93 replies to this topic

[Belorion]

Converting maps to be procedural would make drops more dynamic, and be somewhat of an equalizer for people who don't play as much. Dropping on the same maps over and over you start to understand the strategy of the particular map, if they changed this would add an element of newness to each drop.

[Nightbird]

Great idea, can you pass them 5 million dollars to do it?

[martian]

Belorion, on 20 October 2020 - 07:00 AM, said:

Converting maps to be procedural would make drops more dynamic, and be somewhat of an equalizer for people who don't play as much. Dropping on the same maps over and over you start to understand the strategy of the particular map, if they changed this would add an element of newness to each drop.

Generally speaking - yes.

Practically speaking: What if it happens that such map has some inaccessible spots that must be passed through to achieve the objective (for example to cap the base or destroy HPG station or so)?
Or even worse, what if you spawn on some such spot or get stuck there?

[MW Waldorf Statler]

1.you can not converted handcrafted Maps to Procedual Terrain .thats like a Handcrafted Mona Lisa Pic, and a Pic from a AI created from algorithms

2.not with a Cryengine 3.5 Version personal coded from Guys thats long gone ,and without each Programmers and coders ,PGI have only a UE4 Crew

Edited by MW Waldorf Statler, 20 October 2020 - 06:02 PM.

[LordNothing]

MW Waldorf Statler, on 20 October 2020 - 05:59 PM, said:

1.you can not converted handcrafted Maps to Procedual Terrain .thats like a Handcrafted Mona Lisa Pic, and a Pic from a AI created from algorithms

2.not with a Cryengine 3.5 Version personal coded from Guys thats long gone ,and without each Programmers and coders ,PGI have only a UE4 Crew

im not to sure of that, sounds like something a neural net could do. given arbitrary geometry, create a seed that will generate said geometry. i think the minecraft community does stuff like that trying to crack seeds of worlds from even grainiest of screencaps. that said it probably far to computationally intensive to be viable.

i only know what cry engine is capable of from crysis. and its far more than what pgi has utilized. i think the living legends crew did a better job in leveraging those features. day night cycle, gravity, dynamic terrain features, underwater areas. pgi has leveraged very few of these things.

frankly i dont think id miss the static maps. having to learn the terrain for every game you play means no more exploiting known flaws in map geometry. would certainly do away with the home field advantage so to speak. pros have gone over the maps millimeter by millimeter to find the ideal positions on a map. something thrilling about being on an alien world youve never seen before with a hand drawn map and having to complete some mission.

Edited by LordNothing, 21 October 2020 - 04:47 AM.

[Lanzman]

Procedural maps would be cool, but if that's a bridge too far then release the mapmaking stuff to the community and let the player base make 'em. Yeah, most of 'em would suck but at least there'd be more maps and exploring them . . . even bad ones . . . would be fun.
You guys remember fun, right?

[Absaint]

I also thought about a procedural map when i saw one in the latest pubg, Might be hard to do, but this would definitly bring new life to this game!

[martian]

Lanzman, on 21 October 2020 - 05:27 AM, said:

Procedural maps would be cool, but if that's a bridge too far then release the mapmaking stuff to the community and let the player base make 'em. Yeah, most of 'em would suck but at least there'd be more maps and exploring them . . . even bad ones . . . would be fun.
You guys remember fun, right?

Read this:

Matt Newman, on 07 August 2020 - 09:10 AM, said:

No go on the map editor.

[Belorion]

MW Waldorf Statler, on 20 October 2020 - 05:59 PM, said:

1.you can not converted handcrafted Maps to Procedual Terrain .thats like a Handcrafted Mona Lisa Pic, and a Pic from a AI created from algorithms

2.not with a Cryengine 3.5 Version personal coded from Guys thats long gone ,and without each Programmers and coders ,PGI have only a UE4 Crew

^ This isn't necessarily true, maps can be build from components which helps mitigate inaccessible spots. Testing is the name of the game here. proc maps are becoming the norm in games not the exception. There are strategies to help make sure there are not any major flaws in the resulting maps.

Edited by Belorion, 21 October 2020 - 06:50 AM.

[The pessimistic optimist]

I LOVE procedural generated maps if there done well but that's a pipe dream

[RickySpanish]

Nightbird, on 20 October 2020 - 07:02 AM, said:

Great idea, can you pass them 5 million dollars to do it?

Your cost estimates, as always, are hilariously off base

[Nightbird]

RickySpanish, on 21 October 2020 - 09:33 AM, said:

Your cost estimates, as always, are hilariously off base

Go ahead, write down yours...

[Thorqemada]

SirSmokes, on 21 October 2020 - 06:51 AM, said:

I LOVE procedural generated maps if there done well but that's a pipe dream

Minecraft does that really well - they should make a Minecraft-Battletech/Mechwarrior-Game!

I mean it not only has destructible Buldings, the Terrain also is destructible...

Edited by Thorqemada, 21 October 2020 - 10:19 AM.

[The pessimistic optimist]

Thorqemada, on 21 October 2020 - 10:18 AM, said:

Minecraft does that really well - they should make a Minecraft-Battletech/Mechwarrior-Game!

I mean it not only has destructible Buldings, the Terrain also is destructible...

That is a very odd idea

Edited by SirSmokes, 21 October 2020 - 10:28 AM.

[RickySpanish]

Nightbird, on 21 October 2020 - 09:43 AM, said:

Go ahead, write down yours...

About 500k for the development work including two developers for two years at 100k. Let's assume it also takes two artists the same amount of time, so another 500k (them some highly paid artists but whatever). We're at a million now, but I imagine we will need a level designer and tools developer - we'll give them a year, allowing the developers and texture artists some time to get going, so that's 200k. Now we just need to ship it off to QA and let's be honest, they get paid in food, so we'll call the whole thing 1.2 million even. These are some overpaid guys, but what the heck.

[Nightbird]

RickySpanish, on 21 October 2020 - 01:46 PM, said:

About 500k for the development work including two developers for two years at 100k. Let's assume it also takes two artists the same amount of time, so another 500k (them some highly paid artists but whatever). We're at a million now, but I imagine we will need a level designer and tools developer - we'll give them a year, allowing the developers and texture artists some time to get going, so that's 200k. Now we just need to ship it off to QA and let's be honest, they get paid in food, so we'll call the whole thing 1.2 million even. These are some overpaid guys, but what the heck.

Well, here's the thing, you can't just make people and money magically appear. A workable plan has to have a very good chance of generating more income than it costs. Your approach assumes that you have or can hire cryengine v3.5 devs, which you can't today, that it's ok to put work into a dead-end engine, which it's not, and that people will magically pay money if the procedurally generated maps appear, they won't. One way trip to bankruptcy.

This version of Cryengine is dead, PGI has no cryengine devs, but they've worked on UE and have UE procedurally generated maps already. So, while you don't need to create the map tech anymore, you need to build the entire game around it. From UI to game modes, to net code, to inventories, and game flows. Finally, the monetization strategy, since selling mech packs will not work again. Basically building MWO2 which comes with work for a full team for 2 years. Lastly, marketing, to pull people in. 5 million is the bottom of what it will cost.

[LordNothing]

Nightbird, on 21 October 2020 - 02:02 PM, said:

Well, here's the thing, you can't just make people and money magically appear. A workable plan has to have a very good chance of generating more income than it costs. Your approach assumes that you have or can hire cryengine v3.5 devs, which you can't today, that it's ok to put work into a dead-end engine, which it's not, and that people will magically pay money if the procedurally generated maps appear, they won't. One way trip to bankruptcy.

This version of Cryengine is dead, PGI has no cryengine devs, but they've worked on UE and have UE procedurally generated maps already. So, while you don't need to create the map tech anymore, you need to build the entire game around it. From UI to game modes, to net code, to inventories, and game flows. Finally, the monetization strategy, since selling mech packs will not work again. Basically building MWO2 which comes with work for a full team for 2 years. Lastly, marketing, to pull people in. 5 million is the bottom of what it will cost.

minus inevitable budget overruns when their plans fail to work.

[Belorion]

Is there any reason to think an engine upgrade is off the table?

[Nightbird]

It'll take 2-3 years to do. I think it's guaranteed to happen but the new game will be called MWO2 and nothing from MWO will carry over.

[Belorion]

The last time they did an engine upgrade it did not take 2-3 years.

That is a bit of a heavy estimate.