Future of our game

[freaknigh]

I actually just finished cloning two classic games of this era, Z and Diaspora. So I decided to stop by to see what the state of this game was and if it was something I should clone as well. All I can say is subspace is actually a very simple game for someone who has refined coding skills, it would take me no time to clone if I knew the protocol and had all the graphics set aside.

 

The thing that is worrying me though is I'm noticing all games of this era are dieing and each game's community is desperately trying to keep it alive. To me personally, I think it's just the nature of software and the industry of creating games. This game is over 10 years old but it is just is playable as it was then. A game that is created now has to actually compete with games created over a decade ago and every game up to that point and vice versa. The amount of playable games out there is simply becoming over saturated. Every time I look at a new game though I think that it is cool as hell but I could careless because the games of my era are all I need.

 

I would rather see games like this be expanded upon instead of being remade new. Firstly at this point the whole game needs to be open source if it wants to have a chance moving on in its classic state.

 

Otherwise, to move on and bring in new players etc, I'd say it is going to have to be something like Brady is talking about and that is a subspace like game that is modernized and commercial (of course still having the base elements such as making bots and running your own server etc).

 

Here are my other projects though...

The Zod Engine (the Z remake)

Project Diaspora (the Diaspora remake)

 

But looking at it all, I think the only games that really keep on are the ones who keep making new releases of themselves to keep up with hardware capabilities etc. Maybe if the game was made fully open source, and then the mechanics were kept the same but the 2d models were replaced with 3d models that would help?

[Dr Brain]

The sad fact is that computing, latency and bandwidth still limit multiplayer real-time open-source games. You can't ignore the cheating aspects that can spring up around an open source game. Cheating nearly killed this game a decade ago, and no open sounce proposal has sufficiently addressed this (saying we'll fix it when it comes up is not addressing the problem).

 

In a nutshell, no open source is a great idea, but focusing on it detracts from the ultimate goal of a fun game.

[freaknigh]

The solution to that is very simple. Each server will have its own encryption key and a client will be compiled with that key as well. The only client that will connect to that server is the one that was precompiled for it. Otherwise it is dependent on how much of the game is regulated in the server or the client, if the server can regulate 100% of the gameplay then it doesn't matter if the client is open source. If that requires a change in the protocol of the game then that should be fine just make a matching open source server to go with it.

 

If you can't regulate the gameplay 100% then you should be able to stick in a lot of checks depending on what kinds of cheats are still possible to see if a player is likely cheating and then temporarily suspend them and let a mod review them to see if they can put cheat check ignores on them.

Edited February 3, 2011 by Freak Nigh

[Dr Brain]

The solution to that is very simple. Each server will have its own encryption key and a client will be compiled with that key as well. The only client that will connect to that server is the one that was precompiled for it. Otherwise it is dependent on how much of the game is regulated in the server or the client, if the server can regulate 100% of the gameplay then it doesn't matter if the client is open source. If that requires a change in the protocol of the game then that should be fine just make a matching open source server to go with it.

 

If you can't regulate the gameplay 100% then you should be able to stick in a lot of checks depending on what kinds of cheats are still possible to see if a player is likely cheating and then temporarily suspend them and let a mod review them to see if they can put cheat check ignores on them.

Latency is the problem for server-side everything. Also zones like trench wars probably exceed today's computing capabilities, especially when you consider the physical server is almost always hosting several zones at once.

 

And an encryption key doesn't prevent memory hacking. Also, disassembling an open source program is significantly easier, since you know what everything looks like on the inside. Once you've disassembled that key, there's no reason you can't make your very own "legit client".

 

And server side checks only get you so far. You can only check what comes back to the server. There's a whole slew of things you can't check. Bots also become a huge problem.

[Cheese]

thats why you stick in memory honeypots

[freaknigh]

Well just having a unique key is only a simple approach, you can just swap entirely the whole encryption code so that for that client / server it is an entirely unique encryption method.

 

Another solution is make it so you have to pay $1 for the server to allow you to play normally and you'll say bye bye to all the hackers who aren't willing to lose their paid for account.

Edited February 3, 2011 by Freak Nigh

[Dr Brain]

Another solution is make it so you have to pay $1 for the server to allow you to play normally and you'll say bye bye to all the hackers who aren't willing to lose their paid for account.

 

True, but since you're also saying goodbye to 95% of the population, I doubt this will help the game survive.

[Cheese]

dr brain: parry, thrust, then in for the kill

[L.C.]

What about daily automated updates where every time the client is launched a unique key is attained. The key would be generated based on true randomness. If I recall correctly, random.org uses a physical device(s) of some sort that are designed to attain a true randomness, but it has been so long since I last tread upon this topic (so you would have to do the research yourself). I have always thought that it would be very cool if an encryption and security mechanism was based upon true randomness versus algorithms and code mortal man has created.

 

I imagine though, to do it just right, would probably require an excelling and masterful proficiency in mathematics (and enough programming with it to be able to code it).

[Kilo]

Well anyway, look at Discretion if you want some ideas of how some fundamental anti-cheating will be evaluated.

There is no easy solution for the whole problem.

Edited February 4, 2011 by Kilo