67 replies to this topic

[o0m9]

Anjian, on 24 June 2017 - 06:28 PM, said:

If you want to know, mech type games and their game engines:

Hawken --- Unreal 4

Hawken runs on UE3, unless they've actually got off their asses and done something in the last year, which I kind of doubt.

...

I miss Hawken.

[MustrumRidcully]

MechaBattler, on 23 June 2017 - 11:39 PM, said:

Anyone with programming experience, is it possible for them to use their work on MW5 as a base for recreating MWO in Unreal 4?

Everything is possible. Practical, however?

Well, look at the Battletech Kickstarter. They are actually using MW:O assets in part in a different engine. It obviously works. But the project is now 2 years in the making. You can expect that it would also take MW:O a year or two to make such a transition. A year or two where they do nothing else but this transition. No new maps, no new game modes, no new mechs, no quality of life changes, absolutely nothing else. They would be busy recreating what they already have, instead of making new stuff.

[LordNothing]

porting becomes a better possibility closer to the mw5 launch date. i bet how well sales and preorders for mw5 do the higher the probability of them porting. i dont want to see an mwo2 as another f2p game where i have to rebuy all my stuff again.

if the port is accompanied by a lot of mech packs, and a lot of new big features you might be able to reboot the game without needing to make everyone rebuy. bring in a lot of new features that ue4 uses. like that procedural map feature. do weapon manufacturers and more rpg elements. spare no expense and make it awesome. thats how you get the players back.

Edited by LordNothing, 24 June 2017 - 11:23 PM.

[Anjian]

o0m9, on 24 June 2017 - 10:12 PM, said:

Hawken runs on UE3, unless they've actually got off their asses and done something in the last year, which I kind of doubt. ... I miss Hawken.

You're right, its UE3.

Hawken is getting a PC relaunch soon, on Steam.

Its Heavy Gear Assault that is on UE4.

[Anjian]

MustrumRidcully, on 24 June 2017 - 10:50 PM, said:

Everything is possible. Practical, however?

Well, look at the Battletech Kickstarter. They are actually using MW:O assets in part in a different engine. It obviously works. But the project is now 2 years in the making. You can expect that it would also take MW:O a year or two to make such a transition. A year or two where they do nothing else but this transition. No new maps, no new game modes, no new mechs, no quality of life changes, absolutely nothing else. They would be busy recreating what they already have, instead of making new stuff.

You can't use the Battletech kickstarter as an example of a game port. The Battletech game had to be made from the ground up. Actually I thought it went faster than I expected. Crafting the game on the game engine, which is Unity 3D, is likely the least of their bottlenecks.

[o0m9]

Anjian, on 24 June 2017 - 11:19 PM, said:

You can't use the Battletech kickstarter as an example of a game port. The Battletech game had to be made from the ground up. Actually I thought it went faster than I expected. Crafting the game on the game engine, which is Unity 3D, is likely the least of their bottlenecks.

Porting to a new system would probably be about as intensive. You can't just copy/paste old game code into the new engine and be done with it, you have to redevelop everything over again.

This is partly why Hawken has languished so long, the new company working on it is doing major updates to the back end (or was, apparently the relaunch was just last month) to get it up to date. Moving to anew engine would probably require PGI to rebuild most of their infrastructure from scratch, depending on how much they can transfer over from the single player game.

[Anjian]

o0m9, on 24 June 2017 - 11:47 PM, said:

Porting to a new system would probably be about as intensive. You can't just copy/paste old game code into the new engine and be done with it, you have to redevelop everything over again. This is partly why Hawken has languished so long, the new company working on it is doing major updates to the back end (or was, apparently the relaunch was just last month) to get it up to date. Moving to anew engine would probably require PGI to rebuild most of their infrastructure from scratch, depending on how much they can transfer over from the single player game.

Old games have a lot of these dependencies, sometimes, rooted in bad coding practices.

Its really how the game is coded.

Modern games using 'modern' game engines that feature a high amount of portability --- e.g. Unity and Unreal --- might be something else.

Hawken languished but that's more the fault of its parent company failing to develop a proper business model, then don't know what to do with it, as the founding capital runs out.

Often its better to rebuild infrastructure from scratch, this way, you remove the ills from the past. You don't want the new game to inherit problems and dependencies from the previous infrastructure.

What you preserve is the overall game design, the content, such as the maps, the models, the stats on weapons, avatars, etc,. --- a lot of the soft things that had been worked out as part of the game's balance and requires their own and extensive play testing and game experience.

[Valhallan]

The problem is, is it worth it for them to relaunch. Frankly unless they plan to actually do CW properly it's really not worth porting over MWO to just get a bit shinier. There is 0 new/old people draw in that. Not to mention the increased cost of using the engine, cryengine was probably chosen in the first place because it's so cheap, 9.90$ a month. Unreal is only cheaper if you make less than 1200$ gross per month, since they use a 5% royalty for funds after 3000$ per quarter. Unity meanwhile has a fat subscription fee based on gross that caps at 125$/month. Sure for current players a bit shinier in general sounds better than the likely 1 year of mechpacks. But it makes 0 sense for their end if they can't leverage that into more players (which they can't without some magnificently awesome endgame CW to make people care).

[Anjian]

The initial cost of CryEngine might be cheaper, but due to the poor documentation, lack of support, small developer community, you end up with higher costs maintaining and developing the game. It also makes feature adding much more difficult and that's already one of the problems in this game that afflicting it deeply.

HBS already chose Unity 3D, and so they may already have worked out their revenue model vs. game engine licensing costs.

Another way to move the game into Unreal or Unity is simply not to do a port, but a clean room recreation. It might even be faster that way.

Instead of giving the source code, and having it reworked line by line to the new engine, you simply give the developers a concrete idea of what the game looks like and how it works. Like a simple running example of MWO in their machines as a sample reference. All they need is to recreate that with clean from scratch code. Of course you let them know what your existing datasets are, so the new code will tap into this. You tell them what they can do, what they cannot do, give them a schedule and a deadline and make sure everything is well documented. Then let them loose and work this among themselves. Like a clean lab room. Not a port, a recreation.

Edited by Anjian, 25 June 2017 - 12:51 AM.

[Major Tomm]

Unreal Engine is for doing interiors like in Quake. You start with the level as a completely filled single object and then you start hollowing out your level. Not sure it would do a MechWarrior game justice. Unreal exteriors have boundaries like being in a box. maybe it's gotten better, but technically you would lose the distant vistas you get in Cry Engine.

The reason you don't get tanks and destructible objects in MWO is not the Cry Engine, it's that the state of each of those objects has to be constantly updated to 24 players. It has limits before the game becomes completely lagged out. Unreal doesn't fix that.

[Koruthaiolos]

Major Tomm, on 25 June 2017 - 01:13 AM, said:

Unreal Engine is for doing interiors like in Quake. You start with the level as a completely filled single object and then you start hollowing out your level. Not sure it would do a MechWarrior game justice. Unreal exteriors have boundaries like being in a box. maybe it's gotten better, but technically you would lose the distant vistas you get in Cry Engine.

The reason you don't get tanks and destructible objects in MWO is not the Cry Engine, it's that the state of each of those objects has to be constantly updated to 24 players. It has limits before the game becomes completely lagged out. Unreal doesn't fix that.

Beg to Differ... http://www.greybox.com/dreadnought

[MustrumRidcully]

Anjian, on 24 June 2017 - 11:19 PM, said:

You can't use the Battletech kickstarter as an example of a game port. The Battletech game had to be made from the ground up. Actually I thought it went faster than I expected. Crafting the game on the game engine, which is Unity 3D, is likely the least of their bottlenecks.

I think it's pretty apt, because pretty much only the graphical assets will be reusable with little work. THe entire user interface will need to be reimplemented in a different engine. I don't really know how far Unreal and CryEngine go into certain things, like providing basic abstractions for game entities with game stats like weapons or something like that, but I suspect there is some, and it will need to be reworked. Basic stuff like network code will also be based on whatever CryEngine already had, and will need to be reworked with the Unreal engine, requiring to learn its idiosyncrasies and to apply them correctly, and fixing any issues that come up.

If I imagine trying to rebuild the software I am working on with something different than Qt as underlying framework, it would be seriously big deal, akin to rewriting it entirely. (With Qt, aside from the UI, there are also a lot of underlying helpful abstractions. Like, a generic List class. Eliminating that everywhere we use them would probably be already a task for days. Certainly some components would be reusable, with some refactoring for whatever the new hypothetical framework (or maybe even C++11 have to offer), but a lot of fundamental things aren't.

[Major Tomm]

Raasul, on 25 June 2017 - 01:43 AM, said:

Beg to Differ... http://www.greybox.com/dreadnought

Looks good, much better than Unreal Engine 5 years ago. The distant ship models look convincingly huge, but your own ship has the apparent scale/perspective of a 3rd person shooter handheld rifle and at least the planetside battles seem to be confined to that bounding box. I haven't played it yet. Do any battles give you the full space to planetside battle or do you always warp to the planet surface?

I am still looking for that in a space game. Battles that convincingly cross from space to planet or moon terrain. RSI has some nice procedural planetscapes in their engine, but too expensive and they haven't finished the game yet.

Edited by Major Tomm, 25 June 2017 - 06:27 AM.

[Cy Mitchell]

DAYLEET, on 24 June 2017 - 12:21 AM, said:

They are already doing the work for mw5, they are learning. I remember that townhall where Russ said it would stop development for 6 months. Well fk it im willing to stop everything to change to an engine that will survive the test of time at better framerate, better eyecandy and support for actual features.

It is easy for us to say, "Let's stop everything for 6 months and port the game to a new engine." The problem is that if everything stops then so does the money coming in to PGI. That leaves the company with a bunch of hungry developers who have house payments and car payments etc. and no way to pay them. I do not know very many companies who can cut off their income stream for 6 months and still survive. Especially not ones with under 50,000 customers many of whom do not even pay to support the product.

There is not enough potential benefit of moving the game to the UE4 engine to outweigh the development expense and loss of revenue during the transition at this point in the life of the game. Perhaps if there was another reliable income source for the company (MW5?) then it would be possible to absorb the cost and loss of income related to the transition but, for now, I do not believe it is feasible.

[JC Daxion]

MustrumRidcully, on 24 June 2017 - 10:50 PM, said:

Everything is possible. Practical, however?

Well, look at the Battletech Kickstarter. They are actually using MW:O assets in part in a different engine. It obviously works. But the project is now 2 years in the making. You can expect that it would also take MW:O a year or two to make such a transition. A year or two where they do nothing else but this transition. No new maps, no new game modes, no new mechs, no quality of life changes, absolutely nothing else. They would be busy recreating what they already have, instead of making new stuff.

It all depends on the size of the team.. But yes it would mean some things are put on hold, but honestly i think it could be worth it. You are thinking of the future. Especially if the new engine can do lots more, like Replay, Knock downs, better opitmization with DX11-12, Multiple core use ext.. There are lots of reasons why to do this.

I mean people complain that development is slow at points anyway, Do what is a hold for a while going to actually mean?

they could also get the art team making assets for future maps as well. So all development doesn't need to stall. They can still make mechs which is the main income, along with the paint/premium/camo, mech bays, cockpit items So income is not going to stop. they can still do balance tweaks.


So changing engines doesn't mean everything just stops, It just means base game changes are put on hold. And honestly at this point i don't think that would be a bad thing. It really all depends on how much the upgraded engine can help things.


As i said in another thread, We should really push PGI for a responce on this and see where they are now on the subject. If the game is giong strong as they say, and if they want to think of the future, maybe this is a good time to make this step.

Look at HBS, They pushed back development how many months? was it 3-4? because upgrading the engine was the right thing to do. Even though it broke a ton of crap and set them back. But my guess it was the right choice.

[Mechwarrior1441491]

You have to update the game if you want to extend its life and keep it relevant. If it means being able to do more with the engine, which they most certainly will, I say do it.

#Weshallforgetcryengine


If they can put together some solid info on what they'd be able to do with UE4 over now, they should do that. It isn't just to make the game prettier.A

Start off 2018 going into a virtual relaunch of MWO. Add Solaris to that. A spattering of attention to CW. A pass of achievements, adding, altering. A new game mode. Map.

It can be done.

Edited by Mechwarrior1441491, 25 June 2017 - 08:21 AM.

[STEF_]

Skanderborg, on 24 June 2017 - 05:46 AM, said:

i hope so , MW5's gameplay looked A LOT better than MWO's. It seemed to have that "heavy" feeling that the past games had and felt like you were in a mech.

and do u know why?
Because in that preview mechs are slower.

And that's the reason why I would like pgi to nerf mech speed.

[STEF_]

Major Tomm, on 25 June 2017 - 01:13 AM, said:

Unreal Engine is for doing interiors like in Quake. You start with the level as a completely filled single object and then you start hollowing out your level. Not sure it would do a MechWarrior game justice. Unreal exteriors have boundaries like being in a box. maybe it's gotten better, but technically you would lose the distant vistas you get in Cry Engine.

The reason you don't get tanks and destructible objects in MWO is not the Cry Engine, it's that the state of each of those objects has to be constantly updated to 24 players. It has limits before the game becomes completely lagged out. Unreal doesn't fix that.

I suppose you've never heard about ..... a lot of other fps multi, with dozens of players each side......

Ops, I must correct....not dozens, hundreds.
and this is quite old:

[Mechwarrior1441491]

With the size of the units we use, Mechwarrior is committing yourself to an engagement before you are likely to even use your weapons, because getting out of an engagement isn't likely possible. Disengaging seems to be something that is allowed by your opponent or something out of fiction so the hero can save the day later.

[Lightfoot]

Stefka Kerensky, on 25 June 2017 - 08:26 AM, said:

I suppose you've never heard about ..... a lot of other fps multi, with dozens of players each side......

Ops, I must correct....not dozens, hundreds.
and this is quite old:

That's not the Unreal Engine. It's a cinematic. I wasn't impressed with Planetside 2, (I remember trying it) just not my interest in games. EVE puts 30,000 players on the same server, but when they all show up at one spot it gets a bit laggy. I don't think you can put hundreds of mechs on a single server due to the number of systems and targets a mech runs, but I don't know.

I'll say this much, if PGI is going to go through all the trouble of porting to a new Engine they should pick one that creates the correct perspective for giant stompy robots. I don't see Unreal doing that yet. Any examples of mechs in Unreal 4? Basically all the past mech games have handled the perspective by making the world around the mech miniature and the mech human size. MWO is pretty good for scale, but still when you walk through river city the windows in the buildings look about one foot tall and your mech's arms do not look like they are 7-10 meters away from the cockpit. So the new engine needs to be able to project curvilinear perspective.

Basically, the mech has one perspective which is kind of a forced perspective, curvilinear perspective (2), and the world around it is normal from a set distance outwards. This creates the illusion of the mech being a much larger scale than the normal view. If done right. You know, like when a giant space frieghter passes right in front of your viewpoint and then into the deep distance. Think Star Wars or Alien.

Edited by Lightfoot, 25 June 2017 - 02:44 PM.