Real Time ORPGs and Construct 2

Discussion and feedback on Construct 2

Post » Mon Jun 24, 2013 3:02 am

So before you all go off on me, let me just lay out a basic game plan for my two-man team so you guys don't think I'm a "im making mmorpg where do i start anyone have plugin?" kind of person.

1. We are going to determine what kinds of engines and platforms are available on which to plan for future game development.
2. We will be creating everything from arcade games to RPGs to puzzle games, and every game will feature a different element of the distant MMORPG we plan on creating.

For example, aspects of platforming/sidescrolling, storytelling, upgrading, magic/ability systems, multiplayer and server architecture, player shapeshifting, character creation and paperdolling, 2D perspectives, & al until we have the proper amount of experience.

And this is what I would like everyone's input on:
  • Is Construct 2 a fit choice for creating and designing the client for a 2D MMORPG using solely HTML5 export options and TCP to handle data transfers? Mostly this is because I really enjoy the idea of cross-platform HTML5 support, so that anything from phones to desktops could play

  • If not, will it ever be? Is there enough of a future for both C2 and HTML5 to where MMORPGs will be feasible enough to pursue? Would working with another engine such as Unity3D be a better choice for developing a web-based MMORPG? Or Adobe Flash, for that matter?

  • If it is possible, what kind of native support for real-time connectivity is there? I was reading that the current vanilla packet functionality would only support at most enough consistency for a turn-based ORPG
Thank you all!

Edit: and to add what other engines I already know about:
Any (every terrible) VB6 Engine
NetGore
Every 3D MMO engine
BrowserQuest
DatapawWolf2013-06-24 03:07:48
ImageImageImageImage
B
55
S
19
G
51
Posts: 633
Reputation: 30,621

Post » Mon Jun 24, 2013 2:45 pm

You could try using Construct 2 as the front-end, with WebSockets to communicate with your backend.

WebSockets don't quite cut it for real-time connectivity (where latency is a significant concern, such as in "twitch" shooters). I think it might be OK for a slower-paced RPG providing you have high-quality, geographically placed servers like Bombermine does (which uses WebSockets as well AFAIK). WebRTC solves that since it supports UDP instead of TCP, but that may be a way off wide support, especially on mobile.
Scirra Founder
B
378
S
220
G
84
Posts: 23,868
Reputation: 188,111

Post » Mon Jun 24, 2013 8:05 pm

@DatapawWolf
You could make the game you want with C2. It all comes down to time and investement. In some cases you are going to have to do a lot of ground work to fortify the parts that C2 doesn't cover. Custom plugins etc. But likely not so much.

I like you am designing a overall game, by developing it in small parts. Good luck :)

As for WebSockets. You can make Realtime game play with Websockets just fine. However, you need to keep in mind idea usage and expecations. TCP is not ideal for high paced competitive gaming. Mostly shooters(Call of Duty, Halo) and Fighters(Street Fighter). This only really applies to competative gaming. In non compatative gaming you can usually get a way with it. And games that aren't high twitch you will be ok too. I month and little bit ago I wrote a websocket top down shooter in C2. Played fine as long as the ISP and connection was good. Played people from West Coast Canada vs European players. no problem. yet on occasion someone in the same city would be laggy as hell :|..

in an ideal world. TCP is no different than UDP. TCP however has data safetry protocols. And when data is lost, or fragmented TCP attempts to fix this. The problem is that TCP forces an in order handling of data. So even if TCP get's data, it will hold on until it fixes a lost message.

so UDP is better, for twitch games due to this. As often game messages can ignore lost messages due to the fact that some of the same data is in the next message. and more up to date.

but you can still do online gaming fine.


as for what tools you have at your disposal
1. Scirra WebSocket plugin.
This will require you to build the entire infrastructure for your games. A lot of work, but you get what you design.

2. Rexrainbow's Bottleneck. This is fairly fleshed out, and if you have programming talent. You can do the server logic in JS for NodeJS. However, bottleneck already covers rooms, connections, disconnects, join/leave... so on so forth.

3. PubNub, a recent discovery is that PubNub has an old out of date C2 plugin. However, if people use it and prove a community for the PubNub plugin. They might end up supporting it more effectivly.

4. VelotJay networking plugins, he has a few. A socket.io plugin and a photon behaviour. I'm not sure how much they are fleshed out for practical gaming.but i'm sure you can always tinker it up :)

5. My unfinished photon client plugin. It mostly needs to have the loadbalancing features added. ie room lists. however, I did have a shooter game using it.

6. And then for those with patience. With awarmenhoven and I are working on a network game api. While it's a JS api, it's built with C2 developers in mind. So the feature set is to be easy to use. but again. that's for those who are patient.

B
88
S
18
G
9
Posts: 2,455
Reputation: 14,928

Post » Thu Jun 27, 2013 7:43 pm

I appreciate both your responses, thank you for taking the time to reply! I've been kind of busy, so please forgive this late reply. ^-^"

@Ashley, Our team probably will only have enough for one at the start, but yeah, we'll definitely look into WebSockets first, and go from there.

@jayderu thank you for the detailed reply! Hopefully TCP will be enough for what we want to do, namely, a Sidescroller. I don't know how twitchy the combat will be, but I've played with a few that got away with it just fine.

1. The WebSocket plugin, was, I thought, documented as unsuitable for real-time MMOGs. But I'll check again.

2. I'll take a look!

3. I've been hearing on the forums that support might start up again, which means it sounds like it could be worth the effort.

4. Sounds good.

5. Hmmm, rooms.

6. Oh, we'll have lots of patience, since we will be focusing on our other game ideas first before beginning multiplayer games. :)


Thanks again, both of you! I will be doing some more research while playing around with Construct 2. I have to say I am very pleased with the documentation and support, overall, with this engine. :)
ImageImageImageImage
B
55
S
19
G
51
Posts: 633
Reputation: 30,621


Return to Construct 2 General

Who is online

Users browsing this forum: No registered users and 7 guests