Code inspiration and rendering resource

[FluffyFreak]

What is your antique computer, it's not a 486 with an S3 Virge is it? :D

Efficiency is an odd way of putting it, and it depends on the shader or GPU in question of course, but for something like heightmap generation it should allow us to exploit parallelism and do much more work than the raw speed of a CPU could achieve.

If we've got a simple 32*32 patch for example then that's 1024 fragments that a GPU can process in parallel, on my works nVidia GTX 670 that in theory means that I'd actually be leaving 320 of it's 1344 "Cuda Cores" idle :) compared to it's 4 cpu cores. Now that's not actually a far comparison as there's a cost to doing work on the GPU in the setup, transfer, synchronisation, etc blah blah. You can however see how even if the GPU's shaders aren't as efficient as the CPU due to branching and other problems they outnumber the CPU's cores...

...unless you're on an ancient computer :)

[FluffyFreak]

Damn, I hate these talented bastards (Space Engineers).

I mean look at how awesome that looks! I know the planets not "realistic" but it still looks fantastic and that technology could easily be turned to making realistic planets, it's bloody glorious.

[FluffyFreak]

Well, I've started implementing the SoA terrain test thing in C++ so as another type of "terrain". Much like you can specify a heightmap you can specify a JSON file and it builds list of calls to various noise functions to be done in C++ code.

This will let me verify things, test ideas out and isn't dependent on me writing multi-platform code for doing crazy shader shit. It's also why I was asking about adding a node-graph editor using the new-UI last night in IRC ;)

The part I'm struggling with right now is dealing with child nodes, the explanation and intent are really difficult to understand, it looks simple but actually turning it into operations to perform is proving quite complex. I am slowly making progress.

[FluffyFreak]

This is not an inspiration but might be useful information, especially as I was looking into Uniform Buffers recently!

Apparently OSX OpenGL sucks. Well to be fair, I already knew that it sucked.

[jpab]

This doesn't work on my laptop, but perhaps it might be interesting for someone with a more WebGL-capable computer...
http://wwwtyro.github.io/planet-3d/

[FluffyFreak]

jpab, this probably won't work on your laptop either but try to give it a go http://oskarstalberg.com/game/planet/planet.html

[FluffyFreak]

Two great discussions / presentations about procedural methods are:

Creating FPS Open Worlds Using Procedural Techniques (video)
http://www.gdcvault.com/play/1020340/Cr ... rlds-Using

and
Procedural Snake Eyes
http://www.rogueprocess.run/2016/03/03/ ... nake-eyes/

[impaktor]

So he's comparing the two new X-Com remake games, and their combat systems. Not having played them here's where I get a bit lost:

with a few small exceptions [XCOM 1], is deterministic. Attacks never miss, hacking always succeeds, vision is always explicit and consistent. Some effects are randomised – rewinding a turn where a guard spawned can cause it to spawn at a different location, for example...

...But in general, when I am planning my turn, I know what the effects of my actions will be. XCOM 2 builds itself a different way – its combat is famously built around percentage chances to hit, it hides knowledge from the player, and most problematically it also unintentionally hides knowledge that the player is supposed to have, through patchy UI design.

XCOM 2 builds itself a different way – its combat is famously built around percentage chances to hit, it hides knowledge from the player, and most problematically it also unintentionally hides knowledge that the player is supposed to have, through patchy UI design.

So, you have zero chance to miss in the XCOM 1 remake? Seems a bit weird to me. My three favorite games might very well be Civilization 1, X-com UFO and X-com TFTD. All three have dice roll attacks. (And I cheated shamefully in all three with save-reload when outcome was too bad). But I very much have a dice-roll system in mind for a future person-to-person combat system in pioneer.

[FluffyFreak]

No, he's comparing "XCOM 2" and "Invisible Inc", two different games but they both feature procedurally generated content for levels, missions etc.

As for Dice rolls, I think it depends on the kind of game, XCOM wouldn't be "XCOM" without the dice rolls, but it's always worth questioning whether you should be using them.
Imagine having dice rolls in an FPS, you take aim, you shoot, the bullet flies through the air, hits the enemy on the head! ... and then: "Miss! 0 points!" appears next to where it hit, you'd uninstall that crap immediately.

I think that many RPGs take the dice roll thing too far as well, if I've just sliced 6 feet of steel clean through my opponent, and seen it happen right before my eyes, then no, I did not "miss".
Now if they blocked or parried the blow on a dice-roll then that makes more sense.

Some games are about the skill required, in Pioneer it takes effort to hit a ship, that should never be dependent on a dice roll. If we did isometric space stations etc though, then having chance affect the outcome of person-to-person event might make sense.

[impaktor]

No, he's comparing "XCOM 2" and "Invisible Inc", two different games

Doh! I though Invisible Inc. was/could be the name of the company that did XCOM 1 remake, and that he, for some reason, referred to it by the name of the producers of the game.

Sorry.

Well, I'd assume you roll the dice _before_ doing the animation, like in X-com:
- roll dice: will it be hit or miss?
- do animation based on die outcome