400 billions of procedural generated or 100 handcrafted?

[PavelKCZ]

They've yet to add exploration to Elite.

Right now its all jump, scan, screenshot and repeat for as long as your addled brain can handle it.

Problem is that nobody can say what the proper Exploration in game like ED should be.

Personally, I think that it is not possible to implement thing called "Exploration" in ED in the age of YouTube and Internet. Simply because if FD will add thousands of new objects/events somewhere in the Galaxy, sooner or later someone will find them and will make a YT video about this subject. Subsequently, all players and their mothers will flight there, will see the same thing that they saw in the YT video and will post at the forum something like "Meh, exploration sucks because it is always the same. Fly somewhere for a long time and find the same dull thing that hundreds of other players found before you".

Problem is that in the shared Galaxy, any exploration will be always a bit dull for anybody, except for the first pilot who discovered something new. For example, I can imagine that it must be a real thrill for the CMDR who found the Jacques station. For the rest, it was not exploration, but simple travel to somewhere.

 

[aRJay]

True! 100 crafted systems is not ED is another game. I think you misunderstood the intent of the op. It's not a critique or a change request. It's an analysis to see if after 5 years of playing the game the procedural generation is still considered the best choice compared to handcrafted contents in terms of gameplay.
Imho procedural generation gives you the vastness of a galaxy which is good for the realism but it also proved many times to have many limitations for the gameplay because of lack of interactivity between the generated systems and the lack of variations of the contents. Yes you can add many more variations but the effort it takes in the end is the same as generating handcrafted contents so in the end does it worth the effort?

For this game the procedural generation route is the right choice, for something with a much smaller game space hand crafting could well be the way to go, but then I am quite content with the gameplay we have now.

 

[Sysmon]

Problem is that nobody can say what the proper Exploration in game like ED should be.

Personally, I think that it is not possible to implement thing called "Exploration" in ED in the age of YouTube and Internet. Simply because if FD will add thousands of new objects/events somewhere in the Galaxy, sooner or later someone will find them and will make a YT video about this subject. Subsequently, all players and their mothers will flight there, will see the same thing that they saw in the YT video and will post at the forum something like "Meh, exploration sucks because it is always the same. Fly somewhere for a long time and find the same dull thing that hundreds of other players found before you".

Problem is that in the shared Galaxy, any exploration will be always a bit dull for anybody, except for the first pilot who discovered something new. For example, I can imagine that it must be a real thrill for the CMDR who found the Jacques station. For the rest, it was not exploration, but simple travel to somewhere.

Exploration is the act of investigating an unfamiliar area, or examining something in detail. Surely people can find something unfamiliar among the 99.9+ % systems that still haven't been visited.

S

 

[Sysmon]

They've yet to add exploration to Elite.

Right now its all jump, scan, screenshot and repeat for as long as your addled brain can handle it.

Jumping is required for all travel. Scanning is a feature most developed for exploration gameplay, but is useful elsewhere as well. Screenshots can be taken any time, anywhere, and is not required for exploration. If what you describe above is all you do when exploring, maybe you need to look around for ways to expand on it? Otherwise you are just doing a very barebones way of playing this game, and that I can imagine would become boring really fast.

Here are some potential things to do:
Find life forms
Visit unusual stars and systems, including black holes and WR-type stars
Visit tourist beacons for game lore and unusual system features
Visit nebulae
Look for orbital bodies in interesting constellations: Shepherd moons, trojans, fast orbits, eccentric orbits, overlapping planets, strange rings
Look for planets with interesting features: High gravity, proximity to stars/other planets, interesting landscapes
Look for specific types of planets for profit
Look for specific materials
Find Raxxla
Find other alien sites
Visit Colonia/Beagle Point/any of the other remote outposts.
And so on...

S

 

[BlackSpaceCowboy]

Both. 100 handcrafted systems surrounded by billions of procedurally generated systems.

Most of the content in the 100 handrafted systems would also be procedurally generated. The cities or outposts would be handcrafted but the deserts and vast landscape would be procedural.

A good balance of the two is key.

 

[Abysinth]

Game development took lately this procedural generation road

umm, Elite was doing procedural generation for its galaxies in 1984...

 

[Lestat]

Game development took lately this procedural generation road. I'm wondering from a developer point of view of it's really better to have a useless high number of systems (useless because they will never be discovered entirely) or maybe 100 handcrafted systems with several handcrafted contents suitable for all game styles proposed.
After playing Elite since so long what do you players think about this trend?

You know SenseiMatty. I think if we get landing on earth like worlds. I think they will have a few handcrafted worlds maybe 50 to 100. Frontier already has some Mechanics on other games that they might be able to use on Elite Dangerous.

 

[Sapyx]

For me, one of the attractions about exploration in ED is the chance of finding things that even the developers are not only unaware of, but were unaware could even happen. Discovering and examining the emergent patterns that arise out of procedural generation. Now, some of those "emergent patterns" are what would reasonably be called "bugs" - things that, when they occur, cause game-breaking glitches. But others are not only hamless, but, once discovered, have been embraced by not only the community, by by FD as well.

Case study 1: Glowing Green Giants. There's no way that FD deliberaely planned for the existence of these extremely improbable worlds. They just created rules about how gas giants will look, and an extremely rare possible outcome of those rules is "bright green". They are very rare, yet completely harmless. Not only did FD not patch them out of existence (like they did with the Glowing Pink Giants and Glowing White Giants) , they embraced them by giving them unique planetary identification text and separate entries in the Codex.

Case study 2: Trojan planets: where one planet is locked into the Lagrange or "Trojan" point of another, much larger planet, separated by 60 degrees. FD apparently believed that Trojan orbits would not be stable in ED, so when the FE2 prequel planets were copied across to ED, the two Trojan planets in Eta Cassiopeia were "normalized" so that they are not Trojans any more. However, Trojan orbits can actually be formed by the Stellar Forge; they're even not that hard to find, if you know what to look for (I've found two, personally). There's even an inhabited Forge-generated Trojan Earth-like in the Alpha Caeli system, plus another dozen or so known throughout the galaxy.

Case Study 3: Earth-like Groupings. If you take the bulk amassed data of all the stats for Earth-like planets, some interesting patterns emerge. My favourite is plotting the atmospheric pressure versus the surface gravity: one would "logically" expect a more-or-less linear arrangement: bigger planets tend to have thicker atmospheres. Instead, three distinct "lobes" appear, with clear gaps between them with few if any worlds foujnd in the gaps. Why? There's no "logical" or "rational" reason I can think of why the worlds are like this; there must be some underlying interplay of the Stellar Forge equations that creates ELWs, that prohibits certain combinations. FD can't have deliberately designed things this way, otherwise, I suspect, they might have tried harder to make planets with exact-match-for-Earth conditons (the 1:1 point on the graph) fall in the middle of a Group, rather than be ouside in the Gaps.

 

[graflex22]

For me, one of the attractions about exploration in ED is the chance of finding things that even the developers are not only unaware of, but were unaware could even happen. Discovering and examining the emergent patterns that arise out of procedural generation. Now, some of those "emergent patterns" are what would reasonably be called "bugs" - things that, when they occur, cause game-breaking glitches. But others are not only hamless, but, once discovered, have been embraced by not only the community, by by FD as well.

Case study 1: Glowing Green Giants. There's no way that FD deliberaely planned for the existence of these extremely improbable worlds. They just created rules about how gas giants will look, and an extremely rare possible outcome of those rules is "bright green". They are very rare, yet completely harmless. Not only did FD not patch them out of existence (like they did with the Glowing Pink Giants and Glowing White Giants) , they embraced them by giving them unique planetary identification text and separate entries in the Codex.

Case study 2: Trojan planets: where one planet is locked into the Lagrange or "Trojan" point of another, much larger planet, separated by 60 degrees. FD apparently believed that Trojan orbits would not be stable in ED, so when the FE2 prequel planets were copeid across to ED, the two Trojan planets in Eta Cassiopeia were "normalized" so that they are not Trojans any more. However, Trojan orbits can actually be formed by the Stellar Forge; they're even not that hard to find, if you know what to look for (I've found two, personally). There's even an inhabited Forge-generated Trojan Earth-like in the Alpha Caeli system, plus another dozen or so known throughout the galaxy.

Case Study 3: Earth-like Groupings. If you take the bulk amassed data of all the stats for Earth-like planets, some interesting patterns emerge. My favourite is plotting the atmospheric pressure versus the surface gravity: one would "logically" expect a more-or-less linear arrangement: bigger planets tend to have thicker atmospheres. Instead, three distinct "lobes" appear, with clear gaps between them with few if any worlds foujnd in the gaps. Why? There's no "logical" or "rational" reason I can think of why the worlds are like this; there must be some underlying interplay of the Stellar Forge equations that creates ELWs, that prohibits certain combinations. FD can't have deliberately designed things this way, otherwise, I suspect, they might have tried harder to make planets with exact-match-for-Earth conditons (the 1:1 point on the graph) fall in the middle of a Group, rather than be ouside in the Gaps.

up vote for graphs and words.

 

[Cmnd Fulsom]

Rivers and caves on vacuum worlds...

I would have had an issue with it if there were there now.

Anyway, we have both now, so why chose?

Airless worlds have caves, in fact, they probably have much larger caves because airless worlds tend to small and have low gravity such that they can't retain an atmosphere. Without erosion from the elements or high gravity to pull things down you would expect things like lava tubes and similar to be much larger and last much longer.



Stay Frosty,



Cmnd Fulsom

 

[dxm55]

They just need to handcraft the few major worlds, especially when atmo planets are up. Earth, Mars, Achenar, Facece, Lave, Alioth, Cubeo, and all the other Powers' capital planets. Everything else can remain procgen.


And even in handcrafting these worlds, especially the cities, can be done at a more cosmetic level. Because unlike Star Citizen, we are never truly gonna be able to visit every street, nook or cranny. We'd be seeing these cities from the air, and at a distance most likely.

I'd take it that FD will try to conceal the practical lack of details in these cities with a minimum altitude restriction over populate areas. Since it only makes sense, just like IRL you can't fly too low over cities.

 

[Quantis Trap]

Game development took lately this procedural generation road. I'm wondering from a developer point of view of it's really better to have a useless high number of systems (useless because they will never be discovered entirely) or maybe 100 handcrafted systems with several handcrafted contents suitable for all game styles proposed.
After playing Elite since so long what do you players think about this trend?

This procedural generation vs hand crafted debate will go on and there is only one real solution which is an all of the above approach also enabling user driven content.

In fact tapping the user base for content generation (to varying degrees) is the trend with several existing and many new space games such as No Man's Sky, Stationeers and Starbase just to name a few.

Yes, I know that these games I mentioned don't have to level of detail and realism of Elite Dangerous, but the approach is still valid. I hope that is where Frontier will is going with the game in the next couple years.

 

[IndigoWyrd]

What's wrong with both?

I have nothing against hand-crafted systems, but at the same time, when you stop and really think about it, our own, actual galaxy, indeed the entirety of the universe itself, beyond the visible universe, is, for all intents and purposes, procedurally generated. We may not ever fully understand those procedures, but we CAN see, as we expand our knowledge, more and more of those very procedures at work throughout the galaxy, throughout the universe. As we grow our understanding of how things work, how planets and moons and stars are formed, we unravel those procedures.

And in a game, such as this, we can see those same procedures reflected in how our digital galaxy is formed, especially as we expand our pool of data into the astronomical.

A bit of hand-crafting sprinkled out here and there is fine, but the sense of scale and wonder are greatly diminished too as the data pool is made smaller.

 

[varonica]

Airless worlds have caves, in fact, they probably have much larger caves because airless worlds tend to small and have low gravity such that they can't retain an atmosphere. Without erosion from the elements or high gravity to pull things down you would expect things like lava tubes and similar to be much larger and last much longer.



Stay Frosty,



Cmnd Fulsom

The vast majority of caves on earth are formed from the dissolution of limestone in carbonated water, that is rain, yes there are some formed from other actions but most of these absolutely must have active vulcanism and most need a body with a liquid magma core. Lava tubes won't form on most of the small bodies we visit because the vulcanism is of a different type, it's vulcanism caused by gravitation stress rather than vulcanism from a retained liquid magma core. Without a liquid magma core you can't have large volcanoes, you can't have caldera's, you can't have magma chambers so you aren't going to have large amounts of lava flowing to form lava tubes.

But you can still have some, there are in fact some caverns on the moon but these are likely caused by evaporation of underground water leaving empty voids, and there may be a few fossil lava tubes, but the moon has been pummelled by big rocks so much that in the past it's unlikely there will be many. So there should be some, but I suspect primarily the reason we don't have caves is for the same reason we don't have overhangs, because the surface generation system simply can't handle them. I hope this will be fixed in the 2020 update with a new surface generation system and then we may see caves.

 

[Muzz UK]

Useless is an opinion... I'm sure explorers mostly would disagree OP.

However 20,000 odd players over 100 systems would be quite cramped.

The thing is with the pro gen you can just keep adding stuff over time and its easily implemented.

 

[Barking_Mad]

100% on 100 hand crafted!

You'd be complaining in record time I fear!

 

[Slàine Sìdh]

Game development took lately this procedural generation road. I'm wondering from a developer point of view of it's really better to have a useless high number of systems (useless because they will never be discovered entirely) or maybe 100 handcrafted systems with several handcrafted contents suitable for all game styles proposed.
After playing Elite since so long what do you players think about this trend?

400 billion procedurally generated systems. Every time.

Sure, you can handcraft a hundred intricately designed systems, fill them with exciting scripted events and narrative questlines and such, and you might have built a really cool game. But it's not one I'd be playing. I want the openness, even if there's repetition and sameness that might be apparent at a high enough level. But while I don't claim any expertise on how these things are created from a technical standpoint, given several layers of interlaced generation I don't see any reason why procedural content can't be complex enough at an everyday enough level to camouflage the patterns. I always think it's like soundwaves: interlace a few simple, regular sine waves at slightly different frequencies and you get a massively complex waveform. Build an algorithm for the bodies in the galaxy, another determining the factions, one for the surface maps of planets and the distribution of cities (another for their street maps and population movements), vegetation and animal life, others for the placement of stations and settlements, the density of traffic, the aggressiveness and motivations of NPCs, the dynamics of the trade markets, the space weather, the assembly of parts making up the structures, and so on and so on, and I'd have thought you could create something pretty intricate.

I'm guessing this is kind of what Frontier have done with ED, and this isn't me saying "this is how you design a game". I have no idea how you design a game. But I'm working on my understanding of how simple patterns can turn into complex ones with no obvious pattern to them. Yes, it's still there - it's still regular - but at a resolution that isn't immediately obvious. And all this depends on the player being someone like me: someone for whom the ideal game isn't a game so much as a toy (a difference Maxis used to emphasise with their Sim- games). Some people want to play through a story, and that's absolutely cool, I have no problem with that. Others, like me, want to be put in an environment, given the means to navigate it, and told to go off and do our thing.

I want as big a playing area as possible, and no 'win' condition except any I set for myself.

 

[KarmaCop]

Game development took lately this procedural generation road....

Yoda...is that you?

P2P instancing a problem might cause...mmmm!

 

[VR SenseiMatty]

Problem is that nobody can say what the proper Exploration in game like ED should be.

Personally, I think that it is not possible to implement thing called "Exploration" in ED in the age of YouTube and Internet. Simply because if FD will add thousands of new objects/events somewhere in the Galaxy, sooner or later someone will find them and will make a YT video about this subject. Subsequently, all players and their mothers will flight there, will see the same thing that they saw in the YT video and will post at the forum something like "Meh, exploration sucks because it is always the same. Fly somewhere for a long time and find the same dull thing that hundreds of other players found before you".

Problem is that in the shared Galaxy, any exploration will be always a bit dull for anybody, except for the first pilot who discovered something new. For example, I can imagine that it must be a real thrill for the CMDR who found the Jacques station. For the rest, it was not exploration, but simple travel to somewhere.

This leads again to the topic that in the galaxy there's not enough things to be found for everyone

 

[VR SenseiMatty]

Both. 100 handcrafted systems surrounded by billions of procedurally generated systems.

Most of the content in the 100 handrafted systems would also be procedurally generated. The cities or outposts would be handcrafted but the deserts and vast landscape would be procedural.

A good balance of the two is key.

I agree but that's would be a titanic project that no one would do