Jump to content

  • Log In with Google      Sign In   
  • Create Account


Member Since 02 Nov 2012
Offline Last Active Today, 03:52 PM

Posts I've Made

In Topic: how to differentiate performance on android devices

30 June 2016 - 11:40 AM

This is a hard problem, and one of the reasons that more people make their games free on the Play Store than on Apple's app store. There's just so many devices out there it is effectively impossible to avoid selling your game to a customer who won't be able to play it.


One tool in the manifest to add to Nanoha's suggestion is the screen size: <supports-screens android:smallScreens="false" /> ditches tiny phones which are probably underpowered. But in general it's hard to do much in the manifest beyond OS version, screen size and GLES version.


At runtime, I assess device's graphics capabilities mainly by proxy, I look at information like:


* Number of CPU cores

* Amount of RAM

* Whether highp is supported in fragment shaders

* Whether GLES3 is supported

* Whether depth textures are supported


And I choose a graphics option based on that - I let the user override my choice.

In Topic: In depth articles about fast paced multiplayer network architecture

28 June 2016 - 12:40 AM

This might be a useful video which talks in detail about the techniques they use for rocket-like projectiles in Overwatch (which are much harder to do well than bullet-like projectiles):

In Topic: Generating Provinces From A Map

18 June 2016 - 11:24 PM

Is it possible to flip it around and create the land map from provinces instead of trying to break-down a pre-existing land map?


That's what I ended up doing in my title to avoid having to solve this problem. My landmass generation looks something like:


for (number of provinces that I want)

1. Pick a single sea cell as a start point (after rolling a dice to decide whether to start attached to existing land, or start a new island)

2. Grow the new province, adding one extra sea cell at a time until its a suitable size

3. If we managed to grow large enough, then accept it, otherwise back to #1 to pick a new start point.

In Topic: how to solve z-fighting

11 June 2016 - 05:31 AM

I always avoid depth bias style approaches because they're never very portable across different rendering APIs. I prefer to use a different projection matrix with an ever-so-slightly reduced field-of-view for decal rendering.

In Topic: New Cryengine/Unity/Unreal 3rd/2nd/1st person shooter

28 May 2016 - 03:28 PM

Possibly the most relevant piece of research you could do for your project:

I think this is the more appropriate link: https://en.wikipedia.org/wiki/Poe's_law