but it ran more smoothly than the previous, more intensive setup.

3rd Attempt: No Camera

Next, I tried playing Stardew Valley with no camera. This worked better than before, but I'd still be cursed with crashes and unpredictability. After a few tries at making this work, it was time to go back to the drawing board.

4th Attempt: Video Game Novel Readthroughs

I picked up a companion novel to one of the video games we were doing on stream and did a dramatic reading of that story during simple Just Chatting shows. Since I typically create voices for all the characters on video game streams, this fit the channel identity well, and tied into the games we played. But I wasn't passionate about this idea, so I ended up scrapping it too.

5th Attempt: Walking Around IRL

I then tried doing streams where I would walk around outside the hotel where I was staying. Sometimes I would be in really cool places and could do virtual tours of the area. But the schedule of my work while traveling made this unsustainable: if my plane landed at 11:00pm, it would throw a wrench in any plans to walk around on camera.

6th Attempt: Coloring Book Stream

Getting to rekindle my love of art was a major source of
happiness in my life, and it's all thanks
to my Twitch streaming career!
One day I found a Harry Potter coloring book at a used book store. I thought it would be fun to try a coloring book stream, and after attempting this once I knew it was a huge hit. Everyone loved watching, and enjoyed the relaxed tone. I quickly realized this could be my idea for travel streams: all I'd need was my phone, a set of colored pencils, and this coloring book. It was easy to carry, fun to watch, and I was given an outlet for a major passion of mine. One that I never thought I'd have time to do anymore, before incorporating it into my Twitch channel. For the first 20 years of my life, I had been studying the fine arts, but ever since I moved away from that in my career I haven't had time to draw, paint, or create much at all. It was amazing to regain this creative outlet!

For the past several months, I've been doing an ongoing 'playthrough series' for the entire coloring book. Just like a story-based game, we are aiming to finish the entire thing from cover to cover. WHEN that task gets done doesn't matter, because it's a fun thing to look forward to every time I go away on business. I don't dread the extra work of setting up a stream away from home, or avoid streaming while out of town and then stress about how I'm missing  potential opportunities. I've met all sorts of new people who are interested in watching artwork-focused streams, who have then joined my normal video game shows after enjoying my personality while coloring. I now know there's a thriving art community on Twitch, but at the time I had never seen an artwork stream before. I didn't even know whether anyone would be interested in watching my coloring book show. All I knew was that I would enjoy doing it. And that's what fueled me. Even though the coloring stream concept seemed craziest to me out of all the ideas I tried, it ended up being not only the one I liked the most, but the one people enjoyed watching the most too. Who would have guessed?


➣ UTILIZE YOUR PASSIONS


I'm not suggesting you find some gimmick for your Twitch channel, or attempt streams you know won't work. I want you to try the streams that you WISH would work- the ones that would make you so happy to be able to create, even if right now you don't know who would want to watch them. Don't mistake this for me advising you to buy better tech either, to make a 'dream version' of your existing streams- I'm talking about trying your completely outside-the-box ideas that incorporate the passions that you haven't ever shown on your Twitch channel.

Do you love cooking? Video editing? Talking about sports? ASMR? Archery? Woodworking? Watching YouTube videos? Petting your cat? Learning new languages? It doesn't matter what the idea is, as long as you're passionate about it. Imagine if there were a way that could be incorporated into your Twitch channel. How pumped would you be to go live on Twitch when you were making one of those shows? Look at my example- every idea isn't guaranteed to work for you, but once you find the one you love doing, you'll be glad you chose to 'color outside the lines'. So get out there and try your worst idea for a Twitch stream. You might find that it was actually your best!

Friday, May 3, 2019

Finding the Flow of Your Twitch Chat




You now have people chatting in your streams and having a good time. Using some of the concepts we've discussed in the previous entry, you've been crafting your community into something you can be proud of, and viewers feel more welcome in your shows. When it comes to your personal interaction with the chat, the typical rule of thumb is that more engagement is better. But I've found that this is not always the case. You should certainly be getting to know everyone in chat and responding to comments during your shows, but every stream has its moments where the action on screen should take precedent. In many instances, engaging with your chat too much can actually hurt your performance in a game, or even damage your channel's identity. I want to help you identify the flow of your own Twitch chat, and then help you to refine the way you interact.


➣ SETTING A CADENCE


There are many different kinds of Twitch channels, and each will have its own flow of chat. Someone who is painting with oils on canvas will likely only look at their chat during set intervals, not wanting to constantly break their concentration. Someone doing requests for songs to play on guitar can't interact with chat at certain times even if they want to, because they're in the middle of playing. These kinds of streams are blessed with a natural cadence. The audience implicitly understands when the streamer is capable of even looking at their comments, and it's within these established windows of time that the host chooses what to read, and how to respond to it.

Whether your instrument is real or fake, your chat
implicitly understands that you can't read
comments during a song.
If you're playing a video game, walking around in IRL, or doing a Just Chatting stream however, there are virtually unlimited opportunities for you to both look at and respond to your chat. And while 'maximum engagement' is the generally accepted strategy, I find that a truly refined show in these categories will still have its own style for when to engage with comments. This means setting a cadence for responding to chat on your shows, establishing times when the chat knows that you will or will not be responding to comments, and finding interesting ways to keep people engaged throughout.

All of this will first depend on the kind of show you're trying to make. Are you focused on playing very skillfully? Then you probably shouldn't attempt to read chat in the middle of combat in Apex Legends. If you're walking around and doing an IRL stream, certain moments could be better for talking to chat than others- don't interrupt someone you're talking to in real life in order to respond to chat, for example. For me personally, since I focus heavily on story-based games, I never talk or even glance at chat during cutscenes or story-heavy moments. This creates less engagement at certain times, but it strengthens the overall brand of the channel. People can rely on a natural rhythm for chatting, and can trust that on my channel they'll always be able to watch the cutscenes uninterrupted. This ultimately makes the shows more enjoyable for the kinds of viewers who tune in. Identifying what about your streams is crucial to the identity of your overall channel is the first step. You want to establish what can be interrupted by engagements with your chat, and what can't.


➣ NEW WAYS TO ENGAGE


If you're playing Dark Souls for the first time, you're
probably going to want to concentrate when
you fight these guys.
One of the biggest mistakes I see streamers make in their engagement cadence is attempting to read a comment at the wrong time, and then blaming the commenter for having spoken. For example, someone fighting a tough boss in Dark Souls. They glance at chat while in battle, read a viewer's comment, then look back over at the game just in time to see their character get smashed with a giant mallot. They then either get mad at the chatter, upset that the person broke their concentration, or they make a playful note of how that chatter 'jinxed' their boss run. Neither of these is OK. You as the streamer are responsible for when you read chat, and shouldn't make someone feel bad for having engaged.

This same concept applies to any skill-based game. If you're playing Overwatch, you should have the discipline not to look at your chat during a high-stakes battle. In Fortnite, there are certainly more relaxed moments that are perfect for discussion, but when you're furiously building and fighting other players you shouldn't have your focus split. People shouldn't be discouraged from chatting during moments of intense concentration on your part, but instead given different opportunities for engagement.

Returning to the live music stream example, a guitar or piano player could have different chatbot commands for viewers to activate while they play their instrument. A command full of fire emojis, excited faces or crying faces could be used during different kinds of songs, or even different moments within those songs- a sad refrain, an exciting chorus, or a skillfully played solo. An IRL stream could encourage a sort of 'prop hunt' concept within their chat, asking viewers to post a certain emote set every time they see something in the background: an animal if they're streaming from a zoo, a boat if they're at the beach, or a certain kind of hat if they're on a crowded Tokyo street. On my shows, I actively ask people during boss battles to 'send hearts' as a show of encouragement, knowing that I typically won't be able to read normal comments, but could see the hearts coming in from the corner of my eye and thank the chat for them. In these examples, by having moments where you as the streamer are not expected to engage with chat, you're also not shunning people who are interested in actively chatting on your stream. Instead, you're giving them different and more varied ways to become a part of the show.

Like in some video games, the world has lots of hidden objects.
Why not let your chat join in searching for them?

Experiment with different cadences for your chat interactions- you might find something unexpected that suits your channel well. This will not only help you to concentrate more on what you're doing on-stream, but will give the audience more opportunities to get involved. Some people watching don't always want to come up with questions or comments to write in chat, but they do want to support you as the streamer. You may find that people who don't normally chat during your streams will come out to post emote sets during tough battles, if you make that a part of the experience. Talking to your chat and answering questions is always important, and you should be trying hard to build connections with your community. But despite what many others might tell you, it's not necessary to be available for immediate response to chat at all times. In moments where it's difficult for you to engage, you shouldn't fight against the current and attempt to respond to everything. You need to find the flow of your Twitch chat, and sometimes that means engaging less in certain moments in order to ultimately create more engagement overall.