|
The Cheshire Cat posted:I feel like it would be more thematically interesting if combat itself was resolved as a game of "full contact" baseball. You run into a biker gang in the postapocalyptic wasteland, oh poo poo! Set up the diamond, it's time for a throwdown. I'll think on it, fortunately I've not even start that chunk of it, and I would like to integrate things. I feel like part of the issue though is that Baseball is just so far from a contact sport. Basewars is a fun example and I loved it, but the fun was it being 2 player. I am not sure if I am going to have baseball even be playable. Instead being something you get a highlights run down on after the fact. The instinct right now on contact baseball is that it becomes slower and thinner, than a combat scenario where your pitcher can toss a ball at a Guarded batsman who will hit the ball at the enemy, or your catcher blocking damage while another batsman swings the crap out of people. Fortunately I have time to mull these issues over while I finish the Oregon Trail systems. [edit]To clarify, I want something that is closer to the OT hunting, where it is a fun little side diversion that is technically risk/reward, where you have the chance for food at the cost of bullets. Except I'd like combat to be a way to 1. Gain stat upgrades to participating players, 2. Gain resources from fallen enemies. So, as long as it is a quickish experience, maybe around 5 minutes, and fun that's what i want to do. JossiRossi fucked around with this message at 01:41 on Mar 21, 2016 |
# ? Mar 21, 2016 01:21 |
|
|
# ? Jun 10, 2024 03:14 |
|
Very violent sport, isn't it, baseball? https://www.youtube.com/watch?v=Ty_QxHOeErw
|
# ? Mar 21, 2016 01:22 |
|
The Cheshire Cat posted:I feel like it would be more thematically interesting if combat itself was resolved as a game of "full contact" baseball. You run into a biker gang in the postapocalyptic wasteland, oh poo poo! Set up the diamond, it's time for a throwdown. https://www.youtube.com/watch?v=Cr2tvMguUiA
|
# ? Mar 21, 2016 01:36 |
|
xzzy posted:I think the rock trees could become tree trees with some quick fixes.. make a smoothing group to get rid of the hard edges, and make the trunk more brown. Presto, stylized tree. Way easier than that, even; I'd set a low threshold for auto-smoothing to let me retain some hard edges, just unchecking that makes it look properly smooth. The color thing though... I need to make a reference folder of neutral-lit images of various objects, I can't tell most browns, reds and greens apart. If I had to take a guess of the current colors, I'd say the trunks are grey and the leaves are green, but it could be light blue and orange for all I know. I should just run with it and not care if the colors make sense to everyone else, but that's been the hardest thing to overcome.
|
# ? Mar 21, 2016 01:39 |
|
Adding movement and some UI elements to my space tactics game.
|
# ? Mar 21, 2016 01:40 |
|
WaterIsPoison posted:Adding movement and some UI elements to my space tactics game. Love this.
|
# ? Mar 21, 2016 02:26 |
|
WaterIsPoison posted:Adding movement and some UI elements to my space tactics game. This is good.
|
# ? Mar 21, 2016 02:29 |
|
So I've had an idea for a game for a while, and I've had many failed attempts at learning programming (this is the furthest I've gotten and I can actually see something from my efforts.) The game is a management game, where you have a team who fight it out with another team. You have no control over the "match" portion, where they fight, and try and complete objectives. What you're doing is training them, setting out tactics, buying and recruiting manthings for your team, and eventually breeding their replacement. I haven't gotten near the management side yet. I've been concentrating on getting my game/match scenario set up. This involved a lot from me, having never really done anything like this before on the programming side. I started out with this. Just getting a manthing to explore a map. This was kind of funky and redundant, with dodgy exploration code so I made it somewhat better. What I noticed with this is that even a minor change, of a simple instruction can make the behaviour appear a lot more natural. Here I added health boosts, and put two of the manthings on the same board. The man things fight if they get near each other, seek out health boosts if their health gets too low, and eventually fight it out if they're doing better than the other person, or run away if they're doing worse. And then just tonight I added a shooting mechanic. If they're not next to each other they shoot in one of eight directions towards the opponent. (This is kind of buggy.) And if the opponent is next to them they attack, push them back and attack again, pushing them away a few squares unless there's a wall in the way. I'm going to keep working on this game type aspect, maybe introducing more objectives, and the logic of what the manthing does. After that I'll have to add a team of manthings, and then I'll be altering their behaviour so they don't all act identically. Once I have that done I'll work on the management part of the game. At the moment it's mostly an exercise in AI, which is quite enjoyable, but a lot to wrap my head around. I will say that it took me two days to get the exploration done. Then another full 12 to 16 hours to get the exploration, health boost and fighting done. And finally it's only taken about four hours to get the shooting done, even if it is buggy. So I seem to be getting better, or at least faster at what I'm doing.
|
# ? Mar 21, 2016 02:48 |
|
I'm hitting the point where I either need to start learning about materials in UE4 or rewrite my story to explain why everything looks like undifferentiated plastic... is the general rule of thumb to create one material per literal, real-world material (e.g. one for wood, one for metal, one for skin, one for cloth...), or is there a completely different convention that I'm unaware of?
|
# ? Mar 21, 2016 04:07 |
|
Omi no Kami posted:I'm hitting the point where I either need to start learning about materials in UE4 or rewrite my story to explain why everything looks like undifferentiated plastic... is the general rule of thumb to create one material per literal, real-world material (e.g. one for wood, one for metal, one for skin, one for cloth...), or is there a completely different convention that I'm unaware of? It depends what you're doing with the material. For instance: If you're applying the material to a floor or wall or a landscape you'd want one for each different type of surface. (Ladscapes often have blended materials that allow you to paint different surfaces and blend from say dirt to grass) But a complicated mesh like a prop or a weapon will often just be a single material with multiple texture maps that define things like metalness and AO to make different parts of the mesh look more like metal or more like wood or cloth or skin, etc, etc. Some character models will have multiple materials, but generally in that case it's just one for the skin, one for clothing and one for hair. Amstrad fucked around with this message at 04:36 on Mar 21, 2016 |
# ? Mar 21, 2016 04:32 |
|
It's pretty complicated, I'm still trying to figure out the intricacies myself. You can use multiple materials per object with the caveat that each material will be a separate draw call, or you can use blend layers with the rgb channels of a special UV map to assign submaterials, which will be one draw call yet still let you have different kinds of materials/textures for all the subs. But, it's a bit more complicated to set up. A really good example of the submaterial stuff is the default gun in the first person shooter template. It's got different layers for the plastic, rubber, and screen materials, plus a normal map, and it's easy to pull apart to see how it works. E: I do wonder what the performance limit is on modern machines, if you're just aiming for PC. I've been building up a level with as much crap as I can shove in, to get an idea what the max of everything is. I have a bunch of meshes with 4-6 materials, even more with 1-3, with some instancing but a number of unique ones. Any slowdown I've found has been from post processing or poorly optimized code, not the amount of materials, so I figure I'd have to really go crazy and use no instancing to start getting slowdown from it. The Gasmask fucked around with this message at 06:01 on Mar 21, 2016 |
# ? Mar 21, 2016 05:46 |
|
Colon Semicolon posted:https://www.youtube.com/watch?v=mtbCtwgI440 Thanks for posting that it looks amazing.
|
# ? Mar 21, 2016 05:55 |
|
Omi no Kami posted:I'm hitting the point where I either need to start learning about materials in UE4 or rewrite my story to explain why everything looks like undifferentiated plastic... is the general rule of thumb to create one material per literal, real-world material (e.g. one for wood, one for metal, one for skin, one for cloth...), or is there a completely different convention that I'm unaware of? If you're going for a realistic look (Which I believe you are), then you'd be best off going with a PBS shader, as you can control whether something looks metallic, wooden, dirty, etc entirely through shader values. While it's for Unity, this article goes over how the various textures interact with each other, and should be portable to UEs equivalent: http://blogs.unity3d.com/2015/02/18/working-with-physically-based-shading-a-practical-approach/ As for deciding when you use different materials, it can come down to use cases. Objects (or parts of objects) that require unique shaders such as refraction or scrolling textures need their own materials, but apart from that just split them up where it makes the most sense. Like for a character there are multiple routes you can take: Having the entire character as a single texture (Most convenient, least adaptable), one material for skin with a subsurface-scattering shader and a different material for clothing, or different materials for each piece of clothing (Which makes sense for games where characters can change clothes). Another thing to take into account is whether or not you want to have looping textures, like having a wall texture that can be repeated for the entire wall. You can use shaders to do the looping using a texture atlas, or you can add edges to the mesh and do the looping manually via the UV coordinates, or you can have one material per wall type and just use standard hardware wrapping. Also take into consideration that more materials generally means less batching; ie more draw calls. So if you have something that really needs to have high performance, say a large crowd of people, you'd be better off having a single material per character, or even using the same material for multiple characters. Although, as usual, unless you're absolutely sure it's going to be a problem, don't optimise until you've identified that it's actually an issue.
|
# ? Mar 21, 2016 06:44 |
|
That's all really helpful, thank you! I shouldn't be astonished at how deep material design can get, but as I go I'm getting increasingly grateful that they did me the courtesy of building a visual editor instead of making me learn HLSL or something.
|
# ? Mar 21, 2016 09:42 |
|
I spent 3 hours playing around with Cryengine 5 today. Initial verdict: stay away for now. The docs point to items that don't exist, the engine crashes in hilarious manners. There's no documentation about the C#/C++ bridge, although I can answers some questions. Only 1 of the 3 examples works, I found no way to package a build. Instead you "export to engine" which makes a .bat file that runs the game in engine. I found no asset pipeline, instead you unzip everything into specific folders, performance was not great. There is very little interest apparently, the answerhub is a wasteland with almost every question a question about not about technical issues. Some things are impressive coming from Unity but not nearly enough, but it really seems like an unpolished turd which is too bad. It's lacking all of the nice utilities form UE like the frontend, the light baking. In general it feels like a mix of the worst parts of Unity and Unreal with a little bit better of a default rendering than Unity. FYI in Unity 5 esp. 5.4 (they made several massive performance enhancements in 5.4, probably tripling performance from 5.3 for VR and doubling for non-VR) you can get some really nice looking output without too much effort. They haven't changed the defaults and probably won't till 6 but Unity is a fully capable engine now. If they could just update the .Net runtime so I don't have to simultaneously live in C# 2006 and C# 2016 every day. Here is a Twitch/Youtube archive of the stream, if the youtube isn't ready the Twitch archive should me. https://www.twitch.tv/beep2bleep/v/55733176 https://www.youtube.com/watch?v=vQwP_sGcSFQ
|
# ? Mar 21, 2016 10:50 |
|
Oh, Unity 5.4 is in public beta. Probably there's no hurry to download it since I'm not developing on the bleeding edge?
|
# ? Mar 21, 2016 11:41 |
|
supermikhail posted:Oh, Unity 5.4 is in public beta. Probably there's no hurry to download it since I'm not developing on the bleeding edge? Yes, 5.3 seems really stable 5.4 is not, if you're working on an active product I'd only suggest upgrading to 5.3 for now. Even when they take a product out of beta it tends to have bugs for a couple of weeks so while in beta it's even more of a crap shoot.
|
# ? Mar 21, 2016 11:47 |
|
Yay, got our website up and running. Thanks again to all the goons that helped with suggestions for the video teaser!
|
# ? Mar 21, 2016 16:05 |
|
Can't stop the occasional sparkle from lights dunno why. I thought color correcting / tonemapping the post bloom image was supposed to stop that? Also even shadow casting lights bleed through my voxel meshes for some reason.
|
# ? Mar 22, 2016 03:06 |
|
FuzzySlippers posted:Can't stop the occasional sparkle from lights dunno why. I thought color correcting / tonemapping the post bloom image was supposed to stop that? Could it be the specularity of your material?
|
# ? Mar 22, 2016 04:57 |
|
FuzzySlippers posted:Can't stop the occasional sparkle from lights dunno why. I thought color correcting / tonemapping the post bloom image was supposed to stop that? I guess it depends on what the color correction/tonemapping pass actually does. I had this issue on a previous project I worked on, and the issue seemed to come from ultrabright pixels being rendered into the PBR lighting cubemap. Adding a pass that clamped the color values to 1.0 before doing the bloom pass seemed to fix it. edit: There's a discussion and an example of a clamping shader in this thread: http://forum.unity3d.com/threads/bloom-pop-under-hdr.281656/ AntiPseudonym fucked around with this message at 05:50 on Mar 22, 2016 |
# ? Mar 22, 2016 05:38 |
|
Okay, before I get used to this idea, could I take a Unity NavMeshAgent, disable it, put it manually (as in, via the player actions) on top of another NavMeshAgent, then enable the first agent and have it navigate off the second agent? I don't mean, convert the second agent into a part of the NavMesh, but just not get too freaked out that an agent is between it and the mesh. Obviously, the pure agent is going to freak out, I've witnessed warping and stupidity, which is why I want to disable it updating the position and rotation and instead use root motion... I guess the second agent should be disabled, but I still don't know if that's going to help, and I can't test it right now, if only because I don't have a root motion character. For clarity, I think that if I'm going to have armadillos, it should be possible to use their rolled-up form as a stepping stone (that can be reproduced).
|
# ? Mar 22, 2016 08:05 |
|
Guns, Guns, Guns.
|
# ? Mar 22, 2016 09:04 |
|
Here is the game ape and I have been working on. Going to be released in the next coming week/s. Thanks #sagamedev for all the input over the last few months. dhw fucked around with this message at 22:17 on Mar 22, 2016 |
# ? Mar 22, 2016 09:40 |
|
dhw posted:Here is the game ape and I have been working on. Going to be released in the next coming week/s. You like this video, right: https://www.youtube.com/watch?v=er416Ad3R1g
|
# ? Mar 22, 2016 09:50 |
|
dhw posted:Here is the game ape and I have been working on. Going to be released in the next coming week/s. Is the car supposed to wobble around like that or is that just the way it's being driven?
|
# ? Mar 22, 2016 10:51 |
|
Megazver posted:You like this video, right: Great sound track! chiefnewo posted:Is the car supposed to wobble around like that or is that just the way it's being driven? Just the way I was driving it. dhw fucked around with this message at 22:17 on Mar 22, 2016 |
# ? Mar 22, 2016 10:54 |
|
dhw posted:Here is the game ape and I have been working on. Going to be released in the next coming week/s. This better have a syntwave soundtrack! E: oh its a video, its loading so slowly I didn't even notice.
|
# ? Mar 22, 2016 16:43 |
|
Zaphod42 posted:This better have a syntwave soundtrack! Yeah the game has about 25-30 different tracks.
|
# ? Mar 22, 2016 16:45 |
|
dhw posted:Yeah the game has about 25-30 different tracks. Holy poo poo that's a lot - did you do it, or was it commissioned or something? E: might as well toss in some more goon love - game looks awesome, and really tweaks the current nostalgia for neon-drenched '80s! Hope you get some good exposure for it! The Gasmask fucked around with this message at 19:02 on Mar 22, 2016 |
# ? Mar 22, 2016 17:10 |
|
The Gasmask posted:Holy poo poo that's a lot - did you do it, or was it commissioned or something? Licensed, presumably.
|
# ? Mar 22, 2016 17:12 |
|
AntiPseudonym posted:So I'm taking a short break for Mystery because I came dangerously close to burning out completely. I think this is really cool. Homebrew engines high five. Regarding FBX. I've written my own Collada importer already (it's well rounded with support for skinned models, etc.) But since then FBX got really popular so I'm thinking about supporting that too. How did you find working with it? and do the various annual updates make much difference? Did you look at the official SDK ?
|
# ? Mar 22, 2016 17:36 |
|
Shalinor posted:- Never change the player height, and anywhere the player needs to be able to crawl through, fill the gap at the top with a special-case collider type, whose collisions you ignore VS the player pill if the player is crouch mode. Good for character action games that need crawling through gutters. I tried changing the height, but I feel like this option is better. I haven't messed around with second colliders, how would I do that? Would I just make multiple colliders for the player object? Could I group them together in some way to be able to reference/check them all at once? 22 Eargesplitten fucked around with this message at 20:32 on Mar 22, 2016 |
# ? Mar 22, 2016 18:32 |
|
dhw posted:Here is the game ape and I have been working on. Going to be released in the next coming week/s. This is pure weaponized 80s. Congratulations for everything you've done and I hope the game's well received.
|
# ? Mar 22, 2016 18:38 |
dhw posted:Here is the game ape and I have been working on. Going to be released in the next coming week/s. Could you post this on youtube or something? I'm getting not found/bandwidth errors and can't watch whatever this is.
|
|
# ? Mar 22, 2016 21:22 |
|
oops. out of bandwidth. https://www.youtube.com/watch?v=prEUK4tFjiw
|
# ? Mar 22, 2016 21:45 |
|
dhw posted:oops. out of bandwidth. The city setting looks good, proper pink/blue neon Miami city looking background, but oh man the lava cliff and the digital tron grid are both even better and super cool. If spinning polygonal pyramids in the sky are wrong, I don't wanna be right
|
# ? Mar 22, 2016 21:54 |
dhw posted:oops. out of bandwidth. Thanks, your game looks rad!
|
|
# ? Mar 22, 2016 23:43 |
|
Thanks everyone! We might have to quote a few of you on the website, haha.
|
# ? Mar 23, 2016 01:23 |
|
|
# ? Jun 10, 2024 03:14 |
|
dhw posted:oops. out of bandwidth. Neon lights, radical tunes, and awesome explosions. It's everything I loved about the 80s.
|
# ? Mar 23, 2016 02:38 |