Feedback request: Matchmaking services.

[keith.lamothe]

But I don't care what you do, I am fully satisfied with the singleplayer at the moment 

This is an important sentiment to factor in, because there's a large player population that only ever plays singleplayer anything, and then there's a significant chunk of the multiplayer folks who are quite happy with direct connect as is already supported and wouldn't care much about matchmaking.

Matchmaking should be done, but any costs associated with a particular option should be evaluated in that light.

Of course, a really good matchmaking implementation *could* attract a large number of customers and/or get a lot more existing customers interested in mp.  So it's not a linear thing.

[x4000]

But I don't care what you do, I am fully satisfied with the singleplayer at the moment 

This is an important sentiment to factor in, because there's a large player population that only ever plays singleplayer anything, and then there's a significant chunk of the multiplayer folks who are quite happy with direct connect as is already supported and wouldn't care much about matchmaking.

Which is precisely why I never did anything before.

Matchmaking should be done, but any costs associated with a particular option should be evaluated in that light.

Of course, a really good matchmaking implementation *could* attract a large number of customers and/or get a lot more existing customers interested in mp.  So it's not a linear thing.

Yes, that is somewhat of a factor now, I think if the cost is less than $100/month it would probably pay for itself just with the few players who might join because of that feature being there.  And if not, it's not exactly a terrible gamble, the gamble is more the time cost of building this out in the first place, which is a lot more costly than $100/month.

On the other hand, for Tidalis, this is pretty much going to be a must.  Going along with that, we want to let players share custom puzzles (as with Boom Blox Bash Party or similar), and actually custom levels and adventures is something that Alden Ridge (and A Valley Without Wind) are also going to be needing to support.  And when AI War eventually gets a Scenario Editor, then we'd need that there, too.  So an infrastructure like this is something that Arcen is going to need in general, for all our games, eventually.  And if the cost of such a service is effectively amortized across those services, it could be really worthwhile.  Free is better, of course, as we're not made of money, but assuming that our games continue to be at least as popular as AI War, we should be okay with something like this.

[x4000]

I keep forgetting, there is that new Amazon RDS (http://aws.amazon.com/rds/), which could do well.  It's really hard to know what that would cost.  It's possible that even their SimpleDB thing would work for our purposes here, because it's not exactly going to be needing complex RDB functionality.  Though if there's not cost differential, then RDS would be the way to go.  But then there's still all the development time for using it.  The bright side is that it can work exactly like we want it to, and it can do more than matchmaking.  The downside is that we have to effectively reinvent the wheel for this, and it will cost us money on an ongoing basis, and it will definitely require Keith to work on this for both AI War and Tidalis.

[dumpsterKEEPER]

The only other matchmaking service that I currently use is Games for Windows Live (beyond what you listed in the first post). It's ok, and while not my favorite, has improved significantly since it's first inception. I have no idea how the developer side of it works however, so I don't know how feasible it would be.

Probably my top pick would be Impulse Reactor (in large part because it wouldn't require any additional client), but as you say, I don't believe the current version is capable of what you're needing to do.

[x4000]

The only other matchmaking service that I currently use is Games for Windows Live (beyond what you listed in the first post). It's ok, and while not my favorite, has improved significantly since it's first inception. I have no idea how the developer side of it works however, so I don't know how feasible it would be.

Oh yeah, I did look at that some, too.  Right now we are now GFWL certified, and there are a lot of hoops that we have to go through to get that working.  It could be worthwhile, but I don't know what their SDK is like, either.  I'll have to look at this more closely again to see.  I was a Microsoft Partner for years, and part of their ISV program, and they did a case study on my product at one point -- from all of that, I know how fast they move, and it's not exactly blazing.  Blah.

Probably my top pick would be Impulse Reactor (in large part because it wouldn't require any additional client), but as you say, I don't believe the current version is capable of what you're needing to do.

I talked to Stardock about it just today, and basically the word is that it will do everything we want, but just not until later in the year.  That would be acceptable to just wait in the case of AI War, but for Tidalis if it is going to really have multiplayer, it needs matchmaking, which is the instigator of all this.

[Collic]

I don't want to sound as though I'm shooting down every potential solution, but there is also a lot of animosity toward GFWL as well.

I know a number of people who try and steer clear of it. Partly because it's Microsoft, with all the hangups and mistrust that fosters in a lot of people, and partly down to the GFWL UI, particularly among the hardcore pc gaming community. Some of that is almost certainly down to the gui and feel of the thing, as well as the appallingly bad start the service had. There are also a lot of negative DRM connotations (though I know those aren't elements of the service you would be using).

As has been noted, I also  doubt very much they're 'Indie friendly' so to speak 

Disclaimer: I should note that I'm also not a fan.

[x4000]

Yeah, I really don't think we have time to jump through all the GFWL hoops.

[Black]

Your service requirements seem to have two components:

1) Getting people together to play games

2) Sharing content

It seems to me that if you only had to focus on the first requirement, the development cost wouldn't be nearly as high. Perhaps you can delay the second requirement by exploring alternative methods of content sharing.

For example, the forums are a great surrogate for content sharing if the demand is small enough. A forum where people can post their content could hold up temporarily to ease the burden of development.

Another example: A friend of mine develops small games for the iPhone. In addition to setting up a small server that allows users to share their high scores, favorite games, and other stuff through the app, he also wrote a facebook app so that people could use facebook to share that information as well. Certainly his needs are far less than yours, but the idea merits examination. Are there services out there already available that you could integrate into at little cost to help meet your goals in the short term without investing so much so quickly for the long term?

[x4000]

Your service requirements seem to have two components:

1) Getting people together to play games

2) Sharing content

It seems to me that if you only had to focus on the first requirement, the development cost wouldn't be nearly as high. Perhaps you can delay the second requirement by exploring alternative methods of content sharing.

For example, the forums are a great surrogate for content sharing if the demand is small enough. A forum where people can post their content could hold up temporarily to ease the burden of development.

Yep, that is exactly the conclusion that Keith and I came to this afternoon, too -- great minds think alike!  Basically, the game will just package the content, and then consume packages, to make the burden easier on the user but still something that they have to get through the forums instead of some other mechanism.

Another example: A friend of mine develops small games for the iPhone. In addition to setting up a small server that allows users to share their high scores, favorite games, and other stuff through the app, he also wrote a facebook app so that people could use facebook to share that information as well. Certainly his needs are far less than yours, but the idea merits examination. Are there services out there already available that you could integrate into at little cost to help meet your goals in the short term without investing so much so quickly for the long term?

Right now my main point of interest is in Impulse v1.  High scoreboards and other things of that nature are not even on the radar yet for me with Tidalis, although of course that's something that would be nice in an ideal world.  Please don't misunderstand my intent with this post: my various requested things that I'm looking for are not the minimum requirements for implementation.  Those were my requirements plus my entire nice-to-have list.  When considering vendors, I generally feel like it's good to look at all the possibilities based on the "pie in the sky" features, and then when it comes time to implement scale back where needed.

At this stage, I've pretty much ruled out building it ourselves, except possibly for high scoreboards or something.  But that's a wholly different issue than the matchmaking, which is a different issue from the content sharing.  I had hoped for a service that I just hadn't heard of that would be perfect for all three, you know, but evidently it does not exist.  If Impulse doesn't pan out with their v1 stuff, then there might actually be no matchmaking at all until whenever their v2 stuff is out and ready, which likely would not be nearly at launch for Tidalis.  So it goes, I guess, though that's not what I want to happen ideally.

[Black]

I understand.

Seems like there's an untapped potential in the game services market. Just a note to any would-be entrepreneurs

[x4000]

Seems like there's an untapped potential in the game services market. Just a note to any would-be entrepreneurs

Indeed!  I think that's what Impulse v2 is moving toward, honestly.

[Vaos]

Please don't use GFWL, it is truly horrible on my opinion (forced to use it for DoWII). It may be fine for xbox players, but I feel like it doesn't have it's place for computers games. Gamespy is pretty terrible as well, it was fine in 1999, but now ... 

Never heard of that Impulse v1/v2 thing. What it is exactly?

[x4000]

More on Impulse Reactor:

http://www.impulsedriven.com/developer/Platform (Fairly old)
http://www.impulsedriven.com/downloads/Impulse_Reactor_White_Paper.pdf (Fairly old)
http://www.draginol.com/files/ipv2.pdf (about v2, much newer)

[Goekhan]

Hey guys,

I'm trying to get a good handle on the various matchmaking services that are out there, so that I can evaluate their pros and cons for both AI War and other upcoming Arcen titles.  Google turns up surprisingly few results, so I thought I'd ask the community here what you guys have seen and used in the past.  Here's what (non-defunct) matchmaking services I already know about, and/or have the SDK for:

- Steam
- Gamespy
- Impulse Reactor

What other ones do you guys know of?  I found two or three others, but those seemed to be (or were definitely) abandoned in the last few years.

About matchmaking services;

Gamespy: I really dislike Gamespy. I've played some games that uses Gamespy matchmaking services, Borderlands, Titan Quest: Immortal Throne, and some more that I cannot recall at the moment. As far as I can tell, server browser offers no valuable information, especially on the games that I recall (Borderlands and Titan Quest). But I think this is based mostly on games, not Gamespy.

Steam: I play a lot of games on Steam. In fact, that's how I found this game, and I thank Steam for that On local hosting, one sometimes (mostly) needs to enter some commands in console in order to eliminate problems, especially on Left 4 Dead 1&2. Server browser gives enough information. However, especially on L4D1&2, matchmaking sometimes becomes annoying. But you already said no-no anyway.

Impulse: Have no idea to be honest. No idea.

GFWL: Stay away please... Please! Batman: Arkham Asylum, Bioshock 2... I need to sign in in order to save? Yeeeeah right..... That thing made me an anti-console, hurray-for-PC kinda guy.

IWNet: I'm writing this to tell how it sucks. CoD:MW matchmaking system, designed by Infinity Ward. They removed dedicated server support, and made some changes on MW2, so that now you cannot really play with someone from America (I'm from Europe) unless they're on your friend list. 7 games out of 10, I mostly play with people that I've already played with. Annoying. Drop-in and drop-out is possible, but outs are being punished, whether you got pwned and rage-quitted, or your internet connection gone bad, you get the same punishment.

eehm... can't think of anything else. Freespace 2 had a system they used. After some time passed, they opened their sources, so that open-source community could reconstruct another matchmaking service on their own for Freespace 2. By the time I've played, there weren't any issues.

[x4000]

Thanks for the info, Goekhan -- IWNet is another that I've not used.  Not that it's available, and not that I would.