What are you working on?

[AntonioR]

[QUOTE=Tamschi;50340204]That sounds like something that would be totally illegal here... let me check.
Yes, looks like party (non-)membership is a protected criterion at least for normal state-funded jobs, and probably most jobs in general.

[editline]17th May 2016[/editline]

Was it a specific party or just "any party"? Either way it's seriously shady in my eyes.[/QUOTE]

The current ruling party of course. Don't get me wrong, it wasn't requested on a interview, an ad or something. It's just that my cousin knows a guy in power who could get the wheels rolling in my favor, but me joining the party is the first step in the whole process. She did it for her daughter already, but I don't want to owe these people anything. I bet the whole internship program in government agencies works that way here.

[Rimbas]

[QUOTE=proboardslol;50340361]It'd be cool if there was a keyword in OOP languages which automatically made a get method for private variables. something like:

private readonly int x;

Is there anything like this?[/QUOTE]

int x {private set; get;}

In C# only

[Falcqn]

[QUOTE=proboardslol;50340361]It'd be cool if there was a keyword in OOP languages which automatically made a get method for private variables. something like:

private readonly int x;

Is there anything like this?[/QUOTE]

C#'s properties work kinda like that;

private int x { public get; private set; }

[editline]17th May 2016[/editline]

[QUOTE=aurum481;50340481]In C# only[/QUOTE]

Damn ninjas

[thatbooisaspy]

Working on the editor's UI:

[t]https://dl.dropboxusercontent.com/u/357850863/ShareX/2016/05/2016-05-17_18-19-42.png[/t]

The imgui docking is actually taken from [URL="https://github.com/nem0/LumixEngine"]LumixEngine[/URL], it's slightly modified to disable the annoying tabbar border to causes a bug where the first tab's menu is unselectable. I also converted all of it's saving/loading to simpleini! His original saving/loading mechanism used a combination of lua and his own implementation of ostream. Here's a [URL="https://gist.github.com/thatbooisaspy/b8bd5a9abab5ae5d2c8f4c7f8dd6788e"]gist of the modified source files if you're interested[/URL]. I've also added some other (physics-related) things but i'll save those when I get around to cleaning them up. Colors are still WIP

[Tamschi]

[QUOTE=AntonioR;50340407]The current ruling party of course. Don't get me wrong, it wasn't requested on a interview, an ad or something. It's just that my cousin knows a guy in power who could get the wheels rolling in my favor, but me joining the party is the first step in the whole process. She did it for her daughter already, but I don't want to owe these people anything. I bet the whole internship program in government agencies works that way here.[/QUOTE]

So essentially nepotism, or corruption 'light' :s:
Situations like that are total bullshit, but I suspect it's even more difficult to do something about it where you live.

[editline]18th May 2016[/editline]

On a lighter note, I just overloaded my [I]where[/I] implementation to support pattern conditionals (patterns that result in [I]Nothing[/I]s), which gets rid of all those useless variables:

        internal static Pattern<ITracker<long, byte>, StructureDna> Pattern(Endianness endianness) =>
            from _identifier in FixedByteArray("SDNA")

            where FixedByteArray("NAME")
            from nameCount in Int(endianness)
            from names in nameCount * ZeroDelimitedString

            where Align(4)
            where FixedByteArray("TYPE")
            from typeCount in Int(endianness)
            from typeNames in typeCount * ZeroDelimitedString

            where Align(4)
            where FixedByteArray("TLEN")
            from typeLengths in typeCount * Short(endianness)

            let types = typeNames.Zip(typeLengths, (n, l) => new Type()
            {
                Name = n,
                Size = l
            }).ToArray()

            where Align(4)
            where FixedByteArray("STRC")
            from structureCount in Int(endianness)
            from structures in structureCount *
                from structureType in from st in Short(endianness) select types[st]
                from fieldCount in Short(endianness)
                from fields in fieldCount *
                    from fieldType in from ft in Short(endianness) select types[ft]
                    from fieldName in from fn in Short(endianness) select names[fn]
                    select new Structure.Field()
                    {
                        Name = fieldName,
                        Type = fieldType
                    }
                select new Structure()
                {
                    Type = structureType,
                    Fields = fields
                }
            select new StructureDna()
            {
                Types = types,
                Structures = structures
            };

It (like those multiplications) deviates quite a bit from the relevant mathematical notation, but it's a lot more readable now. Unfortunately you can't start a LINQ query with a [I]where[/I] though.

[Map in a box]

[QUOTE=thatbooisaspy;50340567]Working on the editor's UI:

[t]https://dl.dropboxusercontent.com/u/357850863/ShareX/2016/05/2016-05-17_18-19-42.png[/t]

The imgui docking is actually taken from [URL="https://github.com/nem0/LumixEngine"]LumixEngine[/URL], it's slightly modified to disable the annoying tabbar border to causes a bug where the first tab's menu is unselectable. I also converted all of it's saving/loading to simpleini! His original saving/loading mechanism used a combination of lua and his own implementation of ostream. Here's a [URL="https://gist.github.com/thatbooisaspy/b8bd5a9abab5ae5d2c8f4c7f8dd6788e"]gist of the modified source files if you're interested[/URL]. I've also added some other (physics-related) things but i'll save those when I get around to cleaning them up. Colors are still WIP[/QUOTE]

using imgui for something this large is kindaaa iffy IMO

[proboardslol]

[vid]http://s1.webmshare.com/VNKov.webm[/vid]

Got basic tile rendering down :toot:

[editline]17th May 2016[/editline]

[vid]http://s1.webmshare.com/m7Qnq.webm[/vid]

Basic collision

[helifreak]

[QUOTE=Rocket;50342245]so this really bad flash game i made two years ago now has 220,000 plays
[url]http://www.kongregate.com/games/PlasticContainer/youtube-lets-player-simulator-2014[/url]
[img]https://fi.somethingawful.com/safs/smilies/4/1/shrug.001.gif[/img][/QUOTE]

Holy fuck I'm still 83rd for highest income in a day.

[proboardslol]

[QUOTE=Rocket;50342245]so this really bad flash game i made two years ago now has 220,000 plays
[url]http://www.kongregate.com/games/PlasticContainer/youtube-lets-player-simulator-2014[/url]
[img]https://fi.somethingawful.com/safs/smilies/4/1/shrug.001.gif[/img][/QUOTE]

Did you make any money off of it?

[naelstrof]

[QUOTE=Rocket;50342245]so this really bad flash game i made two years ago now has 220,000 plays
[url]http://www.kongregate.com/games/PlasticContainer/youtube-lets-player-simulator-2014[/url]
[img]https://fi.somethingawful.com/safs/smilies/4/1/shrug.001.gif[/img][/QUOTE]

aaaaaaaaAAAA don't remind me please, of the garbage game I've made 8 years ago with [url=http://www.addictinggames.com/action-games/ragdolltoss.jsp]2,891,883 plays on addicting games alone...[/url]

And I didn't make any money off of it because the ad revenue system asked if I was 18. :dead:

[bonickhausen]

still working on that shitty mspaint game about internet forums

[t]http://i.imgur.com/LVjn35E.png[/t]
[t]http://i.imgur.com/PsOiJOZ.png[/t]
[t]http://i.imgur.com/LAd0PEv.png[/t]
[t]http://i.imgur.com/ZQVTY2q.png[/t]
[t]http://i.imgur.com/ih0jOCt.png[/t]

funny thing is, this is the first screenshot i have from the project:
[t]http://i.imgur.com/jVMqtAo.png[/t]

i love mspaint

[hentie]

we made it to the top 5 finalists in our category yeey 8-)

[media]https://www.youtube.com/watch?v=JlQ_hz40cjU[/media]

[Map in a box]

the sound on that video is so horribly mixed, but otherwise nice job

[editline]18th May 2016[/editline]

(no offense :v:)

[Darwin226]

[QUOTE=AntonioR;50340407]The current ruling party of course. Don't get me wrong, it wasn't requested on a interview, an ad or something. It's just that my cousin knows a guy in power who could get the wheels rolling in my favor, but me joining the party is the first step in the whole process. She did it for her daughter already, but I don't want to owe these people anything. I bet the whole internship program in government agencies works that way here.[/QUOTE]

Hey, if you're able to work in Zagreb I could recommend you to a place I used to work at.

[Fewes]

One big hurdle I've been trying to get over for some time now is that of "outside" lighting. I faked this for a long time by placing point lights around the environment but I knew this wasn't going to cut it in the long run, so this week I spent some time trying to figure out a dynamic system for sun lighting.

What I was looking for was a system where no extra stuff had to be added manually, where it is clear what is "outside" and "inside" and ideally with some way of projecting shadows. Because I'm not doing 3D rendering in the traditional sense, tutorials on how to do shadow projections were of little use to me. Finally I ended up with what I suspect is a similar way of doing things however: Shadows are rendered to a fullscreen 2D buffer which is then projected into the scene using depth information.

This [B]severely[/B] limits their usage as one shadow covers all of the z axis, however I do plan on splitting it up into several buffers thus dividing the z spectrum into smaller cells. Another drawback is that only "hard" geometry (which can be expressed with vertices only) can cast shadows (only world geometry such as walls for now). It does however work wonders for clearly defining outdoor and indoor areas and is very fast, so for the foreseeable future it is a solid way of doing it.

Using the new shadow data I spent some time reubilding my lighting pipeline to better accommodate ambient + direct lighting. Here's what an outdoor scene looks like with [B]0[/B] light entities in it:

[img]http://puu.sh/oWb6k.jpg[/img]

Another new feature which I use to grab high-quality screenshots is the ability to set a rendering resolution different from that of the window resolution. This also helps a lot when working on lower-end systems and a really nice side to it is the ability to have a full-res UI and a low-res renderer.

[Map in a box]

The top seems a bit uncomfortably bright but otherwise, purdy :dreamy:

[Tamschi]

[QUOTE=fewes;50343317][...]

What I was looking for was a system where no extra stuff had to be added manually, where it is clear what is "outside" and "inside" and ideally with some way of projecting shadows. Because I'm not doing 3D rendering in the traditional sense, tutorials on how to do shadow projections were of little use to me. Finally I ended up with what I suspect is a similar way of doing things however: Shadows are rendered to a fullscreen 2D buffer which is then projected into the scene using depth information.

This [B]severely[/B] limits their usage as one shadow covers all of the z axis, however I do plan on splitting it up into several buffers thus dividing the z spectrum into smaller cells. Another drawback is that only "hard" geometry (which can be expressed with vertices only) can cast shadows (only world geometry such as walls for now). It does however work wonders for clearly defining outdoor and indoor areas and is very fast, so for the foreseeable future it is a solid way of doing it.

[...][/QUOTE]

If you have depth data, you can definitely use all of the traditional shadow rendering methods (but yours is probably a bit faster and more precise within the current limitations you set).

If you want to keep the hard shadows, I'd say use plain stencil shadows which will work fine without needing multiple additional textures or passes and are very precise, supporting arbitrary geometry. You just need your scene's depth data beforehand. (There may still be a patent on the more efficient way to do it, but I don't think it makes much difference here.) If you want them smooth, you can use shadow mapping from the directional light's perspective, but that's less precise since it doesn't pixel-perfectly match the screen or shadow casting geometry.

[editline]18th May 2016[/editline]

Will you support shadows of transparent geometry? If you go with stencil shadows (or another kind of screen space shadowing) then that should be easy using that dithering system you use for shrubs and so on.

[Nigey Nige]

Not really programming (sorry) but does anyone have thoughts on this article about designing game worlds? [url]http://www.theastronauts.com/2014/03/secret-immersive-game-worlds/[/url]

[Tamschi]

[QUOTE=Nigey Nige;50344006]Not really programming (sorry) but does anyone have thoughts on this article about designing game worlds? [url]http://www.theastronauts.com/2014/03/secret-immersive-game-worlds/[/url][/QUOTE]

It definitely depends on what you're trying to do. For something like The Vanishing of Ethan Carter yes, but that one actually feels really guided and quite game-y if you ask me ([sp]which does make sense considering the story, but it still contradicts this article in my opinion[/sp]).

If I'm not mistaken, both Call of Duty and Bulletstorm are spectacle shooters, so a less linear approach would probably be out of place there.
The same goes for Half-Life 2, but that one does feel very alive I think because the gameplay is a lot less 'filtered' than with many AAA shooters. It doesn't feel like a film at all, and only a small part of that is due to the way the levels are structured.

Dead Space 1 has definitely a huge horror focus, so there the somewhat unpredictable and confusing [I]world[/I] design is apt.

For something like what cam64DD is making... I don't think it matters at all.
Depending on where the rest of the setting goes, it can very well work either way. It might even benefit from a somewhat explicit 'tourist' feeling.

[HappyHead]

Fat rocket launcher(can detonate rockets with right click), knockback, breakable walls, 2 power ups(one that gives 4x weapon firerate and one that gives faster weapon switch and faster walk speed).
[vid]https://a.desu.sh/jsmpoe.webm[/vid]
Is this stuff interesting? should I post these update vids with more substantial content?

[Tamschi]

[QUOTE=HappyHead;50344369][...]
Is this stuff interesting? should I post these update vids with more substantial content?[/QUOTE]

I'd like to see a little of the code and/or process that goes into making this. To me that's usually the most interesting part of dev blogs.

[proboardslol]

Why aren't a lot of games made in Java?

[Foda]

[QUOTE=proboardslol;50345056]Why aren't a lot of games made in Java?[/QUOTE]

I have a feeling everyone here is going to say "because of speed" (which is kinda true given how the GC works), but the real answer is because of available frameworks/libraries, and lack of console support.

[profan]

[QUOTE=Foda;50345084]I have a feeling everyone here is going to say "because of speed" (which is kinda true given how the GC works), but the real answer is because of available frameworks/libraries, and lack of console support.[/QUOTE]

libgdx for example has a lot of games written with it though, I think there's more out there than one imagines: [url]https://libgdx.badlogicgames.com/gallery.html[/url]

though the gallery doesn't rank by popularity so it's a bit of a crapshoot trying to browse it :v:

[thatbooisaspy]

[QUOTE=Map in a box;50341645]using imgui for something this large is kindaaa iffy IMO[/QUOTE]

No it works great for editor ui's and other non-gameplay stuff (that's the entire reason imgui was created no?). If you look at the screenshot thread in the imgui github lots of people are using it for their engines and other tools. The argument would be valid if you were trying to use imgui for something like game ui :v:

[Geel]

Java, to me, is just really....clunky.

[proboardslol]

[QUOTE=geel9;50345402]Java, to me, is just really....clunky.[/QUOTE]

I used to think so, too, but in my Computer Organization and Architecture class my professor gave us different code to run in Java, C, and Python, and we found that Java was better at optimizing code most of the time compared to GCC, and had faster execution times, believe it or not. Now, these weren't graphical programs, but it made me rethink java

[Nigey Nige]

[QUOTE=proboardslol;50345056]Why aren't a lot of games made in Java?[/QUOTE]

I loved developing in Java, but I had three main problems with it (just my experiences though, pinch of salt):

1) For lots of reasons, the average quality of online Java tutorials is pretty poor. Stanford's Engineering Everywhere course was the very best I could find, but it was limited in that it wasn't really designed for online learning. (Indebted to it though for giving me my first real start in programming.)

2) Finding an engine/library to use was difficult, and the learning paths for Java programmers are treacherous because of the low standard of tutorials. I was constantly running into half-finished documentation and indecipherable forum threads from developers

3) Deployment. I can't remember who but I vividly remember someone on this forum saying "Java: proving that 'cross-platform' means 'only works on one machine' since 1995". My inexperience might have been the reason I always struggled with deployment, but the lack of accessible engines and tutorials certainly didn't help.

I never had issue with the automatic garbage collection or things like that. 'Java is slow' might be broadly true, but I doubt the actual reasons for it being slow affect more than 5% of users.

I'd recommend anyone doing game development to switch from Java to C#. Can't think of any reason not to.

[Geel]

[QUOTE=proboardslol;50345476]I used to think so, too, but in my Computer Organization and Architecture class my professor gave us different code to run in Java, C, and Python, and we found that Java was better at optimizing code most of the time compared to GCC, and had faster execution times, believe it or not. Now, these weren't graphical programs, but it made me rethink java[/QUOTE]

I'm not talking about runtime speed, I'm talking about actually programming in it.

[proboardslol]

Specifically OpenJDK
edit:

MY AUTOMERGE, TOO