That is what I have gotten with everything - from ops, to auto-combat, to anomaly explorations: the first reload creates a new outcome, but this outcome is then set for every reload after that. However, I don't think I've ever tried save scumming in this game beyond that example, so I don't know how consistent that is. That could be bad RNG, but it sounds pretty consistent with what MrNo is talking about. So I reloaded another 5-6 times, and all of those were the same result of success but detected. I reloaded and tried again, and the second time it succeeded, but was still detected. Originally posted by Midas:I recently had a situation where I used a covert op, which both failed and was detected(55% success, 45% detected). you *could* reload and get a different result out of those). On AoW 3, this was certain events related to AI behavior at turn end, and the spawning of cosmic events on the next turn (i.e. I will note that there are probably cases in this game (there certainly were in AoW 3) where a random event is generated independently of the saved seed. In the past, when asked, Triumph devs have said they do it to make testing bugs easier - after all, it's harder to reproduce a bug when the save file someone provided doesn't produce the same outcomes every time. Sometimes, games do this to prevent save scumming. Each time you save the game, the state of the RNG at that point is included when you load it, it loads that RNG state.Ĭause random events to occur in a different order, and the RNG will produce different results. ![]() This is known as Seeding - providing an initial state for the RNG to start off of that guarantees a certain sequence of random events will produce the same outcome. ![]() maybe someone with a better understanding can make it clear.Ĭomputer Random Number Generators can be set up so that they are *effectively* random - as in, on first contact, you cannot predict what will happen - but on repeat will have the exact same sequence of events. Really this is just speculation and I can't prove any of it. But if the AI changes nothing, then I get the exact same world event every time as the new turn begins. When the results of events at the start of the new turn are determined they are revealed as different. So when the save is reload, the marauders and other AI make their moves (but differently) and the turn ends. Maybe it doesn't attack or it changes the order of the attacks, or if there are any combats it moves slightly differently in the combat so the random numbers in the list get used less or more as units die at different times. So if it's saving the random number seed (or what ever it is) with the random number list predetermined how did the result change? I think its tied to the AI doing something slightly different when moving on their turn. When I reloaded I got a different world event than the previous time. My only other save to escape this loop of crashes was the Auto save at my previous turn's end. Example- Combat kept crashing over and over and reloading into the combat auto save just got the same result. Now I have reloaded a turn for various reasons. ![]() It's been a long time since I did any programming, but I think it requires some specific programming command to generate a new seed for the RND. I can't prove this and I hope someone would show me I'm wrong because I'd rather have new random numbers generated for every event. I believe it does not generate new random ones for each game or even each RND event. From my experience with Age of Wonders 3 and Planetfall I think the game saves a queue of random numbers. Sorry I don't know the definitive answer to your question.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |