Besides video quality, there's another huge downside to doing "smartphone or tablet in the back of the sanctuary" live stream. Getting graphics onto the live stream is going to be difficult. Sure, you could pan over (or tilt up) to the projector screen, but that's going to give you so-so results at best.
Additionally, some of the systems that might seem like a good next step once you want to grow your capabilities don't make it easy, either. You might be able to go to a full screen by treating the computer output as a camera or do a PIP (picture-in-picture), but that's it, and it's not available on all systems.
So, what can you do?
Start with your objectives
Do you want a lower third with both transparency and translucency? Then, your only real option is to use an alpha channel, which requires either a more sophisticated encoder or a video switcher. Are you okay with full-screen graphics only? Then maybe you can capture the computer as a video source and switch to it. Want something in between? You'll need to have something more than a base system.
Look at what you have and your budget
Here's where you need to be realistic. Going from a single smartphone in the back of the room, you're not going to jump to a multicam live stream with animated lower thirds that have translucency for free.
On the other end of the spectrum, don't settle for a $500 upgrade that won't actually meet your needs if you have a budget of $100k.
That doesn't mean you can't dream big and do upgrades in stages. Just be realistic about what you can accomplish and what your needs are in relation to your budget.
If your church already does IMAG, and you have a 2 M/E (or greater) switcher that you're just using one M/E of, the addition of graphics to your live-stream might be pretty simple, and maybe free.
Don't forget to include current equipment in your list, too. If your church already does IMAG, and you have a 2 M/E (or greater) switcher that you're just using one M/E of, the addition of graphics to your live-stream might be pretty simple, and maybe free. In that case, it's just using what you have, better.
Consider your options
Now that you have a realistic idea of what you have and what you're trying to accomplish, let's look at the different ways of doing it.
At its most basic, adding graphics from a computer, whether it's running worship software like ProPresenter, Proclaim, MediaShout, or something like PowerPoint or Google Slides, can be just adding a full screen version into the live stream.
If you like everything about what you have now and you're encoding separately from capture, it might be just as simple as adding in a Roland V-02HD or BlackMagic Design Web Presenter (with the optional front plate) so that you've got the ability to switch back and forth. Other options include capturing a duplicated feed from your presentation computer directly into encoding software via a capture card or over the network with NDI. With the SDI module, ProPresenter 6 can send NDI directly, but other software might need additional add-ons like NewTek's NDI scan converter (which is a cross-platform piece of software that's included with the company's free NDI tools suite).
Either way, this approach comes at a cost. While people watching the live stream will see the same full-screen graphics that people in person see, they don't have a view of the people on the platform. So, if you're showing nature footage during a song about the majesty of God, those online won't see the facial expressions of the worship leader and won't see how she's really connecting with the song in her worship. If the pastor insists you leave the sermon notes up while he's preaching, people online will only see the sermon notes and won't see the expressions on his face.
Another option is to do a PIP. Think of this as the boxes of information that show up over the anchor's shoulder on the news. You might already know the problem with this approach. On a smartphone or tablet, the amount of information you can convey in a small portion of the screen will be pretty limited.
One strategy to overcome these limitations would be to combine the two. If you can do both a PIP and a full screen, maybe show each new slide fullscreen so that it's up long enough for people online to read the information, but go to a PIP until the pastor is done referencing the information on the slide. A similar strategy is to use lower thirds or side thirds instead of PIPs.
If you can do both a PIP and a full screen, maybe show each new slide fullscreen so that it's up long enough for people online to read the information, but go to a PIP until the pastor is done referencing the information on the slide.
As most video people know, good shot composition places the most important information on an imaginary line at the top third, bottom third, left third, right third, or the conjunctions of those lines on the screen. So, if you put the pastor's eyes on the top third line, you could put a smaller graphical element on the bottom third line. This would both balance the shot and provide a good compromise between PIPs and full-screens. You could also do the same with graphical elements designed, not for the bottom third of the screen, but for either the left or right third.
That's the artistic idea, but technically, how do you show part of the graphic or format it so that only part is shown?
Here you have some options. If your system can only do a PIP, but you can crop that PIP to show only part of the image, perhaps you could create a hybrid between a PIP and a lower third, showing the graphic full-size, but cropping the top two thirds.
Normally, lower thirds (and side thirds) are created with other means, though. In a professional production environment, they'd probably use an alpha channel to key out the parts of the image created by the CG (character generator) to make a lower third.
Some worship software will allow you to do this using additional hardware. You take two HD-SDI signals into a switcher from that hardware and your switcher interprets one as the main signal and the other as a mask. This can be a pricey option, though.
So, you might consider using a luma key or chromakey instead. You've probably heard of chromakey and know that it's how the local meteorologist can stand in front of a computer-generated weather map. The video switcher just removes everything that's bright green (or blue) from the video signal, leaving everything else to be layered on top of the background. There's no rule that you couldn't use a green graphics background instead of putting a person in front of a green wall.
There's no rule that you couldn't use a green graphics background instead of putting a person in front of a green wall.
Another option is similar, but doesn't use a color. Instead, it uses the brightness information in the video signal. Luma (or luminance) keying removes either black or white from the signal, instead of green.
Depending on your switcher or software, this might cause a problem when a white shirt, for example, is a similar enough color to white text to make the text keyed on top of it illegible. If you could add a dark gray outline to the text, that the switcher ignores, that solves the problem. Not all switchers, particularly the cheapest ones that most churches consider first, are capable of distinguishing between black and dark gray for keying purposes.
So, it's best to sort out what your equipment will allow and what will give you the best key.
Putting it all together ...
If you have all these tools at your disposal, you'll get the best combination. Perhaps the pastor wants to show a full-screen image. Switch to that, then for the following scripture slide, use a lower third keyed over the live video.
No matter which technique you use, they'll all be better than just panning over to the projector screen.
Maybe you'd add in a PIP from time to time, when the information allowed it and you wanted to show the live video as well. No matter which technique you use, they'll all be better than just panning over to the projector screen.
