No Man's Land

[Code]

Hey guys
We are currently working on No Man's Land, which is a fast-paced arena-based combat game about claiming resources and surviving the changing environment
It's in its early concept stage, but we have a lot of ideas for it. Our main idea is to create an arena game that can be played in 10 minute sessions - for all those action-arena loving people who simply don't have time to play DOTA for the usual prolonged sessions.

You can check our page on Facebook here! https://www.facebook.com/MissileMoose

Here's some sketches:

[s_u_m_s_a_r]

Elsker Missile Moose i Action-pose

[jonaz.dk]

hehe.. dejligt crazy. Glæder mig til at se mere. Stay 2D art.

[vonknut]

Elsker Missile Moose i Action-pose

Hvis den laver poop hver gang den skyder, så kan du godt signe mig op for pre-buy

[Code]

Elsker Missile Moose i Action-pose

Hvis den laver poop hver gang den skyder, så kan du godt signe mig op for pre-buy

Hah! Det skal vi helt sikkert overveje!

[DoomCaek]

I love it! 8D

[Code]

Some nice things are happening. We have talked to someone who really likes the project (Motivation!) and now we are going to get all the documents and a small prototype of the gameplay done that we will show them. Things are looking good for this game!
Here's a little update to our map picture: http://imgur.com/nC42k5V

[Pegepinden]

Looks fun and awesome!

Just out of curiosity, did you think about a network solution yet?

[Code]

Looks fun and awesome!

Just out of curiosity, did you think about a network solution yet?

Yes we are making our own. We have an awesome programmer who knows his shit

[Pegepinden]

Ouch! But fair enough

[Code]

Actually we just announced this on our facebook:

"Yesterday we reached another major milestone in the implementation of our prototype. We got an internet lobby, server and a play session up and running. In other words, when we'll need testers (which we will!) it'll go smoothly over the wide web."

[Pegepinden]

Its just, no matter how good a programmer you are, if you server needs to work very well and scale well it will take a LOT of time (load balancing, data compression, 'reliable' UDP and so on). Its not that hard to create a basic socket server, its more having a polished scale-able solution that is the problem.

[Code]

Its just, no matter how good a programmer you are, if you server needs to work very well and scale well it will take a LOT of time (load balancing, data compression, 'reliable' UDP and so on). Its not that hard to create a basic socket server, its more having a polished scale-able solution that is the problem.

True that. This is only for our prototype at the moment, we'll see what other types of things we'll need later on - and then we might need something we can't just do ourselves xD

[ThomasLund]

I would never ever code my own server solution - its suicide for a small team to even attempt that, almost no matter how "good" the programmer.

Many very good solutions out there that have taken years to build and that scale well - and that are affordable.

Just my advice from someone who has been there, done that and moved on to a 3rd party solution ages ago

[jonaz.dk]

Indeed.. networking is quite a science in itself.

A few tips on that.

We've been happy with Unity Park Suite on the Star Pirats project. Free loooong evaluation.
http://developer.muchdifferent.com/unitypark/Downloads

Photon - Highly battle tested
https://www.exitgames.com/en/Realtime

[Pegepinden]

Worked a lot with Photon, both server and cloud. Its really great.
Unity Park Suite should also be awesome.

However as an outsider: https://www.assetstore.unity3d.com/en/#!/content/18358

Loading Balancing is not in at the moment but should come. However this is something you could fix yourself. Its quite cheap. It offers so much actually. Only played with it a bit as the author more less offered it for free to me( ) but was impressed.

[Code]

Thanks for all the replies. We already know about Photon, and it's good to know about all these other network solutions and your experiences with them
I'll make sure to ask the lead programmer about this.

[ThomasLund]

Tell him, that he should be making games and not reinvent the wheel

[zhennTil]

Hi y'all, lead programmer here.
Figured this would be a fine time to stop lurking and step in.

First of all, I agree with all of your concerns about the networking. We are well aware of the complexity of the issue.

Let me just make it clear that we are not writing our own networking engine here.
For the prototype, we are using Unity's fine RakNet-based engine as a peer-to-peer(ish) solution.
The only central server we are running is therefore Unity's slightly-less-fine-and-extremely-poorly-documented Master Server/Facilitator/Connection Tester triplet, again purely for the prototype -- what we end up using in production is an entirely different discussion.
This system has only the (relatively) simple job of arbitrating the connections between peers, so it sees very little load compared to an actual game server.

We'll see how well this setup actually behaves in wider testing, and at that point we'll certainly consider solutions like Photon and what-have-you, but for now it is having no problems.

Thanks for the tips, much appreciated!

[ThomasLund]

Hi lurker programmer @ Red Ocean



Just simply drop using the shitty RakNet Unity stuff in the first place and swap to Photon, SmartFox or similar solution right away. Even for prototype usage. Some of the "real" solutions require a different (and better) way of separating client/server code and you will end up having to rewrite a shitload of code from scratch.

So do it right the first time!

/me steps down from the sandbox

[zhennTil]

Just simply drop using the shitty RakNet Unity stuff in the first place and swap to Photon, SmartFox or similar solution right away. Even for prototype usage.

I'm curious as to why Unity's RakNet stuff is so shitty -- I'd love to hear your reasons, the main one I can think of is lack of maintenance.

Some of the "real" solutions require a different (and better) way of separating client/server code and you will end up having to rewrite a shitload of code from scratch.

I assume you're talking about authoritative server stuff, of which I am well aware.

[Pegepinden]

I think one of the thing's is that why start all over again? Also its not just 1 to 1 from a P2P solution to a authoritative setup, it is two different beasts.

For Unity's RakNet, well for your case it might be OKAY, but try and give it a spin on Google and you will see all kind of problems people ran into. There is a reason that almost no one (serious) uses it now a day.

[zhennTil]

For Unity's RakNet, well for your case it might be OKAY, but try and give it a spin on Google and you will see all kind of problems people ran into. There is a reason that almost no one (serious) uses it now a day.

Yeah, performance problems seem to be the main concern. But then, UnityPark Suite boasts full compatibility (API-wise) with the built-in RakNet interface, so making the switch to uLink et al if/when we acquire funding seems like the best option here.