Spring RTS Engine

edgarwaen you know as well as i do that change will never be in AA...right? ;.;

Min3mat wrote:edgarwaen you know as well as i do that change will never be in AA...right? ;.;

Yeah. But even without it, that thing could now have a role.

I didn't get owned, mat, I'm just too lazy to design stuff from scratch right now and there's nothing suitable that I could find.

May have found the Maverick problem. Since Spring operates in rather.... interesting ways.... compared with OTA, tolerance needs to be rather high. Maverick's 2.1 aiming tolerance was 500, while the engine defaults to 3000 (which is a good value for pretty much everything)... So in the next version I'll be eliminating any tolerance value that's below 3000. This should eliminate all jitteryness some units display.

Ehhh... maybe.

Just played a game and witnessed the maverick firing bug. I got pwned, and while speccing witnessed a maverick sitting there refusing to fire at a moho metal extractor, it moved away a bit and then fired. A con bot then started to reclaim it and it sat there, its owner obviously paying attention to something else.

Looks like mav's firing animation is just generally fubar in Spring....

~~~~

Ah, ripped off XTA's. Much better. Still jitters, but at least the shots come from the proper barrels now. I'll see what I can do about fixing the jittering now...

~~~~

Fixed!

Caydr, I haven't had any problems at all with aiming scripts never reaching "go" in Spring, except for one thing- the script has to get to aiming and then stay there long enough for the unit to not cycle back to pre-aiming conditions. The way I handled that for the Holder in NanoBlobs is an (exaggerated) version of what you want- basically, a pretty darn long sleep delay before trying to return to start. Also, I've been finding that doing a series "now" steps and using Spring's interpolate, until I've reached aiming animations, is better than using "turn", for whatever reasons (mainly, because Spring doesn't handle "turn" very well, something I was going to bring up with Fnordia before he disappeared into the aether).

Changes so far:

2.1 --> 2.11

Many weapon tolerance values increased, this should result in better
targeting
Fido now defaults to gauss mode
Maverick jitteryness when firing at close targets fixed or at least
greatly improved

If I was to release this as a patch, it would be only 155 kb. The maverick thing is bugging me now, so I might post this by the end of the week probably... If anyone has anything to add to it by then, be sure to post.

Hmm.. It seems shockwaves are only used when weapons impact the ground, not when they hit a unit, could this have to do with their use in air?

Not sure, but for some reason when a unit is set to patrol and is reclaiming or helping build, when that action is completed and there is nothing more for it to do, it will sit there and the start/stop movement sound will play (I dont believe the model actually shows it moving, it just sounds like it does). This gets rather annoying after a while, and you have to search out the units and move them manually to stop it.

The new explosions look sweet. Had a big land battle on Xantheterra last night and the new exposions, combined with the ground flashes really make it seem more destructive!

well, the moving circle from center of explosion to border shouldn't have a linear speed. It should be fastest in middle and slow down and fade out to border. Currently it looks .. hmm .. too artificial.

Anything else nice.

Rayden wrote:well, the moving circle from center of explosion to border shouldn't have a linear speed. It should be fastest in middle and slow down and fade out to border. Currently it looks .. hmm .. too artificial.

Anything else nice.

eh? what do you mean?

When constructed by swarms of construction planes

(not sure if it was reported before)

2k4 wrote:When constructed by swarms of construction planes

(not sure if it was reported before)

I think you may have forgotten something.

Caydr wrote:

2k4 wrote:When constructed by swarms of construction planes

(not sure if it was reported before)

I think you may have forgotten something.

i didn't want to say anthing in case i looked the fool

2k4 wrote:When constructed by swarms of construction planes

(not sure if it was reported before)

Yes. Logically.

The new Explosions look kickass. Any way to give a huge mushroomcloud of death to the nuke?

Caydr wrote:

2k4 wrote:When constructed by swarms of construction planes

(not sure if it was reported before)

I think you may have forgotten something.

Look up at the top of his post. The full thing is:

Displacing of nanoturrets when constructed by swarms of construction planes.

unpossible wrote:

Rayden wrote:well, the moving circle from center of explosion to border shouldn't have a linear speed. It should be fastest in middle and slow down and fade out to border. Currently it looks .. hmm .. too artificial.

Anything else nice.

eh? what do you mean?

Ok after watching 20 guardian shots in slowmotion and pause ... i guess i know how to describe better ...

- the light effect around the explosion has a constant radius and intensity independent of the shock front ... i think this is not optimal and should be adjusted to the size of the shockfront
- the shockfront itself maybe a bit more stringily to inner side.
- the broadening speed of the shockwave should slow down in a exponential function
- and finally the fading out the shockwave should be 2 or 3 times slower in the end or at least start earlier

... i think this would make a much more better effect.

I would experiment by myself but unfortunally i have no idea how such an effect is done.

ok, the constant radius threw me - you can't have a shock wave travelling outwards with a constant radius