Auto-diving Marines
Posted July 25th 2026, expanded later
Associated movies
Back in October 2024 I was working on BCM520, a movie concerning a Warthog which was very oddly suspended high off the ground by having its nose seemingly impaled by a tree trunk. This related to the interesting topic you can read about in my article Tree in a Warthog.
The movie included a few clips showing that the two Marines I had with me didn't like the state of the hog at all. I was able to hop into the driving seat (not that I was able to go anywhere) but they wouldn't board. Moreover, when I used a Banshee to make them dive onto the snow underneath it, they then endeavoured to dive clear of it. However, sometimes it took them multiple dives, because they seemingly weren't using much intelligence about it.
An idea forms
That 'auto-diving' as I came to call it (automatic diving; diving without being manually prompted) was already quite amusing, but it gave me an idea. By hemming them in with Ghosts to some degree, maybe I could get them diving for longer, in their bids to escape. And if they were actually fully enclosed, or to put it another way, penned in, perhaps they'd dive indefinitely! To my mind that had some very appealing comedy potential, and would make a rather striking sight. Plus, I suspected that nothing like it had ever been done before.
I made a start by bringing two Ghosts along to block them on one side of the hog (see first pic), and was soon able to get them diving for longer than before. That is to say, they were tending to take longer to escape from proximity to the hog. I recorded footage of that, but realised that the topic really merited a devoted movie. So instead of trying to cover things in BCM520, I put the topic aside for special attention later.
That time finally came in July 2026, and here I point you to my first movie on it, BCM660. After a recap on the oddly suspended hog, it shows some of the aforementioned footage I'd recorded using two Ghosts, but then it continues into new efforts, in which I'd freshly brought along a whole bunch of Ghosts, with the vision of full enclosure in mind. I didn't quite get to that, but showed some other amusing results where, in conjunction with Ghosts, I was doing some blocking with my own body.
Full enclosure
The goal of full enclosure was focused on in the sequel BCM661. There I showed a succession of failed attempts, then finally a success in which the Marines were fully enclosed by five Ghosts, and diving together for minute after minute (see second pic). A definite success, on which to end the movie. When it looked like they were stuck in a never ending rut on one side of the hog, I eventually broke it up. But by then they'd been diving together for about nine minutes.
Perched Marines for easy insertion
Some further ideas were featured in BCM662, where I used two new ways of initiating auto-diving. In particular, having a full enclosure already in place, and Marines perched on barrier Ghosts (see third pic), made initiation very easy. The Marines merely needed to be shot, to send them into the enclosure. There was no need to fill a gap, something which had often resulted in failure. It makes a nice set-up, as long as you're willing to invest a bit of time fiddling around getting your Marines perched.
Synchronized diving (and grunting)
For BCM663 I got into a new aspect: the possibility of getting Marines rhythmically synchronized in their diving and associated grunting. It turned out to be quite interesting.
Using the 3-Marine set-up I'd ended with in the preceding movie, I initially focused on the two AR guys I had. I found that if you first get one guy diving, then shoot the other one off his perch just as the first guy is grunting, that can potentially give synchronization from the outset. But even if not, shooting one guy to interrupt his rhythm can subsequently get him in sync. At first I was just doing the shooting randomly, and it tended to take perseverance. But possibly the thing to do is to try and shoot him at the time the guy you're trying to match gets to near vertical. I haven't checked into this fully however.
When it came to the other Marine, namely an Aussie with a plasma rifle, I was able to get him visually in sync too, i.e. in terms of diving animation, but his grunting actually doesn't work on the same rhythm. Unlike the AR guys, he grunted at a slightly slower rate. Perhaps due to the length of the audio snippet? As such, even if he sounded in sync with them for a few seconds, that effect would soon be lost - until it temporarily came back again, some while later.
When you've got some guys in visual/animation sync, something you can do is watch for sequences in which they also match in terms of their dive directions. Looks nice! See how many times in a row they match. With one situation seen in BCM663, featuring my two AR guys constrained to rolling on the spot, I actually got a 10-dive match (coming just after a 5-dive match). That was the longest I counted.
Some technical talk
If you want a few specifics, consider the case where a Marine is are repeatedly doing a roll. In my PAL version of the game, which operates at 25 fps, he takes 38 frames per roll, which amounts to 1.4 seconds. But whereas my AR guys also took 38 frames per grunt, matching their rolling exactly, the Aussie took 41 frames per grunt (1.64 seconds).
When Marines have enough room, they sometimes do a bellyflop dive rather than a rolling dive. A bellyflop seems to last exactly twice as long as a roll. As such, doing bellyflops doesn't get things out of sync. The basic rhythm is still there; it's just that the bellyflop uses up two beats instead of one.
At the current time, I've only ever had three Marines for testing things. At a later date maybe I'll again get a suspended hog, but in a situation where I have more Marines available. So there may be further findings to add, giving a fuller picture of things.