250 replies to this topic

[Mister Blastman]

ScarecrowES, on 13 February 2016 - 02:17 PM, said:

I've read your posts before. What you always describe is pointless for a game like MWO. Your argument has never made sense for THIS game. Games like these require a smaller number of well-tuned maps. Procedurally-generated maps won't work here.

I'm not against them in general, and I understand how they'd work alright in place of real maps for games like Elite: Dangerous... but NOT in competitive multiplayer games.

They would provide brilliant gameplay for competitive multiplayer. This is coming from someone (not to brag) who has held top competitive titles in StarCraft, QuakeWorld Team Fortress, Team Fortress 2, Mechwarrior 2: Mercenaries, Mechwarrior 3 and Mechwarrior 4. I don't like to ever bring that up. But some folks seem to feel this or that without ever coming from a top level position and recognizing the benefits.

Comp play thrives on increasing challenge.

Some like to memorize maps and exploit their nuances--but that's memorization and repetition. Constantly new maps and terrain pushes the player to fall back on their true skill and nothing more. My thread elaborates how procedurally generated maps can be successful. We have the technology now to do it.

I must also be honest in saying I have zero time for competitive play at the moment or the foreseeable future due to life challenges. But I love (and hate) MWO and want to see it thrive.

Edited by Mister Blastman, 13 February 2016 - 06:38 PM.

[Ratpoison]

Mister Blastman, on 13 February 2016 - 06:37 PM, said:

They would provide brilliant gameplay for competitive multiplayer. This is coming from someone (not to brag) who has held top competitive titles in StarCraft, QuakeWorld Team Fortress, Team Fortress 2, Mechwarrior 2: Mercenaries, Mechwarrior 3 and Mechwarrior 4. I don't like to ever bring that up. But some folks seem to feel this or that without ever coming from a top level position and recognizing the benefits.

Comp play thrives on increasing challenge.

Some like to memorize maps and exploit their nuances--but that's memorization and repetition. Constantly new maps and terrain pushes the player to fall back on their true skill and nothing more. My thread elaborates how procedurally generated maps can be successful. We have the technology now to do it.

I must also be honest in saying I have zero time for competitive play at the moment or the foreseeable future due to life challenges. But I love (and hate) MWO and want to see it thrive.

Praying to RNGesus for positional advantage takes a lot of skill, yeah. Your e-peen waving isn't going to impress me or make me think your idea is any less awful.

[Mister Blastman]

tortuousGoddess, on 13 February 2016 - 06:42 PM, said:

Praying to RNGesus for positional advantage takes a lot of skill, yeah. Your e-peen waving isn't going to impress me or make me think your idea is any less awful.

My epeen isn't important. I could care less. Notice no tier badge. Anyways, per my thread and previous posts, procedural does NOT equal random.

Procedural can be equal and balanced for all players and spawn positions. Mathematics is fun.

[KrazedOmega]

People will complain about anything.

To anyone who thinks this was a waste of time, well lets just remove all your mech camo, and paint colors, and cockpit items, because those have no effect on game play at all and are just a waste of resources.

A new pilot texture has been needed for a while. The one we currently have looks like something from 2005.

[Ratpoison]

Mister Blastman, on 13 February 2016 - 06:47 PM, said:

My epeen isn't important. I could care less. Notice no tier badge. Anyways, per my thread and previous posts, procedural does NOT equal random.

Procedural can be equal and balanced for all players and spawn positions. Mathematics is fun.

No. There will still be the "good" pieces of terrain cover and the "bad" pieces of terrain, and since you aren't going to make the maps symmetrical(since that would defeat the purpose of implementing it), it's going to have uneven sides at random. The only difference is that now instead of having a couple dozen maps to tweak and balance, you have HUNDREDS of terrain segments to try and balance, having to keep consideration of ALL the possible combinations that could result around it. It's a horrendous balancing nightmare.

[Johnny Z]

Well theres two kinds of procedural generation that are claiming to be that. One is just sections of larger maps arranged in a random order like how X-Com 2 did it. That would actually be possible for MechWarrior Online eventually and if enough maps are made like they are doing it now it can in some ways achieve the same thing.

True procedural generation like how "No Mans Sky" is doing it would be worthless for MechWarrior Online for every good reason. The quality isn't there, the AI would have to be zombie AI. It may work with mixed success for "No Mans Sky" for exploration reasons. And believe me the success will be mixed.

Its not impossible that one day MechWarrior could join all the snow maps in random ways creating a type of procedural generation. X-COM 2 did this using roads as the map joiner and randomly generating cars and stuff on the roads. This could be an extremely long way down the road but not even close to impossible. This is yet another reason why the way they are doing it is the right way.

Edited by Johnny Z, 13 February 2016 - 07:38 PM.

[CMDR Sunset Shimmer]

Requesting ability to choose gender.

[Johnny Z]

CMDR Sunset Shimmer, on 13 February 2016 - 07:51 PM, said:

Requesting ability to choose gender.

Your in the wrong topic. This topic is about maps and procedural generation. Um wait.

[CMDR Sunset Shimmer]

Johnny Z, on 13 February 2016 - 07:53 PM, said:

Your in the wrong topic. This topic is about maps and procedural generation. Um wait.

*reads the prior topics*

Wow, this thread got WAY the heck off topic.

On the OP's base topic, this new model looks awesome, frankly I've been tired of the starlegue era cooling suit for a long time now.

[Mister Blastman]

tortuousGoddess, on 13 February 2016 - 07:24 PM, said:

No. There will still be the "good" pieces of terrain cover and the "bad" pieces of terrain, and since you aren't going to make the maps symmetrical(since that would defeat the purpose of implementing it), it's going to have uneven sides at random. The only difference is that now instead of having a couple dozen maps to tweak and balance, you have HUNDREDS of terrain segments to try and balance, having to keep consideration of ALL the possible combinations that could result around it. It's a horrendous balancing nightmare.

I don't see the problem. The maps we have right now are full of bad pieces of terrain. Procedural generation, when the algorithm is properly coded, could do better.

Unless you're saying the maps we have are perfect--because they aren't.

Johnny Z, on 13 February 2016 - 07:31 PM, said:

True procedural generation like how "No Mans Sky" is doing it would be worthless for MechWarrior Online for every good reason. The quality isn't there, the AI would have to be zombie AI. It may work with mixed success for "No Mans Sky" for exploration reasons. And believe me the success will be mixed.

MWO has no AI. Your point is irrelevant. MWO is player versus player.

[Johnny Z]

Mister Blastman, on 13 February 2016 - 08:25 PM, said:

I don't see the problem. The maps we have right now are full of bad pieces of terrain. Procedural generation, when the algorithm is properly coded, could do better.

Unless you're saying the maps we have are perfect--because they aren't.



MWO has no AI. Your point is irrelevant. MWO is player versus player.

I wasn't even going to reply because you must be trolling. But if not, have you taken a moment to consider what your saying. Procedural map generation cant even come close, not even within light years of hand done maps at this time. In fact procedural map generation isn't very good at all in comparison. Like really bad.

I cant see it coming close to what can be done by hand for many decades. If then.

I could find screen shots of both types of maps but I'm not going to. One isn't even remotely close to the other. Even the theory of what procedural maps can do in unreleased games isn't even close.

Hand done sections of maps joined in random order is something else entirely.

Edited by Johnny Z, 13 February 2016 - 08:39 PM.

[Mister Blastman]

Johnny Z, on 13 February 2016 - 08:32 PM, said:

I wasn't even going to reply because you must be trolling. But if not, have you taken a moment to consider what your saying. Procedural map generation cant even come close, not even within light years of hand done maps at this time. In fact procedural map generation isn't very good at all in comparison. Like really bad.

I cant see it coming close to what can be done by hand for many decades. If then.

I could find screen shots of both types of maps but I'm not going to. One isn't even remotely close to the other. Even the theory of what procedural maps can do in unreleased games isn't even close.

I'm not trolling, I'm serious. You keep bringing up irrelevant points to defend your position. Procedural maps... work. This is terrain. By nature terrain is random and organic. It is not perfect. There are imperfections in it. That is what we need--we need maps that are fluid. We need maps that induce variable playstyles instead of... march to I9 and push the other team off the mountain first.

Or... head to the center of the volcano before they do or we lose.

Or... get the Citadel! Hurry!

Or... go camp in the basement, it is safe there.

Or... Suicide alley! Hurry! We can flank them in the narrow passage and nothing bad will happen to us if they take the high ground and shoot at us like fish in a barrel...

Or... cave rush!

You get the point.

Procedural maps will break us free from a mold we all are comfortable with and get us to focus on piloting battlemechs--big, giant robots and do what they do best--kill other robots.

It isn't about perfect terrain--it is about improving the often stale gameplay we already have. Procedural maps are the best way to do this as PGI can't afford to hire someone to make a map every month. It isn't sustainable.

[Nightshade24]

GRiPSViGiL, on 12 February 2016 - 04:43 PM, said:

Looks good but totally unnecessary to waste resources on but PGI gonna PGi.

In the end nothing is really 'necessary' for MW: O anymore.

We got a reasonable ballanced game that has mechanics and attractive aspects to it and hundreds of mechs where you can fight with or against your friends and so on and on. The game is already a game that works well. For the past year all there have been was improvements and this was one that is more than welcoming.
It isn't that hard to do it either... no collision meshes, no hit box, not many animations or colour / camo patterns in the dozens required, no economical changes or what ever. It's easier to do this than making a mech model by a long shot.

[MoonUnitBeta]

Mister Blastman, on 13 February 2016 - 08:25 PM, said:

Procedural generation, when the algorithm is properly coded, could do better.

/cringe

I hope you don't think that PGI can take on coding a fool-proof procedural generator? See you in the real-life year 3052.

To just quickly touch on this topic though, assuming it does get implemented wonderfully, I can't say that it would have really been worth the time... I mean, there comes a point where procedural generation will become stagnant (for a game of this type). Procedural Generation, while allowing infinite combinations, it's done so in infinitely incremental changes. It's tough to argue the inherent flaws of procedural generation when the counter-arguement is based on the assumption that it will be flawless. And honestly, I think procedural generated maps, "when the algorithm is properly coded", would probably be the biggest waste of coding resources this game would ever see.
By the time it's up and running as a beautifully coded generator, we could have had 8 times the game we do now, plus a pile of new maps that the art team spat out in the mean time.

Anyways. I just wanted to chip in. I think I remember your thread on it (I would have to re-read it though because it's been a while), but I do recall these thoughts whenever the topic came up.

Ciao for now.

[MauttyKoray]

n o x i D, on 12 February 2016 - 04:25 PM, said:

Every book I've read says they wear a cooling vest shorts and army boots! Still a big improvement!

I'm 100% okay with 'breaking lore' in this respect. Also, the only instance of this which I remember is the Clans from the animated show. Otherwise the books I've read (what few I have) talked about combat attire but never shorts.

[Mister Blastman]

MoonUnitBeta, on 13 February 2016 - 08:58 PM, said:

/cringe

I hope you don't think that PGI can take on coding a fool-proof procedural generator? See you in the real-life year 3052.

Well I believe anything is possible if you put your mind and effort to it. It sure beats waiting three years for them to fix an existing map with glaring problems on it.

[Johnny Z]

Mister Blastman, on 13 February 2016 - 08:43 PM, said:

I'm not trolling, I'm serious. You keep bringing up irrelevant points to defend your position. Procedural maps... work. This is terrain. By nature terrain is random and organic. It is not perfect. There are imperfections in it. That is what we need--we need maps that are fluid. We need maps that induce variable playstyles instead of... march to I9 and push the other team off the mountain first.

Or... head to the center of the volcano before they do or we lose.

Or... get the Citadel! Hurry!

Or... go camp in the basement, it is safe there.

Or... Suicide alley! Hurry! We can flank them in the narrow passage and nothing bad will happen to us if they take the high ground and shoot at us like fish in a barrel...

Or... cave rush!

You get the point.

Procedural maps will break us free from a mold we all are comfortable with and get us to focus on piloting battlemechs--big, giant robots and do what they do best--kill other robots.

It isn't about perfect terrain--it is about improving the often stale gameplay we already have. Procedural maps are the best way to do this as PGI can't afford to hire someone to make a map every month. It isn't sustainable.

Well now things are getting interesting. A fairly old argument. Not nearly as old as units vrs independent by a long shot but fairly old none the less.

In what world is it that almost everything is still done by hand. Not machines, that's right, the real one.

The industrial revolution was blown way out of proportion. Really it was. That auto generated maps will be even remotely close to hand done simply isn't going to happen. Ask anyone actually making games. Or those playing that bother to actually compare the two.

The tools that peeps use to build maps will improve. Nothing more. Improved by the peeps and for the peeps doing the building.

Edited by Johnny Z, 13 February 2016 - 09:23 PM.

[CMDR Sunset Shimmer]

Why don't you guys make a procedural generation thread, instead of discussing it on a topic about the coming MWO pilot model? because this is not at all what I, or other's, have come to this thread for.

[Mister Blastman]

Johnny Z, on 13 February 2016 - 09:11 PM, said:

Well now things are getting interesting. A fairly old argument. Not nearly as old as units vrs independent by a long shot but fairly old none the less.

In what world is it that almost everything is still done by hand. Not machines, that's right, the real one.

The industrial revolution was blown way out of proportion. Really it was. That auto generated maps will be even remotely close to hand done simply isn't going to happen. Ask anyone actually making games. Or those playing that bother to actually compare the two.

The tools that peeps use to build maps will improve. Nothing more. Improved by the peeps and for the peeps doing the building.

Are you arguing for or against yourself? Because it looks like against...

Industrial revolution--automation won. Look at all the stuff we have now! Thanks assembly lines! Thanks CAD! Thanks CNC machines!

We wrote the algorithms... the machines do the rest.

We now have a computer algorithm that can beat a human at Go! That's saying something.

[Johnny Z]

CMDR Sunset Shimmer, on 13 February 2016 - 09:12 PM, said:

Why don't you guys make a procedural generation thread, instead of discussing it on a topic about the coming MWO pilot model? because this is not at all what I, or other's, have come to this thread for.

Well its slightly related because this guy thinks that the pilot model would have been improved with procedural generation. LOL