Jump to content

The White Guardian

Members
  • Posts

    1,683
  • Joined

  • Last visited

Everything posted by The White Guardian

  1. @Disparia Books the cloud problem was fixed because in the previous version the configs used textures from the default EVE setup (the BoulderCo folder). This new update comes bundled with noise and particle textures, removing the need for the BoulderCo folder. Thank you for confirming that CKAN updated automatically by the way, that was a big worry for me. On a side note: Disparia is still in beta, it'll get some overhauls in the future. The final version will be somewhat like this: (Image of Glacia, as it'll appear in the next version of Planet Cyran) For everyone experiencing lag in the current version, that's because I'm an idiot. In the EVE configuration Cerillion uses the same cloudtexture about 4-5 times. The problem is that this texture is 8K. With WaterEye enabled that means that your PC has to load 6-7 8K maps plus Scatterer plus noisetextures plus moons plus skybox. I'll release a hotfix today adding a 2K texture for stuff like sandstorms and mist to be a bit more performance-friendly. That should leave all of you with only 1 8K map loaded by default. Because of university however (current subject: quantum computers) I won't be home until six P.M. local time (as I'm writing this it's 9:30 A.M. so, simple math: in eight-and-a-half hours I'll be home. When I arrive I'll upload the new version immediately so expect the hotfix in nine hours. Since the time zone difference with the USA is roughly six hours, if you live in the USA the hotfix will likely go live around 3:30 P.M. I'm also working on a sister mod for Cerillion, currently named 'Kovilon' that will make some changes to the inner solar system, beyond the orbit of Eve. Release date will probably around early May, though I'll try to get it done before that. If anyone has any ideas for the moon system, feel free to share! I might end up adding the best ideas. Fun fact about Cerillion by the way: it was designed for the Kopernicus Creativity Marathon, and ended up winning the second KCM. It's been through many design stages though. First concept was an alien-ish ringed planet with purple oceans, red terrian and purple trees going by the name of 'Trogavel'. Second design was a canyon-ish sandy desert with the eventual green oasis named 'Nisma', which got really close to the final design of Cerillion. Maybe I'll revive these old designs one day in some form. @Gameslinx for Cerillion I use VertexPlanet. The geological anomalies on the planet were pure luck, so I went like 'OMG I GOTTA USE THIS!' when I saw them, though you can replicate the effects with a MapDecal. Still, try messing around with VertexPlanet, it's basically an all-in-one mod - in-depth procedural generation of planets. For inspiration on procedural generation try taking a look at @KillAshley's work. Uncharted Lands is fully procedural, as you can see there you can do pretty amazing stuff with it.
  2. Currently uploading the update to Google Drive and SpaceDock. How CKAN will handle the update I do not know. Version 2.0 released. Again, I don't know how CKAN will respond to this. If it doesn't update automatically - manual installation is very, very easy. As easy as dropping a single folder in KSPDirectory/GameData.
  3. Nevermind, I got it. For some reason I cannot select a texture as both the color and heightmap. Darn PQS systems. Oh, well. Expect moon 4 to be done soon.
  4. Okay, moon number 4 will likely not be in the next update and so won't half of the easter eggs be. Because *bleep* do I hate working with MapDecals. Right now it's offsetting the entire terrain instead of applying the deformity of the decal, which is frustrating as heck. Another option would be a heightmap, but it's really annoying when something like this just screams 'a-nope!' for no reason. God, I feel like shooting something right now out of frustration. *Eyes copy of Ratchet & Clank* That'll do. @Gameslinx could you please give me a hand here? radius = 90000 position = 53, 95, -120 angle = 70 absolute = true absoluteOffset = 0 useAlphaHeightSmoothing = false heightMap = Planet_Cerillion/PluginData/Canyon1.dds colorMap = Planet_Cerillion/PluginData/Canyon1.dds heightMapDeformity = 10000 smoothHeight = 0 smoothColor = 0 cullBlack = true enabled = true order = 600 This is the code I'm using. For some reason it's applying the deformity as an offset however - it raises a huge chunk of terrain but this chunk itself has no deformity. So, what obvious mistake am I overseeing here? Because I feel like the answer is right in front of me.
  5. @Disparia Books watching your recentmost stream right now (I planned on watching it live but it started at midnight where I live...). You misunderstood a thing about procedural generation: the planet is generated by the computer on the go, no maps have to be loaded. All 'PQSMods' responsible for altering the terrain however have a value for the 'seed'. This value randomizes the placement of the noise. It's simliar to Minecraft's world generation algorithm - when creating a world you can enter a specific generation seed. Using the same seed results in the exact same world. For KSP planets this is the same. Entering the same PQSMod on different planets with the same seed results in the same noise generation. So if I would transfer Cerillion's mod, VertexPlanet, to the Mun for example, it'll look similar. Not identical because the Mun has Noisemods as well that will be applied on top. A planet that's different for everyone though... that's an interesting concept. It'll probably require a custom DLL to generate a new seed for the surface on the go, but I'd love to give it a try one day. Other ideas for future addons by the way: - Take the Spacecenter and push it to Cerillion. - Terraformed Cerillion.
  6. That is because Kopernicus is the engine behind custom planets and rings. No Kopernicus -> no rings at all. Try rotating the ring texture 90 degrees clockwise.
  7. First off - please don't quote the entire OP for those that have subscribed to this thread. The issue you are having could be because of the way KSP handles normal maps on gasplanets. Unlike rocky planets, the normal map is not ovelayed over the planet in full but is instead used as a noisetexture. Jool, for example, uses the texture cloud_normal, which is the same normal map Kerbin uses for it's grass. For the normalmap to display correctly, add the following line to Material{} bumpMapScale = 1,1 So, it should look like this: Body { ScaledVersion { Material { bumpMapScale = 1,1 ...
  8. Your answer is in the logfile. As you can see it errors right after it tries to read your heightmap. [LOG 16:55:47]: Parsing Target map in (Kopernicus.Configuration.ModLoader.VertexHeightMap) as (Kopernicus.Configuration.MapSOParser_GreyScale`1[MapSO]) [LOG 16:55:47]: Exception Was Recorded: Object reference not set to an instance of an object 'Parsing target map' -> reading the value of 'map' 'MapSOParser' -> texture loader So, in normal English: [LOG 16:55:47]: Loading file requested by 'map' in mod VertexHeightMap [LOG 16:55:47]: Error loading the map. Cannot proceed adding PQS to non-parseable terrain. Conclusion: the problem is your heightmap. Judging by the error message either the filepath is incorrect or Kopernicus can't read the map.
  9. @Thomas P. I think MeshScatter needs some more work. MeshScatter { materialType = DiffuseDetail collide = true science = false Experiment { } Material { color = 0.2,0.5,0.3,1 mainTex = BUILTIN/distantground mainTexScale = 0.5,0.5 mainTexOffset = 0,0 detail = Planet_Cerillion/Textures/Cliff detailScale = 0.5,0.5 detailOffset = 0,0 } scatterName = Misc seed = 1721 maxCache = 512 maxScatter = 20 map = Planet_Cerillion/Object/Scatter.dds mesh = BUILTIN/boulder minSubdivision 1 countPerSqM = 1 verticalOffset = 0 minScale = 1 maxScale = 5 castShadows = false receiveShadows = false enabled = true order = 100 } Unless this code is incorrect, it does not seem to work. I checked the logfile - Kopernicus doesn't acknowledge the mod's existence.
  10. @Disparia Books That's probably because the current cloudsystem uses the default BoulderCo textures. In the next update this dependency will become obsolete. Cloudsystems are integrated with Cerillion - all it needs is the EVE plugin. Same goes for Scatterer.
  11. WEATHER WARNING: INCOMING STORMS! You can now find sandstorms (that lower the light by ~85%...) and mist banks (minus ~15% light) drifting on Cerillion. All moons are nearing completion. Stuff like biome maps and science defs are planned for the update after that. So far three out of four moons are complete, although the fourth one may be revamped later.
  12. I thought there was something called the 'solarPowerCurve' for that but I'm not entirely certain. I thought RSS used it to account for the new system size.
  13. Introducing the first official add-on for Cerillion: Watereye - Ultra-HD planet maps - Ultra-HD surface textures It'll come at some point after the next update.
  14. Thanks a bunch! Time for me to start integrating the new ring feature into my mods then! I thought it was a photoshopped image of noodles planet...
  15. @GregroxMun driving a train is always tiring. I hope you're not full zombie mode behind your computer though... I personally have mixed feelings about the spreadsheet being permission locked. I respect @OhioBob's decision to protect his work and find it quite a handy solution, but on the other hand, I can't make any edits myself (such as adding gases to the table). Oh, well. I can always still make an atmosphere by hand if necessary.
  16. Just a note for everyone: in the spreadsheet '1360' is defined as the 'solar constant' by default. However, this value is for G-class stars. For other star classes you can use these values: O 40800000 B 6800000 A 23120 F 4080 G 1360 K 408 M 108.8 Color legend: O = Blue B = Blueish white A = White F = Yellowish white G = Yellow K = Orange M = Red Should you have trouble remembering this order, astronomers use the following sentence to help them should they forget: 'Oh Be A Fine Girl Kiss Me'. As you can see, the first letters form the order: O-B-A-F-G-K-M @OhioBob the following function can be added to the spreadheet to make this process automatic: =IF(X="O",40800000,IF(X="B",6800000,IF(X="A",23120,IF(X="F",4080,IF(X="G",1360,IF(X="K",408,IF(X="M",108.8,0))))))) With X being the cell that takes the 'star class'. I've tested this method using the formula for effective temperature given in your thread 'modeling atmospheres in KSP' and it seems to work as intended. So, this 'star class cell' will only accept the entries O, B, A, F, G, K and M. Anything else will result in a value of 0. (I created a 'star spreadsheet' for personal use that automatically computes stuff like lightColors. Instead of giving a value of 0, it shamelessly calls me a 'scrub' if I try to enter 'Taco!' as a valid star class... ) Also, can Xenon be added to the table of atmospheric composition? Here's the info: Molar mass: 0.1313 kg/mol Density: 5.9 kg/m3 For the heat spec. constant (adiabatic index) it uses the same value as the other noble gases (Krypton, Argon, Helium) Also, there's a typo in the spreadsheet: Krypton's symbol is 'Kr'.
  17. @blackrack did you use a horizontal texture by chance? Kopernicus automatically rotates them it seems. Rotating the texture makes the ring work regardless of new or old shaders. @Red Shirt Sarnus' rings were never gone, the first pixel on the texture is just transparent.
  18. Imported my old stuff from SpaceEngine 0.9.7.1. Ah, good old projects that I worked on until three years ago when I made the jump to KSP.

    I'm actually gonna import some of my best SpaceEngine works. Now, expert Kopernicus modders, before you come across this message and go like "NOOOOOO!" I gotta show you all what I mean.

    xS3cMoG.jpg

    THAT is the kind of stuff I made in SpaceEngine.

    1. Show previous comments  8 more
    2. Whirligig Girl

      Whirligig Girl

      With regards to transporting Space Engine stuff into KSP.

      The main problem that importing Space Engine stuff into KSP is generally the simple fact that unless you add noise on top of them, you end up with mushy or blocky (or both!) height. This is also a problem for color maps (Especially with ocean worlds. Look at @Galileo Planet Pack, for instance, and you'll see bits of blue on the shore. If that's been fiixed since I last looked at it I do apologize) but it's less of a big deal because they can be at a lower resolution without losing important details, sometimes. The problem is if you add noise on top of a fully finished heightmap, you get too much noise and the whole thing becomes a mess.

      So I would recommend that to put a SE world into KSP, try to remove all small scale procedural detail from the SE world, and do your best to replicate a similar look with PQSMods. Alternatively, export the map and then manually iron out the small scale noise from the map and add some with PQSMods.

      Note that this doesn't just apply to Space Engine worlds. It also applies to Real Solar System worlds, and worlds that use real world detail. (SSRSS, RSS, Galileo, to name a few)

      Now there are some super-ninja tricks to get less blockiness out of heightmaps, but it works best for hand-painted ones instead of things like SE or RSS maps. The trick is to have lower resolution maps for higher scale things, and work your way up and down respectively to add details.

      For instance, imagine you have a large lobed asteroid that you want to add, but you also want really fine detail. If you tried to implement it without using any noise, you would end up with a pixelated effect on the terrain. If you use a lower resolution heightmap, the pixelation goes down because KSP blurs it out better. But then you lose small details like craters and cliffs and such.

      The solution is to use a low resolution (512x256) map for the lobes of the asteroid, and then put the mountains and cliffs and craters on a higher resolution map. (depending on the scale, 1024x or 2048x) The technique can also be applied to large bodies. Use a 1024x map for large scale topology, and put small scale details on a higher resolution map. Now the smaller details will still be mushy and/or blocky to some extent (Not obviously so, but still noticeable), but the terrain in general won't.

      Now, @The White Guardian, that volcanic crater looking thing in the picture would be a shame to leave mushy, but it would also be a shame to have the whole terrain look blocky. Try using a low definition heightmap for much of the planet, erasing some of the important high detail land formations, and then copying them onto a high definition map.

      This method is a lot of work, but it gets good results, especially in conjunction with procedural PQSMods.

    3. Gameslinx

      Gameslinx

      There's a neat nifty trick to try and remove some noise 'errors' in heightmaps without manually smoothing them by setting 

      PQS
      {
      	minLevel = 2
      	maxLevel = 8
      }

      to this:

      PQS
      {
      	minLevel = 2
      	maxLevel = 6
      }

      That actually reduces the quality of the heightmap in relation to the terrain (It has the same effect. I don't really know how to describe it... It lowers the octaves of the entire terrain, if you know what I mean). It's good for smoothing out jagged areas, and generally keeps the rest of the detail relatively well...

    4. Whirligig Girl

      Whirligig Girl

      Level changing is tricky, to be sure. First of all, it decreases the definition of all terrain on the planet, including procedural. Second of all, under certain conditions having too high or too low a Level can result in the planet being partially intangible.

  19. @JacobJHC That right there in the background is a WIP skybox (still hand-painting the stars) that will likely be shipped with Cerillion in a future update. You can also spot the two in-dev moons in the background if you look close enough. The last image is a WIP noise texture by the way, it'll be more fine-tuned in the release version.
  20. A note to everyone BEFORE YOU UPDATE: From the next version of Cerillion and onward they will be anchored to a specific 'flightGlobalsIndex'. What this does: Kopernicus automatically assigns each planet an 'index' number based off their position, for example if the Sun is 0 then Moho is 1, Eve = 2, Gilly = 3, Kerbin = 4, etc. Previously Cerillion had no index specified so that Kopernicus could automatically assign it a number upon loading the mod. This was done for optimal compatibility, because when two planets have the same index number both will not load. Therefore Cerillion could adapt no matter which mods were installed. However, if another mod had no Index number, there is a chance that they would be indexed before Cerilllion. An example would be KerbolOrigins (tagging @amarius1) which adds several objects to the inner Kerbal solar system that have no FGI (if I remember correctly that is). This would mean that the index numbers would shift. So, what does this mean to gameplay? Every craft in KSP does not have an orbit specified by the name of the object that it is orbiting. Instead orbits in KSP work using these index numbers. The save file of a craft orbiting Kerbin would read '4' as a reference body if we'd assume my previous example (Sun = 0, Moho = 1, etc) to be true. Since it is a number that is not tied to the name of the planet or any of it's properties for that matter, orbits could shift when installing multiple planet packs with no FlightGlobalsIndex. Therefore, from the next version and onward Cerillion and it's moons will be tied to a specific FlightGlobalsIndex. This means that, when updating to the next version, crafts landed on Cerillion or orbiting Cerillion for that matter may shift to orbit it's innermost moon, Mavis, instead. This is a one-time thing. So, what should you do when updating? - Try to finish any mission to Cerillion before installing the update. - Make a backup of your save file just in case. - Install the new update. The next update will only contain Mavis and Jacob for now, the last two moons are still in-development and will be added over time. Since these will have an FGI from the start they will not affect the orbits of spacecrafts, that is, unless the spacecrafts are orbiting in the vicinity of these new moons. When I get home (and can get a look at the configs) I'll post the orbital info of Mavis and Jacob so any orbiting crafts can be adjusted in advance. I'm terribly sorry for this FGI inconvenience but I better make the jump now before there is a whole bunch of moons that can complicate things further. Again, this is a one-time inconvenience. On a separate note, I'm looking into the asteroid systems to see if I can add little moonlets to Cerillion's ringsystem as possible outpost locations. Minor Kopernicus-moons (read: moons with gravity) are an idea that I have to work out further, the idea being several tiny objects in the outermost regions of Cerillion's system as scattered asteroids.
  21. @Gameslinx check on the main Kopernicus thread for the KopernicusExamples that @KillAshley created, you'll find the link in the OP. One of the example configs is moving the KSC.
  22. AFAIK it adds the temperature on the y-axis to the craft every frame, so negative could work, but I'm not sure if it's a good idea. The curve you posted here means that people can just hover over 10m and cool down immensely fast. Your best bet would be to gradually decrease the temperature-per-frame taking heat radiation into account. I'm willing to write the curve for you if you'd like, I made one for Cyran to simulate it's superheated atmosphere and heat radiation.
×
×
  • Create New...