vossiewulf
-
Posts
805 -
Joined
-
Last visited
Content Type
Profiles
Forums
Developer Articles
KSP2 Release Notes
Everything posted by vossiewulf
-
Still using the 1.30 release, I see quite a few of these in the logs, trying to get my game running without exceptions everywhere. Do you know what is going on here? These are in the sections where we are loading EVE clouds and GPP textures for them (I assume). Let me know if there's something else you need.
-
Possibly a bit on the memory front but otherwise it shouldn't slow down rendering. Considering the insanely low number of polys being rendered at their highest LOD in any KSP scene, the rendering pipeline has to be a minority fraction of the time being taken to render frames.
-
Looks like I happened to be at the right place at the right time to say the right thing Really cool that you've taken that basic suggestion and turned it into this awesome implementation, as others have noted, with correct environment reflections and a few weeks to noodle around with the materials to exactly nail the surfaces you want to use, this is going to render everything else obsolete from a visuals standpoint. You might start thinking about how you could make this shader and texturing workflow available to other modders, it will be really cool when everyone upgrades to this level, think BDB's Atlas A with full mirror finish, slightly ripply-surface glossy-painted rocket bodies everywhere, materials being properly shiny or matte with correct specular levels depending on their real characteristics, it's a brave new Kerbal world P.S. Understand how much fun you're having but remember to sleep!
-
Ok, that's a good clear description, and I would have noticed it had it been there so I assume new users will see it and understand. And ok, I'll keep it in mind that there is no guarantee of the sorting of parts on the tech tree.
-
Yep, that's a slight visual upgrade over what we have currently Very nice job. Default settings for substance materials are usually very good, if they don't look right there's something wrong with the scale of your model. That's where I was saying to flesh out the material library, because other than color you generally use them as is, only messing around with that which you're intended to, like thresholds for procedural materials. Or at least beyond that, I found messing with them to being heading down a rabbit hole from which I was unlikely to emerge. Besides free, check out Paweł Łyczkowski's substance material offerings. I've used those constantly when working in SP. For some reason can't add the link to his name, so have to paste it raw: https://gumroad.com/plyczkowski
-
Ok, but 468kN at .3 tons is ~1/4 the mass of the next lightest engine I'm aware of that is capable of similar thrust, and I think I have almost all mods with engine parts. So at a minimum, I suggest moving it to later in the technology tree, I think it is placed at the 45 science level right now, that is very early in the tech tree. Otherwise it makes lots of other lifter engines obsolete. Ok on the other two, but can you add that to the description? I don't remember having read that.
-
Two more that are meant to be LH2/LOX but are LF/O: RK-0-100F and RK-0-120F.
-
Three engine config issues leading to accidentally overpowered engines: This one has a thrust of 468kN but the size and mass of a much smaller engine (mass .3t), makes it a killer 1.25m-1.875m lifter. The next two are supposed to be LH2/LOX and have very high ISPs but are set as LF/OX.
-
Perfect, thanks. Please put it at the top of the OP so other folks who may be interested get a good clear explanation
-
Can I ask what exactly IS an HLV? The OP informs me that the first one had horrible Cg problems and you're much happier with this one, that's about it as far as I can see. Is it just a giant lander, or is it meant to be an all-in-one mini-base? Can it be built for cargo?
-
The ones that mean something are reflected in changing thrust and burn time numbers. Also setting positive gimbal angles the nozzles outward, giving up some thrust for rocket stability. There are some functional nose cone options in ones that are designed to be attached to something, otherwise just various aerodynamic options. Nozzle options purely aesthetic but been trying to convince Mage to make them mean something.
-
[1.2.2] SSTU Expansion Pack (31/12/2016)
vossiewulf replied to JoseEduardo's topic in KSP1 Mod Releases
I've been having lots of problems with this in 1.3, currently if I try to use any of the BDB pieces in the VAB the VScale field becomes NaN and I can't dismiss the right-click dialog, leading to having to restart KSP as that dialog is very modal. So unfortunately going to have to remove it, which is too bad as this has lots of very good parts. -
Also, a request that should be quick to do - can you put an arrow or some other indicator on your probe core that shows which way it's facing? At the moment it seems completely symmetrical, no idea if it's aligned for polar or equatorial orbit. Unless there's something there already and I'm just missing it.
-
Awesome! Really glad you think it's going to work for you, I think with a bit more learning and some fleshing out of your material library, you're going to wonder how you ever managed without it, it hopefully will accelerate your process quite a bit and that means more cool parts for KSP with more cool functions and textures. Yep, similar to what I've seen but this is reasonably flexible. Your base maps sound like basically high-res AO maps, if you haven't already look up the function that allows you to isolate a map on the model, so you can see each of the maps you've baked standalone. First it's a useful way to check your baked textures and to also see certain things about the model, one of them should be Ambient Occlusion. I have to think the AO map has to be real close to what you need for you base maps, or at least the process of creating those maps would involve blending down the AO map into it. Start with a simple flat drawing of the details, burn the AO map into it, should have grayscale version of your drawing with all the correct lighting highlights and shadows.
-
That sounds like you've almost worked out a new workflow with SP? As for the shaders, little bit of a difference between 5 and your PBR, primarily in the glossiness, otherwise very close. How does the recoloring work? One way I've seen that done is the actual maps applied to the objects have only a series of pure color values applied, and the recolor logic knows (e.g.) for material 1 it alters any pixel with a hue of 255,0,0.
-
There ya go. Yes, SP is in essence a procedural texturing system, all materials are procedural to some degree, and usually to a great degree as you see from the number of controls available on simple materials and smart materials get even more complex. And you probably haven't fiddled with the filters yet, that's a whole other way to procedurally generate effects. A good way to get the basics of materials and surfaces is to just cover something in the basic paint material, and then start playing with the surface parameters and the metal/roughness, especially the scale on your generators, you can get wildly different results by controlling the scale of the effects. The paint material has all those basic controls and not much else, and it's really easy to see the changes as you mess around with the controls, and you'll very quickly begin to understand how to get anything from a smooth, shiny, slightly rippled surface to rough paint to dead flat very rough surface. However, SP is a tool where although it's fun messing around with materials, you can get much more done by downloading free ones and there are also people selling material collections, some of which are very much worth the money. The fact is they rapidly get super-complex and you're never going to understand enough to be effective unless you're willing to spend lots of time learning how their models work and messing around in Substance Designer. So you flesh out your material library until you have multiple options for all the types you're going to use, and at the same time make lots of alphas of the kind of surface details you need, things like interstage ribs can just be alphas you stamp into the normal map with a few clicks. I don't like bleeping over something like that but I spent some time with Designer and the people making materials are just uber geeks and you need to be very close to a physicist specializing in EM radiation to even follow the discussions. Once upon a time we did everything, now the field has started to develop a number of specialties where the complexities rule out learning much more than one.
-
A minor bug, with the SRB-A when you mirror them in the editor and then fiddle the settings, all of the settings will be propagated to the mirrored instances, except the Variant control. If you want to switch variants, you have to do it individually to each one. Not sure what the intent is there as there's only one variant and it at least appears to just be shorter. A useful thing for that control to do would be to give you heavy, medium, and light versions; no change to the rocket other than setting the various settable controls to give the player a max power version, one in the middle, and something close to as small/light as it can be. That would let the player quickly get within the ballpark of what they want and have only minor tweaks to do. Also my game died last night when the UI locked up in the editor trying to use the MFT-BDB Bluedog tank. If I remember what happened correctly, I tried to change the diameter, as soon as I did the VScale control turned to NaN and the right-click menu froze, I couldn't dismiss it. And although you can get a couple mod windows to pop, otherwise that dialog appears to be very modal and if you can't dismiss it, you're stuck. I will try it later tonight, even with a relatively fast machine my many-mod install takes like 10 minutes to load, doesn't make it easy to test such things, If I need to try it repeatedly I'll spawn off a clean install with BDB and SSTU and see if I can still reproduce. Also there's still something clearly wrong with the fuel tank that grew to like 10m once in space - when I tried to use it in the editor it looked the same, no top or bottom, empty tube. I tossed it aside before trying to place it, I have a feeling it will break things too. I'll try to test that as well and get some info.
-
[1.10.1+] Contract Pack: Field Research [v1.2.2] [2020-09-20]
vossiewulf replied to nightingale's topic in KSP1 Mod Releases
Thanks, yep that's it. -
[1.10.1+] Contract Pack: Field Research [v1.2.2] [2020-09-20]
vossiewulf replied to nightingale's topic in KSP1 Mod Releases
Can anyone help me with this one? Wish I'd looked at it more closely, but it said 30 science and 196 reputation for completion and my greedy index finger clicked before I could stop him. It says it is a WBIResearchContract and it is called Research Materials Exposure Space Study at Gael (running GPP). It says to refer to experiment details but none are listed, so have no idea what to do, tempted to fly a materials bay into orbit and then land it, but that seems not much effort for the reward. And I have to find a solution now as the penalty for failure is -90 reputation, stupid greedy index finger. -
Um, much easier way to deal with this running the Steam version - don't run Steam, and launch the executable manually from the folder, or with a shortcut, no issues running it that way. If you're not playing multiple Steam games that works fine, and that is what I'm going to do. But your tip is a good idea to do anyway, and as soon as someone upgrades to 1.31 they should do the same steps and lock themselves to 1.31 so even if you want to run Steam, it won't update until you tell it to. Only issue I have with the patch is that there doesn't seem to be much groundbreaking there, and they should be accounting for the fact that most people (at least it seems so to me, don't have stats) are running heavily modded games and a change of any kind by Squad is going to trigger a few hours of updates for most people, and several weeks of waiting until their game is running with the mods they want again. So in this case a slower update rate is called for, with more important changes implemented prior to scheduling a release.
-
Yep, as noted the learning and setup curve are significant, and setting up models for SP texturing is a completely different output setup than you have been using. But there is an actual logic to the UI and workflow, and once you do get it you can do work very quickly. I remember flailing around working on it daily for a couple weeks before I started to get it, and by the third model I was able to do the work in a tiny fraction of the time the first one took. Best thing to do is just bury yourself in the tutorials and watch the key ones more than once. For unwrapping, the best option is to not do it by hand- SP doesn't work well at all with unwraps where the sizes of objects are not normalized - i.e. you can't have one piece relatively larger than another, at least as far as I remember- I've not been using it for last several months so the info is out of the hot and warm DBs and is in cold storage, so I need to go run through it again to de-archive the knowledge. There are many tools, most are standalone, in the $40-$80 range now that could easily handle unwrapping models destined for KSP.
-
@hraban I flipped a coin whether to post it here, SSTU, or Contares Add Nertea's stuff, SpaceY, OPT, a couple other things + Venn's stock upgrades, that should be KSP's default install "movin' stuff through space" parts set. if you can't come up with multiple cool solutions to any lifting/flying/landing problem with that library of great parts, you should hang up your slide rule and drafting machine, give up any idea of being a designer, and move into middle management like the rest of the failed designers do
-
Also you need to pay considerable, and I mean considerable, attention to the tutorials on properly setting up models for easy SP texturing. First, doesn't matter what material you assign to the model in the modeler, it will get blown away, so I just use flat standard materials roughly in the colors they'll be just for visual purposes in the modeler. And you want to minimize the number of materials, each one will turn into a big stack of RAM-eating layers. And in some cases as long as you're not under a tight poly budget, it makes sense to actually cut your model into separate elements based on your color pattern, it makes applying said basic pattern very quick to do and also allows individual control of those colors so you can churn out lots of color combinations as mentioned above. And when you bake your maps in SP, remember to bake a materialID map, it gives you another tool to quickly select what you want and mask things for your color patterns.
