Huck’s guide to making machinima in Second Life part five: working with other people

In part four of this series I looked at some of the technical uses that alts – alternative accounts, running at the same time as your main SL account – can be put to. In part five, I’m going to take a look at a more social element of machinima-making: working with other people.

Before I go any further on this topic, there’s a biggie that has to go first. Filming machinima by yourself in SL is very time consuming, and filming machinima with other people collaborating is more time consuming.

There’s just no getting around that. It’s going to take more time because there are more things to do. You’re going to have to communicate what you want the other person (or, if you’re really brave, people) to do, and you’re going to have to wait for them to do it. For you, this will be experienced as anything from a small to a mindbogglingly large increase in time, depending on how quickly, perceptively and skilfully your actor responds. But this message isn’t about that. The most important thing is that your actor understands in advance just how much time this activity is likely to take, because however long a period of time it feels for you, it’s likely to feel a good deal longer for them.

There will be long periods of time where they have nothing to do whilst they stand in a pose or animation and you film them from different angles and distances (sometimes, they’ll try to fill this time by chatting with you, which you will be unlikely to have the time or mental capacity to respond to) and there will be periods of endlessly repeating the same movement or action so that you can capture it the way you want to. Be mindful, then, that when you’re asking someone to play a role in your movie this is a very big ask of their time, and make sure that they understand this too. It can be exciting (and flattering) to be asked to take part in such a project, and it can sound like fun. Make sure they understand that the shoot (or shoots) is likely to be long and boring. If you don’t get this understanding in advance, then one of two things is likely to happen. (1) You (both of you) will get frustrated with the other person and your friendship (if it’s a friend you’re filming with) will be stretched. (2) You’ll try to rush the shoot and end up with footage you’re not happy with.


So: communication. There are three key things you will need to communicate to your actor:

  1. What you want them to do.
  2. Where and when you want them to do it.
  3. Stuff that you know about SL and the thing you want them to do that they might not know.

When I was talking to Strawberry Linden about my movie STÖMOL on LabGab, I explained that I give my instructions to actors in IM chat, not in voice. This might sound counter-intuitive, since surely it’s faster and easier to simply say what you want the other person to do rather than to write it – and I joked on the show that this meant the other person couldn’t hear the irritation in my voice. In all seriousness, however, there is some logic to this method. First, the act of writing out the instruction makes you think about it more carefully: hopefully, the result of this will be that the communication is then more clear to the other person. Second, having a written record of what you’ve asked for means that the other person can go back and re-read it as many times as they want if they forget what you said or missed an important part of it. You don’t know what other things your actor has going on around them in their RL and writing things down for them will mean that whatever distractions they have they can still go back to that and check what it is you want them to do.

Of course words alone aren’t always enough. Sometimes you will also need to show the other person what you want them to do. This is particularly important when it comes to movement. One approach I use is to rez a prim, turn it into a long, flat rectangle and then place it on the ground along the route I want them to follow (obviously, this can only be done in regions where rezzing is permitted). In Firestorm, you can derender objects so that they don’t appear to you, so if I derender this pathway once I’ve laid it down it’s no longer visible to me, but it is visible to my actor to walk along. If you’re filming in someone else’s region, don’t forget to teleport out and back in again once you’re done, so that it becomes visible to you and you can pick it up.

Timing is also important. Actors need to know when you want them to do the action you’ve asked them to perform. One way of doing this, of course, would be to use voice; I choose not to do this, because then I’m in voice and I’m going to be tempted to switch to using it for all my other instructions. Also, I really don’t want to encourage chat whilst I’m concentrating. So my approach instead is to use good old, “action”. However, I actually don’t want avatars to start doing stuff the moment I enter this into IM because before I start filming I need to get rid of the SL user interface and any HUDs I have on screen (see part one on how to do this). Also, I might want there to be a short period of still on film before the desired action is performed. For this reason, then, I always give my avatars a time to count down after I say ‘action’ before they start (eg, “After I say, ‘action,’ count down five seconds and then go”).

And don’t forget, it’s highly likely that you will need to repeat each take several times – either to capture the action from a different angle or to film again where a mistake happened the first time (or the second, or the third…). For this reason, it’s important also to tell the actor what you want them to do once they’ve finished the desired action. For example, “When you get to the end of the route, I want you to go back to the point you started from and wait there.”

Here’s a top tip about filming an avatar walking. Walking in a straight line in SL is actually very hard. You really want avatars to do a straight-line walk, however, because mid-walk adjustments just look unnatural. You might think it’s just a question of positioning your avatar at the start of the walk so that it’s facing the end point, but it doesn’t quite work like that. When avatars turn on the spot they don’t do so in individual degrees: press repeatedly the right or left key whilst you’re standing still and it will take a little bit of time before your avatar responds, and then it will turn by a whole 20 degrees or so (my estimate). But even though your avatar isn’t visibly turning during these key-presses, your viewer is still registering them: if you press it just once then when you walk forward your avatar will follow a different route than if you had pressed it three or four or five times – even though there was no visible turn before walking. So it can take a few goes at a straight line walk before finally you hit that sweet spot and make it all the way to the end without bumping into something or having to make an adjustment mid-walk. If you then need to do a repeat take, however, you have to go back to the start and experiment all over again. But, if you reach the end point and then press the back key (I use the arrow keys for my movement in SL, so this is the down arrow key) then your avatar will turn on the spot and walk back along precisely the same line that you just took: when you get back to the starting point and stop walking, it will then turn back in the original direction, all ready for the next take and perfectly lined up.

Another top tip about walking. Try doing the walk in mouselook mode. Often in this mode your movement will be a lot smoother. This also makes more complex walking routes (ie, routes that aren’t a straight line) much easier to carry out.

This is the sort of information I’m referring to in point 3 above, things you might understand about SL which might be of help to your actor. Don’t assume that everyone knows the stuff that you know. Whilst it might take a little time to teach an actor who’s having some difficulties a tip like the one above, you’ll get that time back several times over once they start using it.

And, of course, the opposite is also true – sometimes your actor will know stuff about SL that you don’t. Talking to them about what you want to do and the obstacles you have to overcome will give them insight into your knowledge and give them an opportunity to teach you when they spot that you don’t know something helpful that they do. This happens very frequently with me.

Positioning another person

So communicating what you want an actor to do in SL is hard. There are some circumstances, however, where you don’t necessarily have to tell a person what to do because you can do it yourself. Where it’s possible to do this, do this: it’ll save a lot of time and pain.

The most obvious of these circumstances is where you want to show the other person in a static position. This might involve them sitting on a chair, for example, or standing in a fixed spot. You might want to keep them in that position for a still or moving camera shot, or you might want to focus in on a particular aspect of their situation in that spot, for example a hand animation that’s playing whilst they’re located there.

In many cases, it’s enough to simply say to your actor, “Please sit on that chair.” You might also want to direct them to a particular animation on the chair animation menu, if it has one: “Please select the animation, ‘Sit 1.'” As we all know, most furniture animation has to be fiddled with in order for it to ‘fit’ perfectly for your avatar, and most furniture menus these days come with an ‘Adjust’ menu that allows you to make small, medium or large adjustments to the X, Y and Z positioning, as well as rotational adjustments. In most cases, however, you have to be actually sitting on the furniture in order to access these menus. This means that your actor – rather than you – will have to make these tweaks.

But if you’re filming in a sim which allows rezzing, you don’t have to go with just the animations available in a piece of furniture there (you might well not even like these). You can bring your own. Many regions these days permit temporary rezzing if you’re a member of the region group – so always check this by clicking on the land information at the top of the viewer window and then selecting the ‘Options’ tab – and even the ones that don’t might allow you to do so if you contact the owner and ask them (nicely). So long as you (the director) own the animation, it’s then just a question of rezzing a poseball containing it (most animations come with poseball versions; if they don’t, you can buy an empty poseball to load it into here for just L$1) and then moving the poseball over to the furniture. Once your actor sits on the poseball it will turn invisible, but you can make it visible to you again by activating transparency mode (Ctrl- Alt-T) so that you can make fine adjustments to its positioning.

But what if it’s the actor who owns the animation rather than you? Obviously one way around this is to simply buy the animation yourself. Aside from making things unnecessarily expensive, however, animations do disappear sometimes from the Marketplace or from inworld vendors, so it’s not always possible to do this. Luckily, there’s a trick you can use to get around this problem: rez a pine cube on the ground, ask your actor to sit on it, and when they’ve done so ask them to activate their animation (ie, the non-poseball version; double click on the animation in inventory and then click on ‘Play inworld’). This should then activate the animation, and you can then position their avatar by moving the pine cube. Once you have it exactly where you want it, edit the cube to make it transparent and then film your shot. By the way, ask your actor to disable their animation over-rider before activating the animation, since sometimes this can interfere.

Can’t find the right static pose for what you have in mind? Lately I’ve been using with increasing frequency AnyPose by Phate Shepherd Creations. This inworld tool allows you to manipulate each of an avatar’s joints, and recently this has included Bento hands (though not, at the time of writing, Bento faces). I currently use the basic version (available here at a cost of L$1,500, and you receive upgrades for free), which serves all my needs for creating static poses. I am tempted by the ‘BVH’ version though (L$4,500) since I understand it includes an animation creator.

Using another person as a behind-the-scenes technician

Very occasionally, there will be shots you want to make which you just can’t do by yourself, even using an alt. For example, the clip shown below and discussed in part four, in which I move Caitlin Tobias’s head with an alt, would not have been something I could have done by myself if I’d also needed the camera filming to move in some way.

For that shot, I got my alt to rez a pine cube and asked Cait to focus on it (with her AO turned off). I then switched back to my main account, composed the shot how I wanted it and hit record, then switched back to the alt account and moved the pine cube (which moved Cait’s head) whilst the recording was going on. I then switched back to my main account and stopped recording. If I had wanted the camera to move during that shot (for example, to zoom in on Cait’s head) then this wouldn’t have been possible. I can’t be in two accounts at the same time (and even if I could, I couldn’t move the cube and simultaneously do a zoom).

In this situation, then, you are going to need another person to perform a task for you. This arose in the filming of STÖMOL just the once, in the clip below. Here we see the camera moving around Stomol, who is standing in front of a unicorn in a dream sequence: whilst the camera is behind him, the unicorn disappears.

The technical issue of vanishing the unicorn was easy. I rezzed it, moved it under the ground, then moved it back up into position. All that needed to happen for it to disappear was a Ctrl-Z key combination, which would undo the last edit and immediately return the unicorn to its last position – underground and out of sight.

But, simple though this was, I couldn’t do it at the same time that I was panning around my avatar. So I recruited Cait to do the Ctrl-Z (of course, I had to give her the ability to edit my objects first) and she performed the action whilst I was filming. On this occasion, because timing was so crucial, I did use voice rather than text communication – a single word: “Now!”

In the next part of this series, I’ll take a look at filming speech in SL, one of my biggest challenges when creating STÖMOL.

4 thoughts on “Huck’s guide to making machinima in Second Life part five: working with other people

  1. “I can’t be in two accounts at the same time”
    No? I did it, even with up to 5 accounts. Me and my 4 alts spread out across 3 computers in order to reduce workload. But that was on DSL 10 and veeeeery laggy. Arranging my family for a portrat took forever, and filming a machinima would’ve been impossible … but technically it can be done. ;)

    And operating two avies on two machines + driving a vehicle … forseeable end result. :o

    Liked by 2 people

      1. Well, the co-piloting younger “sister” didn’t have anything to do and I could concentrate fully on driving and crashing. :o

        Liked by 2 people

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s