The 1.31 Open Beta has been steadily rocking along. Few little wobbles at the start there, and I’m not sure if Ramp has fixed the gargoyles yet (infantry “statues” that appear when a trooper dies out of your line of sight but inside your vis range; the trooper animation system surrenders the body to the ragdoll system, which promptly disowns it because he’s not in LOS; leaving a death-throws statue).
Started out rough: In short, the debug build of the servers I was building couldn’t handle 20 people simultaneously, the world-update loop was frantically dialing back its capacity and then treating everyone to the “wow, long time no hear” treatment.
I’ll explain the technical issues in a little more detail after the bump, but this post is primarily going to be about the gameplay feedback we’ve gotten…
Specifically, regarding the feel of the new capture system; picking up a discussion I fired up a little while back.
Lot of factors to the update issues… I’ve humanized some of the inner workings of the update loop. With optimization, it’s actually slightly more efficient. Without optimization? Crash and burn.
We’d elevated some of the hard-coded capacity settings, increasing the amount of memory some of the critical loops were working across, causing lots and lots of CPU stalls. Unfortunately, stalls look like idleness.
But fortunately, factor #1 helped me go through and find lots of redundant code. I used references to eliminate a whole bunch of unnecessary validation-branch operations (which it turned out were all generating expensive branch misses), some “future feature” junk and various other odds and ends.
And with the replay tool (it’s not a feature, it’s not remotely near to being a feature at this point) I’ve been able to get fairly good confidence that we’re back on track there.
But back to the main topic and opinions from beta… Specifically, the matter of Capture.
We’re not done with the capture system yet (please note: I’m talking about the back end, not the client presentation of it, which I think needs a lot of work): the current beta implementation is trivially simple. There are some minor tweaks we definitely want to make, specifically to ensure that the system best rewards “fair” play styles – with somewhere between 5-8 people capping a depot, perhaps more for a military facility.
But that means that Mr Solo Cap should expect a long, slow capture (perhaps not in a softcap town?) and if the BKB all pile into one single capture depot, it will still take a minute to capture: if you have more than X people in a depot, your “reward” should be that you’re harder to kill off, not that you cap the depot in 1/10th of a second.
Someone, somewhere, suggested that capture be replaced with attrition: create an mechanism that gradually deletes units from the enemy spawn pool unless they outnumber you. So – choosing not to come fight lets you win faster. Pulling the brigade(s) out, lets you win nearly immediately.
Iffy suggestion. But one thing I’d want to add: You can’t move a brigade into a town without first physically outnumbering the enemy presence…
The ruleset we’re using for capture right now is fairly simple. Don’t get excited, that is prototype code.
Interestingly: When I mentioned that I was going to add capture announcements to chat on the beta server, everyone grumbled. I mean, everyone. I explained that I only meant “to your team”. But people still grumbled.
Given that I’m a douchebag, I went ahead and did it anyway.
And apart from a few bright sparks who were there at the time spotting that I’d done it anyway, I’ve not seen a single complaint yet.
Right now it’s sending an “LSTR_GENERIC”, which means it’s not localized and the client can’t do any smart processing of it. But if the producers don’t beat me to a pulp for feature creeping it, and decide they want it included in the actual code, I’m hoping I can convince Ramp to make it flash on the UI or something.
As I said in the previous post, this is all about encouraging team work.
Capture is pretty slow in beta. It takes 500s for a single player to cap a depot. However: If he dies, or if he fires his gun, or if he walks out of the building, that progress isn’t wiped out.
If you’re talking about a stray depot 2km away from anything else – yeah, it makes little difference. But in most of the battles we’ve had in beta, it has really changed the play of depot battles.
You have to apply a certain amount of skepticism to anything that happens in beta tests, but the simple fact that one dude has to spend nearly 10 minutes in a depot before it gets captured means that he’s more likely to get company. And then once there are two guys, it starts to cap twice as fast. And the new arrival doesn’t have to play ei bait. Both of you can work together, shoot out of the windows, etc. The stuff that normally only happens when a couple of players decide to defend a hard earned depot. Now that hard work starts during the capture too :)
Beta players are generally trying to have fun. Whereas as soon as the new capture system meets the live game, people will be trying to win. The current ruleset would fall flat on its backside as people took depots in moments by zerging.
I demonstrated a formula for the capture-points-per-player curve to Dana and he was not keen on it because as the number of players went up, it actually started to come back down again (i.e. get slower).
Trouble there is it might get griefed. “Damn it guys, leave the depot, you’re slowing down cap”. But we’re talking about short times here. And my thinking was: If you have 30 guys in a depot, you’re not so much pwning it as cowering in it. And if Bob steps into the depot and slows down capture, well (a) one of you could step out or (b) you could wait the extra 5 seconds: with 30 of you in the depot, it’s not a big deal.
We’ve had a few people complain about the dreary days. In 1.31 it is either bright and sunny for a day, or it’s overcast and dreary with a chance of rain. The rain doesn’t last more than a few minutes at a time. But people keep asking to see it, so we keep switching to the dreary days. And then people complain that it’s overcast all the time :)
The rag doll stuff and the new building models are being pretty well received. The new clutter and eye candy make such a huge difference. Some of the towns are unrecognizable. The new bunkers – all of them – are a real different combat experience. My thoughts of making military facilities take longer to cap dissipated after I’d tried attacking a couple of bunkers :)
I’m very much looking forward to playing 1.31 when it goes live :)