Elder Game: Grab Bag Update: Web Player, Skill Swapping, Languages

This Blog Post was part of the Elder Game blog. It was posted by Citan on April 13, 2013.

Previous Post: Elder Game: 4:30am Update

Next Post: Elder Game: The flying animal fiasco

Additional Blog entries can be found on the Developers page or in Category:Game Blogs

A quick update on lots of topics:

Unity Web Player Performance Sucks

It comes and goes, the web player. It depends on so many things: the browser, the OS, and the version of Unity. Right now, it’s pretty godawful for lots of people. If you’re on a Mac, it drops key presses, which is infuriating if you’re trying to chat. And on a PC it can get bottlenecked on mouse events, which is slightly less torturous, but also annoying.

It also loses frame-rate on all platforms, which causes the game to downgrade its graphics to Butt Ugly Mode. It runs significantly better in the stand-alone version.

So this is a big issue for me. I wanted to always support a web version, but if it’s going to suck, I don’t want people seeing the game like that. I can optimize performance in the web player, but it takes a LOT of time, and is it worth it?

But that’s a less-pressing issue than what I should do right now. I don’t have time right now to try to optimize the web player on every OS/browser combination.

So maybe I should switch to standalone game clients for now. The problem? There’s no patcher. That means people will have to manually download a new zip file every other day they play.

Fortunately the zip file is pretty minimal, because the game will still stream most of the assets. So we’re talking about a 100mb zip. Most people won’t have too much trouble with that. It’s just the annoyance factor of manually doing it.

(There are patcher solutions, but they’ll require a lot of prep work. I think it would be premature to set up a patcher right now.)

NaCl Builds Don’t Stream Scenes

The other thing I can do is provide a NaCl build of the game. That is Google’s Native Client, and it runs well in Chrome. But I think you have to manually turn on NaCl support, and anyway not everyone uses Chrome!

But I’d at least provide it as an option, except that it doesn’t stream-load like the web-player does. What I mean is that in the web-player version, you’re busy making your character while the world downloads in the background. With the NaCl version, you have to wait for the 100mb download to finish before you can begin playing at all. That’s not the best first impression. Actually, it’s kind of worse than making you download a zip file, because at least that’s a better-understood process. If you’re looking at a loading bar for 100mb of download, you’re likely to start wondering if it’s broken, or if you’re insane or what.

So anyway that’s why I haven’t used a NaCl build so far. I assume it would perform better once it got loaded, but I haven’t actually tested it out much.

(I can eventually fix this by streaming scenes a totally different way, as Asset Bundles, but I’d like to wait on that for a little while. It’d be even more variables added to the mix.)

How Bad Is The Framerate?

So I think the framerate issue is making the game un-fun for a lot of people. But I don’t really know for sure. I’m going to wire up the client so it tells the server its framerate every few seconds, so I can see how bad people have it. That should tell me if framerate is causing new players to abandon ship or not.

I’ve only had a couple hundred people try the game, and of those number, the gameplay times are pretty reasonable. In fact, if these were random people off the street, I’d say the game is retaining players at a really great ratio!

However, most of these people have been fans of the game, so I actually expected even higher average playtimes. There’s lots of possible causes of that, though…

More Direction Tools

After talking with some players, I’ve decided that there really needs to be some better hints to the content. I mean, it’s an exploration game, and I want you to just go poking around and find cool stuff. But a lot of people have been aggressively trained not to do this by other games in the MMO genre.

So I’m implementing the “Stuff To Do List” real quick. Basically it’s just a check-list of stuff to find in the first couple of zones. It’s not a complete list, but enough to hopefully teach players that there’s tons more stuff out there.

The danger is that once they complete that list, they’ll assume they’ve done everything and quit. So do I keep making more checklists? I guess that’s reasonable. The later ones will be kind of vague, just so you know there’s stuff out there.

But I still don’t want to put everything in a list… so I need more advanced ways to convey that there’s lots of stuff out there for you to find.

More Communication Tools

The average player never speaks, and most players don’t even open the Chat window for very long — it’s clunky and large, and if there’s no chatting, what’s the point?

This makes the game feel even more barren than it is, so I’m working on that a little. First I’m doing the obvious thing and adding a “ghost” mode for the chat window. So you’ll see player chat appear in a translucent box briefly, even if you don’t have the full chat window up.

Fiddling With Combat

Right now I have a little experimental GUI addition: there are two tiny dropdown boxes, one next to each skill bar. These let you change to other combat skills really rapidly.

Earlier in the blog, I said that you wouldn’t be able to change skills except at “Designated Changing Stations”. The reason is that I didn’t want you to feel like you had to constantly swap your skills… but practically speaking, it doesn’t seem to be a big deal, at least right now.

So I figured I’d see what happens if you go to the other extreme and make it super trivial to change skills. This feels to me like it goes too far… seems to tell new players “you’re going to need to change skills all the time.” But early on, that might not be such a bad thing. Most players right now are sticking with the two default skills even after they get new ones. That’s not necessarily the optimal way to play. Or at least not the most fun.

Eh, I dunno what’s best. Thus the experiment! I’ll see if I can gather some feedback about this new GUI approach.

Racial Languages

And now back to unimportant minutia!

I’ve been mulling “racial languages”. This stems from the idea that a Werewolf should be able to understand another Werewolf, even if they have low Beast Speech skill and just sound like growls to other players.

First, I’ve sped up how quickly you learn Beast Speech, so this isn’t even an issue for as long. (It was never supposed to take too long.) But eventually I’d like to get it so that other Werewolves always understand each other (and other Cows understand each other, etc.) When I do that, I’ll also have the tech for racial languages.

At this point, racial languages don’t make sense to add, because they divide players up rather than helping them communicate. But nevertheless, elven players have expressed interest in being able to speak Elvish. Maybe if I can think of a clever way to use the idea… any thoughts?

(But if Elves get Elvish and Rakshasa get Rak, what do Humans get? It’s assumed the default language is the Human language. So do Human players just not get a special language? Well, I could always give them an alternate human language. Say, French! Heh.)

Anyway, things are still chugging along. I’m hoping to start drumming up some more playtesters soon… in fact, I was going to do that today, but decided I needed to figure out the web player issues first.

A content update is in the works, with some new armor appearances, quests, dungeon areas, and more, so stay tuned!