Warhammer drops 5 MAY, and the F1 game is coming up. Draw your own conclusions.
And the Warhammer forum is on fire!
The bugs are more fun than the game...
Raxxla !Hi, just for a laugh I checked Part of the centre immage with a Bar Code reader,It came back with. 723811451. Is this some mystry code ? or someone saying, Ah f**** it I`m done with this bit.
Just a coincidence
Given that it’s the planet generation system (Horizons version or Odyssey version) that generates the surfaces, expecting the Stellar Forge to do it is massively unrealistic.
And what trade offs and negatives would be the consequence?
It is unrealistic. If continuing with the Horizons planet gen tech was realistic then why would they invest in a full rework of the planet gen system? And given that they’ve made that investment it’s not realistic to think they’re just going to discard it.
Changing the Planet Gen system isn’t shortcutting the Stellar Forge. All the parameters from the Stellar Forge are still being generated and fed into the Planet Gen system - unless there’s anything to say different?
As the change was from one planet gen system to a different one, all is presumably fine then.
Exobiology is likely a factor. In Horizons, it works by the Planet Gen system generating the entire planet, and then the entire planet being scanned for suitable locations for localised bio sites. With Odyssey they’ve changed to much broader biomes.
There’s more to it than just the bio though. They’ve moved to a ‘geome’ based model as well. Part of the reason for that was to better handle different macro-terrain types. Horizons only worked on a full body basis, so worked for Rocky Worlds and Ice Worlds, but Rocky Ice worlds were never properly handled. - This obviously becomes more and more of a consideration as more complex atmospheric bodies bring more and more macro-terrain types. There’s probably more as well - I’m just covering a few things Dr Ross talked about in the stream about the new tech.
Do you mean having mouse headlook? If so, there’s a keybind for it.
If you mean that you’re using mouse headlook but other people can’t see your head moving when you do, then no idea.
It's not twisting anything when you're wrong and continue to be.
In what way is the post I made suggesting that the planets are no longer procedurally generated? I said tiling is circumventing the stellar forge. By skipping the aspect that creates the surface from the existing algorithm, and instead replacing it with pre-built tiles. they're both technically procedural, but one is the stellar forge, and the other is something new..something technically less.
What made the stellar forge interesting and attractive was that it didn't do those kinds of things. That was something that games already leveraged and used to make such environments. What sets the stellar forge apart is that it doesn't use those short cuts to create it's environment. That's why circumventing it in this way rubs those players who value the technical demonstration aspect of the game the wrong way. And that was a big deal to many who come from the kickstarter days.
Pivoting? is that your attempt to make it look like i'm changing the subject, because you made the statement of players opposed to odyssey "tiling" as unrealistic in their desire to utilize a more horizon-ish approach to planets. I'm just responding to your post, not making up new topics.
Exploring is just as boring and pointless now as it was in horizons. I know this topic likes to be prefaced as a exploration ruiner, but i think most of the players complaining about it do so because it feels like a retreat of existing functionality in a context that has other instances of such retreats happening in the game all coinciding with odyssey or leading into it. Exploration is mostly only good for screenshots and videos. It made the most "hardcore" feel accomplished when they found something generated by horizons to actually look interesting, because nobody had a hand in it. That has been taken away from them with much more adulterated planets in odyssey.
Repeating patterns isn't tiling though. Certain kinds of repetition in nature is entirely expected and a consequence of fractals that are often leveraged in precedural functions. What makes one better than another is in how seamless and natural it looks. Odyssey happened to have some fairly unnatural examples of nearly exact copies of the same large scale patterns used showing up on not just the same planet, but even across planets in close proximity to eachother to be noticed by players in a very short amount of time. Something that didn't happen with horizons. Yea horizons had the crater feature that was an option it used in places, but it wasn't used in a way that made you go "oh wow, someone copied and pasted the same image a bunch of times here and it looks fake." If horizons had a couple samples to supplement it's procedural surfaces, odyssey appears to have moved to another level of sampling and it skipped into easily noticeable territory. This is an issue of relying on samples less in horizons to come up with the surfaces it did is seen as a better technical achievement than the usage odyssey leverages. To many people who signed on during the kickstarter and such, the lack of hand-creating was a major selling point and these larger, more frequent samples erodes that for them.
You can't authoritatively say if something looks or appears one way or another? It's basic observation ability and awareness of the context it's happening in. I didn't say they definitively did, i said it looked like they did. The perception of the defect is what players see, not the factual circumstances of the reality of the situation because fdev never tells people that so those are always at best educated guesses. We should all easily have the authority to see why this is a bigger annoying problem in odyssey than it was in horizons and why the explanation of how it will be addressed (it wont be) will be received the way it is in the current context of the players and the game. Because we're all here experiencing it. And that's what I was doing, commenting on why this matters to players but what existed in horizons did not.
You want to paint people upset about this as being unrealistic, when you've not addressed the fact that the game was more than capable of not producing such obvious copies before odyssey and the developers never really explained why other than an attempt at better aesthetics, this new tiling was needed. So to players this is either an aesthetic choice that sacrifices some existing experience for some ostensibly new one or it's an attempt to deal with a complication brought on by a new feature (that such players probably dont really care about). That's why this appears like a short cut. In the context of the rest of what's going on in odyssey, this looks like either some means of avoiding the need to rework a complicated piece of software like the stellar forge to get better terrain without making it extremely worse in other areas or a means of reducing terrain building overhead by increasing the size of samples the terrain generation uses.
It's not unrealistic to think a more elegant solution should have been entirely within the realm of not needing large scale assets to copy and paste (or at least, not leveraging the use of those assets to occur so frequently and close together to be noticeable). It's not asking for something technically out of reach of what the software or hardware has been proven to do. The only thing unrealistic about it might be being able to get someone with enough expertise on how the stellar forge works to be able to make the changes needed.
That's not true, if something actually good comes about I dont neg it out of some limitation to only be negative. You're just observing a period of time in which fdev has been exceedingly good about putting out bad releases while making a mess of things. I said positive things about carrier interiors despite their bugginess and odd choice of interaction controls. It just gets lost in the sea of bad things that have been done or released this past year. I was also positive about carriers when they were released. And before that, the mining and exploration mechanic changes, the rift narrative (other than the ending) and jaques (at least leading into it) and powerplay (when it was released) and the FDL and vulture. it's not my fault their balance sheet of successful updates/ changes/ content is overshadowed currently by their unsuccessful or periods of nothing. Different play styles will obviously vary in how they receive what content has been made in the last bunch of updates as their impact is varied.
If you want to talk about repeating patterns and are not happy about the negativity surrounding the game and fdev in general, you should look into things like play testing before release and using beta cycles that document bugs to fix said bugs before release instead of just cash opportunities and things like that. Doing something about that pattern the game has been stuck on forever should go a long way towards having better updates and as a happy side-effect, resolve the pattern of negativity you have a problem with. Complaining about people being negative with what's happened in elite this past year+ is like walking into a ghetto and complaining that the people there look upset or depressed. in any case, i look forward to the next update where this all starts to turn around and everyone is talking about what's going on in the game rather than what's going to happen to the game. Like we used to when it seemed like fdev was actually excited and invested in it.
the planet generation system is part of the stellar forge. in the venn diagram, it's inside it.
It's not like the stellar forge is just a mechanism for placing items. It's the name of the system that predurally creates all of the stellar objects you see in the game. From the stars, to the content of the systems, to the roids, and the rings etc. It puts them there, it determines what they'll look like and what they'll consist of. That's all the stellar forge.
The planet generation system is just a subset of that.
edit: I was rethinking and I think when they talk about the planet generation system, it's not just the part that has to do with the stellar forge (the part that makes up the look) but also what dictates the object content on it, the physics changes from space that occur and gravity effects and the changes to any flight models or atmospheric features. So it's heavily overlapped with the stellar forge circle because it's a name given to the system of not just creating the planet surface, but creating the environment your client is currently in. So it includes the part that makes the planet come alive / interactive / distinct from space and not just be a static object and that part is outside of the stellar forge's wheel house. But I would still certainly categorize everything static about the planet as being part of the stellar forge - holistically.
That really depends on the real reason for why tiles were put into play. If it's just an aesthetic need, then there could be no tradeoffs. If it's something else than not being able to rely on a certain terrain for a given area if a certain terrain feature you're looking for exists might be a tradeoff. I could see that being a problem if the exo biology feature that requires a certain number of populations of a given organism to exist in order to fully scan doesn't have the habitats it is associated with ...or they are extremely far apart.
But you could just use the horizons way of doing things and work it, you could also end up with some truly unique new features and landscape that you wouldn't have been able to see if you were using large scale samples. But you may have to re-think the exo biology mechanic.
That really assumes fdev had chosen to do what they did because they couldn't achieve their goal after trying (or that fdev makes choices based on what's best for the game or players - all of which we could easily find examples of that not being the case). There could have been other reasons for doing what they did that had to do with say, gameplay priorities in new game mechanics being introduced in odyssey. Gameplay priorities and mechanics that are entirely choices and optional and potentially the players who have a problem with the tiling dont care about and wouldn't be upset if they had not existed or existed entirely differently.
Obviously they're not going to throw it away now. Just as they wont trash odyssey and start over. It's not unrealistic to do so, in the sense that they can totally decide to do and actually do it. If you're going to define unrealistic based on what fdev is likely to do rather than what's possible for someone similar but not fdev to do then you're going to find almost anything players want changed can be labeled unrealistic.
Generating the planet is still part of the stellar forge. It's just the final step for objects. What is dictating the use of the samples vs non-samples and where those samples go is unknown. Obviously it diverges at some point in the process from what horizons did. How far along it actually diverges is less important than how far along it looks like it diverges. It's hard to imagine it being more procedural with larger and more frequent use of pre-made assets - but it could be. It's just that reality wouldn't matter because it looks less.Changing the Planet Gen system isn’t shortcutting the Stellar Forge. All the parameters from the Stellar Forge are still being generated and fed into the Planet Gen system - unless there’s anything to say different?
As the change was from one planet gen system to a different one, all is presumably fine then.
The problem here still would be what are they leveraging to create the differences in the new way planets are generated vs the old ... are they leveraging the use of pre-built assets to predetermine the limitations of what the stellar forge can create or have they reworked the procedural creation to result in the creation of the geomes and other odyssey alterations?
Forcing the procedure to limits pre-determined from these assets would be seen as a step back rather than a step forward in a game built around a proc gen galaxy. Even if it looked prettier. As that is taking things closer to hand-created and being hand created is not what made the stellar forge impressive.
It would be interesting to see what odyssey planets would look like without the use of large scale tiling assets.
Last edited:
Citation, please, because as far as I can recall, the devs never spoke of terrain generation as being a part of Stellar Forge, using the latter name for the algorithm that generates the astronomical objects with all their bulk data.
AFAIK SF creates the seed which Hor or EDO then translates into an actual surface. Hor used less variables and applied only 'math', EDO uses more info from the seed and uses tiles + Pg.
PS4 owners can play the Bedrock Edition instead, and I am sure XB1 owners can, too.
The very first picture in the “New Planet Tech is KILLER of Exploration” thread shows the kind of terrain features that occur frequently in nature but are so unlikely to occur in EDH’s terrain generator that they are effectively impossible.
Also, using templates does not mean the terrain is not procedurally generated. There is still quite a lot of procedural generation, especially at smaller scales. Yes, I wish there was less repetition and more randomness, but as I already stated, I consider EDO’s terrain generation system to be decisively superior to EDH’s plasma fractal mountains and canyons.
“Seed” in the context of procedural generation usually means a random number generator seed. That’s not what Stellar Forge does. SF (as I understand the way FDEV uses that term) creates the galaxy down to individual astronomical bodies and provides parameters (mass, size, composition, atmosphere, volcanism etc.) of the relevant body to the terrain generator.
I think just to cover the SF vs Planet Gen thing. I’m happy to treat the Planet Gen as part of the SF, but the issue with that is that the solution you were suggesting is that the SF handles the planets. And if the planet gen system is part of the SF, then that’s what’s already happening.
Anyway, I think your edit covers quite a bit of it, but the long and short of it is that the SF runs a simulation of system development from an original gas cloud, which generates a whole load of info about each body in the system (mass, composition, orbit, tectonics, etc, - basically all the info from the Sysmap plus a whole load more). It then passes all those parameters to the planet generation system which based on those parameters produces the surface of the planet, whether that’s a 3D landable surface or the 2D bitmap wrapped onto a sphere that’s used for non-landables.
None of the main SF stuff has changed AFAIK.
What’s changed is just the Planet Gen system. So all it’s doing is taking the same info and using it in a different way to generate the planet surfaces.
So what the Horizons one does is to use the SF info to pick from a very small number of templates (or archetypes), and then produces a heightmap using that archetype. As I mentioned in the previous post, one of the issues with Horizons is that that works for bodies which are all one type - icy worlds and rocky worlds - but it doesn’t work properly for bodies that have different environments across them - rocky ice worlds being the main example from Horizons landables.
What the Odyssey tech is apparently doing is being a bit more granular, and rather than doing the templates at a whole planet level, the planet gen system is taking the SF data and working out what the various ‘geomes’ are on the planet and applying templates at that level. - and what that means for example is that a body can have both rocky geomes, and icy geomes. That in principle can then be extended to all kinds of geomes (and biomes).
Anyway, I think the other key thing is that Dr Ross said that a major reason they’d changed the planet generation system was to make it expandable and extendable in the future without needing another full rework. It’s that which suggests that continuing with the Horizons planet gen tech isn’t a realistic option - because if the Horizons tech was already of that nature it could have been expanded and extended and they wouldn’t have had to invest in doing the full rework of the planet gen tech that they have done.
But they dont have any fundamentally different attribute that didn't pre-exist. You have valleys, canyons, plains, mountains all relatively the same physical limits.
What i was referring to was the lack of unique new physical elements not just unlikely to occur, but simply can't occur in horizons. Things like cliff overhangs. Canyons with bridges. Caves. Broken planets. Etc. (not counting features that may require liquid, deforming or weather)
When you have a system that is determined from the smallest scales to the largest by a singular seed and function ...and then move that to something where now that determination is augmented by some hand-created assets along the way ...that is the "not procedurally generated" that people are referring to. people aren't saying everything about it is now not procedural. That's not what people are referring to. The surfaces of environments in many games are procedural noise but nobody would classify a world like zelda's as procedurally generated like they do in Elite Dangerous.
What the template represents is a failure of the stellar forge and planet generation procedural function ... a handicap that a mathematical solution couldn't (or wasn't attempted) be reached to create the desired environments. Some people would rather have the purity (perceived or real) of what horizon had than give that bragging right up and have the game be that much closer to any other game.
Personally, i found the horizon planets to be boring the same way I find the odyssey planets to be boring. They're barren, uninteresting, unutilized and so far they've only been good for taking screenshots on. I've been to many ...many dozens of tens of thousands of systems and have seen many times that more planets and none of them have mattered, been important to my game, uniquely helpful or interesting or memorable beyond the attributes that have nothing to do with look (size, gravity, temp).
So to me, i'd have preferred they stick with the horizons method and tried working the algorithm to create better surfaces just to retain the technical aspect of not having a human hand in what it looked like. Because that would be the only differentiating factor between the two systems that had any value.
I just think the players who do care about what the planets look like, do so in the context of this looking like a downgrade from a technical identity that feels less true now given what they're seeing.
I consider it an element of the stellar forge because there are depictions of all planet types available right now (not simple place-holders but how they look barring another re-creation of this part of the procedural generation). The game creates them all. But there are only a small subset that the client supports creating the surface details, atmosphere, physics, flight model and gravity etc for.
in the same way that we dont break apart different functions in spawning systems vs spawning the galaxy and spawning roid fields ...it's all the stellar forge. From initial seed, down to individual asteroid. How it's divided up and organized doesn't make it not part of that system.
I just also think when they are referring to the planetary generation system, they're not just talking about that, but also the environmental simulation that the surface is part of, and that's the part that is definitely not something you can consider the 'stellar forge'.
If only rocky or icy made a difference, i could get interested in different geomes. It sounds less like it's being more granular, and more like it's using hand-built assets to create that granularity where there was none. If so, that's where i would see it as a downgrade even if it results in more interesting output.
I wish we could rely on what fdev has decided to do to mean that the best choice was made. But they gave us CQC, they gave us the bio scanner mini game. They attempted to say they chose not to give us walking out of a ship onto the surface because it would be too time consuming and tedious. I dont see much reason to think that they made a decision based on what was best for the game or players or that they even made one that was in their best interests and not a total mistake. So I dont trust that a similar outcome wouldn't have been possible without relying on premade tiles simply because fdev spent a bunch of money and went a different direction.
And they gave us a slow on foot GUI to get on foot missions and gear!
Still, I don't understand what's wrong with CQC, I've been there and thought it fun, even though I blew up all the time..
Well in theory the EDH terrain generator might somewhere create erosion features like seen on that picture, but the probability is so small it’s not worth mentioning — and again, that kind of feature is ubiquitous in real life.
Well, overhangs/caves/bridges are really hard unless they are totally hand-crafted (rather than partially hand-crafted like EDO’s features).
As for broken planets, if you imagine something like this reddit thread the comments below the picture explain why that is not a real life thing.
Small non-spherical bodies, OTOH, might be on the cards — it appears the new system can handle them (in fact, I have seen a couple of small moons in EDO with clearly ellipsoid overall shapes, like those fast-rotating gas giants).
What depictions? The system map? It uses the same terrain generators.
Yes, but I have not seen or heard FDEV referring to generation of terrain as a part of Stellar Forge, and it is clear that doing so confuses people.
(For that matter, I do not think the asteroid placement and shape selection algorithm is considered a part of Stellar Forge, either.)
So it is a matter of trust, then.
I do trust that Dr Ross knew what she was doing and really could not create a better terrain generator without relying on templates (which are not the same as tiles), because she obviously knows more about that stuff than I do (and I in turn, having dabbled in procedural terrain generation myself, clearly know more about it than most of people who post in this thread do).
That's the thing that really gets me scratching my head. You don't need to have any PR experience to know that you are supposed to bury bad news with good news. Frontier, on the other hand, just likes to pile bad upon bad. Instead of having some really exciting news ready to share about U12 to wash the bad taste of console cancellation away, they instead decide to announce that a popular pet peeve with Odyssey won't ever be addressed. I understand that decision, but did you really have to prominently announce that right after the bad news for consoles?!
If Frontier isn't doing all they can to demoralize the ED community deliberately, they sure are doing are doing a mighty fine job of it accidentally!
Last edited:
For the record, I believe that Mojang discontinued one version, then released the Bedrock version which gave it better feature parity across all the platforms. Switch, Xbox, and Playstation players can still play a current version of Minecraft and continue to get updates. I don't see the similarity.
