It was with some discontent that today I checked the results after testing the Steam P2P networking approach.
For those who don't know what this was, it's basically how the game would play out in terms of networking, all data (player movements, shootings, etc) would be exchanged through Steam servers, player to player (peer-to-peer).
So, the news. At first I was head first, implemented the full FPS controllers, shooting and all that and tested how movement, shooting, look-around would be sent through the internet to connected players. It didn't go well. The movements were very jerky and delayed but I decided to scale it down.
After that fail, I scaled down to a 2D test. A cube, with X,Y movement and have it send ONLY that information. It didn't go well either. I would move the cube in one pc, and the other pc didn't receive anything, but passing some time (around 20-30seconds) there was a bit of movement but, again, jerky and (super) delayed.
I guess this was expected, as Steam doesn't want full games relying on their servers, but at least I had to try.
So, this leaves two options which I will think over and over, do yet some more testing to see if I can go around some things.
Option #1: Give up. It's hard, not only for me, that I have a 10 page document with lots of information about what would be this game, and the amount of assets I produced and gathered, but for the people that supported me, with feedback and kind words.
Option #2: Go with Listen Servers. For those that don't know, Listen Servers is a way of saying Dedicated Servers that are player hosted. This is an approach used for example by CoD games (remmember Host Migration?) and it's actually a good approach, but with a down side : Port Forwarding.
If you want others to join your server/game, you would need to Port Forward which is a thing that some people either don't know how to do, or can't because ISP's block that functionality in the routers. With this method, those people wouldn't be allowed to play the game with their friends or other people from around the world, and that's also hard for me to do. I've had that problem with AffordaGolf and it was not pleasant.
These are the options, of course, giving up is easier, but I'm still not sold on that. I'll continue testing, and I'll research on NAT punchthrough. NAT Punchthrough can be the key to make this game playable for everyone. It makes auto-portfowarding as easy as pressing a button.
I want to say again, I'm sorry to those who supported and believed in this project, I will keep you updated.