100 replies to this topic

[Angel of Annihilation]

Bilbo, on 23 September 2014 - 08:18 AM, said:

The question is not whether it's possible. The question is whether or not the maps could be generated with spawn points that don't bias the outcome of the match.

Doesn't matter because in the end it would balance itself out. Sometimes you might get shafted, others you would be able to give the shaft.

Also I have to say that when I first started playing they didn't even attempt to balance the teams by weight. Sometimes you would be facing all Assaults on the enemy team, others all lights and this asymetrical balance was some of the most fun gameplay I have had. I mean it actually felt like what you would experience in real life battles where the sides might never be equal.

My wishlist has always been very large maps with random objectives, random missions, limited respawns and asymmetrical teams fighting on the battlefield. It would be great to have this huge map that takes a light mech 10 mins to cross a full speed then toss out about 5 different victory point locations randomly on the map, then give each team a mission like defend Victory Points A, C, D with secondary objectives of destroying supply depot at Victory Points B and E. The other team would then have a mission like protect your supply points while destroying all enemy mechs. Then you would give each player a drop deck limited by tonnage and allow them to respawn in as many times as they have mechs in the drop deck. Finally the match timer would be extended to 1 hour. That is the game I wanted to play.

[Johnny Z]

Oh look another subject I was right about. Now that the No Mans Sky hype has been found to be all for nothing.

Hand made maps are way better. Period.

It aint easy being right all the time.

Edited by Johnny Z, 24 October 2016 - 03:18 PM.

[Troutmonkey]

Johnny Z, on 24 October 2016 - 03:17 PM, said:

Oh look another subject I was right about. Now that the No Mans Sky hype has been found to be all for nothing.

Hand made maps are way better. Period.

It aint easy being right all the time.

The biggest problem with NMS wasn't the procedural generation, but the lack of things to do. That, and excessive hype.

Having messed with perlin noise and other procedural generation they could do it, but PGI would be better served by creating ~3 spawn and base variants for each map. That alone would change how people cross the map and would help create more varied and interesting experiences on parts of maps that are currently unused

[Johnny Z]

Troutmonkey, on 24 October 2016 - 03:33 PM, said:

The biggest problem with NMS wasn't the procedural generation, but the lack of things to do. That, and excessive hype.

Having messed with perlin noise and other procedural generation they could do it, but PGI would be better served by creating ~3 spawn and base variants for each map. That alone would change how people cross the map and would help create more varied and interesting experiences on parts of maps that are currently unused

Ok. But procedural and hand made cannot be compared. Better to let the tech advance to allow for faster hand made maps rather than vice versa.

Star Citizen is doing it right since it mostly to create filler for planets where their POI (point of interest) isn't located to create a feeling of size, even if it is cosmetic. That's a good way to go.

X-Com 2 did procedural by make hand done tiles and mixing and matching using the roads as joiners.

MechWarrior Online is without any doubt doing it right. Really. The battle will only be getting more massive with air assets and other assets and open world/prodedural would have been like them shooting themselves in the foot at best.

If technology advances, which it will, these hand done maps can be used in other ways to allow for that. They could even do what Skyrim is doing and do texture updates etc. Lots of room to play with the route MechWarrior has taken.


I don't need to point out the best parts of No Man Sky game and Star Citizen's demo are hand made points of interest, or POI's.

Edited by Johnny Z, 24 October 2016 - 03:47 PM.

[LordKnightFandragon]

Yes its possible, Diablo 2 did it, Age of Empires did it, Warframe does it(Sorta), they use the same tile pieces within a tileset to build maps in different layouts every time.

BUt yeah, the system is possible..but people want the chokepoint, meta maps that they can wash, rinse repeat....

[Mister Blastman]

Johnny Z, on 24 October 2016 - 03:40 PM, said:

Ok. But procedural and hand made cannot be compared. Better to let the tech advance to allow for faster hand made maps rather than vice versa.

*sigh*

An entire procedurally generated planet. Beyond perfect for MWO.

Stop trolling...

Edited by Mister Blastman, 24 October 2016 - 04:37 PM.

[Deathlike]

It's doable, but PGI would actually have to hire someone that has that kind of experience.. and that would actually cost money.

So... besides seeing this necrothread exist, it will never happen.

[Johnny Z]

Mister Blastman, on 24 October 2016 - 04:37 PM, said:

*sigh*

An entire procedurally generated planet. Beyond perfect for MWO.

Stop trolling...

Very funny. Would work well as filler like how Star Citizen is doing it, but that's about it. Not even entirely impossible, then use the existing MechWarrior Online maps as POI's.

I hope you didn't mean those pictures as game play maps.

If MechWarrior Online did use the existing maps as POI on massive procedurally generated worlds with actual dropships coming in from space and that sort of thing then that wouldn't be for at least a couple years.

A couple years isn't bad because by then Star Citizen will have as many POI's as MechWarrior Online, but only a 1/3 the size or so. Maybe 1/5 the size? I don't know.

Think MechWarrior Online maps cant be used as POI's on procedurally generated maps for filler to make massive worlds? POI's 10x or 20x better than Star Citizen has? Think again.

It wouldn't be easy to do flawlessly but very doable. I know this from using the construction kit for Bethesda games. Crytek wouldn't be much different. I'm sure Star Citizen is having trouble adding their POI's flawlessly without seams as well. Maybe not I don't know.

Star Citizen will have a lot of things not currently in MechWarrior online though. That's keeping them busy instead of focusing on POI's for their planets entirely. Only 1 POI for Star Citizen has been revealed I know of, and its not to bad. But its mostly centered around a single(most likely broken into multiple parts for performance), quite amazing model(crash landed capital star ship), and a very few other lesser models(escape pods mostly).

Will Star Citizen reveal an entire city map as a POI for one of the planets, battle ready? Or other such amazing POI's? Not sure when, but maybe some day. I forgot that game does have an amazing station planet side interior to explore and land ship at.

Edited by Johnny Z, 24 October 2016 - 05:46 PM.

[jjm1]

nah.

At the end of the day its far easier to use cryengines terrain tools to hand craft some decent maps.

I was making massive terrain maps for UT 2004 out of high school. The Onslaught game-mode if anyone knows it. You capture nodes on a network to build a link to the enemy base. You know, something to do other than base camping.

PGIs problem is just no commitment to destroying and remaking FW so its actually fun to play.

[Ex Atlas Overlord]

Heffay, on 23 September 2014 - 08:23 AM, said:

It would make matches a complete crap-shoot. You can't secure known areas. Just wander around blindly and hope you meet a smaller group of enemies with a larger group.

Yeah, it's like you'd need some sort of mech designed to check out locations ahead of the main force to decide on a course of action.

Maybe we could call them something weird, like "scout".... and maybe make them faster than other mechs, maybe with less firepower or something to give them a specialized role...

Actually nevermind,....that'd be crazy.

Edited by Ex Atlas Overlord, 24 October 2016 - 08:49 PM.

[DAYLEET]

Bilbo, on 23 September 2014 - 08:18 AM, said:

The question is not whether it's possible. The question is whether or not the maps could be generated with spawn points that don't bias the outcome of the match.

Dont matter, no side knows the map, every side plays on a never before seen maps with four respawn at best to "learn" the map.

Back when i played SOF2 with randomly generated maps there was often a side with better defensive/offensive position and you would learn them as you respawn 100times but one side still has to kill the other and being defensive was never a good option for anything but stalling a game.

The people playing the map was the deciding factor. The side with the good people would win, no mather what side of the choke point they would be, assuming the map spawned with a choke point or with higher ground or foxholes on their side. The map also spawned rudimentary building and cover, sometimes out of places and all this made for a unique map and a unique experience every time.

On top of this SOF2 also had created solid MP maps that were also very popular. Both can coexist and both can be good.

jjm1, on 24 October 2016 - 08:25 PM, said:

nah.

At the end of the day its far easier to use cryengines terrain tools to hand craft some decent maps.

I was making massive terrain maps for UT 2004 out of high school. The Onslaught game-mode if anyone knows it. You capture nodes on a network to build a link to the enemy base. You know, something to do other than base camping.

PGIs problem is just no commitment to destroying and remaking FW so its actually fun to play.

2K4 is one of the best fps ive ever played and Onslaught was awesome... except for that reactor death that would wake up the neighbours in a 10km radius. That aside, saying crafting map is easy... well im not sure, ive had great games with awesome map and others no so much. Randomly generated fill a hole no vrafted map can, one where you feel like waging a war rather than play in an arena.

CW with static map isnt interesting in the least and completely break what could interest me in lore/immersion and RP. The only Static map option for CW should have been to recreate lore battle with "known" areas so which planet is always that map but no other planet has that map. Much like how we create WW2 battle in Red Orchestra or battlefield in known maps etc. Liberties would be taken but at the end of the day i could read a book and relive it maybe.

Edited by DAYLEET, 24 October 2016 - 10:45 PM.

[dwwolf]

A tile based approach could work.

The main problems for true procedurally generated maps are balance and deadend starting points.
Those could be coded out by strict rules/limits on the generator but that takes serious effort.
Im also not sure if the current engine supports generating and distributing maps on the fly.

[Appogee]

Guys guys guys.

It doesn't matter whether a procedurally-generated map is biased to one side or the other, because:

• war isn't fair
  
• the downside of imbalance is much less than the benefit of infinite variety of battlefield
  
• it creates a reason for scouting
  
• the advantage vs disadvantage would even out for everyone over time
  
• in FP we'd feel like we were actually taking planets in a diverse galaxy

Hell, even PGI's carefully crafted maps have well-known imbalances. (Most recent example is Terra Therma Redux, where the F7 chokepoint in the middle of the map significantly favors the same team.)

PGI can maintain its carefully-crafted balanced maps for tournaments where balance is more important.

[627]

Holy Necro, Batman!


We won't see proc maps, simply because russ thinks this game should be esport. So we need symetrical, mirrored maps for this. Ever wondered why there's no second map in LoL or Dota?
Yay esport... taking the last bit of fun out of a game....

[Kotzi]

Random maps where people are not doing the same old same 95% of the time because it worked last time somehow? Na, would have to think, not like this is wanted in this game.

[627]

Just curious, is there any MP game where you fight each other on random generated maps?

XCOM2 had proc maps afaik but a shooter?

[Wrathful Scythe]

I don't get the whole "Balance could be shifted to one side!" thing. Is that any different to what we have now? Certain maps with certain gamemodes are heavily one sided 100% of the times. I would rather take a procedural generated map that could be unfair for one side than have the same map beeing unfair all the times.

Though, I don't think this topic is worth discussing. The chances of this happening are even slimmer than expecting faction warfare to be good at any point in the future.

[Vellron2005]

Appogee, on 24 October 2016 - 11:13 PM, said:

Guys guys guys.

It doesn't matter whether a procedurally-generated map is biased to one side or the other, because:

• war isn't fair
  
• the downside of imbalance is much less than the benefit of infinite variety of battlefield
  
• it creates a reason for scouting
  
• the advantage vs disadvantage would even out for everyone over time
  
• in FP we'd feel like we were actually taking planets in a diverse galaxy

Hell, even PGI's carefully crafted maps have well-known imbalances. (Most recent example is Terra Therma Redux, where the F7 chokepoint in the middle of the map significantly favors the same team.)

PGI can maintain its carefully-crafted balanced maps for tournaments where balance is more important.

I agree with the points Apogee made.

Procedurally generated maps are the ONLY way to go if we are ever to have meaningful and immersive FW.

I think that it would be possible to have PGM's (procedural generated maps) of every type of environment. Even large cities, cose' I don't think that the entire map needs to be procedurally generated.

Why not have this approach: Take a large hand-crafted city block, like the whole River City map for example, than add miles of procedurally generated hills, forests or shoreline around it, and connect it to another hand crafted block, like the whole Crimson Strait map?

Hand crating mixed with procedural generation. Using hand crafting where PG can't cut it, and PG to fill the miles of gaps in rough wild terrain, so scouting and tactics in an unfamiliar terrain become important?

Also, I would pesonally LOVE the idea of random maps, that are not always fair for both sides, but neither side gets to know of the advantage until they scout it out.

This type of map making would make MWO a great game.

P.S.

I've gone and done it. Installed Star Citizen just to see what the hype is about..

[Appogee]

In addition: PG maps in FP would make Scouting missions meaningful.

No need to create 'capture' contrived intel beacons. Just run where the map is black, expose the terrain, and get to your escape dropship before the enemy kills you.

Bonus points for identifying key target locations, enemy force strength etc.

Just like *gasp* actual scouting.

Damn, how I wish we had a developer who cared enough to make this game all that it could be.

Edited by Appogee, 25 October 2016 - 12:51 AM.

[Varvar86]

here is my older post about alternative approach to FW
Main problems are:
players population
man made maps always better than randomizer - but man cant create 100500 maps for whole unuiverse

Edited by Varvar86, 25 October 2016 - 02:11 AM.