EDO goea a bit beyond a height map and a shade of purple. The EDO stellar forge simulates the distribution of mass around a star, planet formation, the age and resulting number of impacts, plate techtonics and now sedimentary rocks and ancient dried up rivers. I've seen layers in the rocks, but not eveywhere, that's clever.
I paid for a Space Sim, not a buggy mess with immersion killing pop up menus.
- Thread starter Trailstorm
- Start date
-
- Tags
- state-of-the-game
EDO goea a bit beyond a height map and a shade of purple. The EDO stellar forge simulates the distribution of mass around a star, planet formation, the age and resulting number of impacts, plate techtonics and now sedimentary rocks and ancient dried up rivers. I've seen layers in the rocks, but not eveywhere, that's clever.
That's good. Now you need to to recreate that so the entire planet doesn't look the same all over and, from distance, that planet doesn't look like the next one .. oh and you need to do that a couple of trillion times.
Yeah, it sounds nice but in reality, in the game, that isn't what's happening.
Landable planets. Do you think that a moon in close orbit to a gas giant is going to be the same as a moon with a distant orbit to a small planet? The game says they are. The reality is that the gravity forces from the large gas planet would heat the planet up, would have molten eruptions everywhere, would have large canyon type cracks etc. In the game there's no difference. We get some cut and paste craters of different sizes.
Those planets in the game with the large crevices, how do they form on an zero atmosphere moon again? those crevices are made by weather and water in the universe I am a part of. If they're earthquake cracks then why are they shallow sloped? No atmosphere to weather the sharp cracks.
You pick any planet in Elite and drive around it. Every new square mile area would look the same as the previous square mile area, yet I'm supposed to believe this tech is ground breaking stuff created by experts in planet formations and physics?
The colours.. I was watching their youtube stream last week , a white planet with a green atmosphere orbiting a white star had bright red shadows. Shadows are pastel colours, they're a mix of direct light colours mixed with atmospheric colours + ground reflection ambient colours.
If they're using a mathematical model to generate planets then it's about as realistic as NMS model used to create planets.
If you have a mathematical model which is able to simulate an entire galaxy accurately to the smallest detail, please share with the rest of humanity.
If you went to a restaurant and ordered a fish and I give you turd on a plate would you not be able to criticize due to you not being a chef?
EDO is using a more realistic model though you can argue it's more simplistic than NMS in that it's not decorated like the planets in NMS are. In other words they're not (yet) covered in - independently evolved? - grasses, diplodocats and orange apple trees. By decorating a planet you can add a lot of variety very quickly. However, if you want to add your volcanos - assuming you don't want them all to look like Fuji - you do need a sufficently advanced base geography. We're definietly closer to that in EDO than were in Horizons and the devs are of course - at the same time - developing NPC's, tools, weapons, hitboxes and back end. EDO is really day one.
You could. But you wouldn't go in the kitchen and explain to the chief how to cook the meal you ordered, would you ?
Brebus knows I have many issues with planetary gen ATM, look at my sign. But I don't think I can do better at all.
I don't recall going to FD headquarters and knocking on their door.
Re-read the OP he was giving free lesson for Fdev on how to fix their game.
and I was replying to someone.
Have you not ever said anything on this forum on how you think things should be done before?
Great advice too. Remember, this was posted after Frontier basically went "lol idk" with regard to planet tech.
Ful
Stellar Forge is an incredibly cool piece of tech. However, there's also no doubt at this point that EDO's iteration of it isn't well liked. I don't think the other user was saying that they could do better. I believe they were trying to make the point that procedural generation CAN be very easy. I think that point is especially driven home by the thought that procedural generation, in essence, is all about automation and making things easier for yourself.
Obviously there's a lot more that comes into the equation when using procgen for space sims. But looking at EDO it seems totally legitimate to wonder if we maybe could have had, for example, fewer people on tectonic plates, and a few more peeps on making sure not everything looks copypasted and making sure planetary features don't disappear when you approach from space.
The CM said it was difficult to fix. Not "we don't know".
No ? I don't pretend to know better than them when it come to their tech.
The only thing I said was that I think Odyssey should have been cut into piece instead of one monolithic block. And that was a "I think". Not an absolute, "let me teach you how to do it."
They said:
Anecdotal, but in my experience in the software industry "not workable" means no. They'll keep investigating which means it may not be a "no" forever, but this looks about as close as you can get to "lol idk" in corpospeak.
![[VR] DDM_Reaper20](https://forums-cdn.zaonce.net/data/avatars/m/44/44021.jpg?1588332101){kind=avatar}
There is a vast difference between, "We don't have a workable solution," and, "We don't know." The latter admits total ignorance, the former does not. I do not know ANYTHING about procgen (or, indeed, programming in general), so pardon my ignorance, but I imagine that identifying problems is a lot easier than actually correcting them, even if it is known what is wrong.
No offence, not trying to get under your skin.
There is a vast difference between, "We don't have a workable solution," and, "We don't know." The latter admits total ignorance, the former does not. I do not know ANYTHING about procgen (or, indeed, programming in general), so pardon my ignorance, but I imagine that identifying problems is a lot easier than actually correcting them, even if it is known what is wrong.
No offence, not trying to get under your skin.
Oh I'm sure they can think of ways to fix it, just not any they will get internal approval for from MT. That's what "not workable" means in this context in my experience.
I suspect that 99 of those developed are working on other projects in between updating their resumes given the current state of things.
Despite that the planets look like dog sick and the GPU hits 100%.
Noise generated textures have the problem of looking unique, yet entirely the same. You will get such random features that nothing will look memorable. All you will see is a never ending expanse of rolling hills. Actual terrain looks completely different, as it was formed with real geological processes. For example, water erosion is responsible for some of the more memorable features here on Earth. You can't approximate that with noise, there's a definite procedure to it.
I wouldn't be surprised if the new iteration of the terrain system rather ignores the output from Stellar Forge. It's perfectly possible that while the generation of planet composition data and other such stuff is very detailed, the actual terrain engine that computes what we eventually see uses said data in a whimsical manner.