How hard is it to program a carrier ratting bot?


#1

I am a game programmer myself. I mainly program in C# but i am also learning a bit of C++. I was wondering. How hard is it and how long do you think it takes to make a carrier ratting bot? The bot wouldnt need to be able to multibox. I would like to do it by modifying A-Bot. Is this a good idea?

I suppose the hardest part is the actual fighter combat itself?

I also seem to note that i dont see alot of talking about carrier bots. Any reason for this? Is it that hard? (Biggest problem seems to be to not lose fighters?)


#2

not hard, I do have already working code.

but it’s not worth all the logistics needed to keep that shit safe and ticks are not that greater than good old VNI.
simply isk/risk ratio is kind of not worth the effort.

supercaps, different song, but logistics is even more pain.


#3

Hello Caesar and welcome in our little community,
I cannot fly an carrier, but if you read carefully the entire forum, you can see the subject was on the table, not on dedicated threads, but inside other.
Even so, i can suppose some of bigger and more knowledgeable guys created a bot based on A-bot for carriers, but …you wont see that officially. the reason is simple, to keep an balance on game ( a-bot is multibox with virtual ) . Also, the the problem it is indeed the fighters: losing them, manipulate them, actions, orbits, transfer them etc.
The ratting in carrier is different from an simple orbit, release drones and send them to the target. You have to take in account the timings, managing each fighter, their damages, managing your modules, etc.
first you have to know at best how you rat in carrier, bigger experience is better because you can generalise situations .
Even if it take 1 month , or 2 , if you can make him, better keep him for yourself, my advice.

I have another suggestion, about A-bot. This is a free bot, not so developed, and also not so maintained. if you look a little in history, he was updated after 2 months to keep up with the lasts modifications from game. So at the next change ingame are great chances this one to be forgeted about 3-6 months. Also, the framework is really maintained and updated, so in short time the A-bot will be a piece of ancients times, because the framework dll used in A-bot will not be compatible. It the situation of optimat. a real nasty bot, but … to update him you must update almost 30% of his code.
So in conclusion if you make your own bot make him in framework. Is more fiable more efficient and updated.
also i can tell you, since now Im working at my own script similar and better than A-bot: my combat.cs from abot has 700 lines.the anomaly file has 400, and if I add the rest of files i come at 2500 lines. The script in framework has … 900 , he take asteroids, travel, take the loot, salvage, unload stuffs and more than that. if i add him 300 lines he can mine, and take the anomaly. Not bad, and also Im not programmer :slight_smile:


#4

it’s really not about how many lines code has @kaboonus… and updating A-bot for possible new breakouts from eve will not be that much harsh, as it was done before with help of @Viir

framework is nice, but for simple targets it’s always better have specialized tools not big robus framework which “can do anything” even you are using 1% of it.


#5

IfIm not mistake, the difference is in the api offered by sandy , because the A-bot has same possibilities. so, its the same 1% used.


#6

just reacting on that… :wink:


#7

is my opinion,a-bot will die , like optimat.


#8

Hmm okay. I still find it weird that there is barely any talk of carrier botting. Only real reason i want to make one is the fact i would like the challenge and to fund my ingame activities but you would think there is a lot of demand for such a bot.

I do wonder tho, is processing speed going to be a big obstacle for the fighters? If you would use a super you would need to do alot of actions for the fighters in a very small timeframe. Do you think multithreading is a must?


#9

From all discutions around , no. If I’m not mistaken, the time between 2 clicks is ~200ms. I think the biggest obstacle is the optimisation of code,

LAter: Down you can see the real experience and a good response


#10

It has to be multi-threaded so that one thread gets the memory snapshot while the other does the mouse work, otherwise fighters die while waiting for the snapshot process to complete. I still lose 1 fighter per hour.

I am a programmer and I have to tell you that the logic is super complex, and took several iterations and once starting from scratch (after the post above). The combat logic to manage fighters and targets like a person does is 2k LOC (which doesn’t touch the Eve GUI), and all of that code is lean code broken out into functions so there’s no duplicate code in different functions. The rest is probably 5k-7k LOC (lean).

Does it work now, yes.
Was the time worth it? NO… (But it does make 2-3 times more than a subcap bot. Once every few days Sanderling can’t find a memory element and my bot might die.) My time could have been better spent on making a subcap bot farm.

That being said, if anyone has also gotten a carrier bot working and making at least 100m/hr… Message me and we can swap notes or something.


#11

that’s exactly why it’s not worth to do it. 100m/h is almost the same as in subcap and without harsh :slight_smile:
no need to be so greedy


#12

Well, mine can make more than 100 per hour. That’s just where I set the bar.


#13

If you make it compatible with a super its 300m/h tho


#14

That’s how I designed it. It works fine for any number of fighters. However, Keepstars aren’t cheap…
(You have no idea how gratifying it is to watch your bot use 5 fighters (and modules of course) to kill neutrals who warp to your Fortizar.)


#16

Any luck with this? I too am looking to obtain or create with help something like this for a Supercarrier…