Phew!
We are making good progress!
There have been a lot of challenges over the last two months. We've battled with a diverse array of crashes, both on the server and in the client. We've spent days investigating and solving unusually long compile times.
We've had to expand our back-end tools to track down some of the more elusive issues, and come up with tricks to get abilities and features to function without spending more time building whole systems for them.
It has been hard work, but it's paying off. The game's never been in a better place, and even though we're light on content I don't worry about that. The tools are good. Iteration time is pretty fast. Fleshing out gameplay will follow.
During all this fixing, we've also been iterating the game mechanics to explore what fun looks like. Here's one example that I'd like to share...
Setting Up Camps
During the various tests we've been running, I noticed that the jungle was not a fun experience.
Maybe that shouldn't have surprised me. I placed camps to test their features: respawning, aggro, leashing, and rewarding players who cleared them. I wasn't yet thinking about "designing a fun experience" in the jungle, and it showed.
So what went wrong? My approach to making camps was:
- Pick a unit type for each camp
- Place three or four copies of that unit in a clearing, spaced nice and evenly.
I had to reflect for a while on why this doesn't work. It's because there is no technique for players to apply.
It doesn't matter which unit you hit first, or the order in which you attack the creeps if they're all the same. It doesn't matter from which direction the camp gets approached if the units are evenly spaced.
These symmetric camps were unengaging because they didn't let players express themselves.
Natural Variation
The solution, of course, is to break the symmetry.
This lesson dates back to Warcraft III (or prior), where the design team must have realised the same thing. Having a variety of unit types in each creep camp gives the player some (simple) choices to make when approaching it. Alternatively, having a single creep to simplify the scenario can also work.
Most lane-pushing games with jungles already do this, and now RoW does as well.
This might seem like a really simple game design problem and solution, but it's always good to question the way things are done. If we understand this solution, we can understand alternatives!
One alternative is having creep camps with area attacks that can be dodged. Indeed, it might be better to not vary the unit types in this case as the player is already being offered a challenge.
In RoW, we love knockback and environmental damage, so here's a variation we might try during alpha:
Suppose we placed a couple of powder kegs near a camp, and let players lure creeps into them where they could be blown up to deal area damage.
The catch is: explosions are noisy! The enemy team will hear kegs explode and be alerted to the location.
I started coding it a little after having that idea, but figured I should get the blog updated before diving into experimental features. As always, if you have questions or want to chat our discord is the place!
Softmints
Comments
CMGod
This has nothing to do with the jungle but do you think that by stripping away gold and not replacing it with anything you might not be rewarding players who are doing well enough? For instance lets say player A is 5/0 against player B in mid lane but his only lead is exp and he cannot effectively utilize what should be a large lead due to the game not rewarding him enough? I've noticed this happening in current iterations of DOTA now that kills reward very little in both exp/gold.
Softmints
In reply to This has nothing to do with… by CMGod
CMGod,
In most lane-pushing games, players get experience and gold for doing the same things. There aren't many single-resource revenue streams, and we don't see players strategically choosing to prioritise gold over experience for example.
Since having a second resource isn't contributing much to gameplay, RoW uses a single resource 'experience', which causes heroes to level up and can be allocated to "buy items". Essentially, it's one number which unites the two resources.
This might be split up again later if having two resources would be a useful tool for balancing, or we find that having two resources makes it easier to communicate how the game works to the player.
From there, it's more a matter of numbers to determine how much of a lead players should have. Hope that helps!
Softmints
CMGod
Sorry I guess my question wouldn't really be about gold on further consideration but more about how much a player will be rewarded for kills vs creeps vs being afk while other people get kills.
I wouldn't ask for hard numbers while the game isn't even in closed alpha yet but rather your philosophy on this subject. I was just considering that in a game where exp is the only resource and exp is rewarded to players even if they aren't apart of the kill (not sure if this is actually how it will work please correct me if I am wrong). Will this NOT lead to players who are performing extremely well being maybe under rewarded? Is this something you've spent time thinking about?
I've read through a decent number of these blogs now and I've seen a lot of mentions about teamwork and your interest in team play but while lpgs are TEAM games I'm not sure if individual players really love feeling as if their input is not not enough to decide the outcome of a game. Since you are only 1 of 5 players on a team maybe if the developer prioritizes teamplay you end up being only able to take your team 20% of the way to victory. I think instead dev who sees the game the players as 5 seperate people coming into a pug each with their own philosophy on how to win. Might bring about gameplay that is overall more enjoyable for each unique player. Not sure where that rant leads, I guess just to me asking for any sort of comment you might have on which side you might fall on or some elaborate reponse.
Sorry if this feels sort of jumbled I've played DOTA for the last 8 years and am a 7k mmr player (recently quit because I hate the last few patches SO much, no I don't hate jungle items). So I worry about the game design issues I see in modern dota occurring in another game (TOO team play oriented, lane shoving > all else, LOTS of etcs...)
Softmints
In reply to Sorry I guess my question… by CMGod
CMGod,
I think I understand what you're asking, so I'll do my best to share some ideas in that space.
There are three sets of skills a lane-pushing game player is being tested on:
If I understand your question correctly, it is something like "are personal skills going to matter and be rewarded enough?".
The answer is yes, but it needs context.
A player that performs well on their personal skills can bring their team much more than 20% of the way to victory. They will use their abilities better, earn resources faster, and the game wants them to stand out and have opportunities to over-perform.
The same is true of interpersonal skills. A player who over-performs as a morale-boost, who watches their team's backs, or coordinates build choices and 'advanced teamwork' item usage, is also going to bring their team more than 20% of the way to victory.
So while personal skills will matter towards victory, they are being held in balance by the effectiveness of teamwork. On that point:
You also mentioned this:
Sure. We want to see players consistently go from being five strangers, to five people who are participating in a shared vision. The vision can't be one dictated by the game; it has to be something that belongs to one of the players, and that they choose to share.
RoW's "job" is to provide the environment and tools and incentives so players can champion their own vision of how to play, and get buy-in from their team. It's also the game's job to handle competing visions in a constructive way.
If you don't feel satisfied with that answer, or we don't get the implementation right first time: keep asking us! Thanks for your engagement.
Softmints
(P.S. The Leadership design article also touches on some of this topic.)
CMGod
In reply to CMGod, I think I understand… by Softmints
Thanks. Your leadership article answers all my questions on this subject. Although I have GREAT skepticism that you could make interacting with teammates in a MOBA not miserable since I, and many others, believe this to be inherently miserable.
So a question I think is important. Will your matchmaking system have solo queue only? I truely believe that when your teammates are random it is IMPOSSIBLE to enjoy a vast majority of your interactions with them, assuming you are playing to win. So having a game that focuses on solo queue rather than 5s or 3s/2s would make your goals incredibly difficult in my opinion.
Softmints
In reply to Sorry ignore the post above… by CMGod
Glad I could help, and that the article proved useful.
I think your scepticism is warranted — making teams feel like teams is a hard problem and we will probably struggle with it forever. There will be bad patches; people will tell us so. Hopefully they will also be vocal about the good ones.
If you like, I would be happy to discuss further what leads to those miserable interactions and how you feel this might be addressed. Write me on discord, or leave a comment with your email address and I can follow up. (Your address will not be published.)
On matchmaking, I would say these things:
My guess is that our early editions of matchmaking would adopt a flexible model with flex-queue for most skill brackets, maybe leading into solo-queue only for the highest skill brackets if the numbers support it.
Softmints
Add new comment