Spring RTS Engine

erm

that's why it was made a rapid module?

Super Mario wrote:"MASSIVE speedup in lategame rendering (cutting rendertime upto 1.5-3 times!)"
What exactly have you done to achieve this? Did you guys did an overall parts of the engine?

By removing most std::map, std::list & std::set from rendering code.

Truth is: Vulkan is Mantle. The current non-pulic drafts are 90% Mantle headers, which are now cleaned up & improved by all other contributes. Giving the other vendors the chance to adjust the API to better reflect all hw and not just AMD ones. Never the less Vulkan/Mantle is a monster in code s...

As a rule, run-time configurable behavior is better than load-time configurable behavior. This is what you should have been told too, and if you weren't that would be an omission on the part of the reviewer(s). Wrong. Rules are: 1. Can I remove/merge tags? 2. Can I replace rarely used tags with lua...

I don't understand your `question` Never the less, afaik the original implementation only know 1 hash (result of game + map hash) all splittings of those in the lobby protocol were added later. So all bogus stuff comes from that (instead of cleanly implementing an arbitrary data interface, ppl glued...

THEN DON'T DO IT ok, then shutdown the wiki because of security holes? or will someone else update it? 1. It was FORB, AF & YOU who wanted to switch to wordpress which is a security hole itself! 2. When longterm support runs out for a version, it means they will stop porting back security bugfi...

abma wrote:I don't have the time/knowledge to fix/update the template after the upgrade.

THEN DON'T DO IT

I am pissed this website regresses more and more just cause people who never contributed to the website mean they have to `upgrade` stuff and then are to lazy to reintergrate the old themes.

they clump up -> checked!
they run in lines -> checked!
"they are in the rocks!" -> checked!
you have to count them afterwards -> checked!

You were the one against `release` branches unstable / not working stuff always went into branches... if the development branch is broken, other devs can't continue their work! regardingless of that: the release branch was abandoned because it wasn't used probably and the releaser had to fix merge ...

abma wrote:

jK wrote:could you please limit this thread to RC4, the latest weapon commits aren't finished yet ...

then please use a branch next time. the develop branch should be always in a releaseable state.

You were the one against `release` branches

could you please limit this thread to RC4, the latest weapon commits aren't finished yet ...

Split out engine devs discussion of development/release branches to here viewtopic.php?f=12&t=33451. (Silentwings)

recreating dlist every frame doesn't give any speedup. On ATIs it's even a massive lag reason.

you got the source luke

I'm not really sure how the whole gl.PushMatrix/gl.PopMatrix thing works google it tip #1: gl.texture is a very expensive call (doesn't matter if in dlist or not!). 99% of your calls of it can be removed by moving it in front of the loops tip #2: common coding rule: "never have more than 3-4 i...

it was default 0, then changed to default 1 and then the tag was removed as a whole

Why is this even listed if it's removed?

Cause it's not auto-generated ?

no, I made the most tiny commit of those:
the dds one after I saw the map made by gajop, which used a detailtex that creates distortions when zoomed out. Those can be fixed when `manually` creating the mipmaps.

<_< >_> Found a way to make DevIL build a 16bit grayscale PNG commit: https://github.com/spring/spring/commit/83338372b2a2302eddbb5b25f40340c39788b71b updated widget: function widget:GetInfo() return { name = "HeightMap To PNG16", desc = "Saves current heightmap to heightmap.png"...

@gajop seems some gl states differ in your enviroment, appended a version that explicit sets some states: function widget:GetInfo() return { name = "HeightMap To BMP32", desc = "Saves current heightmap to heightmap.bmp", author = "jK", version = "1", date = &q...