CrainBramp Posted November 17, 2019 Share Posted November 17, 2019 The idea of a Blood Moon every 7 days has always been a bit formulaic to me, and I've been in search of a way to get a random Blood Moon. A17 provided an answer to this in the Blood Moon Frequency/Range, but if I'm not mistaken, something has changed in the way that's calculated slightly in A18. Am not sure if I'm (a) derping and not remembering right [it's quite possible], or (b) something actually changed. Regardless of that answer, the way that it seems to be handled in A18 is a bit counter-intuitive and "spoiler-ish" to me. In A17, I used to run a frequency of 1 with a range of 5, which basically gave me a 20% chance per day, where one Blood Moon must occur every 5 days at minimum. Back to backs, were possible, but of course with a diminishing chance. For instance, the first BM happened between 2-5 days away. This is the way I'd expect it work. Now, in A18, it appears now that the Frequency determines the first day of the Blood Moon, regardless of the Range. For instance, a 1/5, as before will happen on day 1 *always*, then proceed as normal. A 2/5, the first blood *always* happens on Day 2, then proceeds as normal. I've tested this on two RWG maps with about 25 fresh starts of the server: keeping the Generated Worlds folder and deleting the profiles file along with the actual generated county folder. Each time, the Blood Moon always, without fail, starts on the Frequency day. The NEXT BM varies as I would expect, but that first BM always seems to land on the Frequecy. This seems a bit counter-intuitive, and spoiler-ish, considering using this, you can pick the day the Blood Moon begins on. So, is there a way to get a truly random blood moon? TFP? Link to comment Share on other sites More sharing options...
theFlu Posted November 18, 2019 Share Posted November 18, 2019 A little odd behaviour, for sure. As a workaround you could just skip the first one then? Start with freq 1, range however far you want to have the first real one, play your D1 and skip to morning once the BM kicks in. Kill or killall the couple buggers that have time to spawn, or just leg it. It should decide the next BM at 22:00 on the current one, so you can't skip it before it starts, but after that it should be fine. Never tried, but this is how I understood it - not exactly sure on the timing. Link to comment Share on other sites More sharing options...
XCOUNTRY Posted November 18, 2019 Share Posted November 18, 2019 I have mine on 1 frequency and the range at 7. I didn't get on on day 1, I believe I got my first blood moon on day 9. Then a decent period (I think it was on 16) I got my second. I've had back to back days once, but everything else seems pretty random to me. I'm playing on the base Navezgane map. No mods. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 18, 2019 Author Share Posted November 18, 2019 Thanks for the input, both of you. XCountry: are you playing SP or MP? I'm almost exclusively MP: I don't think I've played a SP game in several versions (even when doing Let's Plays, I play on a server). I'll do a bit of testing, myself. Anyone else have any experiences? Perhaps TFP would like to chime in? Link to comment Share on other sites More sharing options...
Limdood Posted November 18, 2019 Share Posted November 18, 2019 if you're just looking to skip that first BM, why not just hop on top of a building for the first one? Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 18, 2019 Share Posted November 18, 2019 Random in computers isn't really random. I'm thinking that the randomness of the blood moon range is determined by your seed and all your restarting of the same generated world resulted in the same random values for blood moons, and therefore a possibly incorrect conclusion in your test results. For example, if you shared your seed with someone, it is quite possible that they would have their blood moons on the exact same days you did in your game provided that they gave it the same range settings. EDIT: Or perhaps this same value that is randomly generated is always used for the first one based on seed, but then after it's something different. Who knows? It would be better to check with different generated worlds. 2 isn't enough to say that the first BM is always the first range selection. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 18, 2019 Author Share Posted November 18, 2019 For example, if you shared your seed with someone, it is quite possible that they would have their blood moons on the exact same days you did in your game provided that they gave it the same range settings. This had occurred to me. However, this x25 two-seed test was only after a separate test where I'd done a dozen RWG's in a row, with the exact same results. This test was actually because the biome.png is the EXACT same biome map, for every RWG, no matter what the seed was: the map is simply rotated. During THAT testing, I just happened to notice the BM. My speculation (among many) was that perhaps the starting BM is actually tied to the seed. That still wouldn't explain, however why a dozen+ different RWG seeds all start w/a Day 1 BM on a 1/5 setting. Link to comment Share on other sites More sharing options...
theFlu Posted November 18, 2019 Share Posted November 18, 2019 I have mine on 1 frequency and the range at 7. I didn't get on on day 1, I believe I got my first blood moon on day 9.That doesn't sound exactly bug-free either (1+7=8...). You sure you didn't get a "surprisingly big wandering horde" on D1? Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 18, 2019 Share Posted November 18, 2019 This had occurred to me. However, this x25 two-seed test was only after a separate test where I'd done a dozen RWG's in a row, with the exact same results. This test was actually because the biome.png is the EXACT same biome map, for every RWG, no matter what the seed was: the map is simply rotated. During THAT testing, I just happened to notice the BM. My speculation (among many) was that perhaps the starting BM is actually tied to the seed. That still wouldn't explain, however why a dozen+ different RWG seeds all start w/a Day 1 BM on a 1/5 setting. I see. Right, it wouldn't explain that. But it would explain why not everybody gets a BM on day 1 with a 1/x setting. Still though, you only fully observed the BM days on 2 seeds after realizing. The likelihood of 2 seeds resulting in day 1 BMs (provided that they are seed driven) isn't that far-fetched. Link to comment Share on other sites More sharing options...
XCOUNTRY Posted November 19, 2019 Share Posted November 19, 2019 That doesn't sound exactly bug-free either (1+7=8...). You sure you didn't get a "surprisingly big wandering horde" on D1? Come to think of it, my first one was on day 5 or 6, I'm on day 45 now and my memory is foggy. The second one was probably on 9. Definitely not a wandering horde. The point I was making was I didn't get a BM horde on day one, and every time I've messed with the option, it seems to work as intended and give me random BM nights. I've only played on Navezgane since A18 rolled out. Gonna wait until RWG gets ironed out more. CrainBramp: Mine is set up for "MP", but probably not what you're describing. I'm the host and the game terminates when I log off. No server. Anyone notice that you get a 12am tip-off to the upcoming Bloodmoon? The sky turns blood red at midnight (0000) preceding the BM night for about 1 minute. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 20, 2019 Author Share Posted November 20, 2019 .... you only fully observed the BM days on 2 seeds after realizing. The likelihood of 2 seeds resulting in day 1 BMs (provided that they are seed driven) isn't that far-fetched. Have done an additional test last night with a dozen RWG maps, different seeds, all with a 1/5 (that took a while). Every map begins it's blood moon on Day 1. Every biome.png is the exact same png file, simply rotated or flipped. The biome.png issue has been an existing issue since A18 first released. I've been reporting it in both bug report threads as well as publicly since B149: Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 20, 2019 Share Posted November 20, 2019 Have done an additional test last night with a dozen RWG maps, different seeds, all with a 1/5 (that took a while). Every map begins it's blood moon on Day 1. Every biome.png is the exact same png file, simply rotated or flipped. The biome.png issue has been an existing issue since A18 first released. I've been reporting it in both bug report threads as well as publicly since B149: I wouldn’t waste your time reporting the biome image. It’s intended as a placeholder because they are still working on the generation. The issue with this is unrelated to the other one. Your test results are interesting though. I wonder what 1/4 or 1/6 would give? It could be a math error that only shows up for 1/5. I’m only saying this because other people are claiming that 1/x for them is not guaranteeing a BM on day one. It would be nice to find out why that is. Another possibility is that the seed names are similar? I don’t know the hashing algorithm, but for example all seeds with no numbers in them might default the BM to day 1. You never know. What kind of seed names did you use? Link to comment Share on other sites More sharing options...
CrainBramp Posted November 20, 2019 Author Share Posted November 20, 2019 Interesting on the biome map. Would have been nice if someone had mentioned that on a bug report So I've done no additional tests other than 1/5 in GENERATION; I've changed it post-generation, even before logging in. This changes the start day to whatever I change the frequency to. I've not tested changing it post-login, although that's an intriguing idea. I'll do a bit of testing tonight. I run a gaming community and have 7 dedicated servers colo'ed, so I can generate maps 7 at a time. Couple of hours, I can have 30 or so maps over a various range of settings. As far as seeds, I do random 16-character alphanumeric mixed upper/lower case strings for each. Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 20, 2019 Share Posted November 20, 2019 Interesting on the biome map. Would have been nice if someone had mentioned that on a bug report So I've done no additional tests other than 1/5 in GENERATION; I've changed it post-generation, even before logging in. This changes the start day to whatever I change the frequency to. I've not tested changing it post-login, although that's an intriguing idea. I'll do a bit of testing tonight. I run a gaming community and have 7 dedicated servers colo'ed, so I can generate maps 7 at a time. Couple of hours, I can have 30 or so maps over a various range of settings. As far as seeds, I do random 16-character alphanumeric mixed upper/lower case strings for each. It's looking more and more like there is something wrong with the setting when using random gen and that maybe on Navezgane, the setting works as intended. When reading the posts in here again, I realize that the same person (XCOUNTRY) who had a setting different than 1/5 also has been running strictly Navezgane. This explains the discrepancy (why not everybody gets a BM on day 1 with a 1/x setting). I would try it with Nav instead of wasting the time on 1/not5. I find it weird that they would design it that way, but I suppose each game type could have its own way to generate the random BM range. It's probably more like some logic error in the random gen code that accidentally overwrites the generated start day with the setting itself. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 20, 2019 Author Share Posted November 20, 2019 It's looking more and more like there is something wrong with the setting when using random gen and that maybe on Navezgane, the setting works as intended. Ah, I was working under the assumption you knew this already, as it was part of the thread. This was why I had only been speaking about RWG, as I'm aware BM calculation in non-RWG maps works properly. Simply don't play pregenerated maps as a personal thing, and have actually never even played on Navezgane. Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 20, 2019 Share Posted November 20, 2019 Ah, I was working under the assumption you knew this already, as it was part of the thread. This was why I had only been speaking about RWG, as I'm aware BM calculation in non-RWG maps works properly. Simply don't play pregenerated maps as a personal thing, and have actually never even played on Navezgane. No, it didn't occur to me. Definitely a bug. It should be checked then if the setting also works as intended for the packaged pregen maps, just to be able to strongly say that it is broken only for games started with maps generated locally... and then should be reported, but making a strong note of these findings. Otherwise QA is likely to read the report and just start up a map in Nav and say it's fine. I've not tested changing it post-login, although that's an intriguing idea. Yeah it is... if you find something about that too, please post as well. If changing the setting after running the game once (while still on day one) results in it working as intended, then the problem is even more likely to be that the random value generated is being overwritten by the frequency variable at some point during the map generation... a lot easier to locate and fix with that extra information. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 20, 2019 Author Share Posted November 20, 2019 Here's my testing matrix. Won't take me much more of the evening. I have scripts that auto-gen maps already, it was pretty easy to alter them to do the BM ranges, do the restarts, parse the logs for the values, etc https://rebrand.ly/rwgtest-a18-bm Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 20, 2019 Share Posted November 20, 2019 Here's my testing matrix. Won't take me much more of the evening. I have scripts that auto-gen maps already, it was pretty easy to alter them to do the BM ranges, do the restarts, parse the logs for the values, etc https://rebrand.ly/rwgtest-a18-bm Looks great! Link to comment Share on other sites More sharing options...
CrainBramp Posted November 21, 2019 Author Share Posted November 21, 2019 Haven't gotten the post-logins done yet, but I think the results speak for themselves: https://rebrand.ly/rwgtest-a18-bm Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 21, 2019 Share Posted November 21, 2019 Haven't gotten the post-logins done yet, but I think the results speak for themselves: https://rebrand.ly/rwgtest-a18-bm Yeah. Still curious about the post-logins though. What I don't like seeing in this report is the Navezgane numbers. Previously we stated that the BM settings were fine in Navezgane maps. I really don't see that.The only variance is that sometimes if you choose freq of 1, first BM might be on day 2. So now I'm thinking it is bugged across the board, Nav or not... and that little variance is caused by some other factor... such as some code intended to prevent the first BM if gamestage is too low, but it's not always successful in doing its job. Link to comment Share on other sites More sharing options...
CrainBramp Posted November 22, 2019 Author Share Posted November 22, 2019 I'll be doing the post-logins tonight, as I've got some Let's Play episodes to record anyway. As far as Navezgane, I had done a couple prior and saw the BM moved one of those, and simply took it, with others' statements, that it was simply working. Was definitely thinking in this bigger sample that it seemed it didn't vary as much as I would have expected. But, I wasn't trying to project my own assumptions. Will update when I get the post-logins done, and then maybe try to put it into a understandable bug report. Link to comment Share on other sites More sharing options...
faatal Posted November 28, 2019 Share Posted November 28, 2019 Haven't gotten the post-logins done yet, but I think the results speak for themselves: https://rebrand.ly/rwgtest-a18-bm World seed is set from game name. On your Navezgane chart you show Map Seed as blank, but it is the game name that you should test. Link to comment Share on other sites More sharing options...
AtomicUs5000 Posted November 28, 2019 Share Posted November 28, 2019 World seed is set from game name. On your Navezgane chart you show Map Seed as blank, but it is the game name that you should test. What do you think about the results for RWG though? Does all that get thrown in the toilet too? Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.