Jump to content


  • Posts

  • Joined

  • Last visited

Everything posted by Ninator

  1. Trying the steam democlient, I found something that is highly annoying: At least the nanocrafter seems to run on the client side. When the player logs out with a long jobqeue in the nanocrafter, the material is NOT passed on anymore to a waiting industry. Only when the player logs in again, then the nanocrafter output is flushed, and the waiting industry starts producing. Has it always been like this, or did they "optimize" something ? A similar problem exists now, when using nanocrafter and beeing in VR. upon leaving the VR, the nanocrafter output is flushed immediately, but the linked container is not ready. Therefore typically the nanocrafter jobqeue is stalled because of missing components, and the time the player spent in VR is lost. Are these bugs new ?
  2. Warning : this thread may contain traces of content.
  3. Floating point numbers are tricky. In your example, you only look at a very small area and assume, that the origin of the "universe" is very near. so you wonder, isn't that resolution overkill ? And in fact, for the sole purpose of rendering, when the origin is close by, single resolution floats are the optimal choice. Thats why all rendering piplines know to man are using 4 byte floats. BUT: Dual universe is a space game. And space is fu.ck.ing unbelievable huge. Single resolution floats are just not good enough for that. A 4 byte floats value part has around 23 bits. Divide the size of the universe by that number. Those steps are huge. And you dont want your ship to jump around in 10000 meter steps. You want it to glide smoothly and express all positions with millimeter accuracy. And thats the simple reason why you need double precision floats. And even with these big numbers, you would typically create a position tree to avoid absolute positions as much as possible (one object's position is always described as relative to a parents position, like the next planet or star). Just DU the math.
  4. I was just trying to convince a friend to play this game with me. Thank you for this information. Of course I will not drag him into this hot mess.
  5. Hey guys, there is a misconception here. Stuff like AvA can not be "added" to the game. AvA requires a centralized server architecture that makes strategic decisions, like who shot whoom nearly in realtime. Comparable systems in E-Sports are usually very streamlined and limited to very few players. The DU infrastructure is completely different. It uses the clients workstation for all physics / collision simulations, and uses the server only for data storage. The server itself is programmed in a very generic high level way, or with other words : it's quite slow. So AvA in DU ? not going to happen.
  6. Well, subscriptions have run out a week ago. For me personally, DU has shut down. And I am not even feeling sorry.
  7. Things start to make sense. Of course the wipe decision is engraved into the egg. Belorion , share your findings !
  8. Exactly my question : does it make sense to enable it know, or should we wait until wipe happened ? It would be a big advantage to have those bonuses AFTER release ...
  9. When flying with a carpet to collect ore, it's not because of speed. It's because of much better terrain overview. Have an ecu on your carpet, and whenever you see something, just jump off the carpet directly onto the pile of ore. carpet will land safely meanwhile somewhere in front of you.
  10. The only problem now is : Flying with a magic carpet to search all those piles of rock is impossible (inconvenient). Because when you activate the harvest tool to see the rocks, one can no more control the speed of the carpet with the mouse wheel. Already reported as bug. But maybe that will simply not be necessary anymore in the future, because it looks like the rocks are spawning now right under your feet when calibrating...
  11. The current collision algorithm simply seems to be broken: - after the patch many constructs reported overlapping elements, although they were all checked before the patch. - simply moving these elements one tick to the left and back fixed the error. Although the elements went back to the original position, the collision error was gone. That's simply a bug. It happened on multiple ships. How can stuff like that be released ? Where is the QA department ? I thought the patch was tested on PTS ?
  12. So for a while now I am encountering strange air brake behaviour. Sometimes they simply refuse to brake. Like totally. I am approaching my target at 100km/h hovering 20m above the ground, and its like I have no brakes at all. Only when I turn the ship around so I am sliding sidewards, somehow they kick in, and I am able to brake with CTRL again. What can cause this ??? I really do not like to crash into a market with 2 kilo tons of ore because of this. Sombody has an idea ?
  13. Sure. Take a piece of squared paper. We are going to construct some slopes. But the rule is, that voxel edges lay only on the grid of your paper. You can now define the dimension of your voxel ( how many little boxes on your paper make one voxel ). You will see that this choice defines the possible slopes. The definition of the slope is : angle = atan ( dy / dx ) with dx and dy beeing integral (natural) numbers. This boils down to the fact that you can only construct slopes wich use factors from the prime factorization of your voxels dimension. Sounds very mathematical, I know. But you can visualize that with your paper hands on. Of course somebody could now argue : wait a second, I can create much more slopes by (ab)using fractions if i am allowed to move the voxel edges in 2 dimensions. Sure. And the DU engine does that excessively. I personally do not consider those slopes perfect, and they drive me crazy inside the game. Whenever you encounter strange ripples in one dimension when you just changed a slope in another dimension of the voxel , then the slopes are not "perfect".
  14. Addendum to my previous post: There seems to be a confusion with the 252 vs 253 points. 252 is clearly based on the prime factors 2*2*3*3*7 . The 7 seems to be a random choice, and personally if you go that way, I would have factored the 5 also into it. The thing is , if you draw a voxel of width 1.0 on paper and divide it by 10. What you have then is 10 deltas with a spacing of 0.1 and surprise - 11 dots in each dimension. So the amount of dots must always be +1. Otherwise your voxel can not be divided in the middle. Sounds strange, but its like that. The 252 vs 253 sounds like a communication error between the devs and the community managers to me.
  15. Thank you Deckard for explaining it. This clarifies so much. I gave Tordan, the creator of the wonderful voxel wedge a really hard time, because again and again the points in the wedge seemed not to be perfect. We all assumed that the engine representation of a voxel's width was an integer value and a potency of 2. Man, have we been wrong. I deeply need to apologize to Tordan for that. Not his fault. In fact it is simply mathematically impossible to create a perfect 1/8 or 1/16 wedge with that numerical system. All I can do at this point is to slowly walk away from this numerical nonsense. I am sorry. But from a coders perspective, I simply can not trust the team anymore. Edit : My statements needs more explanation. So someone could actually argue that it makes sense to have as many prime number deviders as possible in the width of a voxel in order to be able to create certain perfect slopes ... Unfortunately the team that coded the available rotation angles for elements did not talk to the voxel team. Everybody who ever created a perfect slope in a cockpit and tried to align glas windows to it knows what I am talking about. It's a mess...
  16. this is failing due to several reasons. did you ever try to put down a static core on an asteroid ? it says "no territory". putting down a space core does not help either, because it can be shot. sorry ... But : as a hint : your space suit is quite capable of flying some su if you are patient ...
  17. and there my friend, I am sorry to tell you, you are wrong. It is all about psychology. There is a huge difference between - if you define the base yield at 100% and punish the player when he is not calibrating versus - if you calibrate nicely, you are rewarded and get more than 100%. see the difference ? Its called motivation. All game loops should be designed to be motivating. If they are not, the game designer failed.
  18. there is no obstruction for weapon damage so far, which is silly.
  19. as Deckard said : it happended and they realized the mistake.
  20. Guys, the problem is a technical one and quite simple : - Demeter patch hit everybody at the same time. - So tax bill must be issued for the whole player base at the very same time. - This means a permanent stressful server overload every week at the very same time. It is very much preferable from a server perspective, if he can deal with taxes one after the other with some pause inbetween. No big deal. What is really surprising is the fact, that the developer who designed that , did not think about it. Its a classic problem ...
  21. Polls showed numbers. Ridiculously low numbers. They showed how many people participated in the polls. Those numbers were 2 digit numbers. I would also try to hide that.
  22. Oh well, please realize that you are talking to CREATOR. He is not a new player at all. All he did was to point out, that this might be a real problem for actual new players.
  23. Mini game / mining units are boring as hell. Borderline annoying. Before demeter I was warping basically every day around. Supply was no problem, if I needed more, then I just dug more. Warplines were spitting out little warp crystals, and fuel tanks were always full. No more. Everything is rationed. Even slow boating a big ship is now considered to be expensive. All I can do is play with my remaining pre-Demeter resources. All progress is gone. Unsubbed. Game is not worth paying money for it.
  24. Ok, I demand that people who comment ,understood the argument. Good luck with your Bauxite or whatever T1 rocks to create some space fuel. Did you find T2 yet ? Then we can start a real debate.
  • Create New...