Senior Member
not that I don't agree that art assets can be proprietary, however... What you just stated here is extremely wrong and shows a complete lack of comprehension of the gamer community. Gamers as long as you're exlcuding the console kiddies are not mere content consumers. Not to say that there aren't pure consumers in the PC gamer field, however a lot of us are modders. People who do exactly what you claim that we won't or can't.Originally Posted by allquixotic
Bethseda games are actually one of the perfect examples, because you see... the point of Bethseda games, isn't the game itself.. it's that it's a low barrier to entry game engine (we're talking between $20-50 after all) allowing you to create mods to do almost whatever you want, which is ultimately the real point of those games, and they do have their unofficial patches, complete texture overhauls, models, quests, landmasses, etc.
And then we've got things like Planescape Torment, which while it didn't provide an SDK, it has it's own unofficial patches and mods to restore the original content, and finish it up, and otherwise change it.
To be very blunt here... If a game it popular it will be modded , no ifs, ands, or buts. So to say that opensourcing the game engine is useless completely fails to comprehend the PC gamer community.
Senior Member
Talking about open sourcing games: this has been rotting away untouched for well over a year.
By the way, am I the only one who always thought HumbleBundle stats were misleading in regard to the size of the Linux game market share?
(As in, using amount of payments vs. actual count of payers)
Senior Member
https://github.com/turol/shadowgrounds/network hasn't.
Why? The Windows average is the lowest, meaning that the percentage of people who will buy a game with a fixed price will be the lowest as well. What would be the point of percentage of buyers when it includes all those micropayments? For that to be meaningful, you'd need a cutoff slider, that discards all buyers below the selected amount (now that I think of it, looking at that would be very interesting).By the way, am I the only one who always thought HumbleBundle stats were misleading in regard to the size of the Linux game market share?
(As in, using amount of payments vs. actual count of payers)
Senior Member
Dude, I've been an avid player of (and contributor to) mods since as long as I can remember. I'm as much a member of that group you described as you are (or perhaps more). I've modded Sins of a Solar Empire, Star Trek Armada, Starfleet Command, Supreme Commander, Skyrim, Deus Ex, Savage 2, and the list goes on.
But all of the games I mentioned have one thing in common: they're proprietary! And people STILL managed to create huge, complete mods that worked extremely well. So your statement serves my argument more than it serves yours. For all the times people say "I wish I had the source code so I could fix that", 9 times out of 10 it's someone saying it who wouldn't be able to fix the problem even if they had the source. People just aren't that competent with native code. People are dumber than you think.
I'm kind of of two minds on this issue:
1. Yes, for serious mod projects like The Nameless Mod there might be enough resources to do something useful with the source code.
2. 99.9% of the modders out there don't have the technical know-how to do anything useful with a pile of hacked-together C++ code. Most engines are written with one thing in mind: get the product out the door on time. So the coding is sloppy and hard to read. It's a rare talent that is interested in making a mod and could really improve their product with the source code. Most modders only know config files, scripts (Lua or whatever) and art assets. They aren't software engineers.
Also, some additional points arguing against the need to release the source to aid modders:
1. Having many versions of the binaries floating around out there would completely destroy a multiplayer community. Any multiplayer system that wants to stave off cheating in any capacity is going to have to at least ensure that the binaries pass a cryptographic checksum computation (SHA-1 or so). Of course you should also be checking the image in RAM, but on disk is the first step. If you don't even check that the on-disk binaries match, not only do you have cheating problems, but you may have huge compatibility problems. If the binaries don't match, there's absolutely no guarantee that the person you're playing multiplayer with is going to play by the same rules as you.
2. So if you DO enforce binary matching, now you have to come up with a convenient way to let modders send their binaries to people, and let people quickly switch in and out binaries.... oh snap, that sounds like a bad idea; what if some modder uses their nice open source code to build a little virus or trojan or keylogger into their mod? How would you know? So do we now need a team of engineers to inspect the source code changes of every modder? That also sounds foolish because the main time when modders become of use is long after the game's development team has moved on -- after all, if the original devs are still contributing patches and fixes and content, just bug them and get them to fix your problem!
Also, I should add, never assume good faith in the players. You can never assume that players of a multiplayer game are going to play fairly. For single player, I think it's fine to let people cheat and not attempt to safeguard against it. But for multiplayer, people aren't going to stop cheating just by virtue of the fact that the game runs on Linux. There's no ingrained, extra-special-honesty built into the hearts of Linux users. Especially not now that it's coming close to critical mainstream mass; the mainstream crowd are the worst assholes. Just look at Diablo 1 and the townkilling. Or Vampire: the Masquerade and code injection that gives you a virus. OR even Call of Duty 4 with VBscript injection into your key binds. Multiplayer game security is serious business, and source code really can hurt it because, unlike bank transactions over SSL, you have to entrust certain details to the client, such as the actual rendering of the world; otherwise you'd have something like a VNC client and the server would have to render the environment for everyone and it'd be heinously slow and inefficient and unplayable. BTW, a couple cheaters in a multiplayer game situation can completely ruin the game for everyone, so don't pretend that it's not a big issue. It's really the kind of thing that makes people stop playing before they're ready to. They may be interested in the content and not bored of the game, but if they get cheated against, they're likely to stop playing.
While I'm very much in favor of making generally-useful stuff open source, I have to say that I really don't think multiplayer games and open source mix.
Imagine if the rulebook for baseball were printed on a publicly-editable (or at least player-editable) wiki site? So any player who wants to change the rules can just go in and edit it, and *poof* what they wrote becomes law for the entire game? How well would that work?
I just think games are a special case where, while yes I'd like to give upstarts the opportunity to pounce on a good, existing engine and use it as a springboard to make their own game, in general I don't favor open source games because it destroys any hope of containing cheating and it also fragments the community so that everyone is playing a different version, which almost forces you to implement MD5 or SHA1 checking, but as soon as you do that, you run into the above-mentioned version distribution problem.
OK, for single player only games, I can see your point. It kind of makes sense there to release the engine to let people play around in their individual sandboxes however they like. But for multiplayer games I couldn't disagree more. But my main point is that professional, serious modding teams like The Nameless Mod's team come around once every 10 years or so. The rest of the rabble wouldn't have the know-how to fix a NULL pointer bug.
Senior Member
Not necessarily; there are a number of things that can be done server-side to prevent cheating, like not sending the enemies position to the client when they are not in sight, or checking the movement speed of the player’s avatar. It also depends greatly on the game, of course. It’s easier to detect cheating in a Chess game than in counter-strike.
Worst case, there could be a player-maintained blacklist where each player can list players with whom they do not want to play… Or it could be automatic with a matchmaking system. But then again it depends on the game.
Edit: now that I think of it, it’s actually harder to check that a Chess player isn’t using a program to help him play…
Last edited by stqn; 06-10-2012 at 10:44 AM.
Senior Member
Yes, different game types have different methods of checking for integrity and reducing the ability to cheat. I was mostly thinking of the two genres I play the most; FPSes and RTSes.Not necessarily; there are a number of things that can be done server-side to prevent cheating, like not sending the enemies position to the client when they are not in sight, or checking the movement speed of the player’s avatar. It also depends greatly on the game, of course. It’s easier to detect cheating in a Chess game than in counter-strike
Worst case, there could be a player-maintained blacklist where each player can list players with whom they do not want to play… Or it could be automatic with a matchmaking system. But then again it depends on the game.
Edit: now that I think of it, it’s actually harder to check that a Chess player isn’t using a program to help him play…
But even if we completely remove malicious intent, open sourcing the engine and allowing people to connect with non-standard assets or non-standard binaries basically throws all integrity out the window. What I'm saying is it would make zero sense to have two people playing in the same game with different assets and different rules. You would really need a server to arbitrate the game (and the server would have to be proprietary or at least unmodifiable to the participants in the game, which just shifts the open source question to the server). The server would have to decide if two players can play in the same game compatibly under the same rules. Even if both players don't WANT to cheat, it's still possible that their mod would make the game unfair whether intentional or not.
As far as sending an enemy's position only when they're supposed to be "in sight": a few games have tried this, but it's really unreliable and degrades the quality of the game. Anyone with more than 50 ms ping or so is going to be at an enormous disadvantage: they won't see enemies at all until after they've already been in view for a fraction of a second. By the time they have a chance to react, they're dead. For twitch games, there are some things that you just can't take away from the client. You can't make it a VNC client. The latency is too high.
Maybe for a turn based strategy game you could go the VNC route and ensure absolute security. And you could open source the server and let people run whatever mods they want on the server. That could work.
FPSes present real engineering challenges though because you have to balance playability and hackability, and unfortunately FPSes are the most likely genre to get hacked in the first place. Start introducing open source mods into this mix and it's just ugly.
Something like a spectrum of "open source suitability / feasibility" for games would be :
Single player = Very suitable
Multiplayer turn based = Somewhat suitable
Multiplayer real time = Not very suitable
Multiplayer FPS (or other twitch game) = Not suitable at all
Senior Member
You don't get or even need source code to mod Bethesda's games. So this is a horrible example to support the "we need source code" argument.
Senior Member
The assets can be considered art, the engine is in no way art. And it's the engine that should be made free (not open-source, because I don't want to look at the source and not be able to modify it).
What are you talking about? Most of the games in HIB are written entirely by one person, so why would you need a corporation to fix a blunder they made? And no, I don't know OpenGL, but that does not mean I couldn't fix a misuse of other APIs used by the game. Most of the issues are with sound, not video anyway. And I can always LEARN OpenGL, I'm not too dumb to do it, just never needed until now.
Senior Member
Come on, take one look at Q3's fast sqrt routine and say with a straight face that it's not art
Senior Member
It's not art, it's black magic.Come on, take one look at Q3's fast sqrt routine and say with a straight face that it's not art
Posting Permissions
Reply With Quote