Ladders in 1.0

Neutrino,

Don't despair, and don't let them goad you into venting it out on the forums. I'm pretty sure Brad is paid to tell them the same thing but sound nicer. That said, I completely understand the frustration.

Far more of the community is supportive of Uber and we trust you to do it right, but some of us are more cynical than others, and there are a plenty of basement programmers who know they could write something up in 30 minutes. Very few of them have experience working in an environment where you're accountable to thousands of users and every 30 minutes of coding is followed by 3 weeks of testing and documentation.

It's unfortunate the least happy users will always be the most out spoken. There's a whole lot of us that don't have much to say about Uber's implementation of a ladder system because we are comfortable knowing you'll probably do it right.

You've made an AMAZING RTS, and the engine you've built is going to stay relevant for a LONG time. Players don't have to agree with every balance decision Uber makes. PA offers an RTS environment that no other game can substitute. I can mod any balance I want into PA, but good luck modding spherical maps in to Starcraft.

Supcom wasn't economically viable because of the matchmaker?

I get your point but, no, whatever you might say, a matchmaker doesn't take much time to do.
Not 5 mins, but not 3 weeks like you've said earlier when talking of delaying the release.

I disagree. Even if you don't convince us, and nothing changes, it's still good to know that you tried. It might seem like a waste, but it will help in the long run. Many of us act like fools because we care, but in the end, we believe you'll do the right thing. This thread does good, even if it doesn't seem like it. Just try to bear with our silly brains!

Oh indeed.

But that's not the point of the exercise, is it?

The point of you coming on here and having a chat with many of us is that it takes the wind out of the sails of all those "Uber never talks with us" naysayers. I'm hoping that you have noticed that the pressure cooker on the forums has just dropped a few dozen bars since you have been having some full and frank discussion with people around here. If you don't do this, then the most common time to see a spike in dev/forum interaction is when people kick up a shitstorm.

And I really hate the idea of a forum where the best way to get attention is to say unpleasant things.

So, even if this doesn't have a direct impact, I still count the increased interaction I've seen over the past week to be a success.

We just have to get to a point where ya'll are comfortable with getting the information we can impart, to be frank, and using me as a conduit. Jon shouldn't have to wade into fights for ya'll to believe you're being heard. (You always have been and will -- this is a project that revolves us and you.)

I fail to see why the fact that Fa costs too much is an argument against a matchmaker, unless it's part of the reason why it cost too much.

That's a positive thing to talk about. We should talk about that instead of "No feature X at point Y". What are your ideas of such a system? All I know is you're a genius for coming up with GPGnet. It's the best I know that can be done.
I think to get more positive discussions we need more positive topics. We are in a topic were crazy people like myself are trying to defend the features they think is of basic importance to the goal you have. You're obviously thinking that the feature is either not helpful or needs to be much more complicated than we think it is. You're aim however is quite clearly what we want. So instead of just making a notification to us "yeah no feature X at release", it would probably lead to better discussions if we could talk about how you want to reach the goal of appealing to that much larger crowd. Cause so far all I can see is "probably no chat (at all?) and no ladder at release", but those are the best features I can think of to actually gather any sort of community at all. What are your ideas to get better than that?

I am not thinking of you as an idiot and the fact that you're the guy who came up with GPGnet alone makes you an awesome person. You probably know what you are talking about when you say "We cant make feature X now, takes too long". But from the view of a person standing outside and only having his own experience a ladder is not a feature that takes weeks to make. The feature I have in mind would take days for a single person. So we probably either have completely different features in mind or there is something complex about the thing I am missing. What is your estimate on the creation of the first iteration of ladders? What are the big issues with it? What costs that much time? I may just be stupid, please explain

I can easily understand that in a big multi-year project asking a few days very close to end day can already be way toomuch.

A single person modder doesn't have the overhead of a development team. Keep in mind, this will need -

• Someone to (technically) design the feature
• Someone to design the UI
• Someone to implement the feature on the client
• Someone to implement the feature on the backend
• Someone to test the feature
• Someone to manage all these tasks

They aren't likely to each be a unique person, but they aren't all the same person, either. The management of this is complex, and itself takes time.

Not to mention standards to uphold. People forgive modders for some bugs. People don't forgive a company for some bugs. Besides, designing the actual formula to get proper matches is likely going to take quite some work, not to mention figuring out how to track skill to begin with. Standard solutions exist for shooters. Not for something like PA. PA from top to bottom has pretty much required custom solutions to just about everything.

Not to bash FAF, but with such a playerbase it's not that complex to make a matchmaker. With PA, catering to a sufficiently broad audience means that it doesn't just have to work, it has to work for everyone, all the time.

Out of curiosity - @neutrino how come you didn't go with the same idea as GPGNet? ie an external client? Just was wondering because I loved that system, albeit I don't like having extra clients like steam and origin etc etc etc.

probably cost of partnership and Uber wanting sole domain over PA

A client like GPGnet has nothing to do with "partnership" or "sole domain over PA".
"External" only means external program. Just like the Patcher is an external program as well.

[sarcasm]Yeah nobody has every tried to create a ranking system for an RTS game.[/sarscam]
Seriously?!

The audience of PA is less than FAF (it must be as the "small online community" of PA is an argument against the viability of a matchmaker for PA).

- Neutrino said that the PA community is too small to make a matchmaker viable.
- FAF matchmaker is viable.
-> FAF must have a bigger audience than PA?

Add to that:
- FAF matchmaker works for everyone, all the time, and took about 2 weeks to develop for a single person working in his free time.
-> Why is it so difficult for a bigger team with bigger resource?

Without trolling, some facts:

- FAF use the exact same system than Xbox Live matchmaker (the same for any Xbox game that ever existed). If you can think of a broader system, please tell me.
- FAF use a similar system than Starcraft 2. Tell me a broader RTS audience if you can.

The original reasons we had an external gpgnet client were:
- able to use real dev tools for UI / communication instead of being stuck in the game UI paradigm that was cost ineffective for building UI. Building using the game UI was in flux and generally painful and expensive.
- be able to test out the concept separately from SupCom to reduce risk as we weren't 100% confident in the tech behind GPGnet (we started with Chess)
- ability to plug in other 3rd party client and have them boot up the game and not have the game itself care so that it could be driven from anything for longevity (e.g. FAF)

At the time we made this call it was actually considered a risky move to do it this way. There was a lot of concern about how the transition between the client and the game would be handled etc.

In PA we skirted most of these requirements by using a web front end. In fact it really is pretty similar except we just embedded the game inside of the external tool and made that thing moddable. Being able to plug web tech directly into the game seems like an even better solution than the external client. This way you can truly communicate with third party sites and really integrate them into the game.

You guys have already seen some of the results from the PA architecture.

Worst comes to worst, it can be made in a mod. Some mods for Starcraft 2 actually have their own ladder ranking on websites. so it is possible.

PAstats already exists. (http://pastats.com/) And anyway, as they said there will be an official ladder/matchmaker at some point, the question was if there were any good reasons to delay the game in order to implement one before "1.0", or have a "competitive features" patch after launch.

I guess the ladder could be like AoE 2 HD, we start with 1650 points, and if you win, you gain points, etc. The points is lower if the diference between the players are huge, and vice-versa. That can make it without Match-Making (or a bridge to it). If the base players of PA continuing to grow, then we can switch to MM (or not, if the players dont like the change).

Now, about what Uber is saying that the player base is more a single-player focus. This is real in the launch time, and, I dont know, 6 months later.
Besides that, it will vanish.

Games have longer lifes if the competitive scene is alive. See for example Warcraft 3, it is still living, due to DotA, a competitive mod that became a game later on. TA, back in 2010 was still alive (I abandoned in this period) because of the multiplayer, player vs player! Another ex: SC, TF2, Myth.

AoE 2, made a HD version because the multiplayer was still a thing. No one will buy that to play the campaign.
Not one game live long with only players VS AI.

Thus, competitive feature, such as Ladder or MM must start at 1.0 launch.

Necro of the month, Runner-up Award for August.

Also, not this thread again

I believe that honor goes to this.