Jump to content

Better ground modification.


GrayLeader

Recommended Posts

Several people have come together on this idea of having a better ground modification setup. (aka terraforming)

So idea as follows.

We would love a way to use square ground modification instead of just a circle. (i would say more of a smoothed cube for server performance.)

As and example allowing us to smooth out some roads of have better territories and not so much randomized structure placement because of ground leveling.

Benefits

1st. way less cores in the world for getting  roads in the game.

2nd. race tracks!

3rd. getting more realism out of dual universe.

4th. lowering server lag due to jagged edges due to the round tools and flattening tools.

Link to comment
Share on other sites

The original videos show square planetary voxel manipulation. 
 

it was changed to spherical later on. 
 

my theory was that if people could build cool non-turd-like constructs with planetary voxels, they wpuld, and then there would endless terraforming and server load. 
 

of course they never guessed how much mining we would do in any case. 
 

 

Link to comment
Share on other sites

The biggest problem with the terrain voxels is that they're designed to simplify terrain shapes like valleys and mountains at a lower LOD.

 

They aren't designed to deal with more complicated shapes.  Or being edited in any way really.  So as soon as you touch the terrain with the terraforming tools, everything just looks like garbage from a distance.  I would love it if NQ would give us more fine control over the terrain voxels.  But i also wish they would work on helping us make the terrain look halfway decent, after we edit it.

 

The tool that i would really like is a terrain healing tool, that would replace terrain voxels that have been deleted, and revert everything back to the way it was originally. 

 

I would also love it if there was a way to control the blending of terrain materials.  For whatever reason, when we place them, they don't seem to blend at all.

 

And a way to replace ground clutter too.  A lot of the terrain materials have ground clutter built into them, like grass and small flowers ect.  But when we place them, none of that is visible.

 

And the larger terrain elements like trees and rocks.  Even if we were just able to craft and deploy them as an element on a core.  It would be great if we had access to all of those assets to build with.

Edited by Atmosph3rik
Link to comment
Share on other sites

The original planet landscape is cached locally on all player clients so that flying over unmodified landscape with no constructs requires hardly any data streaming.

Good idea, but NQ never figured out how to make a seamless transition from unmodified to streamed in modified landscape. So instead we get large chunks of landscape change just popping in out of nowhere as we get closer. And with the exception of very early pre-alpha tests, it has only gotten worse over time since NQ is tuning for server cost and not game performance.

 

And from a technical viewpoint it is strange that there are no square shaped terraforming tools, since circle shaped tools make more costly voxel changes then square ones.

My guess is that NQ are terrified of anything that could lead to more terraforming. An easy way to build roads for example, would just lead to the same performance and cost issues that mining tunnels caused earlier.

 

So in short. DU is an open world voxel based game with unrestricted terraforming, that is hosted on a cloud server solution that makes it to costly to actually allow all players to do any kind of large scale terraforming. I.e. every time you dig it cost NQ money.

Link to comment
Share on other sites

2 hours ago, JayleBreak said:

All terrain within the boundaries of a static construct should be converted (on core placement) to normal voxels with terrain material types. Thus all voxel tools (copy, paste, delete, smooth etc.) could be used to landscape the build.

That's actually something that never crossed my mind... wow...

But I suspect the problem is that cores can be placed at semi arbitrary angles and the conversion from planetary to core voxels would be problematic when placing a prebuilt core (blueprint).

But for static core placements offer a one time operation:

( ) dig planetary voxels to exclude build extremes (as per the pre "release"(lol) landscape reset dig out rules.)

( ) dig planetary voxels out to out core dimensions. (just remove planetary voxels from the entire core)

( ) keep planetary voxels as is. (current method)

( ) Incorporate planetary voxels into core structure. (JayleBreak's proposed method)..

 

If only this game were still being actively developed.... 😢

This game had so much potential.

Link to comment
Share on other sites

7 hours ago, JayleBreak said:

All terrain within the boundaries of a static construct should be converted (on core placement) to normal voxels with terrain material types. Thus all voxel tools (copy, paste, delete, smooth etc.) could be used to landscape the build.

Good idea. But the usual problem is NQ tech limitations.

 

Such a conversion would be the same as a XXXL static core or something like that, even with 1m3 landscape resolution instead of 25cm3 construct resolution.

And NQ has repetitively said they are unable to handle larger cores then what we already have now.

Link to comment
Share on other sites

I'm not convinced that terrain modification of any sort is practical at scale for most MMOs.

 

The benefits it brings are overrated relative to the complexity and inherent scaling issues. 

 

Even with simplistic cube-based voxels, terrain modification can't scale without hard limits in any MMO -- especially in a single shard subscription MMO where there's no mechanisms to control traffic or monetize users beyond the one flat fee (which can't scale due to increasing costs over time).

 

The problem isn't their implementation of terrain systems, it's the core design...voxels don't solve anything at all with the fundamental problem! 

 

The issue has never been the way terrain is rendered (voxels vs. mesh), it's the scale of terraforming that doesn't work with an MMO like DU, period. 

 

If NQ had asked the question "how will this scale?" early on, they'd have saved a lot of time and money that could have been invested elsewhere....but again, the founder started this project with no game dev experience (and limited career experience outside academia). He cared more about metaverse fantasy than real-life engineering.

 

Honestly, there are a lot of great ideas that would "improve" DU, but finding great ideas that are practical, scalable, and realistic to implement for a team of their size...? Very difficult if not impossible. 

Link to comment
Share on other sites

NQ has painted themself into a corner being depended on cloud services (AWS) for the game to function, meaning they get monthly bills for storage regardless and processing and data transfer rates that vary depending on the user load. The last one being the costly one when you look at how AWS operates.

 

I think self hosting would have been a much better alternative with the limited number of players they have. But NQ was shooting for the stars with millions of players from day one so they went cloud. And the entire AWS system is designed so that you cannot easily replace the service once you are up and running.

 

With self hosting on the other hand, once you are past the initial cost of building infrastructure and IT management, then processing and storage becomes very cheap.

 

The real problem here is the cost of bandwidth in a game like this, where server bandwidth requirements go through the roof as player numbers increase. The only solution I can think of that would scale with players is a P2P network between the clients where players close to each other in the game would share the load. But getting something like that to work robustly and with low latency would NOT be a trivial task. And there are also problem with privacy and leaking IP addresses from other players.

Link to comment
Share on other sites

11 hours ago, CptLoRes said:

The only solution I can think of that would scale with players is a P2P network between the clients where players close to each other in the game would share the load. 

 

Yep, its likely only commercially viable solution right now in games with such type of ambition. Starbase (despite their own long list of bs and failure) was, I think, on relativly right track with this, hybriding p2p and own servers things. Probably if some top tier company with money and talent really go hard on such idead, it will work really good eventually.

 

Overall its bit about balance between bulding and combat sides of game like that (and where cut corners). I still think DU drifted too much to "Landmark" from "EVE".

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...