Outerra forum
Outerra Engine => Ideas & Suggestions & Questions => Topic started by: Player1 on June 12, 2016, 07:33:32 am
-
Hi guys,
which features will be included in the next big update and when is it supposed to come? I am really curious since the last one brought rivers, a more precise road system and a lot of other cool stuff with it and I wonder what there is next. You are doing really well! It must be a lot of hard work, but therefore the engine is fantastic. :)
-
Some of this:
https://www.youtube.com/watch?v=nlJG3llZFR0
-
Those developer highlights of what is coming are the best thing ever. I am probably exaggerating a bit, but that's how I feel when I see them after long periods of silence. I wish we had them on a more regular basis. Doing this in video format could be tedious, so even little written updates would suffice. And also blog, I wish it was more active too :(
I am really hyped about upcoming features though!
-
Those developer highlights of what is coming are the best thing ever. I am probably exaggerating a bit, but that's how I feel when I see them after long periods of silence. I wish we had them on a more regular basis. Doing this in video format could be tedious, so even little written updates would suffice. And also blog, I wish it was more active too :(
I am really hyped about upcoming features though!
Yeah, that's why I mentioned developer diaries a while ago! :D
-
*Improvedcollisionwillnotimproveyouraim Pleasedrinkresponsibly ^-^
??? Aw man, I was kinda thinkin' that was my area for a while.... but I don't think so. :P
Personally, I'm waiting for improvements to the road markings, particularly the issue with plain sections automatically being thinner than the marked sections with margins outside the white lines. Can't work with that... Among other marking issues.
-
Some of this: . . .
The realism of these collisions is really impressive !
Huge thumbs up !!!
You wrote: "Box, sphere and capsule shapes supported so far."
Only if you defined as isolated objects or you can apply this "collision ability" even at a meches of a model ?
I think, for example, to a mechanical arm that extends outside of the bounding box and is able to interact with the surrounding environment ...
-
If the model is made up of animated collision shapes, and you animate them, then it will have collisions as well. However, Bullet physics still does some unexpected things so the reaction often isn't what it should be. It could be seen on the mercenary which is made of several such shapes, when he's interacting with the ball. He wouldn't kick it, it just shifts away from the intersection and loses momentum.
-
And I think maybe those balls were frictionless and liable to go on forever if not stopped. :o :)
-
Yep, they do not bounce, do not see the grass, do not react to water ... can of worms.
-
That looks pretty cool. :) What I personally would love very much is the improvement of the ocean. The coastlines are very buggy sometimes, and what matters more to me are the land areas below sea level, which are all flooded in Outerra, for example the Netherlands or parts of Northern Germany. That plus lakes, different realistic biomes and railways would make Outerra perfect in my opinion. :D
-
Looking great cameni!
first i though was: maybe this will finally change the bad vehicle behaivor during accidents?
As i think actually those crazy bounces are connected with how car chasis collide with the complex terrain outerra has. (normally before bounces, the terrain swallow car as collisions missings, coz i suppouse only tires collide propperly actually)
Yep, they do not bounce, do not see the grass, do not react to water ... can of worms.
But i have hopes in balls. Pills was bouncing good at 3:16
Seriusly. It will be posible ball bounce, or its something that bullet limits somehow at the moment?
PD: Mercenary GREAT ball control at 1:51 xDD Finally he will be able to play football in the vaaast green outerra areas ))
Time ago i had a weird (and i know imposible) idea. Represent diferent stadiums around the world, then connect some football game where people play against each other. So while you fly around outerra you can see those matches inside the stadiums. xD (and similar with other things diferent than football, would be the game of games..).
But maybe by the moment we can teach mercenary a bit. Eventually even adding moves in his leg for shot the ball strong :)
-
Not sure if everyone noticed that, but the video description there says "Also shown alpha OSM road import", which means that road on the back was automatically made from the OSM database. We might finally get miles and miles of roads? That's pretty fu**ing rad if you ask me. As physics improve, we might also get better ground vehicle simulation, I'm not a plane guy, but I do enjoy cars, so this is good news for me :D.
Also, the collision is looking pretty great as it is, are you guys now working on differentiated surface frictions too? I remember reading that surface friction was on the to-do list not long ago.
A bit of a dumb question here, but, would having spherical (or just round) collision model for the soldier help with the way he interacts with the ball? I know bullet works with collision margins and has different algorithm to calculate the interaction between different shapes, maybe it is viable to play around with that to reduce the unexpected reactions, to get more consistent collisions.
-
Hi,
I want to "revive" this topic. :D I am always extremely curious about the upcoming features in Outerra and I guess many others are too. Now, it has already been a month since 16.7057. In which time can we expect the next version, what will it bring with it and could you maybe make something like a dev-diary or to-do list which you regularly update? People like me are suffering from the lacking information as you might notice. XD
-
We don't know which feature will become stable and usable the first, so it's not always easy to tell which ones will show up and when. Here's an incomplete list of stuff being worked on:
- flexible and breakable trees, vehicles able to overcome small ones (this is already mostly done on the physics side, but bending billboards would not be spectacular)
- new object rendering pipeline allowing easier and more powerful usage by the rest of the engine
- subsystem for generated-geometry (for railroads, poles, bridges, tunnels but also vegetation clutter), depends on the object rendering pipeline
- enhanced terrain rendering (support for better materials, multiple terrain types ...), also optimizations in terrain rendering
- new particle system, support for long smoke trails etc
- reworking the water rendering
- enhancing rendering distance for runways and rivers
- further optimizing the physics subsystem
- networking for player-hosted servers, support for custom models/data
- framework for game code usable in single and multi-player scenarios (this is a very wide task that includes deeper changes to how characters and vehicles will be handled)
- fixing OSM road/river import
- working on OSM buildings import
- working on higher resolution imagery data import (processing terabytes of raw imagery data), forest data
- incorporating lithology (surface rock) data, support for different erosion patterns
- support for multiple base terrain types combined procedurally and using lithologic and surface-type data
- working on global water level data, bringing support for lakes and land below the mean sea level (and handling conflicts with river data)
and finally
Some screens from the imagery import: 30m imagery, no vegetation, compared to the current 500m data for area around Madrid:
(http://i.imgur.com/FLN3n6h.jpg)
(http://i.imgur.com/ugq8qQt.jpg)
(http://i.imgur.com/zMG6nB9.jpg)
-
A lot of tasks to perform as always. Squeezing the world into a PC isn't quite trivial. ;)
Hopefully you only forgot to mention better vr support.
With 30m imagery we will see man made structures (more obviously). How is this compatible with a "Anteworld"?
-
and finally
As you finally defined this in the roadmap, I started to feel uncomfortable, as the game looms in foreseeable future. I think to myself "When it is going to release, my daily watch of the forum, which already have lasted for 4 years, is going to come to end. How am I going to fill this void?..."
Never realised Anterworld would cause existential crisis :D
Anyway, thank you for taking you time in describing the current path. It's nice to know the direction in which you are going.
-
With 30m imagery we will see man made structures (more obviously). How is this compatible with a "Anteworld"?
The satellite data come with array of infra sensor channels that allow for determining whether given square is an artificial material. Previously I have done some processing with Modis data that removes urban areas by replacing them with randomized surrounding natural data, as if the vegetation overtook the city gradually from outside. From this:
(http://i.imgur.com/IKNkW8H.gif)
to this:
(http://i.imgur.com/BDeeglH.gif)
Since urban areas will be filled with buildings and roads, and in the imagery they come up basically as desert areas currently, we will have to run this algorithm anyway, whether it's to replace the cities altogether or just to get some greenery into cities.
However, all the available satellite data are obviously derived from current epoch; we can't pretend that Anteworld is the world from before 20k years, not without trying to revert all effects of human activity like the massive deforestation to get back to that world state (the pale yellow color is cropland). Rather, the game will almost certainly use the existing world data, only eroded forward into a semi-natural state. It's actually desirable to keep some more or less faint traces of civilization, the extent of which depends on how good we'll be able to make it look - eroding all the roads and overgrowing cities, rendering broken buildings etc.
-
Nice!
- new object rendering pipeline allowing easier and more powerful usage by the rest of the engine
Result of the benchmarks tests?
The satellite data come with array of infra sensor channels that allow for determining whether given square is an artificial material.
I was about to ask if these areas could be useful for some improvements in cities, like flattening the terrain, but then I realized this type of information is already in OSM, and probably more simple to handle that way :)
Some plans in the future for some area shapes for roads vectors ? With the lanes it's now easier to work on some areas, like airport apron, but still complex to get these shapes right, you have to use several roads and with big surfaces getting 100% perfect merging is really hard.
-
We expect great things,,, and I will be happy when Cameni cite the word Beta. :)
I share the view of the planet as it is.
It could not be otherwise, the data are of our time. although everything has moved even seguimo moving. Poles fact they end up being reversed,
Every year many airport runways have to be painted close to the numerical indications and orientardas the new magnetic north. He is now in Alaska but we also know that moves at 65 kilometers per year to Russia and thus the auroras. Now the magnetic pole is in Bathurst Island (NW Canada) and geographical pole is in Adelaide (Antarctica). this will simulate time today not yesterday.
forgiveness for google translation
(http://upload.wikimedia.org/wikipedia/commons/8/8e/Pangea_animation_03.gif)
-
Looks promising!
-
The satellite data come with array of infra sensor channels that allow for determining whether given square is an artificial material.
I was about to ask if these areas could be useful for some improvements in cities, like flattening the terrain, but then I realized this type of information is already in OSM, and probably more simple to handle that way :)
OSM contains incomplete info broken into small pieces, I don't think it's usable for terrain flattening as such. But yes, we'll eventually want to use these data to remove buildings baked into the satellite elevation map. That's what produces those hills in Manhattan and in every larger city. Then it can be also used to generally level the terrain or to suppress fractal noise.
Some plans in the future for some area shapes for roads vectors ? With the lanes it's now easier to work on some areas, like airport apron, but still complex to get these shapes right, you have to use several roads and with big surfaces getting 100% perfect merging is really hard.
I omitted tons of "small" things like a new road editor integrated into the scenery editor, with auto-merging and multi-node editing etc. Spline roads are supposed to handle the most common pavement types, but for anything more complex there will be a polygon based vector overlay.
-
Cool, thanks a lot :)
-
Wow, you still have a lot to do. And Outerra is great already. You are doing a good job, cameni and who else is working on this. :D Really can't wait for the final result, it is going to be gorgeous! Wish you good luck with every further enhancement and take your time. :)
-
Hello!
It can be some details about erosion? Will erosion be applied locally or to the whole territory? Will erosion be applied after the fractal algorithm generating parts
DEM-Fractal-Erosion.
or DEM-Erosion-Fractal?
-
Will erosion be applied after the fractal algorithm generating parts DEM-Fractal-Erosion or DEM-Erosion-Fractal?
It would have to be DEM-Erosion-Fractal, but the erosion may also be a part of the fractal process. If you are referring to this:
- incorporating lithology (surface rock) data, support for different erosion patterns
That's about applying various erosion patterns to rock types, not trying to simulate erosion on them.
-
That's about applying various erosion patterns to rock types, not trying to simulate erosion on them.
Interesting! Can see from the screenshot :)
not trying to simulate erosion on them.
What's the problem with a true simulation of erosion, it will take a lot of time or the main difficulty in the other?
-
True simulation requires tons of data, not just time. For example, different soils and rocks erode differently, and to simulate it one would actually have to have a more detailed model of the terrain than possible. It would be easier if the whole planet was made of sand or one type of dirt, but it's not (lots of erosion demos work just with that). Means we'd have to get detailed data for rock and soil composition, then also get good enough models for their erosion and then get lots of time to run erosion algorithms to produce good detail.
But we do not need to simulate it to get believable results. What we have got is a map of rock & soil types around the world, and we can get samples of how the erosion looks like on them, and we can (hopefully) devise an algorithm that procedurally generates the desired patterns.
It would be different if we did purely procedural planets, where the large scale erosion pattern isn't given by the elevation data but must be generated. But even there, I lean more towards pattern-matching than to the simulation, because of its efficiency and better compatibility with artistic input. For example, imagine somebody draws a fantasy planet, outlining continent shapes and mountain ranges. You can run erosion simulation to break the mountain range into peaks and to produce the river network, but it requires many passes and the erosion model usually doesn't contain nearly enough information to produce interesting results globally.
Also, artists may want to place major rivers there that suit the world setting, and want the rest of the world be refined automatically. A reverse-erosion algorithm would take these input data, compute river basins and iterate from the river upwards, carving the terrain in the process. Basically saying - if the river goes here, the surrounding terrain must be like this, to be realistic. The reverse process enforces patterns (with procedural alteration) to produce something believable. Doing so it needs a lot less resources and computation passes than the simulation.
-
Just now, the fractal detail often look alien
(http://s017.radikal.ru/i415/1608/af/d73a8a6440b3t.jpg) (http://radikal.ru/fp/35c50d8e1d4c4169a0a26ad446e046d5)
Weak erosion (not to river networks) possibly could not take into account the different types of rock, but would make a fractal more natural details, like this
(http://s019.radikal.ru/i609/1608/3f/146c598474d9t.jpg) (http://radikal.ru/fp/48bc9a3f678744339a60fa92f0259208)
(http://s020.radikal.ru/i721/1608/ef/fa6b07eefa57t.jpg) (http://radikal.ru/fp/9289c30c7efb410fab60ebbb9e5a5f61)
Let's hope that your new erosion algorithm will work even better.
-
After watching Outerra for years, I have the impression that the fractal refinement has a tendency to destroy the type of detail shown in that final picture.
-
After watching Outerra for years, I have the impression that the fractal refinement has a tendency to destroy the type of detail shown in that final picture.
Do you mean the kind of crumble mixture we see on (mostly) mountains?
I think the DEM data is still too low in resolution to unveil finer erosion structures.
-
After watching Outerra for years, I have the impression that the fractal refinement has a tendency to destroy the type of detail shown in that final picture.
Which is were the different erosion patterns (presets) come in. I assume those will not be affected by the fractal refinement (or at least not that much) and will have some detail sculpted beforehand to match the type of rock and erosion we are trying to portray. Or better yet, the erosion patterns will be instructions to the fractal refinement algorithm itself, making it follow certain rules depending on the rock and erosion type.
-
Good work so far, Outerra team. However, I must urge you to quickly fix vegetation zones in Siberia - the sharp transistion from coniferous to savannah vegetation is unrealistic, and there should not be any savannah vegetation in Yakutia anyway!
You should implempt smooth transistions between vegetation zones or transparent (mixed) vegetation zones.
-
The problem is likely in insufficient species coverage (what species should be there?), or in some bugs in climatic data used for the computation of probabilities. Normally the transition is gradual, as the probabilities for one species fall, they will first reduce in occurrence and only then be replaced by other species.
-
Hi, today it has been exactly one year since the last update. Any info on when the next version will be released? Regularly looking at the dev screens over at Twitter, I somehow have the feeling this one will be VERY big. ;)
-
this one will be VERY big. ;)
Certainly at least in terms of data size :)
Which is one of the problems, how to set up reliable servers for data download. I think we are nearing the final revision of the new earth dataset, and the size is ~70 GB for the base datasets @30m, additional 20 GB for the global road network, and assuming 30-50 GB for global building data.
We will put some additional effort into compression optimalizations, but that's 10% tops. Also need a way to turn off some layers ...
-
Hi, my first post here :) but I'm watching the project maany years, and recently i made a purchase.
I'm not an expert, but is it possible to incorporate somehow a bittorrent protocol to Outerra client, that players can shere THAT! amount of data? Yes, you probably need to have some players with full database on hdd, but harddrives are not that expensive. Actually I'm willing to buy separate hdd for that experience of outerra if i need 150gigs.
I just can not wait for that update :D It will be huge :D I just wish to see more vehicle mods and some work on car physics, also more weather events like snow.
Sorry for my english ;)
-
DCS World is using torrent and normal Http download option.
-
this one will be VERY big. ;)
Certainly at least in terms of data size :)
Which is one of the problems, how to set up reliable servers for data download. I think we are nearing the final revision of the new earth dataset, and the size is ~70 GB for the base datasets @30m, additional 20 GB for the global road network, and assuming 30-50 GB for global building data.
We will put some additional effort into compression optimalizations, but that's 10% tops. Also need a way to turn off some layers ...
When you say Base Data Set @30 is 70 Gb, are you already including global road and global buildings?
In case no. Why Dataset is now sensibly bigger than before (about 15 Gb)? Only for rivers and lakes or terrain is more detailed or refined?
In case yes, i suppouse terrain remain the same as before.
-
I'm not an expert, but is it possible to incorporate somehow a bittorrent protocol to Outerra client, that players can shere THAT! amount of data?
We had a bittorrent client incorporated into older versions, though it was never enabled due to insufficient testing and problems it was causing, and later removed altogether.
Then we made a torrent file with http seeds which kind of worked, but there usually was not enough seeders and bittorrent has a habit of ignoring the http seeds after some timeout errors on the http and then waiting indefinitely for someone else to have the data ...
-
When you say Base Data Set @30 is 70 Gb, are you already including global road and global buildings?
In case no. Why Dataset is now sensibly bigger than before (about 15 Gb)? Only for rivers and lakes or terrain is more detailed or refined?
In case yes, i suppouse terrain remain the same as before.
No, that's why I wrote "additional 20 GB for the global road network ...".
It's bigger because although earth2 was made from 30m source data, it was resampled to around 76m to avoid having large data with SRTM sources that were stated to be 30m, but in fact heavily filtered and the actual resolution consequently lower.
There's a factor of 3x when getting to the next resolution level from 76m (to ~38m).
The new earth dataset (logically named earth7) is using some new DEM source data plus additional processing that justifies the need for for additional data, mainly in the nicest parts of the world. However, it's possible to constraint the data to use the coarser level and still have it looking good - and that's what we'll likely do, placing an option to have full resolution in world settings, but default it to a reduced one. That will lower the requirements for both users and servers.
There's now also a new dataset for water - encoding water level info for lakes and ground below the sea level, but that one is just around 2 GB.
Also those 15GB were without rivers (that came later).
-
Yes thats what i understood, thanks for explain it.
Do you plan make comparisson between details from previous dataset and new dataset? Because will be another improvement you worked out for the next big update. Terrain more detailed and real 30m dataset finally sounds great to me. I will be the one downloading the full detail version for sure.
-
Not sure if we find time for comparisons - I'd think that's something people can and want to do. Earth2 should still run in the new version, as well as the original dataset (earth).
-
Funny, as you might have noticed, cameni's avoiding to talk about an ETA for the Next Big Thing (umm, Update).
-
and that's what we'll likely do, placing an option to have full resolution in world settings, but default it to a reduced one.
Guess I'm gonna have to remember that...