|
I could use some advice/feedback on the models I'm making for a naval combat game. I'm making them piecemeal and assembling ships out of parts (the player will be able to do this too; ship design is a core feature) so the resulting ships are going to fundamentally look a little un-cohesive. Right now, a ship might look something like this: (I need to turn off those smokestacks when in the designer mode...) I still have the better part of a thousand models to make before I'm done, I'm guessing. Before I go and make all those models: 1. Stylistically, do you think this looks reasonable? It's obviously not super-realistic, and kind of low-poly (each gun in that screenshot is ~400 tris). I don't have the kind of resources to invest in realism. 2. I'll need to do something about texturing. I have next to no experience with texturing in videogames. I know broadly how UV textures work, but I suck at making them. What kind of techniques should I be researching for improving the looks of these models? Again, I don't have the resources to invest in realism...but I don't think that just using a simple colored material for every part would work well either. I'm modeling in Blender, which I have a fair bit of experience at, albeit mostly with higher-res models. I used to do some character design and animation ~10 years ago. The models were detailed enough that I could just assign materials to different faces, in Blender, and get good-looking results, no need for texture unwrapping or any of the tricks normally used in games. Thanks in advance!
|
#
¿
Jul 31, 2019 18:41
|
|
|
|
| # ¿ May 5, 2024 16:18 |
|
|
Captain Splendid posted:The individual elements look fine so far but the proportions are off. Yeah, at the moment I have literally no superstructure aside from the fore/aft bridges and the smokestacks. I'm still trying to figure out exactly how I want generic superstructure to work, e.g. should you be able to build guns on top of superstructure? That's partially a game design question (what's to stop you from just stacking superstructure arbitrarily high and putting your guns on top?) and partially a modeling question (how do you make the models look decent when it's hard to predict how they'll be put onto the ship). As for the guns, good point on them being too high. Their size however is historically accurate, though putting four turrets onto a single cruiser isn't. In any case, I have to be prepared for players to put "wrong" things on their ships, e.g. if a battleship gun physically fits onto a cruiser hull then the player should be able to put it there, even if it eats up 90% of their displacement allotment. That's part of what I meant about the ships looking kind of piecemeal or un-cohesive. EDIT: fixed the gun heights:  Gearman posted:In regards to the texture stuff, you can probably do a quick and dirty angle-based unwrap and pack for your UVs. I'd probably recommend something like substance painter here for the textures. In blender you can assign your pieces different material IDs, and then in Painter assign different materials to those IDs. You can even use a prefab substance material that adds wear and weathering automatically to make it look a bit more realistic. This can be largely automatic, getting you 90% of the way there really quickly. I would do a quick Google search for Substance + Blender and you'll probably find something useful there. Thanks, I'll research that! What I'm hearing is "use Substance Painter to make a few basic materials, then assign them to meshes/mesh vertices in Unity at import or load time based on the material IDs on the mesh." Is that basically accurate? I'm already making fairly heavy use of vertex colors in Blender to indicate things to the game. For example, ships define their usable deck area with vertex colors, and the particle emitters on the smokestacks are positioned to align with specific colored vertices on the mesh. TooMuchAbstraction fucked around with this message at 20:06 on Jul 31, 2019 |
|
#
¿
Jul 31, 2019 19:49
|
|
|
I was all set to get salty about the complaints about scale, and went ahead and made a mockup of the Admiral Graf Spee using the parts I modeled, so I could overlay it with the diagrams I made the models from. And when I did so, I realized y'all were 100% right: the guns are about 30% too big.  I got the scale wrong on my reference images when I set up the Blender file for the gun.So that's fixed now.  EDIT: proof: 
TooMuchAbstraction fucked around with this message at 16:02 on Aug 1, 2019 |
|
#
¿
Aug 1, 2019 15:51
|
|
|
There's always gonna be someone who's more interested in process than in results. If you wanna be the Cuphead devs and hand-animate everything, that's fine, but there's nothing wrong with using tools to get the job done faster.
|
|
#
¿
Aug 20, 2019 03:21
|
|
|
Any of y'all know of a good tutorial on making explosion effects? Ideally using Unity, but I imagine that most particle systems should have fairly similar capabilities. I've found a bunch that are "make/find a cloudy texture, spawn a bunch of 'em in a circle, fade them out over time", but I'm looking for something a bit juicier than that. Right now my gun flashes and other explosions are feeling a bit anemic: https://i.imgur.com/s3WQ76G.mp4 I'm having trouble finding resources on how to get better because everything I find is just retreading the super-basic stuff.
|
|
#
¿
Aug 27, 2019 23:02
|
|
|
Odddzy posted:You need to watch this to answer your question. That's some excellent 1000-foot-view stuff, as well as a few very specific cool tricks. Thank you for linking it. I guess what I was looking for though was more of a "here's a step-by-step process for designing a specific particle effect", so I could get some practice, because the basic techniques used are not obvious to me as an inexperienced amateur. That said, I discovered that Unity provides a free set of particle effects, including some explosions, which I was able to adapt into an acceptable-looking effect: https://i.imgur.com/9ncre0A.mp4 So my request is considerably less urgent than it was this morning. I'm still interested in learning though, especially since I'll need to make more particle effects in future. The smoke coming from my smokestacks is looking pretty crude now, for example. On a related note though, you might notice that the bullets in the above video are extremely crude, being basically squished spheres with a flat yellow texture. I want something that a) is more visible, and b) conveys motion and a sense of energy. What I'm thinking of trying is making a texture that looks kind of like a meteor doing atmospheric entry -- that is, white-hot front, fading yellow to red and transparent in the back. Make two quads at right angles with that texture and make that be my projectile. Maybe slap another quad with a simple circle texture at right angles to both of those so if you view it from the front/back it still looks like it has volume. Does that all sound plausible? There's going to be a lot of these things flying around, so they need to be readable; especially, the player should be able to predict their paths reasonably well.
|
|
#
¿
Aug 28, 2019 05:42
|
|
|
Sagebrush posted:I don't know if you're going for a realistic look or a more cartoony style, but in real life you can't see the shells, obviously. Maybe just a long translucent streak like smoke being dragged along behind the projectile? A static line also shows direction much better than a small object in motion, in my opinion. Yeah, definitely not realistic; ships are faster, engagement ranges are way shorter and shell velocities are about 1/10th what they would be in real life. My thinking behind slowly-moving projectiles is that it helps build anticipation, and also means that if you see enemies shooting at you from long range, you can steer out of the way without having to take a few hits / see splashes first. That said, some kind of comet-tail effect to help show direction does sound like a good idea, thank you! Odddzy posted:Well, you're the art director. We know nothing about the game you're making and the intent behind those projectiles. Taking you by the hand and making it would be counter productive I think. But you've got the tools in the video that explains all the tricks. Watch reference in games of the same kind of game you're making. That 45-minute video is "all the tricks", really? I figured this would be a deeper topic, but you presumably know more about it than I do, since I know almost nothing. I guess what I'm really looking for is a deep dive on a specific particle effect. Probably any reasonably complex effect would do, even if it's for a completely different genre. My problem is looking at all these tools and parameters and not knowing how to translate them into desired results. I'm not asking for someone to make the effects for my game specifically. It's like, I have a lump of clay and 50 different weird-looking tools. I know I want to make a specific kind of bowl, I know what a bowl looks like, but I don't know how to make one. Presumably it involves some of these tools? And all the tutorials I can find are just "here's a really simple way to make a really lovely bowl". If I could see someone make something high-quality, even if it was a vase or a plate instead of a bowl, I'd be a lot better equipped to make my bowl.
|
|
#
¿
Aug 28, 2019 15:58
|
|
|
cubicle gangster posted:There's a reason why your first day of pottery class they don't try and teach you to make a very specific high quality bowl. Start with the fundamentals and be patient, you'll save time in the long run. Speaking as a potter, on the first day of pottery class they do walk through everything involved in shaping the wet clay (trimming, firing, and glazing come later after it's dried). My results aren't going to be nearly as good as the teacher's, but I will have some idea of what I'm supposed to be doing, by imitating what the teacher does. That's all I'm looking for. I'm not trying to take shortcuts, I'm trying to find instruction that tells me broadly how to approach the art, from someone who knows what they're doing. I apologize if I've failed to make that clear in my prior posts. Buying assets, or using free ones that other people have made, is definitely a solution. I'm trying to keep my budget tight though, and it's a lot easier to enforce aesthetic consistency if I'm at least able to tweak things myself. I've made use of some free PBR materials, for example, but since I have prior experience with materials and texturing I have at least some idea of how to change them to achieve a specific effect.
|
|
#
¿
Aug 28, 2019 18:10
|
|
|
ImplicitAssembler posted:The first many weeks as a pottery apprentice consists of learning how to wedge the clay..it'll take months before you're allowed to throw anything. I'm...a bit dumbfounded by this, to be honest. I can't imagine wanting to learn an art in this day and age and being told that I'd have to spend literal weeks doing menial chores before I'd even be allowed to start practicing the actual artform. Every single class I've ever seen has basically gone like "here's a lump of clay, here's how you wedge it, everyone wedge their lumps, okay let's go over to the wheels, I'll demonstrate what to do, you try to imitate, and I'll walk around and help out anyone that gets stuck." It's not like you're working with a table saw or a welder or something where there's a serious threat to life and limb if you don't do things carefully. Like, that honestly sounds like wanting to be a figure sketcher, going to an artist whose work you admire, and being told that before they'll teach you anything you have to spend weeks sharpening their pencils. It's not impossible to learn things that way but it's pretty student-hostile. quote:It's one thing to keep your budget tight, but if you don't see any sort of relevant link between those tutorials and what you're trying to do, the $10-$15 that some stock explosions cost is going to keep you sane and on track. I do see the link, and I did thank the poster that linked the video. There's useful information there! It's just the kind of useful information that I lack the foundational skills to make effective use of. EDIT: in any case, this is getting pretty derailed and it sounds like there's a fairly basic disconnect between what I (think I) am asking for and what y'all can provide. I apologize for wasting your time. TooMuchAbstraction fucked around with this message at 20:29 on Aug 28, 2019 |
|
#
¿
Aug 28, 2019 20:26
|
|
|
Ah, okay. The divide there is between "I want to learn this art form" and "I want to make a living at this art form", and yeah, standards are going to be very different between those two, especially if in the latter case you're working as part of a company instead of solo. I know a few people at my local studio who produce what I would call professional-tier quality pieces, but they're definitely not the norm, and they have way more experience/practice than everyone else does. Most people (at my studio) are dabblers, and work with clay because they enjoy it, not because it's their job. The ones that do work there mostly make money by leading "experiences" attended by drunk tech startup teams. My own standards aren't AAA-tier, as should be obvious if you look at my work. I'm not going to make a living at 3D art anytime soon. I just want the effects to be of roughly equivalent quality to the models; the models aren't amazing either, but right now they're substantially better than the effects I'm making.
|
|
#
¿
Aug 28, 2019 21:13
|
|
|
ceebee posted:If you want to make something that looks good to the general public, aim for the quality professionals achieve or higher. You can only get close with practice. No arguments from me there, and I definitely will be getting plenty of practice. The thing is, though, that I know I can pour more or less unlimited amounts of time into the art for this project, and it will make a difference in how nice the game looks...but at some point I have to finish the game! And there's lots of non-art things that need to be done to get there. So I have to be able to say at some point "this is good enough". It's not a matter of not putting effort in, but rather of appropriately distributing my effort across all the various different things that need my attention. Hopefully the amount of time I can spend on the art results in a quality level that most players are willing to accept. It may not wow them, but it won't make them think "wow that's an ugly ship/landmass/explosion" or otherwise pull them out of the game. For those that won't think the art is good enough (and no matter how hard I try there will always be at least a few people who will reject it on the basis of artistic quality), well, they should ask for a refund. And to be clear, I don't consider the current quality level of my art to be acceptable. The meshes are mostly OK (low-poly, but that's an acceptable stylization) but the materials and effects are still crude. It's a hell of a lot better than it was last week, but it still has a ways to go. I do need to be realistic about what's achievable though, considering my overall goals, skill level, and time constraints. quote:My advice? Just buy a professional quality FX, tear it apart, learn how it works. Bam, you have self educated yourself to be able to achieve a better look and you might not have to spend money next time.
|
|
#
¿
Aug 29, 2019 01:16
|
|
|
Yeah, exactly. One of the reasons I started working on this project was that I'd been playing Sky Rogue, which looks like this:  (this is an old screenshot, but the aesthetic is the same in the current release) and I was reminded of an old PS2 game I used to play (Warship Gunner 2), and basically wanted a recreation of that game along similar lines. I'm not going for realism; I can't go for realism considering the constraints I'm under. I'm gonna be more detailed than Sky Rogue because I'm stealing designs from World War 2 and it's easier to ape them fairly directly than it is to try to stylize them. But the primary thing artwise is to be coherent, consistent, and satisfying to look at. Sky Rogue looks roughly like what you'd get from a PS1 game if it had modern resolutions; I'm basically aiming for one generation later. Anyway, I worked on my shells some more and this is what they look like now:  It's definitely an improvement from what I had previously. No doubt in a few weeks I'll get dissatisfied again and tweak them some more, but for now I have bigger fish to fry.
|
|
#
¿
Aug 29, 2019 02:40
|
|
|
Odddzy posted:Biggest problem I see considering what you've said, is that you've got a realistic water shader, boats that lean towards realistic, but you don't want realistic considering your restraints. I'd just go fully into a more abstact look and fudge it a little in the mean time. I don't pretend for an instant that I've got any part of my aesthetics nailed down, so this is all subject to change. Indeed, nailing down what the aesthetics should be is something I need to do to get the "vertical slice" of the game done (basically a small-as-possible but fully representative chunk of gameplay). So thank you for pointing out those inconsistencies; it's easy for me to overlook them since I'm staring at this stuff every day and don't have particularly finely-honed aesthetic senses. The water shader is some free thing I downloaded; it replaced boats literally just floating in nothingness. There are a lot of wave shaders out there, plus I need to try tweaking this one. The boats have basically the very first pass at PBR materials that I made (again using some free materials I downloaded). Those materials need to be tweaked, probably making them flatter and higher-contrast. When you say "go fully into an abstract look" I picture something like Sky Rogue or Spectre VR -- no shading, super-bright contrast, overall "cartoony" look. My kneejerk feeling, which I'm having trouble articulating, is that that feels a bit too far. But this is the kind of thing I need to be thinking clearly about -- what kind of aesthetic do I want to go for, and particularly what kind of aesthetic can I achieve? bring back old gbs posted:Yeah even in the little video you posted your ships were zipping around really fast. But it kinda looked fun, maybe work that into the aesthetic. These huge lumbering ships that can actually turn on a dime and tear rear end like micro machines. Maybe that's already a thing though, I've just seen full on commercials on TV about realistic tank and ship games, Arnold Schwarzenegger was voicing one. I think those developers got some money, not that there isn't any room for competition but gad dayum those apps have TV commercials. Set yourself apart. Yeah, there's a lot of competition in this space. I'm hoping that my gimmick of allowing players to design their own ships, as well as the focus on a singleplayer campaign rather than multiplayer arena matches, will set my work apart. That, and the fact that I plan to include ridiculous gigantic warmachine bosses (...and I just realized, WarMech from Final Fantasy would fit right in with some of my planned designs). As for movement speeds and maneuverability, that's something that I fully expect will take a lot of tweaking. Realistic scale would be horribly tedious to play -- conflicts took place over hours or days, ships take over a minute just to turn around. Make things too fast though and the ships start feeling lighter and less impactful than they ought to. The speeds I've shown in some of my videos really ought to be reserved for things like PT boats, while battleships are a little more ponderous while still being vastly faster than they would realistically be. Getting that sense of "you're piloting a 60000-ton war machine" without making it feel tedious is going to require some work. Good sound design would help a lot. That's another entire discipline that I know very little about. But imagine, like, the difference between a two-stroke weed whacker engine and a tank engine -- that's roughly the kind of difference there should be between a PT boat and a battleship. Having really booming, full-throated gun sound effects ought to help as well. quote:The new projectiles look cool, would like to see some shading in the muzzle smoke. If I could direct your focus AWAY from game dev for a moment to watch this tutorial on how to do a pretty decent smoke effect in a video compositor using a particle system and a single animated sprite: Thank you for linking this! You're right that it should be fairly straightforward to apply this kind of thing to Unity's particle system. Interestingly, someone in #unity3d (an IRC channel on Freenode) indicated that you should also be able to make Unity particles receive lighting directly, presumably via normal maps. I'm not sure how easy it will be for me to figure all that out, but it bears at least some investigation. If I can get it working, then I'd be able to get things like the flare of light from the gun illuminating the clouds of smoke.
|
|
#
¿
Aug 29, 2019 22:22
|
|
|
Went and implemented that scrolling-clouds shader from the Blizzard video, to replace my old smoke effect. https://i.imgur.com/Nr2g8uR.mp4 Then I tried to figure out how to implement normals on particle effects in Unity, but goddamn that is way too much vector math for me right now. Shaders are hard. 
|
|
#
¿
Aug 30, 2019 18:30
|
|
|
Big K of Justice posted:You probably want to implement some form of camera alpha fade/culling on the particles, so they'll fade to 0 as they get really close to the camera so it'll get rid of that particle "popping". Depending on the engine that feature should be built in somewhere... It seems like it ought to be easy to implement; Unity provides a _WorldSpaceCameraPos which is the "World space position of the camera", and I'm passing the vertex positions of the particle quads to the shader (using the "custom vertex streams" feature of the particle system). So I added this to the fragment program: code:I don't consider this to be a particularly huge deal since there's plenty of camera angles that don't have a continuous stream of smoke being blown at them, but it would be nice to figure it out. 
|
|
#
¿
Sep 2, 2019 02:21
|
|
|
500 posted:I don't know what custom vertex streams are, but I have a feeling 'i.position.xyz' isn't in world space. So it's possible you're trying to compare your world space camera position (which is probably something like [-55, 12, -34]) to a non-world-space vertex position (which might be something like [0.2, 1.0, -0.4]), and when you clamp to values between 0 and 1, you're probably always getting either 1 or 0, and that's why your particles either display as fully transparent or fully visible. of course the positions aren't in world space. Thanks for pointing that out. Unfortunately, at least when used for Unity's particle system, the position doesn't seem to have much useful information at all -- all three components appear to be constant, as best I can tell if I set them as the fragment's color output. Well, that explains what's going wrong at least.
|
|
#
¿
Sep 2, 2019 19:44
|
|
|
Kanine posted:is it rude for me to ask an indie developer if they're using store assets or if they're producing them in-house? if i do im going to preface it with saying both are valid, and that im asking mostly out of curiosity I feel like the answer for the vast majority of indie devs is that they make most assets in-house and buy some of them. If you're wondering about specific assets, I'd just ask about those, and phrase it more like "can you tell me how you made X model/effect/texture." Then they can say "well actually we bought it" or they can tell you more about the creation process.
|
|
#
¿
Sep 6, 2019 00:13
|
|
|
New ocean shader: https://i.imgur.com/BnMeW22.mp4 It's a basic Gerstner waves shader from this tutorial, with some tweaks. The big one is that the function that calculates the wave position originally took as input a position in object space. I changed that to world space plus an offset provided by sampling a cloud texture. That does a nice job of making the waves more irregular, though the above clip makes them a bit too irregular; I toned it down a bit after that. The distortion does add the cost of occasionally creating some odd "shadows" where the offsets make the waves too steep. You can see that in this screenshot: the top view is Unity's editor view showing the wave folding over itself, the bottom view is the in-game view.  But honestly it doesn't feel that bad; in-game your viewpoint generally can't get low enough to see that kind of thing. On the whole I like this shader. I feel like it's a pretty good stylistic match for my other assets, and it's easy to configure so I can e.g. recolor it for sunset/nighttime missions, or tweak the wave height to simulate calmer or stormier waters. Comedy option: too much distortion of the waves makes for some weird-looking effects: https://i.imgur.com/Xbrs4If.mp4
|
|
#
¿
Sep 9, 2019 21:52
|
|
|
I'm making my very first normal-mapped objects for my game, based on this tutorial. I started with a crate, and I noticed what looks like Z-fighting going on between the low- and high-poly meshes when I try to bake the normal map: I was able to fix this by scaling the low-poly mesh to be very slightly larger than the high-poly mesh. Is this a common problem, and is that the recommended solution? I don't really know how baking normal maps works from a technical perspective, which makes it hard to know what the best practices are. I did get an OK-looking crate though: 
|
|
#
¿
Oct 16, 2019 19:24
|
|
|
Odddzy posted:Were both your lowpoly and hipoly being evaluated to get baked to the lowpoly? The tutorial I followed has you select the highpoly then the lowpoly mesh, and then click the bake button. I assume that this is some mechanism so Blender knows how to arrange the normal map on the texture that you render the normals to. I honestly have no idea how it works though.
|
|
#
¿
Oct 16, 2019 23:06
|
|
|
Aha! Yep, that was set to 0. Setting it to .1 got rid of the obvious artifacts. Thank you! What's interesting is that it gives significantly different results from the "scale the lowpoly mesh up by 1%" image though: Nonzero ray length:  Upscaled lowpoly mesh:  But it looks fine when used in Unity, so
|
|
#
¿
Oct 16, 2019 23:55
|
|
|
Looks like a truck to me. Nice work! Something that can be really handy for modeling from reference material is that you can drag an image from your filesystem into Blender, then model by "tracing" it. For example, drop this image into Blender, scale it up until the cargo box is 20 or 40 feet long (for a standard cargo container; I just googled the dimensions), hit 'z4' to switch to wireframe rendering, and now you can model by just dragging vertices around the edges of the diagram. I've done that a lot for the ships and guns I'm making for my game: 
|
|
#
¿
Nov 4, 2019 03:11
|
|
|
Sininu posted:Does Blender have modifier stack like 3ds Max does? I'm kinda interested in trying it out now that it has a sane interface. Are you talking like mirror, array, curve, etc. modifiers? Yes, it has those, and they work as a stack. 
|
|
#
¿
Dec 16, 2019 01:06
|
|
|
There's an "edge crease" operation you can perform that dictates how closely subdivision adheres to the underlying mesh. On the subject of inner quads: I did the "make a cube, cut it in half, throw half away, add a mirror modifier" sequence so many times that I made a script to give me a mirrored cube straight up. Y'all're welcome to use it. code:Note that the normals on the generated cube are screwy, but that's easily fixed by hitting shift-N in edit mode to recalculate them.
|
|
#
¿
Feb 22, 2020 22:32
|
|
|
I dig it, it feels kind of like sci-fi/fantasy novel cover art from 50+ years ago. Agree that the gate should be bigger, at least twice as big IMO. As for conveying that it's in the process of opening...I think it'd help if it wasn't as open in the shot, just cracked enough that you can see the castle in the distance. It'd probably also help to emphasize the "pocket" in the wall that the gate slides into. You could do something gratuitous like model a chain-and-pulley system that pulls the gate open. I guess what I'm getting at is that the more you visually emphasize that "this is a thing that can move" the more you'll imply that it is moving, I think. Compositionally, can you get your creature in the foreground, the vehicle parked next to the gate in the midground, and the castle in the background? I feel like having both creature and vehicle in the foreground would take up too much visual space, but having nothing in the foreground makes it hard to guide the eyes. Maybe the vehicle could be partially off-frame...a shame because it's a neat vehicle, but it does take up a lot of space. You could also try putting the foreground at the bottom of a hill, the gate in the middle of the hill, and the castle (as it is now) elevated above. Putting more distance between the gate and the creature will give you more leeway to make the gate really huge, too.
|
|
#
¿
Mar 7, 2020 23:32
|
|
|
Something that occurred to me re: showing that the gate is in the act of opening: you can have a road that's clearly wider than the current opening in the gate. If it's a dirt or dusty road you can also kick up some small clouds of dust on the ground to help imply movement.
|
|
#
¿
Mar 9, 2020 04:34
|
|
|
Those look fantastic honestly
|
|
#
¿
May 14, 2020 03:19
|
|
|
500 posted:Thank you! I should add that the heightmaps came from this site, Tangrams Heightmapper. Such an amazing free resource that I only just discovered. I just recently discovered this site and it is a huge lifesaver for me personally because I need lots of height maps of real-world locations for my game. Previously I'd been using terrain.party, which is also good but much more limited in the amount of area you can download at a time. One neat thing I figured out with the Tangrams site -- if you need higher resolution than what you can fit on-screen, you can use Chrome's developer tools to look at the site in any arbitrary resolution. It's intended for making sure a site you're working on looks okay on mobile but you can type in any resolution you like. Just make sure to turn off the auto-exposure calculations first, because the site gets horribly slow if it tries to calculate exposure values for a 4k x 4k height map. This page has a short guide to getting high-res screenshots in Chrome.
|
|
#
¿
Jul 12, 2020 19:16
|
|
|
I think that's a seam or crease? Not something I use much, so I forget exactly what they do. It's either for cutting up UVs or for controlling how much a subsurf modifier can soften the edge or something like that.
|
|
#
¿
Jul 20, 2020 04:20
|
|
|
Another option is to select one of the loops and then hit x -> dissolve edges.
|
|
#
¿
Aug 9, 2020 19:13
|
|
|
A Blender question: I often find while working on a model that I want to recess part of it. For example, I might want to turn a cube into a shape like this:  This is easy enough to do if I know I want it in advance, but often I have a moderately complicated shape already made before I discover I need to make part of it smaller. The only method I've found so far is to do a negative extrusion and then a lot of tedious surgery to fix the geometry. Either that, or extrude everything else except for the part I want recessed, and then adjust scale. Neither of these approaches are really all that great. Any suggestions for a better technique? Most of my methods are self-taught so I wouldn't be remotely surprised if there was a better way. All of my models are various hard-bodied things. For example, this is a work-in-progress bridge for a New Orleans-class cruiser:  You can see a cutout I made roughly in the middle of the image, where I did the "extrude everything else and then shrink down" approach.
|
|
#
¿
Aug 14, 2020 22:07
|
|
|
Warbird posted:Iirc you can inset 0 and that creates a face you can inset/outset (is that even the term) willy nilly. Unless I'm missing something, that doesn't resolve my issue. Here's an example of the kind of situation I'm trying to avoid:  The zero-width faces on the edges there are annoying to deal with; I need to delete them (which deletes the entire left/top face) and then recreate them as six-edged, L-shaped faces. Neon Noodle posted:Use the Boolean modifier, then you can use any mesh to cut into another mesh non-destructively  Doesn't seem to be working  Like, I legit thought this would work, with my main objection being booleans are fiddly to work with, but I'm surprised that it's not doing anything AFAICT. The issue seems to be that I'm working with a mirrored mesh; if I make a plain cube and put a boolean modifier on it, then it works fine -- though it's annoying that it leaves a hole in the mesh that I have to patch. With a mirror modifier though, only the Union and Intersection modes seem to work, not Difference. Maybe it's having trouble telling what's internal to the mesh?
|
|
#
¿
Aug 15, 2020 03:16
|
|
|
Listerine posted:After you apply the modifier, and hide the cutting object, is the object you're trying to cut still intact? The operation worked for me but I couldn't see the result because the cutting object was still visible. Correct, the object is intact. If you select an object with a Boolean modifier on it, all other objects involved in that modifier become semitransparent, so you should be able to see the change in the object or at least the change in its selection outline.
|
|
#
¿
Aug 15, 2020 14:53
|
|
|
Well, I just fired up a new scene and did the same thing I did earlier, and this time it worked. 
|
|
#
¿
Aug 15, 2020 18:27
|
|
|
Haha, no worries, I appreciate the thought and I'm sure there's people here who will get good use out of that video. And yeah, I'm sticking to low-poly both for performance and as an aesthetic choice, in the sense of "it's an aesthetic that I can make look intentional without spending way more time than I can afford on it." I've done high-poly stuff in the past and know a little about keeping my topology under control. In this project it doesn't matter so much (most faces are flat), so I've gotten a bit sloppy, but oh well. I asked the question in the first place because I figured it came up enough that there might well be a solution that I just didn't know about. Like, "oh yeah, just select the face you want to inset and hit ctrl-shift-G" or something (I have no idea what if anything that shortcut is mapped to). Sounds like there isn't, so, tedious surgery and the occasional boolean modifier is the way to go!
|
|
#
¿
Aug 16, 2020 04:07
|
|
|
That looks lovely, but I admit I'm confused by the pods to the port side of the cockpit. They look like jet engine air intakes, but they don't have a jet engine-style exhaust and they don't seem to connect to the rocket plume. Do you have a concept in mind for how this is "supposed" to work? I feel that designs are always stronger when the viewer can look at them and intuit what the major components are for.
|
|
#
¿
Sep 25, 2020 18:09
|
|
|
BonoMan posted:Can't wait for all these game companies to try to get away with cheaper midwestern talent only to produce poo poo games delayed fifteen times and then crash the entire market circa the Atari/80s poo poo-game-shovelware-crash. I feel like games are too big to crash these days like they did in the 80's. Individual companies (and potentially entire niches like VR...probably not mobile though) can absolutely crash and burn but they're not gonna take the entire industry with them. But sure, "let's fire all of our expensive, seasoned content creators / developers and replace them with twice as many cheap inexperienced people" has been tried before, and very rarely succeeded.
|
|
#
¿
Nov 27, 2020 17:29
|
|
|
Kanine posted:I'm making some paper sheet assets in an Unreal 4 environment rn, and I'm wondering what might be the best method for having the back of a sheet of paper blank? would there be a way i can display something else on the back face of a polygon differently using a shader or should i go the simple but dumb route of just duplicating the poly, flipping the normal and making it super close, then shifting the uv? Duplicate the faces and flip the normals. It's simple and it'll work reliably.
|
|
#
¿
Dec 19, 2020 19:50
|
|
|
It looks like you have normal/albedo maps, but everything is very non-smooth, which results in getting no highlights. Which is why everything looks kind of drab. The helmet and visor should be pretty smooth. The skin of the creature would be kind of smooth if it's meant to be a real alligator, but not if it's a cloth doll. The eyes should be very smooth, whether real or marbles. You can see that the black hose connected to the helmet has some highlights on the ridges. That's because of smoothness.
|
|
#
¿
Jan 11, 2021 15:53
|
|
|
|
| # ¿ May 5, 2024 16:18 |
|
|
I'm not really familiar with Blender's material system, but smoothness should be a number between 0 and 1 typically, and you can probably set it for the material as a whole or as a value on a particular node.
|
|
#
¿
Jan 11, 2021 18:33
|
|