58 replies to this topic

[Brody319]

LordKnightFandragon, on 29 January 2015 - 03:17 PM, said:

I kinda wish MWO would go with proper scale, even if it cost us some FoV.....make us feel like were in mechs.....

Strangely enough, only game that ever made me feel like I was beast mode in a mech was BF2142....prolly why I played it so much. Verdun Map, L5 Riesig, rushing tanks and mowin' em down, plus dat dual minigun it had...gun walker....

How it would feel to pilot the direwolf

[LordKnightFandragon]

Brody319, on 29 January 2015 - 03:19 PM, said:

How it would feel to pilot the direwolf

Why? Cuz everything else would be so far away? Thats basically how it would be...being 2-3 stories up and stuff....ofc, I am on a 2nd story of a house, stuff doesnt seem that small really....it would be nice to have the feeling of being in a fairly small, cramped cockpit and feeling 2-3 stories up.

[IronChance]

PvE would be wonderful and definitely needs to happen on some level. However, please keep in mind that the game would be going from no AI code to having some sort of AI bot code. This is "non-trivial" as they say. I know it's on PGI's radar, but even if they hire some top flight AI programmers today and started coding, it'd likely be a year before anything worthy of testing came out. Well, maybe 6 months if they dropped everything else, which I would not be in favor of them doing.

Anyway, I just thought I'd point out that "turret code" is NOT AI code and is insanely hard to get right.

Or even get to a point where, if it's not right, it's at least not horrible.

[Tombstoner]

IronChance, on 29 January 2015 - 03:25 PM, said:

PvE would be wonderful and definitely needs to happen on some level. However, please keep in mind that the game would be going from no AI code to having some sort of AI bot code. This is "non-trivial" as they say. I know it's on PGI's radar, but even if they hire some top flight AI programmers today and started coding, it'd likely be a year before anything worthy of testing came out. Well, maybe 6 months if they dropped everything else, which I would not be in favor of them doing.

Anyway, I just thought I'd point out that "turret code" is NOT AI code and is insanely hard to get right.

Or even get to a point where, if it's not right, it's at least not horrible.

Or worse game breaking.

Any sort of PVE must start with mechs. Simply from a resource perspective. Making an object move through a field of obstacles is going to create a huge amount of collision detection that needs to be done server side or client side where its vulnerable to hacking. if hacking is an issue (see through walls) then network demands increase. Not sure how game security is affected.

Collision detection is done automatically by the player. Automating it is not a trivial task. get this working on mechs first then worry about and AI telling the mech where to go.... then worry about how to dumb the AI down so its not a 100% head shot OP monstrosity.... a Cone of fire perhaps.

[LordKnightFandragon]

Tombstoner, on 29 January 2015 - 03:52 PM, said:

Or worse game breaking.

Any sort of PVE must start with mechs. Simply from a resource perspective. Making an object move through a field of obstacles is going to create a huge amount of collision detection that needs to be done server side or client side where its vulnerable to hacking. if hacking is an issue (see through walls) then network demands increase. Not sure how game security is affected.

Collision detection is done automatically by the player. Automating it is not a trivial task. get this working on mechs first then worry about and AI telling the mech where to go.... then worry about how to dumb the AI down so its not a 100% head shot OP monstrosity.... a Cone of fire perhaps.

WHy is AI so hard? games many years older have made AI with all of the above and it works fine. I wont act like its easy to do, but many games before have done it, why couldnt PGI also? MW4 has AI and they can make for quite an amusing fight.

As for the AI aiming, it seems in MW4, the AI always aims for Center mass. or maybe, PGI adds a sort of TT type mechanic that will determine where the AI will aim, then as the AI is easy/medium/hard, it has a less chance to aim anywhere but the CT. No head aiming. Have it be up to the player to twist damage away as we have it now.

Edited by LordKnightFandragon, 29 January 2015 - 04:20 PM.

[Kiiyor]

Alistair Winter, on 29 January 2015 - 02:10 PM, said:

I wonder how many infantry this game could take, even in a PvE match with only 1 mech (or 2-4 mechs if you allow coop).

I have a sneaking suspicion that a full company of infantry, in addition to tanks, LRM carriers and hovercraft, would set most people's computers on fire.

Not to mention the game servers. There's already a staggering number of calls made to them, and as I understand it 12v12 pushes that to it's absolute limit.

[LordKnightFandragon]

Kiiyor, on 29 January 2015 - 04:29 PM, said:

Not to mention the game servers. There's already a staggering number of calls made to them, and as I understand it 12v12 pushes that to it's absolute limit.

yeah, its why the game should only be a 8v8.....where did 12v12 start from? Sometimes I watch older battle videos of 8v8 and I notice there is no turtling behind a rock, there is no hailstorm of LRMs, there is no deathballing. Rather those games I see the players all moving about, they are spread out, there is 1v1, 2v1 and the gameplay just looks to be alot more enjoyable. Mechs dont simply melt before your eyes, mechs take awhile ot kill.....the game just looks to be 1000x better overall.

[terrycloth]

AI could run on the servers and would cause minimal additional network load -- there'd be no extra data sent to the servers from the client, just a slightly bigger packet sent to the client from the server batched with everything else.

It's nothing like adding more players.

Mech AI would be really hard. Tank and especially infantry AI could be dead simple because they're not trying to be mechs and have fewer options. It'd be a step up from a turret but not a moon-landing type deal.

[Brody319]

I think people are VASTLY over estimating the strain a few troop sprites, and some tanks would put on the servers. Especially since they need about as much AI as base turrets.

[BourbonFaucet]

I may post my "proposal" for PvE and what it could be like soon, been working on it in a word processor for a bit...

[IronChance]

LordKnightFandragon, on 29 January 2015 - 04:19 PM, said:

WHy is AI so hard? games many years older have made AI with all of the above and it works fine. I wont act like its easy to do, but many games before have done it, why couldnt PGI also? MW4 has AI and they can make for quite an amusing fight.

As for the AI aiming, it seems in MW4, the AI always aims for Center mass. or maybe, PGI adds a sort of TT type mechanic that will determine where the AI will aim, then as the AI is easy/medium/hard, it has a less chance to aim anywhere but the CT. No head aiming. Have it be up to the player to twist damage away as we have it now.

Games that launch with a single player component have AI programmers working on them from the beginning. They have entire AI teams, in fact. I doubt there is a single programmer at PGI that specializes in AI currently.

AI is difficult because pathfinding. It's difficult because tanks needs to know how to pathfind AND not run over their own infantry. It's difficult because bots need to learn how to use cover. etc, etc.

Yes, all this stuff is done and done well in other games, but that's because those other games worked on it from the start. I'm not saying it can't be done, just that doing PvE right would require a big commitment that might eat resources we'd like to see devoted to fleshing out community warfare.

[Alistair Winter]

Kiiyor, on 29 January 2015 - 04:29 PM, said:

Not to mention the game servers. There's already a staggering number of calls made to them, and as I understand it 12v12 pushes that to it's absolute limit.

Well, I don't know how this stuff works. In 12 vs 12, there's the constant input of 24 players. Every button pushed, every tiny movement of the mouse needs to be registered and calculated. I don't know which bit is more taxing for the servers. Is it collecting the input, or calculating what actually happens in-game, or communicating the results back to the players?

Because I imagine reducing the number of players to 1-4 should help a lot with server stress. But then, if 24 people decide to play PVE, then that is between 6 and 24 times as many matches for the servers to work out.

Things are looking bleak for COOP, I guess. I would be fine with just a single player PVE experience, whether it's waves or a single player campaign. Maybe PGI doesn't have the resources to do 4 man COOP, like so many other games.

[Kiiyor]

Alistair Winter, on 30 January 2015 - 12:27 AM, said:

Well, I don't know how this stuff works. In 12 vs 12, there's the constant input of 24 players. Every button pushed, every tiny movement of the mouse needs to be registered and calculated. I don't know which bit is more taxing for the servers. Is it collecting the input, or calculating what actually happens in-game, or communicating the results back to the players?

Because I imagine reducing the number of players to 1-4 should help a lot with server stress. But then, if 24 people decide to play PVE, then that is between 6 and 24 times as many matches for the servers to work out.

Things are looking bleak for COOP, I guess. I would be fine with just a single player PVE experience, whether it's waves or a single player campaign. Maybe PGI doesn't have the resources to do 4 man COOP, like so many other games.

All of it - though Matt from the tech side of PGI made a post about it a while back, and apparently the bigger issues revolve around the large volumes of calls that can be made at one time - especially where large numbers of mechs are firing large numbers of weapons at once.

[monk]

I guarantee I could get a lot of friends back into the game if it had decent PVE coop missions, and I know they'd spend money on the game if the PVE was good, as well. Not everyone loves PVP.

[LordKnightFandragon]

monk, on 30 January 2015 - 01:01 AM, said:

I guarantee I could get a lot of friends back into the game if it had decent PVE coop missions, and I know they'd spend money on the game if the PVE was good, as well. Not everyone loves PVP.

Amen....This game's PVP is lame...then again, MW PVP always kinda has....

I hopped into a few PVP in MW4....and it was unlimited heat PPC Cap poptarts everywhere.....its always just meta mechs...

PVE, there is no meta, you play however you want and you go out, have fun and win...shoot some Erobots and have fun....

If ever I could get Mektek working on my PC, id be over playing that over forumwarrioring it up here....

Edited by LordKnightFandragon, 30 January 2015 - 01:16 AM.

[Rebas Kradd]

DEMAX51, on 29 January 2015 - 02:49 PM, said:

I never said they needed to go balls-out and make a ton of new assets. I'm just pointing out how some form of AI/PvE would be a huge boon for the game, and is easily as important as most of the other things you've mentioned.

Nothing is more important than basic playability. The confusing, laborious nature of UI2.0 is actively driving people away and holding others out. The lack of new player tutorials is choking off the new player stream. The problems with hit registration, FPS performance, and declining graphical fidelity are quickly becoming credibility issues for this developer.

These are all priorities over AI/PvE. I acknowledge that AI/PvE has the potential to flesh out new gamemodes, and new gamemodes are absolutely vital to the long-term health of the game. There is in fact a huge untapped market for MWO yet, consisting of people who haven't even started playing because of the lack of a co-op campaign. So it needs to happen eventually.

But there are priorities even higher than that. New gamemodes won't help if their customers can't get into the game or navigate the UI.

Brody319, on 29 January 2015 - 02:51 PM, said:

not like a tank is very complex, and the humans could be sprites which are easy enough to make, and they wouldn't take very many resources from servers.

Everything is complex for this developer. They're undermanned, and it takes forever even for simple things like color-coded death messages to get through the usual development cycle.

Edited by Rebas Kradd, 30 January 2015 - 07:52 AM.

[Asakara]

Rebas Kradd, on 29 January 2015 - 02:14 PM, said:

No offense, but...

* UI2.5

* New player experience

* Hit registration

* Performance optimizations

* Heavy tweaking of original maps

* Ghost Heat revisitation

* Player tools (spectator, social, in-game polling)

These are ALL things that need to pass through PGI's design/review/QA process before they go anywhere near new AI assets. I'm all for new gamemodes any day, but we've got the Steam release to be thinking about.

Rebas Kradd, on 30 January 2015 - 07:51 AM, said:

...

These are all priorities over AI/PvE. I acknowledge that AI/PvE has the potential to flesh out new gamemodes, and new gamemodes are absolutely vital to the long-term health of the game. There is in fact a huge untapped market for MWO yet, consisting of people who haven't even started playing because of the lack of a co-op campaign. So it needs to happen eventually.

...

Indeed there are greater priorities.

All I did was post my suggestion that they do not include Mechs, at least initially, when they create PvE and gave 11 reasons why I thought that. I tend to post suggestions once in a while like Paying MC To Remove My Pants, Infantry And Armor Consumables, CBARS, Elemental Consumables, etc...

Did my OP seem like I was requesting that PvE had to come at a certain time or before anything else?

Edited by Asakara, 30 January 2015 - 08:06 AM.

[Rebas Kradd]

Asakara, on 30 January 2015 - 08:05 AM, said:

Did my OP seem like I was requesting that PvE had to come at a certain time or before anything else?

You weren't really clear. You suggested they start, and I simply wanted to reiterate the community's wishes.

It's true that AI might be good for the new player experience.

[Asakara]

Rebas Kradd, on 30 January 2015 - 08:10 AM, said:

You weren't really clear. You suggested they start, and I simply wanted to reiterate the community's wishes.

It's true that AI might be good for the new player experience.

I apologize for being confusing.

I was suggesting that when they do start working on PvE they do so without mechs, at least initially.

[Joe Mallad]

Belphegore, on 29 January 2015 - 02:09 PM, said:

I doubt we'll ever see pve in the lifespan of this game.

russ already said its coming.