Jump to content

BDAc Challenges: GDJ's Top Gun Extreme Tournament

GDJ

By GDJ
in KSP1 Challenges & Mission ideas

 Share

Recommended Posts

GDJ

GDJ

Posted
  • Author
Posted
32 minutes ago, SuicidalInsanity said:

Are there other factors at play as well governing that behavior? I've had a few planes that have decided to start climbing to 18000m or so forcing their opponent to follow, but they were all flying at no more than 200-250m/s or so.

On a related note, has anyone else run into situations where the AI decides that cowardice is the better part of discretion, and starts flying in a straight line away from the combat area until it reaches physics distance and despawns?

I've run into both, especially with faster aircraft. Propeller driven planes don't seem to exhibit this behaviour as badly.

Link to comment
Share on other sites
SpannerMonkey(smce)

SpannerMonkey(smce)

Posted
Posted
12 minutes ago, SuicidalInsanity said:

Are there other factors at play as well governing that behavior? I've had a few planes that have decided to start climbing to 18000m or so forcing their opponent to follow, but they were all flying at no more than 200-250m/s or so.

On a related note, has anyone else run into situations where the AI decides that cowardice is the better part of discretion, and starts flying in a straight line away from the combat area until it reaches physics distance and despawns?

ALL the above are known odd behaviors of the AI,  climbing forever, or until it stalls.  Permanently evading/running away, and you forgot the other one, diving into the ground :). They don't happen often and when they do,  I've usually found something in the log to indicate that something else had upset the game first and so scrambled the AI.  I've seen all of them so often it's not even a problem, i just revert, check the log, and go again(but then i can put up with a lot of snags before i get mad)  :)     I am wondering if a maximum height slider wouldn't stop the climb issue right away.

RE the rest :)  I have what i think is an OK craft  , I've not built anything else that can beat it, and nothing I've downloaded has beat it, so I'm saying I'm good to go,  it'll probably die double quick to an expert aircraft builders craft but as long as it's not a one shot wonder I'll be happy..  How do you want the craft submitted? KerbalX? PM or?

 

 

Link to comment
Share on other sites
GDJ

GDJ

Posted
  • Author
Posted
16 minutes ago, SpannerMonkey(smce) said:

  How do you want the craft submitted? KerbalX? PM or?

I'm not picky. KerbalX, Dropbox, PM with a working link, etc.

Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)
5 hours ago, drtricky said:

Besides manually running the plane until fuel exhaustion, are there other ways that can determine whether a plane has adequate fuel or not?

ISP*thrust/3600/9.8 = fuel time in seconds. Must be over 900.

3 hours ago, SuicidalInsanity said:

Are there other factors at play as well governing that behavior? I've had a few planes that have decided to start climbing to 18000m or so forcing their opponent to follow, but they were all flying at no more than 200-250m/s or so.

On a related note, has anyone else run into situations where the AI decides that cowardice is the better part of discretion, and starts flying in a straight line away from the combat area until it reaches physics distance and despawns?

Yep. I think it is trying to gain an energy advantage or something. I've seen a dogfight end when both planes, having reached the 800 m/s absolute limit, chased until the front one ran out of fuel and was promptly shredded in a hail of bullets.

More often though, the planes do other stupid things. I just had a battel between my light fighter (1 panther and 8.1 tonnes loaded) and my medium fighter (2 panthers and 14.6 tonnes loaded) where the medium fighter killed the light fighter in a supersonic head-on at a distance of about 50 meters then immediately smashed into it, turning both planes into dust except for the cabin and fuel of the medium fighter.

The overwhelming majority of my 1v1 matches are being decided in an initial, brutal, mach 1 head-on. As such, I'd suggest doing full-on 3v3 matches for testing, else it won't yield reliable data.

1 hour ago, tsoj said:

is it really necessary to have fuel for 15 min or is it just a recommendation?

I think it's a requirement.

Edited by Pds314
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)
1 hour ago, drtricky said:

After a mere 20 minutes of editing and testing, I have a candidate ready!

https://kerbalx.com/drtricky/The-Dorito--competition5

"ONLY" 3 vulcans.

Wow. I thought my medium fighter was heavily armed and it's got 2 Vulcans. I personally see guns as a diminishing returns thing. Even with proper gun convergence and everything, my 10x GAU-8 "fighter" with tweakscaled whiplashes didn't have much better accurate range than anything else. Yeah, it never ever sparks, but at some point, 600+ rounds per second only helps if they are on target, and they can't be on target if you're aiming at moving specks 4 kilometers away. As a point of comparison, it was able to be reliably killed by my medium fighter from 2 kilometers out.

sigh... I just had a 1v1 where they clipped wings at supersonic speeds... The match was decided by the fact that my medium fighter can fly out one wing and with the cockpit bent out of position and my light fighter cannot.

Edited by Pds314
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)
25 minutes ago, drtricky said:

I like to arm my creations with a minimum of 2 Goalkeeper CIWS's. 3 vulcans is nothing to me.

Wow. LOL. Welp, looks like mine will need to win the whole shot-placement thing, because they certainly shall not have volume of fire.

On a related note, just built a heavy fighter. 21 tonnes fueled and unarmed. Utilizes tweakscaled 1.5m panthers as engines. It can do 750 m/s at sealevel without afterburners. 1030 at sealevel with. Its top speed is so ludicrous that it needed structural reinforcement to prevent the cockpit and front strakes from lifting off of the plane.

Did another run on a mild descent from the mountains to the KSC. Melted the engines.

Edited by Pds314
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)
23 minutes ago, GDJ said:

This is going to be interesting. :)

Oh, and you guys and gals aren't stuck with Jets.

SXT.

:)

I may build a prop, but they don't have good top speed or energy generation. Probably just gonna design something that excels at turnfighting and see if I can get my jets to kill them in droves. Also, most of the cockpits are a bit heavy for props. My fake prop has real difficulty against my medium fighter. I.e. if the jet is in front of it, jet just uses supermaneuverability or better TWR. If behind it, the prop pretty much cannot shake the jet. Props would be massively overtiered competing with jets, just like they were when the occasional Korean war fighter met a prop.

 

(Then again, there was a North Korean Po-2 biplane that convinced an American F86 Sabre to stall out near the ground and crash, but we've got a 750 m altitude floor, so unless people's planes stall unrecoverable, that's not likely to happen.)

Edited by Pds314
Link to comment
Share on other sites
GDJ

GDJ

Posted
  • Author
Posted
55 minutes ago, ScriptKitt3h said:

Are your planes allowed to be built utilizing Panthers for your propulsion?

I might have an entry ready.

ANY engine within the allowed mods or SQUAD are allowed...as long as there isn't more than two engines.

Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted

A point about tests: you need to do more of them than you think to truly determine which is the superior fighter.

Let's say I do ten tests?

How many wins do we need out of 10 to be >95% sure that the winning fighter isn't winning by dumb luck?

Answer: 9.

20 tests? Win 15.

50 tests? Win 32.

That's for "this fighter is clearly better."

For "this fighter is clearly much better and will win >67% of combats, we're 95% sure" you need even more tests. 18/20. 40/50. etc.

It might honestly just be better to watch the fighters fight and see what flaws they have and correct them. Taking a rigorous scientific approach that only cares about team wins vs. team losses looks to take unbelievable amounts of effort.

Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)

Interesting cheap tactic I've noticed is to make sure that the cockpit and the BDA stuff are right near each other, that way, if the plane is shot to pieces, the enemy will be going after a cockpit as it falls 10 kilometers or so. Plane falls for thirty seconds and enemies chase it. Meanwhile, you retain the advantage and can get on their tail.

Edited by Pds314
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)

Why don't they teach you the maneuver where you ascend to 8500 meters, fly 45 kilometers away from your ally who's out of ammo against an opponent with plenty of it as fast as possible and dart back into the fight just in time to let them die in combat flight sims?

Sigh.. in the battle between out-of-ammo and missing an entire wing and some of the fuselage idling blindly at 8000 meters...

Fuel is all that matters.

Edited by Pds314
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted (edited)

Who wins if both planes run out of fuel, but one softlands in the ocean and the other, previously stripped of a wing, crashes 20 seconds later? The plane that hit the water second had lost fuel just before the first one landed.

Edited by Pds314
Link to comment
Share on other sites
GDJ

GDJ

Posted
  • Author
Posted (edited)
4 hours ago, Pds314 said:

Who wins if both planes run out of fuel, but one softlands in the ocean and the other, previously stripped of a wing, crashes 20 seconds later? The plane that hit the water second had lost fuel just before the first one landed.

The plane that runs out of fuel first will be considered the loser of the round. Where they land is irrelevant. 
The ONLY exception is that the "glider" manages to destroy the opponent by sheer dumb luck.

 

2 hours ago, notsodeadjeb said:

Can we use three engines if it is VTOL? 

No. Two engines max. There are parts within the approved mods that can take care of this.

To everybody:
REVIEW the RULES.
-No missiles. Guns only.
-Two engines max.
-First plane to run out of fuel is the loser.

 

Edited by GDJ
Link to comment
Share on other sites
Pds314

Pds314

Posted
Posted
3 minutes ago, DoctorDavinci said:

^^^^^This

And although the above is true, as with all things, there are exceptions to the rule :rolleyes:

I'll just leave this here ........

 

Yeah. I seem to remember Azimech or someone made stock helicopters fight and the AI was not great, but sort of halfway sane.

Link to comment
Share on other sites
Guest
This topic is now closed to further replies.
 Share
Go to topic listing
×
×
  • Create New...