Do you have any comprehension exactly how much stuff would need to be changed to get above twelve players in multi? At the very least, you're talking about redrawing masses of interface files and (probably) completely rewriting the way FRED understands the concept of players. And that's without even considering the code in the game itself, which I have zero idea about. It's not an isolated module you can just pull in and out.
Yeah, I do. That was what prompted me to make this post, and this offer in the first place. I'm on GameDev constantly, and have been for years. GameDev was what convinced me I wasn't going to learn game development overnight, and that a mod was the surest way to gain the skills you need to make a full game. I hang out there to learn more about game development, and how to properly make games. That's one of the major reasons, personally, I'm engaged in this project. The primary reason is because I love - absolutely love - Tachyon, and I want to see it live again.
The project is, indeed, massive. I've had several long talks about it with SCP coders, and I've done quite a bit of research on my own. The interface needs to redesigned, every place the interface interacts with the code that assumes 12 players needs to be redone needs to be looked at, the netcode is fairly old, and needs to be redesigned, or replaced with something more suitable for modern broadband/fast dialup players. There's more, but there's a short synopsis.
You have to understand, Freespace is not like a modern game - it's not a glorified multiplayer arena with a shoddily thought out single player campaign tacked on in the two weeks before the thing shipped. It was, and remain, principally a single player, plot driven, mission based combat space simulator. That's why nobody's ever been able to make a decent freelancer style trade simulator, or why nobody's been able to make even a semi random mission generator without masses of work (BHX is more or less the closest anyone's come AFAIK).
I chose the SCP primarily because of it's native mission editor, it's Descent-descended engine (which had the multi-axis thrust Tach needs, natively), it's modding community, and it's activity. I'm not doing a single player campaign, no. That doesn't mean I just want deathmatches. Tachyon had a LOT more than deathmatches. It had a fairly complete backstory, a VERY large fan story continuum which expanded the story tenfold, and, most important, a community that always, always tried to keep the story of their squads, and their pilot characters, in the lead.
Tachyon's Base Wars had, like Allegiance, a lot of tactical/strategic complexity and sophistication, which simply isn't possible with the player limits FS2 has in place. Ever. If you only have 6 people on a side, in Base wars, you really, really are going to have a hard time getting everything done that you need to get done. Without the RTS elements Allegiance's commanders had, though.
Yes, FS2 doesn't have any of that. I chose it anyway, because I like the engine a lot better. Don't think I'm trying to be wholly negative. But, the SCP is about modernizing the engine. If we all just want FS2, and the way it used to be - why are we here? I guess that's what puzzles me about that the most.
The multiplayer aspect was a bonus in a time when not all games had them, and it did fine for years while PXO was running. If it's not perfect for BTRL or any of the new TCs, well, frankly, I don't particularly care.
It's barely working, these days, for FS2, let alone any of the TCs. Especially not for high-poly ships. That's part of the problem. Multiplayer is dead as a doornail. While we're working on multiplayer - and, to be frank, it really needs to be worked on - and, while we're looking for a team to do major/massive projects - let's fix multiplayer, once we've tackled the major show-stopping bugs, like collision detection.
From my perspective, if someone wants to make a space sim mod where the principle aim is to go out and shoot masses people in multi player, well, good for them. They can use any of those engines you've mentioned. If, however, they want to make use of the Freespace engine, with all the freedom of open source, the graphical updates, the ridiculous ease of modding and mission construction and that helpful community you mentioned, they use FSO. Maybe it's an unpleasant choice, but it's one that has to be made. BTRL made it, so it seems will you have to.
I have not played a purely single player game in years. The AI sucks, in comparison to real people. Always. For the folks that like single-player? I'm not talking about anything that's going to affect you. I don't agree with you, and never will - but it won't affect you. You already HAVE one of the best SP mission creation engines and toolsets that exist. What we DON'T have is fully functional multiplayer capacity.
Of course, you know more about coding in general than I do. Maybe you'll look at the problem and come to a different conclusion. Maybe interface art can be dynamically generated, maybe FRED can be convinced to play ball, maybe the FS multi code is that easy to fiddle, and maybe you do an amazing job and we all have 100+ player multi games to look forward to. Personally, I doubt it, but if you can do it, I'll be first in line applauding. Good luck.
Maybe I do, maybe I don't. I do know space sims - I own over 350, from Elite on up. Most of them suck. Freespace is a great game - but for me, personally, I was never a fan of FS2, or FS when it comes to replayability. Planes in space (they fly like planes... why? You have no air resistance - what's to keep you from multi-axis thrust?) is well and good - but it's usually the engaging nature of the multiplayer gameplay that keeps a *multiplayer* game alive for any amount of time. I refuse to waste my time on someone else's story arc, with no ability to change a thing, when I can make my own, with my own character and personality - and *imagination*. Sorry.
For those of us who are singularly uninterested in creating (or remaking) a singleplayer campaign, and enjoy, to a FAR greater extent, the flexibilty and fluidity of strategic warfare with someone *different* every time we play - it's a very, very large handicap. If you want a treatise on the advantages multiplayer gameplay has over scripted single-player campaigns, I'd be happy to provide it. However, if I wanted "MASS CARNAGE111!!", I'd skip Base Wars altogether
Regardless, I never, ever said it was easy. In fact, given the nature of the post, I'd say I was inferring something quite the opposite. I was just... taken aback - by the direct negation of the very purpose I advanced this *entire post* to specifically remedy. The multiplayer netcode, and player limits. I don't mind bumping things that are honestly higher-priority - like collision detection. I don't see, however, why a revamp of the multiplayer code, in both respects, is neither needed nor wanted. What I've heard, and what I've experienced, is quite the opposite.
It was rather rare to get an 8 player game going even during the height of FS2's popularity on PXO, and much more so for 12 player dogfights. I played there daily for years and I've only ever seen two or three 12 player games in progress. The FS2 multiplayer community today is a small fraction of what it used to be, so if increasing the limit is more than just a trivial amount of work, it wouldn't be worth it at all.
Well, with a player limit of 12, and no ability to join-in-progress, I don't see why that'd be surprising. That's the very reason I personally dropped FS2 after picking it up in the bargain bin, in early 2000. It's lack of multiplayer features. I played the SP later, when I didn't have internet, for a short period - but I dropped the campaign after going halfway through it, because all it was was splattering dozens of AI idiot fighters. Besides, we're not talking about just FS2 here. We're talking about an even half-dozen TC's that are all *new releases* - not 9 year old classics - on a revamped engine with all the bells and whistles you can shake a stick at. My shock at the "we don't need any more players" should be rather apparent in my previous posts. That even applies to FS2 mods and campaigns, with the mediavps. Don't you guys *want* to be able to fly with at least a *whole* squadron, when launching from something the size of the Colossus? It could hold like... 10. I don't get it.
TBP and BTRL both had over 100k downloads. Think about that, for a minute. That's *200 thousand* potential players. Half of those, for a DEMO. That's a very, very large group of people to dismiss out of hand as far as new/improved functionality goes. I'm not going to even talk about my mod, or any of the others. Just BTRL and TBP say quite a bit about how much it's needed.
Well before I get into this I'd like to state that I want to see the engine expanded to deal with more players. I do however appreciate that this is a huge undertaking though so I'm certainly not saying it should be a priority.
I agree. It doesn't have to be #1 priority. I do, however, think it should be pretty far up the list for the "massive' projects, however, given it's relatively higher difficulty/scope level.
Yes but is that necessarily a problem in multiplayer? I know it seems odd to hear me arguing in favour of Endorian missions given my longstanding outspoken views against them in the past 
but it's always been the lack of player interaction with the mission that is the foundation of my dislike for Endorian missions.
In your typical Endorian mission the problem is one of balance. Due to the size of the mission the player has little effect on the outcome of the battle. It is very hard to successfully balance a large mission on a knife edge sufficiently that the actions of a single player decides the outcome of the mission. So you typically end up with a mission which the player always wins or always loses with no regard for what the player actually does.
However this isn't an issue that affects Endorian multiplayer missions. It is far easier to simply set both sides to have similar capship forces and then leave it up to one team or the other to get the upper hand in the battle. Sure each individual player still won't have an effect on the game but in this case the team will. Unlike a SP Endorian mission the quality of the players will always tip the balance in favour of the better side.
Exactly. Base Wars was all - and I repeat - ALL about how good you were. NOT as an individual, but as a team - although individual skill did matter, to an extent - the scope required you work as a team. The best teams could stay evenly matched for, literally, days. However, that game mode is unplayable with FS2's current limits. One player can completely decimate the best defenses, if they're good. Two can keep the other team in the stone age, indefinitely. The only counter is to have a half-dozen miners, at any given time. There's the team's half of the 12-player limit, right there. The above scenario is WITH 2-3 defenders on your base. You *need* mro than the player limit has to give, to make Base Wars work - at all.
Yes I do agree that the more players you add the harder it becomes to make a good coop mission but TvT and Dogfight mode wouldn't automatically suffer from having lots of players. Besides I relish the challenge of trying to make a 64 player Coop mission that isn't a simple gauntlet. The 1/6th of my brain that is now totally given over to FREDding is practically screaming at the rest of me by now.
Coop missions with that many people will be... heaven. I agree.
12 people is inexcusably, outrageously small.
Now I have to disagree there. Black Wolf made the point best. With the exception of Dogfight mode FS2 Multiplayer isn't simply about deathmatch games. If you look at the 
missions you'll soon see that what they wanted to do was provide the ability to play something analogous to the SP game but for more than one player. 12 players is more than adequate for that, even in this day and age. Want to disagree? Find me a space combat simulator that lets you play through the main SP campaign with 64 players! 
I hate singleplayer games
However, the ones I have played were hard solely because your idiot AI wingmen are just as dumb as the idiot AI enemy. So, you end up fighting 4,762 enemies. That's not... real. At all. If I'm in a fight for my life - it's *likely* going to happen against forces with, at the least, a rough parity with my own. Not against a force 4,000 the size of mine. What sort of idiot would send a military force against odds like that?
Whichever - that was 9 years ago. This is in the age of MMOs - and only two years after it's release date, it's successors both had much, much more single-server capacity. That was 7 years ago. We can do better now, can't we?
If you want to be able to play massive multiplayer games in the engine that's one thing but didn't screw up making multi only 12 player maximum. You're failing to understand that they had very different goals from the other games you mention.
Okay. That still doesn't negate the fact that it needs fixing, 9 years later.
Bollocks. I'll tell you why Multiplayer is dead for the major TCs and it's absolutely nothing to do with what you say. The reasons are
1. Hard to set up - Having to configure routers is not easy and discourages many players. If there were some way to automate this we'd have a lot more players. But as far as I know there isn't. On top of that some people can't even play because the ports they want to use aren't open on their university/corporate/etc network.
2. Broken standalone servers. These should be fixed in 3.6.10 but for now they aren't helping
3. Lack of games. Partly due to 2. But most people interested in playing multiplayer can't because no one is running a game.
4. Lack of multiplayer mission. WCS doesn't even work in multiplayer. TBP lacks multiplayer either completely or almost completely (I know IPA. I haven't forgotten). BtRL is the only major TC with a large set of multiplayer missions. And we've definitely fallen foul of the factors above.
Lack of massive multiplayer is not the cause.
It will be, once you get out of demo. I do agree with you, those other things are all wrong with it. It's all part and parcel, though. The multiplayer, as a whole, is eons behind the rest of the SCP engine. Until it isn't, it'll be a singleplayer-specific open source platform. It's really that broken. A selection bar to choose how many players you want to host will fix your problem with FS2 compatability, if you really want to stick with 12. Regardless - can anyone really say that it makes sense for the Galactica to launch a *half* squadron of fighters - and that be it's maximum? Or half of these super-giga gargantuan carriers for some of the other mods? It's a serious hamper on what's possible, and on any sort of realism for "team battles" of most sorts.
I overstated the multi limits, in my post. I intended for it to come across as "multiplayer brokenness, in general" - but I didn't express myself clearly. My fault.
Now I'll agree with you that something is causing a lot of lag in the game. Which is why I added the -cap_object_update option as a stopgap. Fixing the current multiplayer code is more of a priority for me than adding more players.
If you're tackling one, you need to tackle the other - or you'll have to tackle it twice. If, what fixes lag problems with 12 players *doesn't* fix lag problems at higher player limits - won't you just be doing it twice? It's going to be a huge project, regardless. If we're going to tear the guts out of the multi code, and redo it - redo the interface/code portions, as well.
On the other hand BtRL achieved 100,000 downloads in under two weeks and I suspect that if we do a good job on it we should be able to equal that figure with the full release. The trick is keeping hold of the players who do want to play multiplayer. I'll bet that even at the height of it's popularity FS2 never added anywhere near that number of players in so short a time.
For 100,000 players:
8,333 servers, at 12 players apiece.
1562 servers, at 64 players.
Granted, not everyone is going to play multi at once. Not everyone will play, or play multi at all. Let's be ultra-conservative. If 2,000 of those play multi on release night - that'll be 167 servers going. With 64? That's 30. That's another major difference. Community *capacity*. I played Tachyon when it was that busy. There would be 30+ servers up, all the time. There was 4 64-man servers hsoted by Novalogic that were perpetually packed. Day and night. You cannot, I repeat, cannot, support that sort of playerbase with 12-man multi servers. Not for long, especially in competition with MMOs.
Flaser/DaBrain: Thanks.
Watch your tone, RazorsKiss. If it hadn't been for your "sorry for the rant" addendum, I would have sent you on a vacation.
Roger.
Upgrading from 12 to more-than-12 players isn't a simple task. It's not even merely a complicated task; it's a massive task. For one thing, the interface would need to be significantly rewritten; and for another, all the mission code that deals with squadrons and communications would need to be tracked down and straightened out.
Thus, it's placement on the "massive tasks of doom" list. I KNOW it's not easy. Or it wouldn't have been the
reason i posted this in the
first place. I was surprised you said something, because I included "netcode", in two separate places, in a thread you replied in, where I mentioned this idea first.
You can't put the multiplayer code on steroids and declare it upgraded.
Didn't say that.
On the list of the top 100 limiting factors affecting multiplayer, "number of players" is probably down around #90. The top five are most likely the ones that kara and CP mentioned. And the vast majority of these limiting factors are infrastructure issues that affect the entire game, not just multiplayer.
Well, then, lets get started. What are they, so I can list them for potential coders?
For example, taylor mentioned collision detection as a significant problem, and collision detection increases exponentially with the number of ships. I wasn't only referring to gameplay and balance when I cited Battle of Endor; I was also thinking of performance. Can you imagine the problems that would arise from dozens of high-poly fighters and hundreds of laser bolts and missiles?
Yes, like I said, I've played games where you have those - multiplied by a factor of 100 or so. It's possible in other games. It's obviously possible here. Take a look at Tachyon's codebase, one time. It's beyond nightmarish. Thankfully, they used the novaworld netcode base, which is fairly decent.
If you're so insistent on these features that you're prepared to ditch FS2 in favor of Eve
How in the world I'd manage to make a mod for 1. an active MMORPG, and 2. a third person screensaver sim - I don't know
or Allegiance or whatever
I already said - this is still the best engine.
I want to improve it, and I want it to have the features it needs to be truly modernized. Yes, it's going to be hard - but it _needs_ the fix - in almost every conceivable multiplayer area.
