vinnyvicious

ping?

I've just finished most tutorial steps on learnopengl.com and I'm really excited about starting to write very small model viewer. Unfortunately, I have no idea how to combine all the shaders in the examples that were given. I mean: how can I do the shadow mapping shader also work with the diffuse/normal one? Is that what render passes are? How is this usually structured in a rendering class?

I'm not really worried about learning curves, or learning a new system. That's not really relevant and usually is fun. What really matters here is that if there's is good money to be grabbed. Does anyone know any public revenue reports from developers on Kongregate or other portals? Are we talking about few cents a day or more Flash-like numbers like 1 or 2 dollars a day, from ads?

If i blend Left and Forward, for example, how can i associate the blending amount from A to B based on the direction vector? As for mirroring, why would i need mirror bones? Maybe just invert the values?

I used to earn a lot of money with Flash games 5 years ago. Either by selling whitelabel games to portals or just publishing them on MochiMedia, i was earning about $500/month. Since MochiMedia died and the popular FGL website all died, and everyone went to mobile, i stopped working with that. I was thinking about returning, but i'm curious about the current market state. Is anyone earning money selling games to portals? Are portals buying that? If so, where's the current hub for people to do business on? Is anyone earning money by putting ads into their games and publishing to portals? Is anyone dedicated to desktop web games?

I'm currently working on a small third-person shooter along with a friend, who is an artist. Today, i've finished the state machine that controls all the possible states that the player can be in. My friend was going to use that as a reference to start animating the player model. Unfortunately, since we are just two guys, i think the combination of states is going to affect his productivity. I mean: Walking_WithRifle Walking_NorthWest_WithRifle Walking_Backwards_WithRifle StrafeLeft_WithRifle Crouched_Walking_WithRifle Crouched_StrafeLeft_WithRifle We have several other weapon types (Melee, Pistol, RPG, Grenade, Mine, etc.). I'm trying to find a solution for this problem with code, but the engine i am using (Urho) is limited on what i can do. For the different weapon types, since they mostly affect the upper body parts, i was thinking about asking him to animate just an idle pose with each of the weapons, and then i can use that to blend with other states (Walking, Running, Crouched Walking, etc.) However, for the NW, SW, and other diagonal movements... what can i do? A rotation of the pelvis bone? What do you guys suggest? Also, for the StrafeLeft/StrafeRight... is that something that i can also do with rotation?

I can't have a single flag. Panel A might fire his own "close" event, but the listener of that event can't simply set IsOverlayVisible to false because he needs to know which other panels are using the Overlay.   I already have events in place, but that only means that all my spaghetti is inside the Mediator (which i call Controller).

So, i'm building the UI for my strategy game and i quickly got trapped into a big spaghetti. I've got a design that works something like this:   [attachment=33183:test.jpg] All panels trigger a dark overlay on the screen. (total black 70%) The blue panels slide from left and right, respectively. They can be present at all times, can appear together, and have the lowest zindex. The red one always appears on top of everyone, but does not force others to close The orange one makes any other panel slide back to their "hidden" position. When the orange slides up, the blue ones slide back and the red one slides up. I also can have centered "popups". Right now, my code is a mess:   Since all panels are independent objects, they share an instance of an Overlay object. They can call the methods hide() and display(). Calling close() on the left blue panel should not close the overlay if the right blue panel is still open, so i implemented a counter inside the Overlay object: I only really hide it when the counter reaches zero. Kinda like a refcounter. I also have to wrap every open()/close() method of my panels into conditional, to check if the panel is not already open, in order to avoid refcounting the overlay incorrectly.   Clicking on the overlay should also trigger all panels to close instantly. So i have more ifs. And i have even more nasty ifs to make a difference between the red and orange panels. When i started doing the popups, which are dynamic, i lost my mind.   So, this mess is already too difficult to maintain and i could not think of a decent solution. Can anyone guide me to a good book on Software UI architecture and proper design patterns? I've been working for too long with Java Web API's and i got too dumb for realtime programming. :P

There's also CDLOD: http://www.vertexasylum.com/downloads/cdlod/cdlod_latest.pdf

This guy managed to do a great job with UDK:     https://krisredbeard.wordpress.com/projects/tfp-demo/

I'm trying to minimize the amount of work involved in my current project, which is a first-person shooter, and the first approach i'm experimenting with is to remove first-person weapon/arms models entirely, relying only on the third-person model. So far, it looks good, with the only problem being that the camera has to be placed on the character's chest, close to his neck. This can ruin the immersion, though, by creating a mismatch between the player eye-level and the camera level. Specially during combat.   I'm sure there are more studies in this area, FPSes being quite popular. Does anyone recommend any article on the subject? Maybe a GDC talk or something? Thanks in advance!

Octrees! https://github.com/gametutorials/tutorials/tree/master/OpenGL/Octree

The loop is based on libev, so i tried the timeout feature, but it didn't work very well. Timeouts were never triggered because, well, there was no timeout going on. I've also tried libev's idle callback, but it runs too many times, to the point where my conditional for the size of the buffer was affecting performance.

So, i'm working on a small multi-threaded game server. Currently, there is a finite amount of threads and each of them are taking care of different tasks. So, the whole loop is naturally thread-safe. Unfortunately, there are a few parts of the application that threads need to share. Logging, for example. I have an unordered_map<string, ostringstream> static variable that i use for logging stuff dynamically. At the end of each loop iteration, i have condition to check the size of the ostringstream, and if it's higher than X, i save it to disk. This way, i'm only offloading logs to disk a few times, not on every frame.   The problem is: when the loop is idle, waiting for commands, the log can be "stuck". If the ostringstream is lower than X, and i go to an idle state, then it keeps there forever, never offloading to disk the last parts of the log. Right now, i've made another thread that runs on intervals, responsible for checking if the size of the buffer has changed or not. If it hasn't changed for a long, it dumps it.   But is there a better way of handling this?

I've been reading a lot on engine development lately, and it's making me question my code. Coming from C#/Java background, i'm used to OOD. I'm trying to understand C++'s data-oriented design and, specially, how it applies to game engine development. My biggest question so far is about the implementation of a scene graph: how can you do that with data-oriented design? For me, scene graphs, and well... graphs, are naturally hierarchical, so OOD matches perfectly.