Register a SA Forums Account here!
JOINING THE SA FORUMS WILL REMOVE THIS BIG AD, THE ANNOYING UNDERLINED ADS, AND STUPID INTERSTITIAL ADS!!!

You can: log in, read the tech support FAQ, or request your lost password. This dumb message (and those ads) will appear on every screen until you register! Get rid of this crap by registering your own SA Forums Account and joining roughly 150,000 Goons, for the one-time price of $9.95! We charge money because it costs us money per month for bills, and since we don't believe in showing ads to our users, we try to make the money back through forum registrations.
 
  • Post
  • Reply
Shalinor
Jun 10, 2002

Can I buy you a rootbeer?

Mr Cuddles posted:

Thanks for the advice guys seems like it's not going to be easy. I don't have any c++ experience only c# and as I said I'm not too keen on being 100% focused purely on writing code. I'm angling more towards the creative/UI side of things. I will take a look at phaser.io and see if I can release a simple javascript based game and how I feel about the whole process. That is surely playing to my current strengths.
UI Programmer a/o UI Artist are your options, then. If your art skills aren't up to snuff, go for UI Programmer, they end up doing a lot of the layout work themselves anyways and need an eye for it. C# would be enough for that position at some studios (especially mobile / Unity-focused).

Jewel posted:

A game like smash bros or a fighting game does the same. They just go through each frame of the animation and add or adjust hitboxes. Usually in a non-Unity situation you'd build a small editor that the artists can step through the frames, add hitboxes, and adjust features about the hitboxes (like what type they are, potentially damage, etc). In Unity I have no idea but perhaps there's a plugin someone's made.
The solution in Unity is pretty much the same - build (or buy) a tool. I built a hitbox system that's invisible, and it's... miserable, just don't do it. Take the time to add visualization.

Adbot
ADBOT LOVES YOU

Rocko Bonaparte
Mar 12, 2002

Every day is Friday!
I don't know if I feel better about hitboxes. I guess I'm reassured that they are as annoying for everybody else as they are for me, but on the other hand I'm sad that they are as annoying for everybody else as they are for me.

I decided to look up "Diablo hitbox" and found tons of people complaining about it. Yeah, it sounds like the kind of thing you tool up to do. But while I just have cubes zooming around, hitting other cubes, I should maybe get some core mechanics down first. I have not even gotten to the animated models within Unity itself yet.

Speaking of that, actually, is there a poor man's tool for getting basic human models and animations down? I was hoping not have to figure that all out up front with Blender or something. I had enough fun in Blender just making static stuff to want to jump into that right away.

ShinAli
May 2, 2003

The Kid better watch his step.

FuzzySlippers posted:

I have noticed this myself. Even jobs focused on Unity and presumably largely C# still want years of C++ experience. Maybe its a way to weed out applicants even if its not directly required by the project? Its one of the reasons I want to try to learn some C++ after this project.

I came into the industry with relatively no C++ experience (some throwaway engines that never got anywhere) and a some hobby experience in C# and Unity, and I've been messing with C++ projects from the get go. First project was Unity based, but lots of C++ involved since a major system of the game was dependent on the native plugin I was making for it. I hopped onto an internal prototype using UE4 afterwards, and now I'm working on another contracted project that's just C++ with one of my dream studios on my dream franchise.

If you think you can pull it off, I'd say apply. Obviously it'd help if you know your way around C++, but if they're asking for Unity devs and if you feel good enough with Unity, just do it man.

Note: While I wasn't super experienced as a C++ dev, I was super confident in my skills with C. That might have helped for my lack of knowledge in C++.

ShinAli fucked around with this message at 18:30 on Oct 29, 2014

Yodzilla
Apr 29, 2005

Now who looks even dumber?

Beef Witch
Current game development status: found out that the Kindle HDX 8.9" for some reason inverts its LandscapeLeft and LandscapeRight returns so I have to handle that case specifically for that device.

Thanks Amazon. Thamazon.

echinopsis
Apr 13, 2004

by Fluffdaddy

Lowen SoDium posted:

\
I really hope that they do something to improve the context sensitive node finder. Ever since 4.5 came out, I am having really bad luck finding what I am looking for. With the context sensitive box unchecked, I would assume that it would show me all nodes and not just the ones that the editor thinks that I want. But I am finding that I often have to search for a node with that box unchecked and then again with it check before I can find what I am looking for.

https://answers.unrealengine.com/questions/111211/having-lots-of-problems-with-blueprint-context-men.html

It's gonna be fixed in 4.6 or a hotfix. You can enable an experimental thing which fixes it but breaks the ability to choose a cast-to which is probably worse.


I use the palette on the right so I don't have to click that checkbox

echinopsis
Apr 13, 2004

by Fluffdaddy
Because I have no shame or self-respect here is a blatent whore-out for my game
http://rodtronics.tumblr.com/

TIA for wasting your time reading my poo poo

superh
Oct 10, 2007

Touching every treasure

echinopsis posted:

Because I have no shame or self-respect here is a blatent whore-out for my game
http://rodtronics.tumblr.com/

TIA for wasting your time reading my poo poo

Post your game:

DEATHSTATE

http://instagram.com/p/uwSmwdrVN9/

http://instagram.com/p/twiwxnrVCo/

echinopsis
Apr 13, 2004

by Fluffdaddy
that second picture owns


Instagram huh? I'm pretty uninformed about the amount of social networks I need to be for my game.

twitter
tumblr
instagram?
LinkedIn?
bebo
MySpace?

Shalinor
Jun 10, 2002

Can I buy you a rootbeer?

echinopsis posted:

Instagram huh? I'm pretty uninformed about the amount of social networks I need to be for my game.

twitter
tumblr
instagram?
LinkedIn?
bebo
MySpace?
All indie devs need to be on Twitter 2+ years ago. It's the kinda thing that's important, but won't be useful until you've spent ages on there building a following.

Tumblr is as important for artists as Twitter is for devs. Though artists are usually on both and do a lot of crossposting.

The rest of those, hahahahaha.

Yodzilla
Apr 29, 2005

Now who looks even dumber?

Beef Witch
Yeah I was mad late to Twitter but I'm slowly building a network. As for other social networks, Facebook and Google+ are recommended (honestly surprising how much traffic each of these gets) and Tumblr is rad as hell. Linkedin is good to have in general for all kinds of business reasons but not really for promotion.


Speaking of which we've been updating out Tumblr on the regular with rad stuff http://ghostcrabgames.tumblr.com/ :smugwizard:

KoRMaK
Jul 31, 2012



I want to rewrite an AS3 game I have as multiplayer, and on a new platform. What platform should I move to? Xna, Unity, something else? It's a 2D game for now, but I want to be able to use the same library later for 3D.

superh
Oct 10, 2007

Touching every treasure
I like Instagram for getting art out there, it's definitely not a serious promotional social network since it's fairly hard to share.

I need to blow up Twitter more, though. I just post the stupidest of stupid things on there.

DeathBySpoon
Dec 17, 2007

I got myself a paper clip!
LibGDX is pretty fantastic and it targets pretty much everything. Java isn't a bad language and I haven't really found anything I can't do with the library.

Angryhead
Apr 4, 2009

Don't call my name
Don't call my name
Alejandro




I'm also a big fan of libGDX. Even got a Mac just a few days ago so I'll see how the RoboVM porting process goes.

WrathOfBlade
May 30, 2011

Hey, anybody remember Pachinko Man? It's a little adventure game I worked on with some friends for an SA Gamedev challenge a few years back and people seemed to like it. Anyway, the team and I finally got together and turned it into a finished (albeit short) product, which you can play online over here if you're interested!

(Incidentally DEATHSTATE looks super cool and I really want to play it)

xzzy
Mar 5, 2009

Anyone ever fooled around with fluid simulation before? I'm tinkering with an idea and am trying to decide if what I came up with is reasonable. I'm just trying to get the horizontal flows working for now, depth isn't being considered.

The basic scheme I have at this point is a 2D grid of points with each point assigned a vector indicating where liquid on that tile is trying to go, then I use a cellular automata style approach to do several iterations allowing the entire grid to interact with itself. The function is literally doing nothing but averaging the vectors of all its neighbors. This produces something that initially seems plausible but considering how easy this was to write and I have no background in this kind of topic it makes me suspicious if I'm missing something obvious.

How it looks at the moment:

http://gfycat.com/GiganticBabyishDaddylonglegs

It doesn't currently handle obstacles which I can imagine being a pain to implement but it's not a goal at the moment.

Am I even in the right ballpark?

Inverness
Feb 4, 2009

Fully configurable personal assistant.
What are the chances that Microsoft will revive XNA? It is a superb library in so many ways and help draw a lot of people into C# for development.

MonoGame's development moves at a crippling pace and they still haven't finished support for a content pipeline.

retro sexual
Mar 14, 2005

Mr Cuddles posted:

Thanks for the advice guys seems like it's not going to be easy. I don't have any c++ experience only c# and as I said I'm not too keen on being 100% focused purely on writing code. I'm angling more towards the creative/UI side of things. I will take a look at phaser.io and see if I can release a simple javascript based game and how I feel about the whole process. That is surely playing to my current strengths.

Just wanted to chime in and say your background is very similar to mine: web dev with UI a specialty. I think you are in a fine position to be programming games in a tiny (indie) team, as you will have both programming skills and an eye for making things look and feel nice (which is incredibly important for games). Don't rule out going on your own. If that's something you are interested in start doing some gamejams (ludum dare is coming up in a month, 1GAM is every month, 7DFPS is like next week) and then think about working one of your gamejam games up into a 'proper' release. You can do all this without giving up your current employment.

With your skillset you will be able to make a nice prototype (particularly in a 2d game), then get an artist involved, and then be able to make their work REALLY feel nice (look up tweening!).

This gif is what I'm working on, having followed this general strategy from your very starting point:
http://i.imgur.com/WHkI6zn.gif

Drastic Actions
Apr 7, 2009

FUCK YOU!
GET PUMPED!
Nap Ghost
For those who submitted for the Xamarin Unreal Engine C# stuff, you should have gotten emails with download links. You have to sign an NDA first though. Also if you just sign up now, you should get an email automatically with the download key.

Let me know if any of you have any issues getting it.

Jewel
May 2, 2009

xzzy posted:

Anyone ever fooled around with fluid simulation before? I'm tinkering with an idea and am trying to decide if what I came up with is reasonable. I'm just trying to get the horizontal flows working for now, depth isn't being considered.

The basic scheme I have at this point is a 2D grid of points with each point assigned a vector indicating where liquid on that tile is trying to go, then I use a cellular automata style approach to do several iterations allowing the entire grid to interact with itself. The function is literally doing nothing but averaging the vectors of all its neighbors. This produces something that initially seems plausible but considering how easy this was to write and I have no background in this kind of topic it makes me suspicious if I'm missing something obvious.

How it looks at the moment:

http://gfycat.com/GiganticBabyishDaddylonglegs

It doesn't currently handle obstacles which I can imagine being a pain to implement but it's not a goal at the moment.

Am I even in the right ballpark?

Yes and no. Yes you're in the right ballpark, no you probably won't be able to get much further without referring to actual papers about 2D fluid simulation. I made this a few years ago in python:

https://www.youtube.com/watch?v=RaaDhAu-QFk&hd=1

It's the same concept, a vector field that is semi-cellular automata. I don't remember the right terms or exact process (you'll have to research 2D fluid simulation, there's one paper in particular which is great) but look up convection. The grid basically uses the average of cells around it and also passes on the motion along with any other properties that you might add (ie color, viscocity). It takes a good algorithm to get it to not absorb or decay anything, and mine built up and kept too much velocity even on empty squares because I must have messed something up. I wouldn't do it in python these days because it was so slow but hey, I used to do everything in python v:v:v

Also the algorithm in the first few seconds of the video is pretty much a 1:1 translation from the C algorithms in the paper.

Jewel fucked around with this message at 23:28 on Oct 30, 2014

Shalinor
Jun 10, 2002

Can I buy you a rootbeer?

Jewel posted:

Yes and no. Yes you're in the right ballpark, no you probably won't be able to get much further without referring to actual papers about 2D fluid simulation. I made this a few years ago in python:

https://www.youtube.com/watch?v=RaaDhAu-QFk&hd=1

It's the same concept, a vector field that is semi-cellular automata. I don't remember the right terms or exact process (you'll have to research 2D fluid simulation, there's one paper in particular which is great) but look up convection. The grid basically uses the average of cells around it and also passes on the motion along with any other properties that you might add (ie color, viscocity). It takes a good algorithm to get it to not absorb or decay anything, and mine built up and kept too much velocity even on empty squares because I must have messed something up. I wouldn't do it in python these days because it was so slow but hey, I used to do everything in python v:v:v

Also the algorithm in the first few seconds of the video is pretty much a 1:1 translation from the C algorithms in the paper.
These algorithms all fall under the heading of "wave propagation." What you're doing in the grid amounts to a 2nd or 3rd degree integral approximation, depending on how many neighbors you involve and the coefficients. One of the most important aspects of the approximation is your boundary conditions, for influencing the amount of energy in the system - and you have a LOT of boundaries in a Terraria-esque world if you think about it.

I... don't actually know how much of that you need to know to do standard fluid sims like you see in games, though. I assume you can find a to-the-point writeup that doesn't touch on much of the "why the coefficients are the way they are" logic.

Truman Peyote
Oct 11, 2006



I want to try out Unreal 4, but my laptop only has a small solid-state drive that's close to capacity already. About how much disk space does the Unreal 4 editor take? Let's assume I want to have the demo projects in there, too.

Unormal
Nov 16, 2004

Mod sass? This evening?! But the cakes aren't ready! THE CAKES!
Fun Shoe

Makeout Patrol posted:

I want to try out Unreal 4, but my laptop only has a small solid-state drive that's close to capacity already. About how much disk space does the Unreal 4 editor take? Let's assume I want to have the demo projects in there, too.

I just installed it yesterday, and the folder it's installed in takes 11.2 GB.

darkpool
Aug 4, 2014

xzzy posted:

Anyone ever fooled around with fluid simulation before? I'm tinkering with an idea and am trying to decide if what I came up with is reasonable. I'm just trying to get the horizontal flows working for now, depth isn't being considered.

The basic scheme I have at this point is a 2D grid of points with each point assigned a vector indicating where liquid on that tile is trying to go, then I use a cellular automata style approach to do several iterations allowing the entire grid to interact with itself. The function is literally doing nothing but averaging the vectors of all its neighbors. This produces something that initially seems plausible but considering how easy this was to write and I have no background in this kind of topic it makes me suspicious if I'm missing something obvious.

How it looks at the moment:

http://gfycat.com/GiganticBabyishDaddylonglegs

It doesn't currently handle obstacles which I can imagine being a pain to implement but it's not a goal at the moment.

Am I even in the right ballpark?

http://www.autodeskresearch.com/pdf/GDC03.pdf

Descrete implementations, as opposed to SPH type will probably be based on this paper. The code is ugly but serviceable, you should be able to work out how to implement it in shaders too. I think there's a GPU Gems article with shaders if you really need it.

I implemented boundaries with this but I can't think what the paper was right now, it wasn't too difficult.

Truman Peyote
Oct 11, 2006



Unormal posted:

I just installed it yesterday, and the folder it's installed in takes 11.2 GB.

Thanks. Guess I've got some poo poo to delete.

echinopsis
Apr 13, 2004

by Fluffdaddy
your project folder can get much bigger than a gig if your baking light maps

snucks
Nov 3, 2008

Try again. Fail again. Fail better.

WrathOfBlade posted:

Hey, anybody remember Pachinko Man? It's a little adventure game I worked on with some friends for an SA Gamedev challenge a few years back and people seemed to like it. Anyway, the team and I finally got together and turned it into a finished (albeit short) product, which you can play online over here if you're interested!

(Incidentally DEATHSTATE looks super cool and I really want to play it)

:frogsiren: PLAY OUR VIDEO GAME GUYS :frogsiren:

ShinAli
May 2, 2003

The Kid better watch his step.
Finally got the Mono UE4 download, might just try to patch it in directly to 4.5.

Pretty dang excited.

OneEightHundred
Feb 28, 2008

Soon, we will be unstoppable!

Inverness posted:

What are the chances that Microsoft will revive XNA? It is a superb library in so many ways and help draw a lot of people into C# for development.
Probably zero. It hasn't been under active development for ~2 years, their main push in that direction lately has been ID@Xbox, which among other things is planning to offer the Xbox One add-on for Unity for free, and apparently longer-term, attempt to make retail Xbox Ones usable as devkits.

Stick100
Mar 18, 2003

ShinAli posted:

Finally got the Mono UE4 download, might just try to patch it in directly to 4.5.

Pretty dang excited.

I tried several times to patch 4.4 and it would not compile (whether or not I compiled first before patching or now). I was using the "terminal" from source tree to apply the patch and it looked like it worked correctly. Please let me know if you got it to compile and what steps you did.

Drastic Actions
Apr 7, 2009

FUCK YOU!
GET PUMPED!
Nap Ghost

Stick100 posted:

I tried several times to patch 4.4 and it would not compile (whether or not I compiled first before patching or now). I was using the "terminal" from source tree to apply the patch and it looked like it worked correctly. Please let me know if you got it to compile and what steps you did.

I think it's against 4.4.3, so maybe try that? There were a few other people on the mailing list who were having issues too. There should be an update soon in any case.

At least my mail server/azure site are working correctly, so my end is all good.

xzzy
Mar 5, 2009


This is a really good read (and I assume is the paper Jewel was mentioning), it's a shame my quick attempts at googling papers didn't turn it up.

Probably because I was searching stuff like "2d fluid sim" or "fluid simulation for games" which yields a completely different set of results.

thanks!

Jewel
May 2, 2009

xzzy posted:

This is a really good read (and I assume is the paper Jewel was mentioning), it's a shame my quick attempts at googling papers didn't turn it up.

Probably because I was searching stuff like "2d fluid sim" or "fluid simulation for games" which yields a completely different set of results.

thanks!

Yeah this is the one. Sorry I couldn't find it! It's really neat though.

xzzy
Mar 5, 2009

I found the source of the demo that went with that talk.. and it still compiles! Yosemite only produces one deprecation warning. I can't remember the last time I tried to build 12 year old code and it just worked. Hell, 1 year old code sometimes blows its brains out all over my terminal.

It's great fun to fool around with, and should be pretty informative as I tear in to it.

darkpool
Aug 4, 2014

xzzy posted:

I found the source of the demo that went with that talk.. and it still compiles! Yosemite only produces one deprecation warning. I can't remember the last time I tried to build 12 year old code and it just worked. Hell, 1 year old code sometimes blows its brains out all over my terminal.

It's great fun to fool around with, and should be pretty informative as I tear in to it.

The way the linear-backtrace in the advection step works tends to wash out a lot of interesting details, there are some simple extensions like vorticity confinement which help a lot with that.

echinopsis
Apr 13, 2004

by Fluffdaddy
I'm gonna fake a waterfall in my game. I think I might just drop some spherical meshes :smugmrgw:

(it'll go well with the rest of the amateur nature of my game)



Also I reckon UMG on UE4 fuckin sucks! Ugly as balls. I've rolled my own menu that is in game, now just gotta find a way to plug meshes onto the front of my camera so they're always there (they don't seem to show up) for my health/energy/fuel etc.

Rocko Bonaparte
Mar 12, 2002

Every day is Friday!
I have a question about setting up hit boxes in Unity. I tried generating a cube as a starter game object so I could easily see it as I play with hitbox mechanics. However, it looks like this cube has some physics to it that I did not establish. My player test cube will collide with it, rather than pass through and trigger inside of it.

Edit: I realized after posting I should just dump the components the generated cube has, and it is sitting on a box collider. I am assuming that's my culprit. Is there an easy way to have the engine generate a cube mesh for me without any other components? All I could online where people manually specifying vertices, which seemed really ham-fisted.

Here's the code:

code:
    GameObject GenerateHitBox(Vector3 position, Vector3 scale)
    {
        var hitBox = GameObject.CreatePrimitive(PrimitiveType.Cube);
        hitBox.transform.position = position;
        hitBox.transform.localScale = scale;

        hitBox.AddComponent<HitBoxCollider>();
        var boxCollider = hitBox.AddComponent<BoxCollider>();

        boxCollider.isTrigger = true;

        if(hitBox.GetComponent<Rigidbody>() != null)
        {
            Debug.Log("Hit box has a rigid body!");
        } 
        else
        {
            Debug.Log("Hit box does not have a rigid body!");
        }

        var allComponents = hitBox.GetComponents(typeof(Component));
        foreach(var component in allComponents)
        {
            Debug.Log(component.GetType().Name);
        }

        Object.Destroy(hitBox, 5.0f);
        return hitBox;            
    }


... The HitBoxCollider

using UnityEngine;
using System.Collections;

public class HitBoxCollider : MonoBehaviour {
    void OnCollisionEnter(Collision col)
    {
        Debug.Log(gameObject.name + " collided with " + col.gameObject.name);
    }

    void OnTriggerEnter(Collider other)
    {
        Debug.Log(gameObject.name + " triggered by " + other.gameObject.name);       
    }
}

Rocko Bonaparte fucked around with this message at 18:43 on Nov 2, 2014

Corbeau
Sep 13, 2010

Jack of All Trades
I'd make a prefab with the components I wanted, then instantiate that.

e: I'm not one of the "everything in code" Unity users though.

Rocko Bonaparte
Mar 12, 2002

Every day is Friday!

Corbeau posted:

I'd make a prefab with the components I wanted, then instantiate that.

e: I'm not one of the "everything in code" Unity users though.

I figure that us where I am going, but I was hoping to have a helper to generate a mesh for the hitbox's bounding box. All I could find is something the crapped out a prefab cube with the box collider already attached.

Given how many people end up specifying vertices, I assume I will just have to man up and do that too.

Edit: I did not realize prefabs were a specific thing in Unity. I'm trying those right now and it seems to be behaving decently. This looks smarter than having to explicitly code and track the buildup of the hitbox GameObject inside the IDE, rather than in Unity with everything else.

Rocko Bonaparte fucked around with this message at 08:49 on Nov 3, 2014

Adbot
ADBOT LOVES YOU

ShinAli
May 2, 2003

The Kid better watch his step.

Stick100 posted:

I tried several times to patch 4.4 and it would not compile (whether or not I compiled first before patching or now). I was using the "terminal" from source tree to apply the patch and it looked like it worked correctly. Please let me know if you got it to compile and what steps you did.

I skipped out on 4.4 and went through applying each patch manually to 4.5. I think they released a new zip in the same download page that might fix things, though.

I've hacked on it over the weekend and only have some weird mono threading crash when the editor exits, but at least I got hot reload working. Compared to last time I've messed with UE4, having new game code reloaded without shutting down the editor is a god send. I know 4.5 has better C++ hot reload, but I like my compile times with C#.

Bummer is no debugger hooks, but I've been thinking of looking into the Xamarin Studio UE4 plugin and adapting something for Visual Studio, maybe messing with adding debugging support.

I hope I'd be able to share my branch soon, there are people probably way more familiar with UE4 than I am.

  • 1
  • 2
  • 3
  • 4
  • 5
  • Post
  • Reply