Elite Dangerous 2 on Unreal Engine 5

That's so that when we are flying our spaceship in to land on said planet we actually see the planet and not some sort of made up placeholder that bears no resemblance to what we will see after we actually land, you know, like we do in Elite Dangerous?

So UE5 is not actually able to render planets in a space ship flying game that allows us to land on said planets, surely that would be a bit of a problem.
Ah you don't even see this planets in ED until you get very close to them, and then you don't even render surface, until you get really close to it. So yeah, i am pretty sure you are seeing as you call it placeholders in ED. And the performace hit in Odyssey surfaces is huge, compared to space. So none renders entire planets.
As I said, you don't get the point.
If you had a point to make, i might have understood it.
 

Robert Maynard

Volunteer Moderator
Reminder to all participants: please discuss the topic. Other participants are not the topic. Failure to comply will result in advisories, warnings, reply bans and / or thread closuire.
 
Last edited:
R type was my poison. So many great space sims from the past.

Galaxians, Galaga, Space invaders, Asteroids :rolleyes:
If I had every quarter that I dropped into these games in the 80's, I would probably be able to own the original game cabinet and afford the warehouse they would fit into right now. I think over 1500 of the games in my emulation cab are space games..err I mean sims ;)
20221127-093443.jpg


20221127-093517.jpg


But back to topic, UE5 is pretty amazing but I don't think a new EDO would in the end benefit from it. Bethesda updated the CE specifically for Starfield and they have like 100 systems and about 1000 planets to explore. I am betting those will all be covered in a month or less by the gaming community with guides and speed run tuber vidoes flooding the internet. I am really stoked for the game but I rdoubt it will replace EDO for me, just a game to add to the collection.
Even with the repetition and grind there's something about the ED universe that brings me back time and time again.
 
But back to topic, UE5 is pretty amazing but I don't think a new EDO would in the end benefit from it. Bethesda updated the CE specifically for Starfield and they have like 100 systems and about 1000 planets to explore. I am betting those will all be covered in a month or less by the gaming community with guides and speed run tuber vidoes flooding the internet. I am really stoked for the game but I rdoubt it will replace EDO for me, just a game to add to the collection.
Even with the repetition and grind there's something about the ED universe that brings me back time and time again.
Starfield 2 would definately benefit from UE 5, since it's full of instances already.
I think Frontier could have invested in to UE5 and started with a simple ED spin off single player - like the story of First Thargoid encounter and John Jameson, and than see how far they will develop UE 5 to take on future of Elite Dangerous.
I don't want for Elite Dangerous to end with this game, i want the franchise to expand, and UE5 seems like a good way to do it.
 
I don't think there is any figure for number of systems that rules out exploration being a major focus for a game's content.
Certainly in terms of physical space a single Saturn-sized ring system contains enough cubic kilometres that an ED-sized player base could never explore all of it, and a single planet at Earth-level detail throughout is also well beyond anyone's ability to hand-generate for a computer game, too.

If 'exploration' continues to imply reaching a star system, doing a cursory scan, and moving on, with travel and scan times that allow one to knock out a system on the order of minutes...sure, you'll need a lot of systems to keep people busy. At the other end of the spectrum, it's taken a good hundred-billion people many thousands of years of history to not even come close to exhausting the exploration opportunities on real-life Earth.

Personally, I think travel is way too fast and far too certain in ED. The content at any given destination is both sparse and repetitive. It's not really a very satisfying exploration experience for me, but that has very little to do with the number of systems.
Sure, but I think whichever way you do with travel times, scan times and equivalents, etc. the maths works out about the same in the end: you either have a small enough region that you can hand-design the content, but a single determined individual can probably see "all" of it, and the player base as a whole will have found "everything" that hasn't been extremely carefully hidden within a few months ... or you can procedurally generate huge numbers of systems / planets / ring system rocks / whatever your base unit of exploration is from templates, and there's a huge amount of stuff to find with even players ten years in being able to see things no-one has ever seen before without having to solve a super-hard puzzle first, but most of it is relatively uninteresting.

I agree that the headline "number of systems" isn't actually the important thing but it's a useful shorthand for available space versus exploration speed.
 
The doesn't matter what graphics engine you use if the game is still a grind. As evidenced by some 8 bit games from a simpler more elegant time. Unreal is open Cobra is proprietary. Apart from Elite D the other games standalone rather than multiplayer . They should like ED PC PZ JP together so you can transport passenger to PC PZ!
 
If it's released in full on next gen consoles, I approve of it and want it, though if it's PC only, it might as well not exist for me! 😯😂😀🤘
 
In sandbox games, you come up with your own purpose. It doesn't change the fact that's it's still a wasted space. This is why exploration in this game is very shallow, and they couldn't even do much with it in 8 years.
Bigger doesn't means better, and Elite Dangerous is number 1 proof of that.
When I go hiking in a forest I don't examine or take special note of every tree. Some trees I will, lets say maybe 10 or 20. If it was decided I only need 20 cool looking trees for a fun walk in the forest we could just get rid of the forest and let people walk in the park. Its basically the same thing right? 10-20 cool big old trees. Paths to walk on. Some birds up in the branches. Its the same thing as a walk in the forest right?

No it is not.
 
When I go hiking in a forest I don't examine or take special note of every tree. Some trees I will, lets say maybe 10 or 20. If it was decided I only need 20 cool looking trees for a fun walk in the forest we could just get rid of the forest and let people walk in the park. Its basically the same thing right? 10-20 cool big old trees. Paths to walk on. Some birds up in the branches. Its the same thing as a walk in the forest right?

No it is not.
Why would you go to the forest to look at the trees? I never hiked my self in the forest, but i imagined people would go on a trip there to change the environment, enjoy the fresh air, the smell of the forest, the sounds of birds, have a nice picnic, or teach your self or your kids some survival skills. A strange example.
On the other hand, with Unreal Engine 5 the forest will look amazing at little perfomance cost.
Ever wondered how Frontier would do Habitable Planets landable, when a small empty planets already destorying game performance? With Cobra Engine it's not possible. They need a new engine to expand this game. Else this is it, they can't add much to it anymore, Cobra Engine won't allow it.
Unreal Engine 5 is the future not just for this but most games.
 
Last edited:
There are only two engines I'm aware of that can do what is needed for Elite Dangerous: Frontier's Cobra engine, and Space Engine. Neither would be considered a general-purpose game engine like UE5.

Think about this: the latest UE5.1 update increases its supported game area to 88,000,000km2. That sounds a lot, but it's actually only ~293Ls. VY Canis Majoris is 1,975,788,000km in diameter - 22 times UE5.1's limit. Then there is the Alpha Centauri system. There would be no hope of fitting the Hutton orbital run into UE5.1.

I agree that it appears that Frontier are having trouble developing their own engine. However, engines like UE5 are designed for small-scale first-person shooters, not space sims.

It's pretty simple, like it or not, without something like the Cobra engine Elite Dangerous could not exist. Maybe they could talk to the Space Engine guys?
 
That sounds a lot, but it's actually only ~293Ls. VY Canis Majoris is 1,975,788,000km in diameter - 22 times UE5.1's limit. Then there is the Alpha Centauri system. There would be no hope of fitting the Hutton orbital run into UE5.1.
There'd be no possibility of simulating an entire star system at 1:1 scale all at once, but it's not as if Cobra does that either (not even in supercruise) - there are plenty of relatively simple techniques to handle things further away than that without players noticing the difference.

293Ls is a big enough distance that nothing smaller than a planet is going to be visible at all, and anything smaller than a star is just going to be a fuzzy dot at best, so you can (simplifying slightly) do two separate renders - one for distant objects with a massive scale factor applied (e.g. 1 megametre = 1 metre), and one for closer-up ones - then use the distant one as a skybox for the close-up one.


(Which is not to disagree with your wider point that making a space game in a general purpose engine is going to at the very least involve a lot of extra bits bolted on to get around standard assumptions)
 
(Which is not to disagree with your wider point that making a space game in a general purpose engine is going to at the very least involve a lot of extra bits bolted on to get around standard assumptions)
Isn't that rather the point, you use Stellar Forge (or something similar) to do the space environment stuff and UE5 to do all the standard game stuff (stop me if I'm being too technical ;))?
 
Isn't that rather the point, you use Stellar Forge (or something similar) to do the space environment stuff and UE5 to do all the standard game stuff (stop me if I'm being too technical ;))?
Sure, but the standard game stuff isn't usually the tricky bit, so if you move all the complicated space stuff outside of "the engine", it then matters even less what engine you use.

Other than maybe some of the "better graphics" ones I can't think of any common feature request or hated bug for ED where the reason it's not added/fixed is likely to be because Cobra doesn't inherently support doing it.
 
Using UE5 will result shrink to "StarField" or "SC" like games.
All those engines expect pre-designed assets with some restrictions. While Elite needs all generated from the set of rules.
All "glitches" we see is because of some rules overlap rarely and they couldn't test initially.
 
Back
Top Bottom