Curse

AexIndex · Jan 19, 2010, 06:00 PM // 18:00

Just to clarify for the naive...

Anet does NOT care

Killed u man · Jan 19, 2010, 06:02 PM // 18:02

Edit: Sup Reverend I never said I wasn't whiny, this doesn't make my statements any less true, tough...

The only different a RBR bot with a pick-up bot would have is a pre-scripted route.

Only knowing the basics of C++/C# myself, I'm not going to use perfect code, but something like this seems very reasonable:

Let's assume the bot controls every movement and skill bar usage, BUT a player can override. If he does, the bot recalculates that override so that it won't RED ENGINE GORED ENGINE GORED ENGINE GORED ENGINE GO up on the next corner.

At all times, the bot checks if anyone is casting spit rocks/blast on him for shield. It also checks when it's adjecent to anyone else, and checks weather or not that person has used ram post-check point. (This is ALL doable through code) If he hasn't, then it gives U a warning that the person next to you could possibly KD you.

While color = (And then give the 6 possible options), do ... (The intial run from start to the first corner) Player can use Shield whenever he pleases.

When "corner is reached" -Use coordinates-, do ... (This is where all 6 routes converge)

When Moral Boost (Or check point, whatever it's called) is true, Use dash/ram (forgot which one you use after the first checkpoint). At the same time, it strafes to the right (As close to the wall as possible)

And thus the code continues... I never said it would be easy, but this and hard are 2 different things aswell. I can write the intire ru down here in pseudo-code, but it would serve no purpose. If you know and understand the posibilities of working through code, you would also know what's possible and whatsnot.

A RBRbot is one of the things that is... (The 200 lines code referred to the intital start. My best gues would be the intire bot is anywhere from 500 to 2000 lines of code)

Martin Alvito · Jan 19, 2010, 06:08 PM // 18:08

Quote:

Originally Posted by Reverend Dr

2. He's whining about an issue that should make the entire playerbase absolutely furious at Anet.

It would make me furious if it really were an issue.

I still haven't seen the dodge/pickup claim proven.

I know that botting RBR isn't plausible.

The bots aren't going to get you gold capes. Champ points, yes. But that title was already wrecked by an uglier exploit. The problem with the interrupt bots is that they are predictable. This isn't an FPS; machine-like precision can be an exploitable disadvantage in a strategy game.

*sigh* Borat, pro tip: you have to precast the Shield to get it to resolve before the opponent's KD earned from a box. Good luck programming the right delay into your bot, as your ping varies.

Killed u man · Jan 19, 2010, 06:23 PM // 18:23

Quote:

Originally Posted by Martin Alvito

*sigh* Borat, pro tip: you have to precast the Shield to get it to resolve before the opponent's KD earned from a box. Good luck programming the right delay into your bot, as your ping varies.

Your ping will always be an issue, but this is just as true for a real player. If your ping is bad, you're not going to be able to rollerbeetle, and neither is a bot... That's common sense at a really basic level.

I never claimed bots can do "superhuman" things (well, aside from detecting incomming KD's), I'm just saying they can perform top notch play every millisecond of the match.

A bot can easily keep track of who exactly had a box, and what skills that person used. (For example, if 6 people get a box, it can monitor all those 6 people, and watch if they use Dlunge/Echo/SRB, if they have, the bot can "decheck" them from possible kd's, as they already used their "box skill".

My point merely being, it's not impossible, not even implausible one could bot RBR.

FoxBat · Jan 19, 2010, 06:25 PM // 18:25

People shouldn't have to get "good enough" to beat a bot. Testing your skill against a computer is called PvE and is why heroes got nuked.

Polgara Val · Jan 19, 2010, 06:30 PM // 18:30

Again I say this to you, not disputing that a sufficient bot could be made to get a reasonably good time in RBR. It would also need to have the perfect circumstances to have the best chance of a good time.

I am simply saying it doesent stand a chance against the top 100 best times during the Canthan New year RBR.

Lets face it though, people wouldent even consider making a bot for RBR if it wasent for a fact there is a mini greased lightning for a reward.

Pol

Martin Alvito · Jan 19, 2010, 06:30 PM // 18:30

You've missed the point.

My ping varies between about 150 and 350. I'm not posting a top overall time if it's north of about 260, sure. But there's still a 100 ms gap to deal with.

Use Harden Shell too early, and the KD doesn't get used until four seconds later. Use it too late, and you get KD'd.

You have to be aware of who will hit a box AND your own latency, and time Harden Shell accordingly, to avoid losing quality runs to Blast.

Killed u man · Jan 19, 2010, 06:39 PM // 18:39

^

But your anticipation is just as good as the bot's. On top of that, I NEVER claimed this was an overnighter (a complete standalone). You, as a player, can always override the bot. If you 'feel' it's necessary to put up Shield, then do so...

Quote:

It would also need to have the perfect circumstances to have the best chance of a good time.

But so would a real player?

Is it that hard to understand? I'm not saying this bot will get you 485K time EVEN IF you have all 5 people spamming all their kd's on you. I never claimed such a thing.

As a real player, you need LUCK and SKILL to get a top 100 time.
Luck here being: Not getting random KD'd, not having ping spikes and getting good boxes.
Skill here being: Cutting corners perfectly, and using your skills perfectly. (Timing of speed boosts)

A bot can take the whole concept of SKILL away, because it WILL cut corners perfectly (On code, unlike a player who relies on visual interpretation) and it WILL use speedboosts at optimal efficiency. (You said yourself, it depends on YOUR LOCATION, so you could pinpoint the coordinates, and when the bot reaches those, it will use his skills). On top of that, as mentioned before, a bot can seriously screw back the part Luck plays in RBR. It can tell you at any given time WHICH player forms a treath, and it renders enemy spit rocks useless. (Due to super human reflexes on shields)

On top of that, ping isn't even that much of an issue. Talking to one of the creators of previously mentioned bots, he says that ALL his bots take ping into concideration. (Low ping = more delay, less ping = more delay)

Shayne Hawke · Jan 19, 2010, 06:42 PM // 18:42

If you don't have anything to say for what ANet can do to combat this "problem" besides "taking the issue seriously", then there is nothing to discuss here.

Martin Alvito · Jan 19, 2010, 06:49 PM // 18:49

So my question to you is:

If the above is true, why aren't you coding it rather than complaining about it?

If it confers that much of an advantage, go do it. Prove me wrong when I tell you it won't, and that I will thrash the bot the old fashioned way. The bot is only as good as its programmer. I will beat it for two reasons. First, I know the course better than the programmer of the bot. Second, I can defend myself more effectively than the bot. The only advantage the bot confers is the ability to do more runs than I can, since I have to sleep.

You are correct that luck and skill are required in any given run to post a top 100 time. Given a long enough series of runs, I'll get the requisite luck. Simple statistics.

HawkofStorms · Jan 19, 2010, 06:50 PM // 18:50

Quote:

Originally Posted by Killed u man

I already emailed Gaile, but my email is getting ignored due to a previous issue I had with support. (I was persistant about my innocence in a perm ban, yet they refused to hear my story)

So a guy who has a known reason to have a vendetta against a game posts unconfirmed crud trying to stir up antagonism against said game.

I'm with Martin. RBR is way more luck and skill based then people are giving it credit. You might be able to get a macro that could assist you in the timing of using Harden, but a full fledged bot would be just about impossible. How the heck would it even be able to chain the skills correctly since the skills are given out randomly from the crates? How would a bot know to echo super beetle, if it can't even tell what skills it has?

Polgara Val · Jan 19, 2010, 06:56 PM // 18:56

Quote:

Originally Posted by Martin Alvito

So my question to you is:

If the above is true, why aren't you coding it rather than complaining about it?

If it confers that much of an advantage, go do it. Prove me wrong when I tell you it won't, and that I will thrash the bot the old fashioned way. The bot is only as good as its programmer. I will beat it for two reasons. First, I know the course better than the programmer of the bot. Second, I can defend myself more effectively than the bot. The only advantage the bot confers is the ability to do more runs than I can, since I have to sleep.

You are correct that luck and skill are required in any given run to post a top 100 time. Given a long enough series of runs, I'll get the requisite luck. Simple statistics.

Basically what he said. Your coder already loses out just by not knowing the optimum route therefore will always fail getting a top 100 time. I still feel that my skill and my ability to interpret any given situation during a race gives me a vast advantage over any botter.

Pol

Killed u man · Jan 19, 2010, 06:56 PM // 18:56

Quote:

Originally Posted by Martin Alvito

So my question to you is:

If the above is true, why aren't you coding it rather than complaining about it?

If it confers that much of an advantage, go do it. Prove me wrong when I tell you it won't, and that I will thrash the bot the old fashioned way. The bot is only as good as its programmer. I will beat it for two reasons. First, I know the course better than the programmer of the bot. Second, I can defend myself more effectively than the bot. The only advantage the bot confers is the ability to do more runs than I can, since I have to sleep.

You are correct that luck and skill are required in any given run to post a top 100 time. Given a long enough series of runs, I'll get the requisite luck. Simple statistics.

Look at my first edit in the OP.

I won't be able to script it myself, because I'm lacking 3 things:

-Insufficient C++ knowledge, tough the code itself isn't that complicated, it would require me to google half my code, and it'll ask more time than I'm willing to invest into it.
-Time + Willingness: I simply am too lazy to go trough the effort of making a bot.
-A private server: Obviously to test it... (I could always ask the GWLP guys, if they still have it)

And on top of that: I don't want the Mini-rollerbeetle reward, I want the ACHIEVEMENT. I want people to say: "Hey, this guy is good at rollerbeetle racing". As I said, I'm definatly not the best, but given a sufficient amount of runs, I can definatly squeeze out a 478-480K run, which is definatly top100 worthy if you don't count in all the bots..

Krill · Jan 19, 2010, 06:58 PM // 18:58

Some players simply have RBR down to an exacting science. In exceptionally hard games such as Ghost N' Goblins you have to memorize every event and execute perfectly, the same is true with RBR to get a perfect run. Once you have the method refined it's just a matter of doing enough runs for the correct powerup sequence and syncing with other players that are going for top times, not to hose each other with KD's and cutoffs.

With all the hysteria over real botting lately I don't think it's really appropriate to be stirring the pot more.

Martin Alvito · Jan 19, 2010, 07:00 PM // 19:00

Which would be true if we botted.

But we don't. It isn't necessary, and as you point out it would be very time consuming to design the bot. I wouldn't want to debug that code, and I'm totally unwilling to waste time during an event. The opportunity cost is too high. Instead of debugging a bot, I could be learning something new that enables me to shave a half second off my runs.

Since Krill brought up syncing: it's dumb. Really. Only one player in a match can have a good run due to rubberbanding. You don't want to be competing with other top players for the same line...unless you happen to have the best ping. But if that's the case, they don't want to be in your runs. You don't want to be in a tit-for-tat situation where you cover each other's runs, either. You can improve on that by playing alone if you don't suck at defending yourself.

I learned this in the first year when they forgot to close down an event. I ended up playing with five other top players for hours. No one could post a top 100 time, even with Echo-SRB. The lag was that fierce.

pumpkin pie · Jan 19, 2010, 07:02 PM // 19:02

So bot in PvP (RBR) do exist. as many of you claim to be able to beat those bot. and you do know it exist, how else you gonna know you are better then the bot, right? and how the programmer of the bot does not know the route any better then you guys do?

This is going to be the same as sync Killed U Man, its a waste of time, ArenaNet do not care about things like this. things that they would much rather not invest time in.

If you send them more emails, and make a lot of noise on the wiki, they will put your post in the still being look at section and it will in times disappear from the face of wiki.

they will promise to look at it. trust me , but weather they do look at the problem is another issue.

Polgara Val · Jan 19, 2010, 07:04 PM // 19:04

Quote:

Originally Posted by Krill

Some players simply have RBR down to an exacting science. In exceptionally hard games such as Ghost N' Goblins you have to memorize every event and execute perfectly, the same is true with RBR to get a perfect run. Once you have the method refined it's just a matter of doing enough runs for the correct powerup sequence and syncing with other players that are going for top times, not to hose each other with KD's and cutoffs.

With all the hysteria over real botting lately I don't think it's really appropriate to be stirring the pot more.

I find syncing in RBR doesent work well enough for the effort of trying to do that with friends and no real point, since most of the top players actually use the same route at the start and more often then not just get in each others way.

Pol

Shayne Hawke · Jan 19, 2010, 07:05 PM // 19:05

Quote:

Originally Posted by HawkofStorms

How the heck would it even be able to chain the skills correctly since the skills are given out randomly from the crates? How would a bot know to echo super beetle, if it can't even tell what skills it has?

The better bots won't be looking to make the most efficient use of what it's given. It will simply be hoping that it gets an Echo and Super out of its first few boxes and then gets to run with it. Trying to make a bot that puts out the absolute best usage of all skills considered at all times would be truly impossible.

zolidamasta · Jan 19, 2010, 07:10 PM // 19:10

lol...guildwars...lol...botdrama.
Gl in 2035 when they ll release gw2...ofc with bots.

Order gw2 now and you ll get free interruptbot in 2035...srs bsns model

Killed u man · Jan 19, 2010, 07:11 PM // 19:11

Who said anything about syncing in the first place?

And I'm pretty sure even a child would realize that when you have 5 friends on your side, it'll make your run a whole lot easier. I'm not saying syncing is effective, or the opposite, I'm just saying that syncing will definatly not decrease your chances. I actually synced with a couple of guildies last year in Korean/Chinese districts. We often got 4-5/6 people teamed up in the same run. We would then see whoever got the best "first box", and whoever did was given carte blanch for the rest of the run. (The remaining ones would just kd the non-guildies so they couldn't KD your guildie)

But this still requires you to be good at RBR.

Secondly, I never said anything about wasting time during the event. Anyone who'se capable of packet sniffing/modyfing and writing complex bots who inject themselves in the process will also be able to create private servers.

GWLP, for example, had those to test on. The interrupt bot, for example, was extensively tested on private servers. To see wether or not it would be detectable, how well it can peform, if this, if that...