|
|
Post by AdmiralObvious on Jul 1, 2017 21:28:34 GMT
Hey Admiral Obvious would you be opposed to me writing a custom AI dogma and using it instead of the pre-sets? The point of this particular contest was to be purely non modded, no modded materials, no "intelligent" AI. I forsee in the future that this'll change, most likely. I'm not sure for certian though. Especially given that different versioning of mods can do weird stuff, like that material modded origonally for railguns having a yield which just happened to be 100 times better than it should have been. Edit: Also, my work schedule has been stuffed full again, so expect fewer uploads. However, the next few videos will be of the second bracket of the tourney for final elimination. Whoever wins it will be given 3rd place, and fight against the primary champion at the end, see if they can win or not. If they do, they get the top spot. |
|
|
|
Post by AdmiralObvious on Jul 5, 2017 21:34:04 GMT
First round elimination bracket. "Buy" bracket. One of these two contestants will face off against the winner of the previous 2 brackets for their final elimination due to being KO'd at the semi finals. |
|
|
|
Post by treptoplax on Jul 6, 2017 19:29:42 GMT
Ok, so assuming a standard double-elimination cross-feed (to prevent early rematches) that'll be lennson/Dhan and nekomata/treptoplax.
My speculation, just for fun:
NDI Wildcat (nekomata) vs. Patrol Carrier C (treptoplax) is interesting; those are superfically the two most similar designs: both fast, compact, skinny carriers with secondary point-defense. Wildcat has flak? missiles, lasers, and solid armor; Patrol Carrier has gundrones, guns, and stuffed armor. I suspect the NDI missiles have the highest acceleration of anything in this match and that'll be the decisive advantage, but nothing would surprise me here - if the missiles end up engaging the drones that'll go badly for them. And I don't think either the lasers or the 50mm guns will be much use... at least this one may not be quite as laggy. Or maybe it will: I'm pretty sure the NDI Wildcat beats the Patrol Carrier's drones in both acceleration and dV, so we could end up with so weird long battles...
FF-HVC (lennson) vs One Trick Pony (Dhan). I think the lennson drones are a real threat to OTP, and the ship itself has enough armor to potentially survive an exchange with OTP while getting off a fatal round of railgun fire. I'm picking FF-HVC for the upset!
|
|
|
|
Post by nekomata on Jul 7, 2017 9:27:36 GMT
It'll be interesting. My biggest weaknesses are the missiles' vulnerability to lasers and the AI's tendency to not dodge gunfire from drones. Assuming your drones fire at the edge of laser range, then the shells will eventually score good hits, depending on how fast AdmiralObvious ends the engagement. The missiles aren't especially good at defeating drones it seems, and if the drones get into tactical range they stand a good chance of shooting though my armor, which has consistently proven useless. So yeah, it'll be interesting.
|
|
|
|
Post by Argopeilacos on Jul 7, 2017 9:38:27 GMT
I made a tournament spreadsheet from a "double elimination 8 teams bracket" template. It puts lennson vs treptoplax and nekomata vs Dhan next, but feel free to adjust/update it if needed. |
|
|
|
Post by treptoplax on Jul 7, 2017 14:52:55 GMT
I made a tournament spreadsheet from a "double elimination 8 teams bracket" template. It puts lennson vs treptoplax and nekomata vs Dhan next, but feel free to adjust/update it if needed. Huh. Well, either way. The opposite way is pretty straightforward; lennson's lasers take care of the missiles and that's that, and treptoplax/Dhan is likely an easy win for One Trick Pony... although I'm not sure at all if it can hit those 10cm diameter drones at range. Best-case I might get a re-run of the first-round lennson/treptoplax match and have a drone take out OTP kinetically. Edit: Oh, wait, I misread. lennson/treptoplax is a re-run of round one, likely with the same results. (Given the right starting position, I might win one with a drone kill and survive the outstanding drones/missiles, but not likely two out of three). NDI Wildcat vs One Trick Pony: well, we haven't seen needle gun vs. missiles yet so that's interesting. Possibly the missiles have a chance? And with the insane acceleration of Wildcat (>1G wet!) it might actually be able to dodge even the railgun fire from OTP if the AI were smart enough to run broadside... but trying to turn than nose-mounted laser toward the opponent would be fatal and the more likely AI move. I dunno. |
|
|
|
Post by AdmiralObvious on Jul 7, 2017 17:03:48 GMT
It'll be interesting. My biggest weaknesses are the missiles' vulnerability to lasers and the AI's tendency to not dodge gunfire from drones. Assuming your drones fire at the edge of laser range, then the shells will eventually score good hits, depending on how fast AdmiralObvious ends the engagement. The missiles aren't especially good at defeating drones it seems, and if the drones get into tactical range they stand a good chance of shooting though my armor, which has consistently proven useless. So yeah, it'll be interesting. I only try to end the engagements when one side officially "loses" it's main capital ship. If there's a chance for MAD, I try and run the game as long as possible, but if the result is MAD consistently, I attribute the win to whoever lasts longest. I'll also stop the recording at least if the frame rate dips below .5 FPS, nobody wants to watch that, and it happens a lot when the cap ships engage with spare drones. I made a tournament spreadsheet from a "double elimination 8 teams bracket" template. It puts lennson vs treptoplax and nekomata vs Dhan next, but feel free to adjust/update it if needed. That makes it a lot easier to keep track of, thanks for making it. |
|
|
|
Post by Argopeilacos on Jul 7, 2017 22:43:03 GMT
I made a tournament spreadsheet from a "double elimination 8 teams bracket" template. It puts lennson vs treptoplax and nekomata vs Dhan next, but feel free to adjust/update it if needed. lennson/treptoplax is a re-run of round one, likely with the same results. I didn't notice that, but it's not really surprising as I swapped the formula for that round to match the state of the tournament (it would have been fgdfgfthgr vs. Hicks in 7 and treptoplax vs. Dhan in 8 otherwise). Edit: fixed that, and added the formulas that got lost in the copy/paste to Google docs. |
|
|
|
Post by AdmiralObvious on Jul 12, 2017 19:07:39 GMT
|
|
|
|
Post by treptoplax on Jul 13, 2017 2:44:36 GMT
Lennson/Dhan: Called it! I'm increasingly convinced that small railgun drones are a strong counter to glass cannons.
treptoplax/nekomata: Huh. Surprised at short range of initial drone/Wildcat encounter, but I'm really happy with performance of those little drones. AI obviously doesn't handle gun range or intercepts the way you'ld want but they're still plenty deadly. Kinda surprised retaliatory missile strike wasn't successful. Not sure if missiles were coming in too fast to make terminal adjustments, or if carrier didn't turn on main engine for some reason (in that case the drones with it have a higher heat signature than it does (!) and they may have acted as successful flares).
|
|
|
|
Post by AdmiralObvious on Jul 13, 2017 3:16:16 GMT
Lennson/Dhan: Called it! I'm increasingly convinced that small railgun drones are a strong counter to glass cannons. treptoplax/nekomata: Huh. Surprised at short range of initial drone/Wildcat encounter, but I'm really happy with performance of those little drones. AI obviously doesn't handle gun range or intercepts the way you'ld want but they're still plenty deadly. Kinda surprised retaliatory missile strike wasn't successful. Not sure if missiles were coming in too fast to make terminal adjustments, or if carrier didn't turn on main engine for some reason (in that case the drones with it have a higher heat signature than it does (!) and they may have acted as successful flares). Nekomata's missiles always tended to miss the mark when they detonate for reasons I couldn't quite figure out. Sometimes they actually hit the mark, most of the time they totally miss. The few times they do hit, they are devastating. I think it may have something to do with the guidance rules, or sheer lack of AI ability in high acceleration environment. |
|
|
|
Post by Dhan on Jul 13, 2017 4:58:48 GMT
Lennson/Dhan: Called it! I'm increasingly convinced that small railgun drones are a strong counter to glass cannons. treptoplax/nekomata: Huh. Surprised at short range of initial drone/Wildcat encounter, but I'm really happy with performance of those little drones. AI obviously doesn't handle gun range or intercepts the way you'ld want but they're still plenty deadly. Kinda surprised retaliatory missile strike wasn't successful. Not sure if missiles were coming in too fast to make terminal adjustments, or if carrier didn't turn on main engine for some reason (in that case the drones with it have a higher heat signature than it does (!) and they may have acted as successful flares). Only against projectile systems. Since laser glass cannons get to arbitrarily set engagement range, they will probably beat drones that aren't just a ball of rubber or aramid fiber. |
|
|
|
Post by Argopeilacos on Jul 13, 2017 7:36:39 GMT
|
|
|
|
Post by nekomata on Jul 13, 2017 9:40:19 GMT
Huh, I think the only time I've seen the missiles act like that was against Vesta Overkill, so it might very well have been the drones spoofing the missiles away. Another possibility is excessive acceleration causing problems with terminal homing. The Harpoon is pulling in excess of 220g at burnout... Still, I'm pretty happy with how she did overall.
|
|
|
|
Post by treptoplax on Jul 13, 2017 16:37:52 GMT
Lennson/Dhan: Called it! I'm increasingly convinced that small railgun drones are a strong counter to glass cannons. treptoplax/nekomata: Huh. Surprised at short range of initial drone/Wildcat encounter, but I'm really happy with performance of those little drones. AI obviously doesn't handle gun range or intercepts the way you'ld want but they're still plenty deadly. Kinda surprised retaliatory missile strike wasn't successful. Not sure if missiles were coming in too fast to make terminal adjustments, or if carrier didn't turn on main engine for some reason (in that case the drones with it have a higher heat signature than it does (!) and they may have acted as successful flares). Only against projectile systems. Since laser glass cannons get to arbitrarily set engagement range, they will probably beat drones that aren't just a ball of rubber or aramid fiber. It's an interesting question discussed on some other threads. Due to wobbling/accuracy cap, I find minimally sized drones are quite laser-resistant at 1Mm and can often close inside 500Km even with modest numbers. Widebeam lasers might solve this, but will decrease intensity. Scaling lasers up makes for a bigger target; I'd guess apophys standard single-laser drones are ultimately a better bet than the (admittedly terrifying) cheaper multilaser ships he's got here because they're so much more difficult to target (a near-miss won't hit the adjacent drone). All topics for future experiments! |
|
