TS Mulitplayer- A healthy discussion

[AndiS]

My crystal ball says that the following factors matter a bit more than our desires.

• Development cost of Multiplayer. For RW, DTG (RSDL?) decided that yanking out the old dispatcher and putting in something that works was too expensive. The UE4 engine comes with multiplayer support right away and they are certainly not considering bringing the AI code that was already dated when it was written a decade ago over to the new thing. Now I have no idea how much of the total development cost for multiplayer mode is covered by these two things alone but it is definitely more than for RW back in 2011 or 2012.
• Cost savings combined with increased sales. If multiplayer is 'it', you can save the development costs for reasonable AI. If you want to see other trains, go online and buy the content you need to join others.
• Trends. If marketing experts tell them that single-player games are dead, they will include multiplayer as they are building a platform for the decade to come. If the same experts say that people already find out that you can run a chat programme in parallel with a game, and/or they are happy enough to show off achievements, things will stay as they are. The introduction of relay scenarios showed that DTG would not mind having their game badged multiplayer.
• The economics of running servers. A while ago, I read that World of Warcraft yields millions from subscriptions without which you cannot log into any server. The Cloud should have made server prices drop, but don't ask me about recent facts. If the hopes of those who promote it provide true, running multiplayer servers could generate a nice little benefit.
• Control over content. When you log into the server, they can check that your content is legit, by whatever mechanisms they establish. When you are off-line, the can not. Also exclude third-party content from multiplayer activities will give them a strong boost in their own sales.

Finally my crystal ball said: Making it an engine collector's 3D chat room where you can invite others into your cab so you can share the experience of driving some engine will max out profits.

Then I kicked it into the corner.

Myself, I can live without multiplayer, but I know that it is coming our way. In that case, I recommend a combination of automatic scoring of drivers to group people into (1) skilled drivers, (2) willing persons not yet up to the job and (3) game spoilers; and humans moderating the shows by admitting and excluding players (semi-automatically), adding and maybe removing services to suit the player count. There may be automatic modes, i.e., the moderator only logs in once a day to see what needs to be done but the show runs 24 hours. There will be fluctuations in player count for different servers and different routes according to their geographic situation but no one can predict them too well.

Shows will be popular routes for a start, and there will be stand-ins for the engines that you don't own but someone else who owns them drives them within sight. If there is a lack of players, AI will run some trains to make it less empty. If a player leaves before his train reaches its target, AI will take over to keep the show going for the others.

I would want to dream of AI that is so clever that it can take over all the roles of all the players without degradation but that would take away half of the motivation for multiplayer and be expensive to develop. In the past, the damned crystal ball won every bet against my heart.

[alexisn]

As someone who's not particularly fussed about multiplayer, for me to be more fussed I'd want a fundamental change in how the sim operates (which could, in theory, happen with the Unreal Engine version). Rather than the current scenario-based system, each route would have a full timetable implemented which can be dipped in and out of by the player or players. In other words, you fire up the sim, see which services are available to drive at that particular time and pick one. Other players online would be driving any other services that they like the look of, with the remainder driven by the AI.

While with the current system of scenario creation would make such a system far too labour-intensive to be workable, it doesn't seem beyond the realms of possibility that a brand new sim could incorporate the ability to read a text-based timetable file and just 'make it happen'.

[gptech]

Good point alexin, so far all our guessing and wishful thinking has been based on what we have now and the constraints of the present game.

In other words, you fire up the sim, see which services are available to drive at that particular time and pick one

We can do that now actually, basically a "free roam with instructions" and I'm sure there's an example in the file library---it may be for CCfE so I'll see if I can track it down.

EDIT:

Slurry in a Hurry [84795 bytes] - Slurry in a Hurry.zip File ID: 29867 Date: 08 Oct 2012 - 399 Downloads

It's moving off topic though, so if there's any more to say on the idea I think we should start a new thread.

[Fenrir]

As an experienced hand in Multiplayer in other games (mostly flightsims) I personally cannot wait.

It's as much about the social interaction as well as the gameplay; I have had the fortune to make a good number of very good friends via my online experiences and still have a great time playing online.

Regards TS? I'd most look forward to being able to accurately crew a steam loco with a chum; whilst having each permanently stuck as driver or fireman could get old quickly, if the code would allow the ability to swap places on the run, it could be not only an interesting exercise in teamwork but a very useful and absorbing training tool for those inexperienced and a bit intimidated by steamers to be proactively taught by a seasoned player how to operate the loco.

Add in 'live' signallers and the opportunity to communicate with them or the potential to double-head with another 'live' loco crew and I'm getting quite an anticipatory itch!

Looking even further forward the potential for gameplay could be very exciting, with an opportunity for almost 'unlimited time' scenarios - servers run a 24 hour sandbox where a general timetable is in operation; you jump in and can either pick an AI service and take over control or select a dormant loco and marshall your own train. In either case you drive for as long as you want.

In addition, if the facilities are in place, a sub-culture based on inclusive programming architecture could potentially allow clever third parties to create a logistical template for each server or route that generates jobs/traffic that users are required to move around the map/route - similar to online campaign systems in the combat flight sim arena. Very immersive to boot.

If - and that's a very big if - this is where we are headed, if the cons revolve around the odd douche bag deliberately derailing trains or ramming others (which given the typical multiplayer controls [bans & kicks] and the generally more mature demographic of the serious players is going to be a small percentage) then IMHO it's a small price to pay