Citan

Hi guys! The next update is coming in just a few days. I've been focusing pretty heavily on the update, but I had some time today, so I wanted to give you a status update.

Remember, this is a dev-blog. It's not an official announcement about future changes, it's just me talking about what I'm doing and thinking. I like to share my current plans, but these plans change often. Speaking of which...


About That Treasure Rebalancing
The last dev blog talked about some major revisions to treasure balance. And then the next two updates came out, and there were no revisions to treasure balance. The revisions aren't going to be in the next update either. What's going on?

Well, after I implemented some of the changes, I wanted to sanity-check them... and realized my gameplay data metrics were out of date. The game server normally collects anonymized combat metrics that let me answer questions like "How long does it take a player using skills X and Y at level Z to kill a level-appropriate monster?" or "If a player is using skill X in area A, how often do they die?" These insights are very helpful in making sure my changes aren't insane. But I don't have that data right now, because metrics-logging was turned off.

After we launched on Steam we had unexpected server stability problems, and the very first thing I did was to turn off all non-essential functions to rule them out as the cause. So metrics logging has been turned off for months, which means the data I have is pretty old and doesn't reflect recent changes to the game or the player-base.

I realized this problem late last month, but at that time we were STILL having occasional server problems, and it seemed dumb to turn the metrics back on while I was still hunting for problems. The server has been extremely stable after the last batch of server fixes (on the 8th), though, so I'm going to turn the logging back on in this week's update. Then it will take about a month to collect new data.

I don't always use metrics data when making changes, because it takes a very long time to analyze that data. For situations like the AoE nerfs (see below), it's pretty easy for me to test what will happen when I make the change: I can create some AoE characters and try them on my local server. But this upcoming rebalance will impact a LOT of skills and treasure effects, and I can't predict all the side-effects. There will inevitably be some weird issues stemming from this change, and that's fine, we'll fix them... but I guess I want more reassurance that this change is moving the game in the right direction. And I need more data for that!

So long story short: there's been some delays. In the meantime, I've focused on more client optimizations and new game features.


Client Optimizations
The next update doesn't have any of the large-scale Serbule optimizations that I talked about in the last dev blog, such as replacing the sky/weather system, but it does have a lot of more traditional optimizations: the game's texture-memory footprint is reduced by over 100mb in most areas, which will help users with integrated video cards (or any video card with less than 2gb of RAM). We've also done a bunch more LOD optimizations in Serbule and Serbule Hills, which will help framerate on mid-tier video cards.

The impact of these particular optimizations will depend entirely on your computer's hardware, so they might make a big difference for you, or they might not. We'll watch your feedback to see what you think. And there are many more optimizations in the works, of course!


Vendor-Stall Searching
The next update will add a way to search player-run-shops. There will be little Catalog Golems in the shop hall, and you can ask them to list all the shops that are selling a particular item.

I'm walking a fine line here, and I wanted to mention the reasoning behind it. The golems' results are intentionally pretty simple: they tell you who sells it, but don't tell you the price they're charging. Why not? For one thing, it can be very difficult to show a price because player shops can sell the same item for lots of different prices at different quantities. But the main reason is that I don't want to make shops compete on price alone. That would drive costs down extremely quickly, instead of the slower market adjustments that I want to see, and it's basically not any fun for player shop-keepers. And it would ruin my long-term plans to let shopkeepers compete in innovative ways.

For instance, in the future shopkeepers might be able to offer frequent-buyer discounts that reward repeat customers. Shopkeepers might also be able to set up raffles contingent on purchase: "Buy any _____ and automatically be entered for a chance to win _________!" The shop system was designed to have little features like this added in pretty easily, and they'll eventually let shopkeepers be really creative. But if catalog golems showed the cheapest price, most players would just go for the cheapest stall regardless of anything else that other shops had.

This is also why we don't have an auction house: it's super convenient for consumers, but it drains every drop of fun out of being a pretend fantasy merchant. For a game with this much crafting in it, I need to protect shopkeepers and make sure they're having fun, even if it means there's a little more hassle involved in buying something.

However, the current shopping process, which I guess can be summed up as "if you want to buy something right away, first you have to browse 100 player shops" is FAR too difficult and irritating. I'm okay with there being a little friction when shopping, but not anywhere near that much. So this vendor-stall-search feature is a step in that direction, and we'll continue to evolve the game's sales tools until we find a good balance of buyer convenience and shopkeeper control.

(If you're interested in this topic from a game-design point of view, I wanted to direct you to some of the old blog posts I wrote about Project: Gorgon's economic plans... but unfortunately our old pre-alpha dev blog is offline right now. So instead let me point interested readers at a great old blog-post by Raph Koster, of Ultima Online and Star Wars Galaxies design fame. Just to be clear, not everything he talks about is applicable to us! Star Wars Galaxies was a very different game from ours, with permanent item loss and a bigger focus on crafting than we have. But he explains some underlying design ideas that will help you understand where I'm coming from. His post is called "Do Auction Houses Suck?" )


AoE Nerfs
Some players felt blindsided by the changes to area-of-effect abilities in the May 25 update. Which is understandable since it was definitely an abrupt change: I decided that it shouldn't wait any longer, I needed to drastically reduce the power of AoEs. Not in a subtle "maybe we'll reduce this one a little bit" kind of way, either -- they were so overpowered that they were creating new farming playstyles that shouldn't exist at all, with players able to safely kill eight or ten monsters at once.

This isn't new -- for a long time a small number of players have been using "AoE Death Builds" to kill dozens of monsters at a time. And that was obviously a problem, but it hadn't really been at the top of my to-do list. To be frank, there's a lot of overpowered stuff in the game right now, and I intentionally try to avoid making knee-jerk adjustments just because somebody found a way to abuse something. Since the game's balance is still changing pretty dramatically, an overpowered ability may suddenly stop being overpowered a bit later. So I try to focus on the big picture.

I do try to tackle balance problems if they're getting out of hand, though, and I should probably have jumped on this one earlier. These AoE builds had started to become really popular, and more and more players were using them. Listening to players talk in global, it's clear that most of these people realized they were overpowered, and were rushing to exploit AoEs before they got nerfed. So I nerfed them.

Let's talk about AoE balance a little bit. Most AoEs are balanced around the idea that they hit 1.5 people on average. That means we expect that half the time they'd hit a single target, and the other half of the time they'd hit two targets. If they hit three or more targets, that's "free" damage, but of course there's a lot more risk when fighting so many monsters.

And that's the underlying balance problem: players were using AoEs to hit more targets than I originally intended, and with less risk of dying than I expected. I'm not just talking about the really crazy builds that could kill dozens of monsters at once. If you talked to any player using an "AoE build", it's clear they didn't plan on hitting a mere 1.5 monsters on average!

The obvious fix would be to nerf their damage, Power costs, and other effects, rebalancing AoE abilities around the idea that they'd be hitting more enemies on average. But I'm resistant to that idea because it turns AoEs into very specialized, very boring abilities. Instead, AoEs were changed so that if you hit more than three targets, the damage is lowered dramatically. If you hit six or seven monsters, the AoE does almost no damage. But if you take care to only hit three monsters at a time, the damage is unchanged, and it's still FAR more efficient than single-target attacks.

I prototyped some other ideas before using this one. One idea was to increase the reset timer based on the number of enemies hit, so if you hit six targets, the ability would be on cooldown for 3 extra seconds. In practice this doesn't work -- the AoEs still do too much burst damage, and if they don't manage to kill the enemy, you have no recourse except to flee because your abilities are on cooldown. I also experimented with increasing the total Power cost based on the number of targets you hit, with similarly crappy results. I eventually settled on the current solution.

And I like the current solution -- it's not perfect, but it's the best I've come up with so far. It means AoE builds are still extremely powerful in the right hands, but they're not the mindless death-buttons that they were before. You need to be a bit more situationally aware, and that takes a little bit of skill. That's nice!

I think the most serious problem with this solution is that having a bigger radius isn't as big a win anymore. There are a handful of abilities with 10 meter burst radii (or even bigger!), and these abilities are much harder to use now. They hit so large an area that you can't always predict how many monsters will get hit. I may end up just reducing their radius, but first I'm trying some experiments to see if I can find a better approach for these abilities.

Anyway, I'm letting the current solution stand for another update or two while I ponder the issue, gather feedback, and prototype some more ideas.

Oh, and one more thing: monsters currently use the same AoE code as players do. If you're in a large group and fighting AoE-spamming monsters, keep that in mind.


NPC Augment Installation
I think the game is a bit unfair at higher level. There's an intentional power jump around level 50, then another one at 60 and 70, where the game starts to expect you to have decent gear and know how to use your abilities. The problem is that the game doesn't give players a lot of feedback on what's going on. Players often report feeling "stuck", unsure of why they seem unable to tackle higher-level areas. The game isn't giving them enough information to figure out how to advance.

So I've been slowly working on making this power transition smoother. NOT easier... just smoother. I think it's fine for the game to demand that you "get good" at high level, but the game has to give you lots of context to understand what's wrong and how to evolve your character. And ideally, it should do that well before you reach the point where you're saying "Huh? I can't kill these monsters in the desert. Why not?!"

One transition hurdle is that by level 60, the game is balanced around you knowing Transmutation and having access to basic Augments. Most players discover Transmutation on their own eventually (although I'll be making it easier to find soon, because it's really important! I expect every player to know and use Transmutation). But Augmentation is made up of several difficult crafting skills that I really don't expect every player to learn. Creating Augments is a pain, but USING Augments should be easy: you should be able to go to a player crafter and buy the Augment you want, then plug it into your gear.

This update moves us closer to that idea by letting NPC vendors install Augments into any piece of equipment for a small fee. This service is available even at low levels: Joeh will happily install low-level Augments into your newbie gear if you want. I don't really anticipate that newbies will buy a lot of Augments (although who knows?), but I want low-level vendors to offer this service so players can discover that Augments exist and learn how they work.

This also means that players don't have to deal with that scheming elf Nightshade. Of course, he will still teach you how to install augments yourself if you have the cash, and I've made the augment-installation recipes cheaper by removing the extra recipe ingredients, like Fire Dust. But I expect most players will just use these NPCs, at least until they're very high-level.

The Augment skills have always been intended to be dedicated crafting skills: Augmenters create Augments, which can be sold to other players. So I'm hoping this update will jump-start the Augment economy a bit! We'll probably need more shopping tools to really make Augment-buying easy, but hopefully this will get things started.



The next update will arrive in a few days! In addition to the stuff I've talked about, it also has the usual bug-fixes and skill tweaks. It also lays the groundwork for the new Gorgon Shop. We'll be bringing the shop back online a few days after this update, so people who missed our Pre-Release Packs will be able to buy them. (But this time they're implemented as add-on packs, so they expect you to already have a Steam key. They just add features to your existing Steam account.)

I'll be back with more dev-blogginess soon!


(Discuss this blog post on the forum here!)