Edited by lsp, 21 December 2014 - 08:14 PM.


Put The "war" Back In Mech Warrior
#61
Posted 21 December 2014 - 08:13 PM
#62
Posted 21 December 2014 - 08:19 PM
#63
Posted 21 December 2014 - 11:21 PM
Community Warfare was going to be about huge epic planet spanning conflict. Not, Here's a funnel to the enemy base, run along it and try not to explode.
So please, for the love of this game PGI, please make some big multi-path multi-objective maps. Go play Mechcommander (preferably one or Gold, though two wasn't bad). That's the kind of stuff we want, but BIGGER!
#64
Posted 21 December 2014 - 11:57 PM
As was stated the current CryEngine is limited to only about 8 Km's radius of terrain.
- After that it starts to lose detailed reference points for locations,
- no longer capable of precise calculations for reference points,
- plotting 24 Mech's for that scale or larger, with all the ray trace LOS checks starts to 'get fudgy'.
Another game had to deal with this limitation and needed to 'enhance the CryEngine Code' to extend the map sizes out from 10's of Km's to 1000's. (now working on 100,000's of Km ranges)
- 32 Bit floating point (current state), to 64 bit double precision calculations.
Equivalent to changing from a 10km map to a 1000km map with all the detail and precise reference points you could ever ask for.
- What is this other game you ask, Star Citizen. So you see why they would need 100's of 1000's of Km of usable space.
Some good news though, is that CryTech has looked at the changes that were made to the code and will be offering upgrades to the CryEngine that can use this 64bit double precision mode. Previously only very high end developers, business, military, government would even need that level of precision in there calculations.
- Will we see PGI look into altering the Engine to allow for these much larger areas?
- Would incorporating this code change break or render all before it useless?
The technology is now out there and available for PGI to allow for nearly any size map area they want, and with the current engine. If this is a direction they want to develop towards and not a game changer, especially for plans that have been previously designed around.
- But it's at least an option now.
Aim True and Run Cool,
9erRed
#65
Posted 22 December 2014 - 01:11 AM
Jack Corban, on 24 November 2014 - 02:28 AM, said:
Jack Corban
Have you tried CW? It addresses some of your isues, and is supposed to the "in character war" instead of the "out of character" fun of normal matches.
#66
Posted 22 December 2014 - 04:50 AM
El Bandito, on 25 November 2014 - 01:57 AM, said:
Hence I said the amount of loot raided/protected should be factored in the reward. Make the loot reward better than the kill reward and all is set.
Time sensitive material in the Convoy! So the Defenders HAVE to get the supplies to the objective instead of getting the kills. To much emphasis is on the fighting. Having a mission that is "This has to absolutely, positively get there by 0800 and not a millisecond later!", will change the paradigm once in a while. Double the pay out for completing the Primary objective over rampant killing.
#67
Posted 22 December 2014 - 05:13 AM
The Wakelord, on 22 December 2014 - 01:11 AM, said:
Have you tried CW? It addresses some of your isues, and is supposed to the "in character war" instead of the "out of character" fun of normal matches.
Yes i have and its the same brainless brawl that i expected it to be. 2 / 3 gates flanked by impassable Terrain with the attacking team stacking on one and the defending team stacking on the other side. No alternative routs. No real tactics involved apart from getting all on one team to push at the same time (head through wall tactic) resulting in a giant ball of Clusterfudge.
Attacks on bases should look like this.
And while i understand this is a Strategy game the rules still apply. Give the Player freedom of choice and stop limiting their movement by ostructing huge parts of the map with giant impassable mountains. Make bases consist of walls that are destructible to make additional entryways. And give the attacker additional points of intrest to attack (ie. Turret,gate,sensor controll buildings.) Add sub bases for generators that need defending and have their own defensive structures instead of cramping everything within 500 meters of each other.
This game could be so much more. It is however just another brainless shooter.
Edited by Jack Corban, 22 December 2014 - 05:48 AM.
#68
Posted 22 December 2014 - 05:15 AM
Edited by Joseph Mallan, 22 December 2014 - 05:27 AM.
#69
Posted 22 December 2014 - 05:16 AM
Jack Corban, on 24 November 2014 - 02:28 AM, said:
Discuss!
Finding any typos or grammatical errors? Keep em.
Cheers,
Jack Corban
On the same subject...
Bring back Tactical Gameplay (Assault mode).
A history of the dumbing down of tactical gameplay to the 'get to the best point and slaughter everyone' routine.
#70
Posted 22 December 2014 - 05:25 AM
They might have a transport system between maps in mind.
Since years ago I've been for planetside2 version of MechWarrior , Living Legends version and Moba version(titanfall).
#71
Posted 22 December 2014 - 08:29 AM
#72
Posted 22 December 2014 - 08:44 PM
The first is what might be the limitations of the Crytek engine, e.g. graphics budget (amount of textures, polygons, etc,.) This is probably the reason why you can't have AI bot mechs since each additional mech in the battlefield takes up that budget. Larger and more complicated maps also consume that budget.
The second is the problem with the map designers thinking they are still designing for a first person shooter. Lots of cramped spaces, bottlenecks, overcomplicated scenery, etc,. This is in contrast to Mechwarrior 4 where maps are designed like they were for a tank game where there is plenty of open spaces, lots of room to maneuver and maneuver fast, and where you have multiple lanes of attack.
Not all MWO maps are bad. I like Alpine, Canyons, HPG Manifold and Tourmaline. The rest seems like maps designed to bring mechs into a big meatball to brawl out. They emphasize mayhem over tactical maneuver. Unfortunately the latest maps, the Bog and the two CW maps are much into the tight spaced direction.
One of the beauties of the game Chromehounds was that their numerous maps are designed for tank warfare, big, expansive, plenty of room for maneuver, for defense and offense in depth. For that matter I would love to see Battlemechs stomping in battlefields like the maps I have seen for War Thunder and the World of Tanks.
#73
Posted 23 December 2014 - 09:27 AM
Anjian, on 22 December 2014 - 08:44 PM, said:
The first is what might be the limitations of the Crytek engine, e.g. graphics budget (amount of textures, polygons, etc,.) This is probably the reason why you can't have AI bot mechs since each additional mech in the battlefield takes up that budget. Larger and more complicated maps also consume that budget.
The second is the problem with the map designers thinking they are still designing for a first person shooter. Lots of cramped spaces, bottlenecks, overcomplicated scenery, etc,. This is in contrast to Mechwarrior 4 where maps are designed like they were for a tank game where there is plenty of open spaces, lots of room to maneuver and maneuver fast, and where you have multiple lanes of attack.
Not all MWO maps are bad. I like Alpine, Canyons, HPG Manifold and Tourmaline. The rest seems like maps designed to bring mechs into a big meatball to brawl out. They emphasize mayhem over tactical maneuver. Unfortunately the latest maps, the Bog and the two CW maps are much into the tight spaced direction.
One of the beauties of the game Chromehounds was that their numerous maps are designed for tank warfare, big, expansive, plenty of room for maneuver, for defense and offense in depth. For that matter I would love to see Battlemechs stomping in battlefields like the maps I have seen for War Thunder and the World of Tanks.
This!
#74
Posted 23 December 2014 - 10:07 AM
Jack Corban, on 23 December 2014 - 09:27 AM, said:
This!
This!
Lets dump the over complicated scenery and fancy doodads in favor of better physics. Lets face it, a lot of these doodads just blend into the back round and are not even noticed after awhile. If you cant interact with it, then get rid of it.
Edited by Deltron Zero, 23 December 2014 - 10:14 AM.
#75
Posted 25 December 2014 - 11:48 AM
Will MWO still have a healthy player base after all that time? I feel like PGI made a mistake by concentrating their initial efforts on Assault and skirmish, when they should of put emphasis on CW and huge objective based maps from the beginning, since setting up Deathmatch-15 min games seems less complex than this whole CW buisness.
Now I would have the patience to see all of this implemented, but would their be enough of a player base willing to wait that long to get all this implemented? I think realistically we'd be very few, so few that the game would die out by then.
That being said, I'm probably being too pessimistic, who knows, maybe we'll see the CW like we've all dreamed of being flushed out in the matter of a few months to a year.
Don't get me wrong I love PGI, I mean they decided to attempt to revive a franchise that seemed to be evaporating from existance, but from the early stages of beta to this very moment, I still can't say I'm playing a legitimate successor/reboot of mechwarrior. I just accept PGI's "error" (who knows maybe flushing out Deathmatch variant modes first turns out to be better in the longrun) and cling on to hope that the player base keeps on keeping on...I mean despite this, the mechs have been reworked to perfection and the combat is rather solid, it's a shame if all of the good stuff PGI has done would go to waste.
I'll keep playing MWO despite it's current arena shooter feel since it's been a franchise that I've loved for so long and I'll keep playing till I'll be the last damned guy in queue waiting to get a match that will never come (hopefully it doesn't end up like this but who knows)
#76
Posted 25 December 2014 - 05:49 PM
o0Marduk0o, on 24 November 2014 - 03:15 AM, said:
I would suspect this be the case too.
But then add in time sensitive, multiple mini objectives spread over multiple parts of the map (not all required to win, but make the win easier, turns base turrets on onto enemies, turns on Radar to give info on enemy side - untill the other side destroy it) and then you have multiple groups split in multiple ways to fight off unknown groups of enemies.
#77
Posted 04 June 2015 - 12:33 AM
#78
Posted 04 June 2015 - 02:34 AM

You had to bring it back from the dead?
It was a good topic, but ran its course.
#79
Posted 04 June 2015 - 02:46 AM

#80
Posted 04 June 2015 - 02:52 AM
8 user(s) are reading this topic
0 members, 8 guests, 0 anonymous users