Sign in to follow this  
Followers 0
jwilly

Wishlist for a New Terrain Tileset

27 posts in this topic

1. Add 10 meters to every height "post". This has the effect of moving the world up 10 meters, and therefore allowing local deformations of as much as 10 meters depth without going into nonallowed negative heights.

2. Copy the existing ground tile set (800 meters square), adding a network of countryside roads/farm lanes that cross tile boundaries only at tile side midpoints. 

3. Increase the off-road Drag Factor for wheeled vehicles; classify mixed type vehicles (halftracks, specialized off-road wheeled vehicles) as "tracked" for Drag Factor purposes, make the Drag Factor speed-dependent for all vehicle types; and make the Drag Factor substantially greater when it's raining, gradually returning to normal after the rain stops.

The point of #2 and #3 being to constrain supply-truck traffic mainly to roads.

4. Enlarge the 800 meter tileset with an extension tileset for shallow, intermediate and deep water, plus channels and harbors.

5. Build a special 50 meter tileset for modeling high-game-significance rough terrain. 256 50 meter tiles replaces one 800 meter tile.

6. Develop and use higher-resolution LOD0 ground textures.

1 person likes this

Share this post


Link to post
Share on other sites

My understanding of CRS's current game-world plan is to extend it (per Phases 1, 2, 3 in Xoom's latest video) to cover more territory in Netherlands, France, England, using the existing technology.

That makes total sense as a first step, since the skills and tools involved haven't been used in several years, and not by CRS 2 yet at all.

This thread is about somewhere in the future, beyond existing plans.

Edited by jwilly

Share this post


Link to post
Share on other sites

This screenshot, showing current state of a game that was first offered in 2010 and runs on iPhones, illustrates good water modeling. By comparison, the primitive state of water modeling here is, well, primitive.

WOTA-Corvette-Storm.gif

Of course, there would need to be a coherent system so that ocean and river-estuary-mouth storms were rough, calm weather was less rough, and rivers were calmer still. And, the sea state would need to correspond to the air state, i.e. storm waves should correlate to major gusty conditions and cloudiness above, and ditto with calm seas and clear skies.

 

Edited by jwilly

Share this post


Link to post
Share on other sites
2 hours ago, jwilly said:

This screenshot, showing current state of a game that was first offered in 2010 and runs on iPhones, illustrates good water modeling. By comparison, the primitive state of water modeling here is, well, primitive.

WOTA-Corvette-Storm.gif

 

 

lmao

 

(cries on the inside)

Share this post


Link to post
Share on other sites

I don't know if Unigine has been considered for future use. Here's a video on its water simulation capability, from Beaufort 0 to 12, with examples of 1, 4 and 9...the game's current sea state of course is Beaufort -1 or thereabouts:

and its cloud capability:

 

Edited by jwilly

Share this post


Link to post
Share on other sites

This by comparison is an implementation of the UnReal engine for ocean water. It's not very good...maybe that's the engine, maybe it was the implementer:

This thread's of course about the tile set. I assume the engine's capabilities are relevant to the water tile set design in regard to how different-depth tiles will work alongside each other if the engine is good enough to take water depth into account in its surf simulation.

A surf simulation will be a key visual and operational element if WWIIOL is to have an ability to simulate beach landings and naval commando raids.

Share this post


Link to post
Share on other sites

Anything can be achieved with enough dev time, I just wish we had more of that. 

Share this post


Link to post
Share on other sites
14 hours ago, jwilly said:

I don't know if Unigine has been considered for future use. Here's a video on its water simulation capability, from Beaufort 0 to 12, with examples of 1, 4 and 9...the game's current sea state of course is Beaufort -1 or thereabouts:

and its cloud capability:

 

First question that arises is : can we implement a similar result in our own engine? I guess it's all about textures and light effects , no?

Share this post


Link to post
Share on other sites

Actually for water terrain I think it's a bunch of Eulerian mathematics to calculate dynamic particles, then dynamic macro surfaces, then dynamic occluding; plus flotation-body movement taking into account the locally dynamic surface and the body's mass and momentum. Pretty much pure fluid dynamics stuff.

My guess is that the existing engine does nothing like that now.

The cloud modeling in Unigine and other engines like it is volumetric and fluid/gas based, not non-collidable solids with textures. The problem with the latter approach would be that as you get closer to the cloud, realism would fail...and WWIIOL's simulation needs to work for aircraft.

Edited by jwilly

Share this post


Link to post
Share on other sites
On 6/8/2018 at 5:04 PM, jwilly said:

1. Add 10 meters to every height "post"

Why?
If i need to go down, i just go down

On 6/8/2018 at 5:04 PM, jwilly said:

2. Copy the existing ground tile set (800 meters square), adding a network of countryside roads/farm lanes that cross tile boundaries only at tile side midpoints. 

 

On 6/8/2018 at 5:04 PM, jwilly said:

6. Develop and use higher-resolution LOD0 ground textures.

Resolution is not the really problem with current.
Problem is the original art from way back was simply scaled up and reused with little improvement, and also does not have a very natural pallet to put it kindly.
The potential of the resolution size unity1.5 could do was not maximized.
Kind of like enlarging 8bit art to 4k resolution, it only gets bigger, not better.
Can possibly accomplish both things in the same art task, depending on how much you want to get out of a farm wagon path.
 

On 6/8/2018 at 5:04 PM, jwilly said:

5. Build a special 50 meter tileset for modeling high-game-significance rough terrain. 256 50 meter tiles replaces one 800 meter tile.

Probably more efficient to build the rough terrain in an entire chuck.
a mod tile if you will, does not have to be designed to the same constraints as a native tile.
Native tiles have to fit the data resolution so they can exist with no human intervention.
Mod tiles have no such limitation since they only happen by human intervention and can be made to do whatever they need to do.
It's simply a matter of being able to dedicate the art time.
 

On 6/8/2018 at 5:04 PM, jwilly said:

3. Increase the off-road Drag Factor for wheeled vehicles; classify mixed type vehicles (halftracks, specialized off-road wheeled vehicles) as "tracked" for Drag Factor purposes, make the Drag Factor speed-dependent for all vehicle types; and make the Drag Factor substantially greater when it's raining, gradually returning to normal after the rain stops.

I was thinking more along the lines of, when it rains, your truck starts sliding, cant get traction and slips on takeoff, cant stop well, cant steer well
AND can not coast for 1+ km or do 50 across farmland and marsh
 

Share this post


Link to post
Share on other sites
3 hours ago, merlin51 said:
Quote

 

1. Add 10 meters to every height "post"

 

If i need to go down, i just go down

We've been told over the years that it's not possible to create a feature or place a unit at an elevation less than 0. So, at a location where the ground surface is +5 meters, certainly you could place a "foxhole" world element that's 1 meter deep, or a bomb crater element that's 3 meters deep, or a fortified building basement element that's 5 meters deep, and an infantryman could go to the bottom of these "holes", If however we're at a location where the ground surface is at 0, we could not place that foxhole, crater or building-basement element, because it would extend to a less-than-zero elevation. Correct or not?

Quote

 

Quote

6. Develop and use higher-resolution LOD0 ground textures.

Resolution is not the really problem with current.
Problem is the original art from way back was simply scaled up and reused with little improvement, and also does not have a very natural pallet to put it kindly.
The potential of the resolution size unity1.5 could do was not maximized.
Kind of like enlarging 8bit art to 4k resolution, it only gets bigger, not better.

 

I was told in the old days by Gophur that the existing textures were scaled up from lower-res because this allowed the size of the compiled world file to be smaller, since the scaled-up textures contained a lot of redundant information.

In any case, everyone would agree that the goal is just to have better ground visual fidelity, particularly for infantrymen.

Quote

 

Quote

5. Build a special 50 meter tileset for modeling high-game-significance rough terrain. 256 50 meter tiles replaces one 800 meter tile.

Probably more efficient to build the rough terrain in an entire chuck.
a mod tile if you will, does not have to be designed to the same constraints as a native tile.
Native tiles have to fit the data resolution so they can exist with no human intervention.
Mod tiles have no such limitation since they only happen by human intervention and can be made to do whatever they need to do.

 

Several years ago I had a discussion going with Gophur about building a series of RA maps of parts of the Ardennes for a set of Battle of the Bulge scenarios. RA of course used a different terrain system, but we also were talking about building RA-like maps for BE battles off the main map, with much more complex terrain, and those would have been terrain-compatible with BE. The problem with building 800 meter rough terrain tiles was that they'd be repetitive. Each one would be recognizable if it appeared somewhere else on the map, even if rotated differently, due to its idiosyncratic geometry. With a set of 50 meter tiles, OTOH, a lot more variety could be achieved.

The cost of course would be the building of that second tile-set.

Quote

 

Quote

3. Increase the off-road Drag Factor for wheeled vehicles; classify mixed type vehicles (halftracks, specialized off-road wheeled vehicles) as "tracked" for Drag Factor purposes, make the Drag Factor speed-dependent for all vehicle types; and make the Drag Factor substantially greater when it's raining, gradually returning to normal after the rain stops.

I was thinking more along the lines of, when it rains, your truck starts sliding, cant get traction and slips on takeoff, cant stop well, cant steer well
AND can not coast for 1+ km or do 50 across farmland and marsh

 

Though the present coasting situation is absurd, and road traction in rain is modeled badly as well and could be considerably improved as you mention, the point of the suggestion was to address the more fundamental issue that in the real world, loaded road-intended trucks with 50PSI ground pressure are fundamentally unable to drive on farm fields with soil that flows at 10 PSI when damp, or much less in rain. As a matter of fundamental realism, such trucks need to be limited to operation on roads or other hard surfaces. That's also the reason for the proposed addition of the quadrant-intersecting farm roads across the replacement tileset.

There had been prior discussion in the old Design Forum of doing this via the wheeled vehicle Drag Factor so that such vehicles wouldn't be entirely barred from offroad operation, but would be very slow.

Share this post


Link to post
Share on other sites
5 minutes ago, jwilly said:

We've been told over the years that it's not possible to create a feature or place a unit at an elevation less than 0. So, at a location where the ground surface is +5 meters, certainly you could place a "foxhole" world element that's 1 meter deep, or a bomb crater element that's 3 meters deep, or a fortified building basement element that's 5 meters deep, and an infantryman could go to the bottom of these "holes", If however we're at a location where the ground surface is at 0, we could not place that foxhole, crater or building-basement element, because it would extend to a less-than-zero elevation. Correct or not?

Can not after the fact deform/modify the living terrain
1) It hasnt the flexibility in resolution, it be a big big dent
2) Have to communicate to all clients "Oh hai, i had make big dent here that not here before" lot of overhead
3) Can not cut a hole in it because you'd have to on the fly change the polys to ones that did not exist a minute ago, and if you can not cut a hole
while you can place said foxhole, you cant get into it.
Would need something that supports arbitrary terrain deformation, like the original red faction did.
most games dont because it is apparently a pain to track and sync or something.

But the on the fly stuff aside
Maybe they did not know how, or maybe unity 1.0 was truely not able to, or maybe didnt explain well...
There are places in the live game map already that are at negative elevations.
Small negatives of course, it is france not death valley.

SO while you can not run up in game and decide Hey ill stuck a tunnel here (because while you might stick it, the ground isnt gonna vanish and let you in it)
You can make one before the fact, and make a ground piece to accommodate it, and the first poor sob who walks over the funny hole falls to -200m
and lands in the bottom, probably breaking all his bones but.

Of course it's all 3d art and takes time to do right

22 minutes ago, jwilly said:

I was told in the old days by Gophur that the existing textures were scaled up from lower-res because this allowed the size of the compiled world file to be smaller, since the scaled-up textures contained a lot of redundant information.

Well, it does.
In the days of 56k modems and early 128k asdl.
Now days, not too much advantage.
Its good to be mindful of bloat for the sake of bloat when there is no real gain but
I think we could do something a lot more real appearing for a few more bytes maybe.

 

29 minutes ago, jwilly said:

The problem with building 800 meter rough terrain tiles was that they'd be repetitive. Each one would be recognizable if it appeared somewhere else on the map, even if rotated differently, due to its idiosyncratic geometry.

That depends
Without going into detail, there are ways not to be repetitive and yet reuse parts over and over

Share this post


Link to post
Share on other sites

@merlin51 is it possible for you to design/edit a town with foxholes and trenches?

These would be permanent structures in the terrain, not PPO objects.

It would be cool to have foxholes and trenches outside the AB and around the perimeter of the town.

Share this post


Link to post
Share on other sites

We already have some kind of trenches around some forests so yes it's possible I guess. My truck don't like it.

 

Share this post


Link to post
Share on other sites

Yes I am aware of those ditches near some forests.

I was hoping for some trenches and foxholes designed around the town. From there, players could setup their defense with PPO barbed wire and sandbags.

Share this post


Link to post
Share on other sites
13 hours ago, rsix said:

@merlin51 is it possible for you to design/edit a town with foxholes and trenches?

These would be permanent structures in the terrain, not PPO objects.

It would be cool to have foxholes and trenches outside the AB and around the perimeter of the town.

Merlin does not do 3d art, you would not like that, merlin is not terribly good at it.
But, at a point where resources allow, we could look at reworking or adding something along those lines.

Perhaps not AB specific as your AB can be anyplace, but perhaps at that point we might be able to explore some additional PPOs
to handle things like that.

It just takes some time.
If anyone secretly knows human cloning, 5 3D and 5 Texture artists would not be viewed in a bad way ;)

 

12 hours ago, Zebbeee said:

My truck don't like it.

 LoL
Well anything surrounding a town would have to be limited and balanced.
You want it to lend some help, but never the amount a defender would wish for, or all the towns become Helms Deep with no magical explosive.
For every defense, there must be a possibility of overcoming it so that both sides have fun win lose or draw

 

Share this post


Link to post
Share on other sites

My recollection from discussions in the old Design Forum is that the test-trenches around some forests had angled sides instead of vertical because that allow infantry to get in and out; made them operate less as inescapable vehicle traps; and had a better visual result when the game engine applied the texture. 

They were considered unsuccessful because the angled sides, necessary for the above reasons, made them look like ditches or inverse-berms instead of trenches.

Fighting holes also were tried, but for the above reasons they looked like craters instead of foxholes and also were considered unsuccessful.

Maybe Merlin could solve some of these problems, though.

Share this post


Link to post
Share on other sites

Well, they do function quite well as vehicle traps, ask any of us that have accidentally rolled into one LOL
But they do present themselves as more of a rural american drainage/irrigation ditch than a infantry dug trench.

Since those are part of the "tile" i would think that they could be redone to resemble more of a trench.
Infantry exits are needed but could be done at intervals maybe instead of the entire thing.
They also do not need to be so deep that infantry can not stand in them and fire over them.
All you need to stop the tank is enough span that it falls in.

Merlin does not make the art directly though, my talents there are terrible.

30 minutes ago, jwilly said:

Fighting holes also were tried, but for the above reasons they looked like craters instead of foxholes

user placeabel fox HOLES need one of 2 things to happen
1) heavily modify the engine to accommodate in the fly temporary terrain deformation
which realistically, being an MMO and not a small game world limited to 16 or 32 players, is probably a bad way to go.
It is also a very extensive and code heavy undertaking.

2) borrow and repurpose the ride vehicle code, and create appropriate art as needed
so you place the foxhole, and it looks like a foxhole, except you can not see 4 feet down, just have it black.
And you hit J to "ride" the foxhole and much like riding on the FMB, you have control of your weapon and can turn.
Something like this makes more sense probably, than managing a complex arbitrary terrain deformation system within an MMO?

1 person likes this

Share this post


Link to post
Share on other sites
4 minutes ago, merlin51 said:


2) borrow and repurpose the ride vehicle code, and create appropriate art as needed
so you place the foxhole, and it looks like a foxhole, except you can not see 4 feet down, just have it black.
And you hit J to "ride" the foxhole and much like riding on the FMB, you have control of your weapon and can turn.
Something like this makes more sense probably, than managing a complex arbitrary terrain deformation system within an MMO?

Interesting 

Share this post


Link to post
Share on other sites
13 minutes ago, Zebbeee said:

Interesting 

Trust me, i probably made it sound 1000 times easier than it is to make

Share this post


Link to post
Share on other sites
14 minutes ago, merlin51 said:

user placeabel fox HOLES 

2) borrow and repurpose the ride vehicle code, and create appropriate art as needed
so you place the foxhole, and it looks like a foxhole, except you can not see 4 feet down, just have it black.
And you hit J to "ride" the foxhole and much like riding on the FMB, you have control of your weapon and can turn.
Something like this makes more sense probably, than managing a complex arbitrary terrain deformation system within an MMO?

This idea was around back in the old days, and I think was modeled in alpha. Back then it was called Virtual Foxholes. 

My recollection is that it was required back then that no part of the infantry character be located below 0 elevation on tiles that are at 0 elevation, and to implement that, some new infantry models and animations would need to be created so that each infantry character position would be "cut off" at the ground plane. For weapons used at face or shoulder height, i.e. pistol, rifle, SMG, LMG, binoculars, hand grenade, panzerschreck, bazooka, this could be done by (in an art sense) cutting off the existing standing-position model of the infantryman at the appropriate height so that the weapon and arms just clear the ground in front of the virtual foxhole. For weapons that normally would be fired only prone, that model could be modified instead.

There would also be a watching position, in which the soldier would have only his eyes and helmet above ground height, and a hiding position with nothing visible from the front, but a 2-D picture of the top of a hunkered-down infantryman visible from above. This picture also would be a 2-D damage model; bullets or fragments that hit it from above could injure or kill the infantryman.

All of the new models would have their terrain collider height correspond to the Virtual Foxhole surface plane.

Weapons that necessarily were fired from the waist i.e. panzerfaust, or that could only be fired from a crouch or while laying down i.e. some rifle grenades that would break bones if fired from the shoulder, would be problematic. So would mortars, and eventually weapons that are mounted on tall tripods such as MMGs or AA MGs.

Virtual foxholes were to be allowed to be dug only on plain ground, i.e. not roads, railroads, airfields, anywhere inside a town/city, not touching a building footprint, etc.

For simplicity, they weren't to interact with vehicle or other-infantryman movement at all. You could run or drive over them without noticing, and your doing so would not injure an infantryman in the virtual foxhole.

There was some discussion that they wouldn't just be wished into existence...that instead there'd be a kneeling-digging-a-hole animation that would have to be done for some period of time before the virtual foxhole would appear.

Back then, nothing was server tracked yet, so if the virtual foxhole was left by the infantryman that created it, or he was killed, it would disappear.

Some of us thought this was a good idea back when it was considered. My understanding was that it was rejected due to not being realistic enough. It might also be that it made the defense too strong, or that the art resources to make it happen weren't available.

Share this post


Link to post
Share on other sites

Well, some things you just dont fire out of a fox hole too much.
A mortar tube kind of comes to mind, doubt i could reach it properly

Share this post


Link to post
Share on other sites
36 minutes ago, merlin51 said:

Well, some things you just dont fire out of a fox hole too much.
A mortar tube kind of comes to mind, doubt i could reach it properly

sara could make it work.

Share this post


Link to post
Share on other sites
Sign in to follow this  
Followers 0

  • Recently Browsing   0 members

    No registered users viewing this page.