Devlog KAG Build 2146 - Siege Changes, Fixes and Improvements for Modding

That would be really convenient but it would make boulders way too abundant. Maybe it takes time to craft one or there is a cooldown before you can craft another.

 

We already have things working on USA I'm afraid.
You might accept unstable mods, but we'd rather not have the official server hosting be unstable, and mods that need to be patched often are a pain. Also, if the mod in question is role-play/persistent I'm not so interested in having it on the server, as it'll be swapped out every few hours (which would break a role-play style game).

That said, I'd be interested to have a look at it and see if it's suitable. You guys are free to host whatever, the official modded server is just to help increase user exposure to mods and help "normalise" them a bit.

As I said right in the OP

Please keep the discussion about that in one place if you can, thanks.

I'm not sure about that. I get the appeal, and we'd certainly see more boulders flung, but catas are actually already very widely used and I'm not sure they really need the buff...

 

I'm not on KAG this week as those keeping score should know - however, I've spent this morning looking at the knight attacking through doors issue and adding messages to the modded server about when it'll restart so as to not take players by surprise.

I've also fixed the issue with modded tiles creating solid/dangerous collapse blocks that @GoldenGuy was complaining about. New build probably tomorrow once I get a bit more time to look at/test the knight attacking through doors issue properly.

 

Do you look at map stuff too or is that just Mazey's job? I updated my map to make it less bad and haven't really heard anything

 

Thanks for following this up!

I'm kind of amazed, as that processor should definitely be able to chew through anything the water is doing, and there's nothing especially complicated going on graphics-wise.
Do you get similar lag if you open up sandbox and flood water from the top of the map?

Turn test mode on in the console (hit home, type /sv_test 1, hit enter and then home again), and then type !spawnwater in chat to spawn some water at your character.

I don't really have time for the map stuff at the moment, but I'll put that specific map on the TODO I guess.

 

A way to get more 'fps' in water maps is to change the priority order in task manager
All this means is that the CPU will do KAG's tasks first instead of others, unless they have the same priority or higher.
To do this, you go into task manager>Details
like so

Spoiler

Find KAG.exe (Like I have on the picture above) and right click it>set priority>High

Spoiler

Now test around in KAG, depending on what windows you have, you might be able to save the priority, which means everytime you launch that app it will start it in whatever priority you placed it in.
When you close Kag you will have to re-do the priority next time you re-open kag (Unless you can save it like what was said above ^)
This is just a temporary fix

 

I think it could be that.
Or water is not optimized?

I have an Intel Xeon E5-2650
It has 8 threads, 16 logical processors.
2 GHz clock speed
+ Multi-core rendering enabled (through Nvidia experience control panel)

Gives me around a good 80 FPS on moving water maps
Gives around 120 FPS on Idle water maps
No water maps- 150-250 FPS

I don't think the GPU has a huge difference on water maps, I've tried overclocking the GPU on water maps, made no difference. Plus last time I played with ann older GPU on kag with a water map I had around the same FPS

 

Water is actually one of the more aggressively optimised parts of the engine, though, like everything in KAG, the rendering could be made less cpu intensive.

More cores/processors don't much help with KAG (all the gameplay stuff is single-threaded, so more cores only serves to help mitigate the effect of other/background processes). FWIW, that processor is intended for server work (highly parallel) not desktop gaming. That clock is fairly low as well.

"Multi Core Rendering Enabled" this only matters at all once the data hits the graphics driver. KAG uses a (very dated, because the engine is very dated) immediate-mode alike rendering - not quite glBegin/glEnd, but nothing but textures and the tilemap vbos are stored permanently on the gpu. For animated sprites, this makes sense since they'll probably change almost every frame. For things that are more static, having more data "live" on the GPU would probably speed up rendering.

However:

Reminder that KAG updates at 30fps and at any point before this, you're spending any "extra" CPU time re-rendering frames rather than rendering "new" frames. This can still help with some people's low-framerate sensitivity up until the point where fps = monitor refresh rate, but it's not really a measure of how well the logic part of the game is running.

Now, if that was all gibberish to you, never mind :)

I'd be inclined to agree, but I'm not sure what on that map cause "spiking" or a sudden drop to 5fps. There's just an oscillator, toggles and a blocker.

That's good to know - do you get lag on the other mechanism TDM maps? Trying to narrow it down.

 

Thanks for that. I'l be searching to learn more about it. I'm still a noob at the details of how hardware and components work.

 

Honestly - pretty much, yes.

I'm happy to do mapcycle updates, I'd like to do more heads and I'm happy to adjust balance to the metagame, but we want to move on to other games. We've been working on KAG for coming up to 6 years - there's not much we want to do with it that has its place in KAG as it stands.

I understand that some folks would probably be happy for us to constantly be adding more content, but this isn't a sandbox game like terraria where you can just drop in another mob or item and it'll work out - anything we add to vanilla has to have its place in team-based multiplayer - CTF at the very least. That means a lot of work and a lot of grumpy players until we get the balance right, and it's not particularly rewarding or satisfying.

We haven't ruled out doing a sequel at some point, but we want to work on other ideas before then.

We mostly means myself here - MM hasn't worked much on KAG for years, though he is looking at doing some development this year.

If you want more content after 6 years - it's probably time to look to the modding scene.