43 replies to this topic

[Nightbird]

LordNothing, on 11 October 2020 - 04:24 PM, said:

there are limits to how much they can milk the franchise. after 2 mediocre games im not sure i want to buy into another mechwarrior game if pgi is the developer.

On the contrary, because everyone has such low expectations, if PGI actually designed a decent game with decent game mechanics, it is easy to create a profitable product. The art is fine, it just the team responsible for game mode size, game flow cycles, and match making that are underperforming.

[The6thMessenger]

Nightbird, on 11 October 2020 - 04:17 PM, said:

MWO is safer to test changes. Keeping a game running only involves paying for the servers. PGI has a lot of artists and such so they are working on MW5 until that money runs out. Even if the tweaks make MWO better, it's hard to draw people back to such an old game. On that front, if PGI learns how to deliver a polished product, MWO2 is the way to attract a lot of new customers. That been said, MWO2 without learning how to polish a game and you have a financial failure since the player base is unlikely to be as patient as with MWO.

Fair enough.

LordNothing, on 11 October 2020 - 04:21 PM, said:

im not sure mods can fix the game. its that bad. its like putting a coat of paint on a poo.

Give it time. You'd be surprised with what modders can do, if given the freedom. FO4 is littered with amazing modders, so much so that they are making FNV on FO4 engine, likewise the Fallout Miami.

Now the problem is that, well, we don't necessarily have the same community of modders, so we can only do is hope and wait.

[LordNothing]

Nightbird, on 11 October 2020 - 04:30 PM, said:

On the contrary, because everyone has such low expectations, if PGI actually designed a decent game with decent game mechanics, it is easy to create a profitable product. The art is fine, it just the team responsible for game mode size, game flow cycles, and match making that are underperforming.

maybe on their 5th or 6th attempt they will get it right.

The6thMessenger, on 11 October 2020 - 04:45 PM, said:

Give it time. You'd be surprised with what modders can do, if given the freedom. FO4 is littered with amazing modders, so much so that they are making FNV on FO4 engine, likewise the Fallout Miami.

Now the problem is that, well, we don't necessarily have the same community of modders, so we can only do is hope and wait.

ive seen/been part of modding communities that exploded rapidly and soon had more mods than players it seemed. this is the other end of the spectrum.

[The6thMessenger]

LordNothing, on 11 October 2020 - 05:27 PM, said:

ive seen/been part of modding communities that exploded rapidly and soon had more mods than players it seemed. this is the other end of the spectrum.

I think this is more of a problem of limited store release, only Epic, MWO launcher, and Microsoft store. Steam, which has a massive playerbase, and even a massive workshop community, is totally excluded.

[LordNothing]

The6thMessenger, on 11 October 2020 - 06:52 PM, said:

I think this is more of a problem of limited store release, only Epic, MWO launcher, and Microsoft store. Steam, which has a massive playerbase, and even a massive workshop community, is totally excluded.

thats another problem. most of the best mod communities are centralized. if you are going to have little isolated pockets of modders on each platform and there is little or no interaction between the groups, then you are going to have problems. you are going to have some modders that refuse to release on other platforms than the one they use. its going to be a mess.

lack of modding forums is also going to be an issue. and dont say discord. you cant discuss complex things with a chat room.

Edited by LordNothing, 11 October 2020 - 08:30 PM.

[The6thMessenger]

LordNothing, on 11 October 2020 - 08:29 PM, said:

thats another problem. most of the best mod communities are centralized. if you are going to have little isolated pockets of modders on each platform and there is little or no interaction between the groups, then you are going to have problems. you are going to have some modders that refuse to release on other platforms than the one they use. its going to be a mess.

lack of modding forums is also going to be an issue. and dont say discord. you cant discuss complex things with a chat room.

Modders do their own thing, they have their own teams, and if they want to collaborate, they will, and will find a way to do so --and many have done so, just consider FO4:NV and Miami. That is what I experienced modding at nexus.

[LordNothing]

The6thMessenger, on 11 October 2020 - 08:42 PM, said:

Modders do their own thing, they have their own teams, and if they want to collaborate, they will, and will find a way to do so --and many have done so, just consider FO4:NV and Miami. That is what I experienced modding at nexus.

some games have awesome modding communities. this is not one of them.

[The6thMessenger]

LordNothing, on 11 October 2020 - 08:47 PM, said:

some games have awesome modding communities. this is not one of them.

Sure. But I'd rather see what happens after it hits Steam. I think it's a problem of who has the game in the first place. Mods are shared all over the internet, and one of them is Nexus, unbiased with any platform.

Admittedly, FO4 is basically filled with attires, and soft-core porn, a lot of realistic guns applauded by guys who don't know **** about guns, as well as assets for settlement building. MW5 isn't any of that, so unless the entire modding team is dedicated in sexualizing an Atlas(#NoRobo), I don't think we'll see that much activity.

Edited by The6thMessenger, 12 October 2020 - 06:27 AM.

[RickySpanish]

Nightbird, on 11 October 2020 - 08:10 AM, said:

Eeesh... keep in mind PvE is available in MW5 which is the unreal engine available for MWO2, so it makes no sense to re-implement in cryengine. As for the cost, write down all the tasks you need to do and ballpark the dev time for each, it's not that hard.

Even in MWO2, with the AI from Unreal readilyavailable, it would cost more than 1 mil to create one game mode that is PvP with AI enemies in the middle for favor. From white boxing to release, there are so many steps.

Nah there's not that many steps, they already have a proof of concept with the simple pathing of the Escort Atlas. Rudimentary AI would need to path around a simplified navigation mesh, determine where flanks are, and understand the height of the lowest weapon equipped. That'd get you a moving, shooting bot that tries to attack from your weak sides and doesn't fire into walls often. That's not a great deal of advanced work. Also once again, a million dollars is ten years of dev time. With two or three developers which is probably the most you would want working on one feature, that's 3 years of development work. If you can't complete a single feature in less than that time with three senior developers I don't think money is the issue anymore.

[MW Waldorf Statler]

Quote

[color=#959595] navigation mesh, determine where flanks are, and understand the height of the lowest weapon equipped. That'd get you a moving, shooting bot that tries to attack from your weak sides and doesn't fire into walls often. That's not a great deal of advanced work.[/color]

Without Coders and Programmers thats can crack the personal modified and coded Engine its a Work over Years and the New Generation have nothing experience with a Old unmodified Cryengine 3.1 thats like a modern Steel Worker , thats now forced to build a Damast Sword in the Style of the 14th Century

[Nightbird]

RickySpanish, on 11 October 2020 - 09:06 PM, said:

Also once again, a million dollars is ten years of dev time.

If you have low paid devs. It is around 3 years including management overhead and logistics.

Maybe 4 years.

[RickySpanish]

Yeah, I meant in total dev time assuming a 100k salary. Newer devs being brought up on more comfy languages is only a problem for them, there are plenty of "hardcore" developers young and old interested in lower level programming. Even the junior dev at my work managed to implement rotation of three points about an arbitrary axis usind direct 3d... Eventually

[Nightbird]

RickySpanish, on 12 October 2020 - 04:38 AM, said:

Yeah, I meant in total dev time assuming a 100k salary. Newer devs being brought up on more comfy languages is only a problem for them, there are plenty of "hardcore" developers young and old interested in lower level programming. Even the junior dev at my work managed to implement rotation of three points about an arbitrary axis usind direct 3d... Eventually

Yes, me too, 100k salary = 130k in total compensation = 200k including management overhead = 230k with logistical costs.

[The6thMessenger]

Nightbird, on 12 October 2020 - 04:42 AM, said:

Yes, me too, 100k salary = 130k in total compensation = 200k including management overhead = 230k with logistical costs.

So when you said 100k worth of development, what is your estimate on flaying the Academy code, recycling some of it to make a game-mode, say:

PVE Incursion -- Incursion base with roving stock-mech patrols?
PVE Onslaught -- army of stock mechs moving through waypoints and then attacking the generators.

[Nightbird]

I don't think it would be interesting enough to be a stand-alone game mode, more an additional scenario for Academy, so no revenue opportunities or potential to increase the player base. Probably doable for 100k, still a hard sale.

[LordNothing]

RickySpanish, on 11 October 2020 - 09:06 PM, said:

Nah there's not that many steps, they already have a proof of concept with the simple pathing of the Escort Atlas. Rudimentary AI would need to path around a simplified navigation mesh, determine where flanks are, and understand the height of the lowest weapon equipped. That'd get you a moving, shooting bot that tries to attack from your weak sides and doesn't fire into walls often. That's not a great deal of advanced work. Also once again, a million dollars is ten years of dev time. With two or three developers which is probably the most you would want working on one feature, that's 3 years of development work. If you can't complete a single feature in less than that time with three senior developers I don't think money is the issue anymore.

that atlas "ai" was just a cheeseball path following algorithm. the mechs in the tutorial are just using turret ai (and thats just a bit of trig and animation code). these are merely components of an ai system. its like having a stick of ram and a hard drive and calling it a computer.

Edited by LordNothing, 12 October 2020 - 08:19 AM.

[RickySpanish]

LordNothing, on 12 October 2020 - 08:00 AM, said:

that atlas "ai" was just a cheeseball path following algorithm. the mechs in the tutorial are just using turret ai (and thats just a bit of trig and animation code). these are merely components of an ai system. its like having a stick of ram and a hard drive and calling it a computer.

Proof of concept as in a server controlled 'Mech being driven by input from a system that is not the player. It's like having a computer without the OS. Except they do actually have the "OS" in the form of AI code from MW5 whose principles could be applied to MWO. You could take the behaviour from MW5 and write a new interface to control the 'Mechs in MWO. Modern AI in games is perfectly capable of handling something in MWO, it won't be exceptional by any means but we did preface this all with "rudimentary".

[LordNothing]

RickySpanish, on 12 October 2020 - 08:35 AM, said:

Proof of concept as in a server controlled 'Mech being driven by input from a system that is not the player. It's like having a computer without the OS. Except they do actually have the "OS" in the form of AI code from MW5 whose principles could be applied to MWO. You could take the behaviour from MW5 and write a new interface to control the 'Mechs in MWO. Modern AI in games is perfectly capable of handling something in MWO, it won't be exceptional by any means but we did preface this all with "rudimentary".

you cant just copypasta code like that, there is work involved in just going between the ue way of doing things and the cryengine way of doing things. different function calls, different data structures. you do a line by line translation and it wont be long till you find a hangup. it many cases its easier to re-write the code from scratch than it is to try and move it.

[RickySpanish]

I know bruv, I've been porting our back catalogue of d3d games to the browser and it's a bit if work. But, writing code is quick, it's the tooling/engine and design work that takes the longest. If you know how your AI is supposed to work that's most of the difficulty out of the way. If you already have the means for your AI to drive a 'Mech that's even better. Writing code is just typing, I spend most of my time coding not typing at all, but thinking and sketching out how it's all going to fit together.

[Nightbird]

RickySpanish, on 12 October 2020 - 10:05 AM, said:

I know bruv, I've been porting our back catalogue of d3d games to the browser and it's a bit if work. But, writing code is quick, it's the tooling/engine and design work that takes the longest. If you know how your AI is supposed to work that's most of the difficulty out of the way. If you already have the means for your AI to drive a 'Mech that's even better. Writing code is just typing, I spend most of my time coding not typing at all, but thinking and sketching out how it's all going to fit together.

If you can implement AI for cheap, pitch that to PGI. You'll make AI mechs work in MWO for X $ in Y month, and if you fail you don't get paid. They won't be able to refuse.