Curse

Killed u man · Feb 16, 2010, 12:23 AM // 00:23

First off, I would like to say "Thank You" to Anet for the rather good state of servers this Canthan New Year, so again thanks for that. (I could actually play at an average ping of 100)

That said, let's move on to the main reason I'm posting this thread, which is to understand how exactly Rollerbeetle Racing works, and more precise how one can get a top 100 score. I've decided to do a "post-analysis" (When the event is over), so I would have a full set of data, aswell as the time for this thread to sink in, (The topic isn't such an issue anymore) whilst still keeping it somewhat related to the event.

I started gathering this data 2 hours after the event started, or atleast when I started playing. I tried to focus on purely entering from outposts, and restricting myself from getting consecutive wins.

These are my data gather over the past weekend, having playing over a total of 1000 games, and thus supplying a reasonable amount of numbers to work with:

Quote:

Total Runs: 1029 (Give or take 50 I missed recording)

Teal: 187
Purple: 207
Red: 158
Blue: 139
Yellow: 163
Green: 175

No KD during first minute: 274
Flawless starts: 143
Flawless starts with only Echo: 96
Flawless starts with only SRB: 20
Flawless starts with Echo RB: 3
Flawless starts with 2 SRB (One from Ox's): 4
Echo RB: 34

Some times reached:

475K: Various times
478K: 3 times (SRB from first 3, and then another one at ox's cage)
479K: 2 times(With Echo SRB and a Ram/Shield start)
480K: 1 time (With Echo SRB and flawless start)

Rollerbeetle Color

So for starters, we have the rollerbeetle color. Everyone with the least bit of experience will agree that Red, Blue, Yellow and to a lesser extend Green are the best beetles. (Green can get 60-70% RRPM from it's cage) But just like you can win a formula 1 race with a diesel, you can also get a top 100 score with Purple, Green and Teal, it will just require ALOT of luck. (You're hoping that the players behind the Blue, Red and Yellow beetle haven't got a clue what they're doing)
And these colors aren't necessarily alot slower, but rather just a bit slower so you'll always get behind a Red, Blue or Yellow player, resulting in you getting blocked most of the times.

My numbers roughly follow these percentages:

Quote:

Teal: 18.2%
Purple: 20.1%
Red: 15.4%
Blue: 13.5%
Yellow: 15.8%
Green: 17%

This is all close to the numeric average 16.7%, and thus I think it's safe to conclude 2 things:

-> The Rollerbeetle you get IS equally random, with apparently a minor variance
-> An average of 50% of your runs will result in a "possible" top 100 score (Not incl Green)
-> The longer the event lasts, the more important your color becomes, as the overal top 100 times will get higher, and it will become more and more unlikely you will get a top score with the "unlucky" colors.

Wether or not getting consecutive wins give you a bigger chance of getting a lucky color is something I dind't test for. It seems unlikely, however, as Anet probably isn't aware of the difference in color, aswell as the fact that entering from an outpost comes close to what you would assume to be the average percentages.

Map, experience and knowledge

The map in Rollerbeetle Arena is always the same. Therefore one can utilize the same strategy, over and over again in order to get a top time. It is often stated that you need a lot of skill to get a top 100 score, but I disagree on that. For me, skill is something active, it's using your abilities and knowledge to manipulate something to achieve positive effects. I agree that you have to KNOW and UNDERSTAND how Rollerbeetle Racing works, and you definatly need to know when to use what skill at what time.

GvG (Old school for arguement's sake) is about skill. You go in with a certain experience, knowledge and "know-how" of the game, and you hope those attributes will overcome your opponent's.
Concidering you often don't know what map you'll play on (outside AT's), aswell as what build and players you'll face, the playing field is dynamic. You can not simply go in with a certain mindset: "This is how we're going to play the game and win", and expect to win every match. (And if you do, you'll either get stomped, or your build is bound to get nerfed (All the ex-spike builds, etc).

Rollerbeetle Racing on the other hand isn't dynamic. It's static. The playing field does not change, the build does not change, nothing changes, aside from the player's names. If you know the strategy to get a 480K+ time, you can effectivly get a 480K+ time.
The true nature of the problem here lies in the fact that Rollerbeetle racing is unforgiving. If you get body blocked or KD'd (Simply put: When you have an unexpected variable) at any point during your race, you simply can't overcome that variable with skill, because you simply can't make up for lost time.

So unlike GvG, Rollerbeetle Racing is about following one playstyle as close as you possibly can, and if you do get kocked out of it, giving up is the only option you have.

Any person who has the ability to use his 2 hands will be able to learn how to Rollerbeetle merely by watching a video of a player playing it. Good luck trying to learn how to Warrior in GvG by watching Chop Chop play, hence point proven.

Boxes and skill usage

The assumed best technique to start a rollerbeetle race is by Dashing when your RRPM peaks out naturally as you leave the gate (about 3-4 seconds in), and then saving your Ram as you hit the upwards corner to the first gate. (As seen in the youtube video)

The more accurate explantion here is rather obvious, yet dodgy due to the "skill anomalies" that exist with some of the Rollerbeetle Skills. Ram and Dash both have undocumented effects which I will try and explain here: (You can find neither of them on the wiki)

Quote:

Rollerbeetle Dash
Skill. For 5 seconds, you move extremely fast.

Unlike what the description makes you believe, Dash does not make you move extremely fast for 5 seconds. Rather, it increases your speed based on your current speed and acceleration. Unlike Super Rollerbeetle, it doesn't increase it for the full duration, but rather spikes out at the first seconds of the skill, and then tries to maintain it. (So you can still loose speed naturally whilst under the effects of dash)

The higher your current speed and acceleration are, the more effect dash will have, and the longer it will maintain that speed. Some anomalies (tough explained here) exist when you dash at low speed, yet still go up to 70-80%. This is because your current speed was low, but your acceleration was high. If you dash after you hit the top of a hill, but still at low speed (<20%), you'll find that dash will still increase your RRPM meter by alot, because the game registered you as having high acceleration (Going down a slope).

Quote:

Ram
Skill. For 2 seconds, all adjacent enemy rollerbeetles are knocked down.
Anomaly: Although not noted in the skill description, it will also boost the beetle's speed to the max.

With Ram, you get the undocumented effect of increasing your RRPM to max, which it effectively does, aswell as putting you back to your intital speed once it's over.
Also, with Ram there is a secondary effect few people know about that exists. Ram ignores outside effects* and temporarily stops natural de-acceleration
*: It doesn't ignore water, but it does ignore every slope in the terrain. So Ram prety much gives you a "bridge" you can build over small obstacles in the terrain.

In essence, Ramming over upwards slopes, aswell as not ramming over downwards slopes will get you the best results. This is also how I improved my technique from the youtube video. (He uses Ram on a few wrong occasions, and I improved it with succes -Cut off 1-2 seconds in certain areas)

So this is why the best possible start (Fences to Gate #1) is a dash once you reach your natural highest RRPM (Preferable BEFORE it dips back a bit again), aswell as Ram about half a second in the upwards corner.

Once you understand the best possible usage for Dash and Ram, it comes down to understanding the mechanics behind top times. For a top time to work, you need a good start. (The first minute) The way Dash and Ram work, as explained earlier, maintaing a high speed becomes an easy task after a mere couple of runs, and then it simply comes down to getting that high speed in the first place. Therefore, the only boxes that really matter are the first couple of boxes, as their speed will carry on through the rest of the match.

It is, correctly, assumed you need an Echo and a SRB (SUper Rollerbeetle) from the first 2 boxes in order to get a top time. This doesn't mean that every other box is useless, or can not gain you a better time. It only means that the first 2 boxes are enough to guarantee you a 480K+ time. Everything else can be concidered unnecessary risk.

This also means that there is a little bit of leeway on the first couple of boxes. If you don't get Echo + SRB from the first 2 boxes, you can still try and go for a top time using a slightly different strategy. Most people go for the first 3 boxes anyways (They pick both the boxes in the river, resulting in a slight time loss due to longer exposure to water), and if they happen to get a SRB from any of those 3 boxes, they can still get a top time by picking up some of the "risk-boxes" you normally don't take in a "Echo-SRB" run. Note that I say "Boxes" here, but in reality you only got 1 more shot, and that's the "Ox-cage box". If that box is a SRB, you can still get a top time if you play the "Echo-SRB" strategy, as you've only lost 1 or 2 seconds by going through the water (In the cave) aswell as strafing in the Ox-Cage. (Rather than going straight with Echo-SRB).

Other players and body blocks

And then comes maybe the most important factor of Rollerbeetle racing, or atleast the most variable one: other players.

As I've seen written in another guide: "Other players are like nitro-glycerine, avoid them at all costs, or they will blow you up", and this is true. Guild Wars simply wasn't designed as a race-game, and further more definatly not as a racing game based off the Z-axis. (It's one big downhill slope)

The problem is that the Guild Wars servers don't check up on the player's position often enough in order to guarantee a nice sync. If I were to explain it to someone who completely hasn't got a clue how servers and clients work, I would do it as following:

Imagine the Guild Wars server as being a teacher who needs to calculate a redicilous amount of mathematical equations, and rather gives some of those equations to his students to calculate for him. And then imagine the client as being one of the students, calculation these equations, but sometimes making mistakes. (Unlike the teacher) Every time they finish an equation, they send it to the teacher, and he checks to see if it is correct, which is still alot faster than calculating himself. If it isn't, he then quickly corrects it, and notifies the student he made a mistake.

So you have got 6 clients running who are each calculating their own version of the game, but sometimes collide with the server because their information is incorrect.

"But why do I specificly lag back so much in Rollerbeetle Racing?"

Well the thing is that the Guild Wars servers are already under so much strain due to the event, and therefore will exchange information slower with the client. This means that if a client makes a mistake, it is unlikely to get corrected straight away (As the teacher is too busy looking at the thousands of other students), which results in more "server lagging".

There is also a difference between other players and the map itself. The map is something static that is there when you load in, and will stay the same for the rest of the match.

The players however aren't. Clients don't ever communicate with eachother and will always have to go through the server. But due to the massive strain on the server, aswell as the fact that there is 6 clients rapidly changing position, your personal client might not always get top-notch information. This will result in you, aswell as other people, being in different locations on your game than where you actually are. (On the server's game)
Thus, Rollerbeetle is the perfect recipe for disaster, as it combines "clusterf*ck" movement with a heavily strained server, both of which aren't priorities on Anet's list when they made GW, and this results in you, and other people, lagging all over the place.

Z-Axis

One of the related problems with the map is the Z-axis, or more precise the non-existence of one. I hope, by now, everyone realizes that any form of elevation in Guild Wars is a cosmetic appearance only, meaning the game doesn't treat a person being higher-up, as effectively being higher-up.
But the problem here doesn't lie in the overal downwards slope of the map, because Anet did an amazing job on imitating that natural slope by adding bits and pieces of code for each part of the map reading: "This is where you speed up" and "This is where you slow down", creating the artificial feeling of gravity and friction, but rather when the track crosses itself.

The true nature of Z-axis problem lies in crossing paths, crossing in the 3th dimension that is. The infamous spiral bridge is the bane of any Rollerbeetle Racer because it can be so random. Sometimes you lag back for no apparent reason: there isn't a beetle anywhere near you, there isn't any obstacle, there is only walls. And the truth here is that there indeed is only walls, but walls that change in height and cross eachother, that is.

Look at the spiral as being a big C and X attached to eachother: CX. You enter the Spiral from the bottom right, from a seemingly higher route. You then cross over the bottom line of the X, which doesn't give any problems, and as you enter the C things go wrong.

The thing here is that for us, a person who understands height, the route seems natural: you make a U-turn, and simply cross under the previous path. For the game, however, you just did something completely different. The game has problems with understanding how exactly you went from one line of the X to the other one, yet at the same time do rather much pathing. (The intire downards corner gives you more distance in the 3th dimension than in the 2nd one) To the game it seems like you teleported from one spot to another.

And if you combine that with the previously mentioned low priority of positioning of the servers, you should be able to understand that the server sometimes forgets to update your client he's still stuck in the spiral. (More precisely the corner area between the crossing paths)

Is there something you can do against this? Not really. At the end of the day, if you're unlucky, you're unlucky. Taking a really big turn (and loosing some precious points) will increase your chances of not lagging back (As you're not in direct touch with a wall and you also give the server a bit more time to update), but it's no definite guarantee.
As Yuris wrote it in his guide, I will do the same thing: definatly don't cut the corner, but rather take the corner at a bigger angle. This doesn't mean you have to go to the complete opposite side of your track, but more towards the middle instead.

The start

Largely related to the "other players" section is the start sequence. The start is the most important part of the run, and more precisely, the VERY START. (Everything before gate #1) It simply comes down to getting out first, not getting KD'd and coming as close as you possibly can to the perfect setup. (Dash and Ram usage)

The biggest issue here is that the perfect setup requires you to activate Dash at a rather late point (Most people activate it earlier and will get ahead of you), aswell as saving up your Ram to be used in the corner. In reality, most people aiming for a top 100 score WILL have to give in some points in order to Dash earlier (So you don't lag behind someone), or Ram earlier (So you're not the one getting KD'ed, as most people by now seem to have realized the corner is the best spot to Ram).
The way the perfect setup works is that there's 3 keymoments which are completely out of your control. There's the opening of the gates, the path prior to your Ram and there's the upwards corner.
Since you've only got 1 shield, you will always have to take a gamble (Unless you know someone's playstyle, which is unlikely with 95% being PvE'ers) and leave yourself vulnerable to kD's at some point. And if multiple people Dash or Ram at the start, you need a good amount of luck to "lagg through them", because most of the time you will just get body blocked behind someone, especially in the upwards corner.

And the truth here is that no matter how good you are, it will come down to luck. (Aside from abusing the speedstart glitch Red and Blue offer) You simply can not know wether or not your opponent will Ram at the start, 2 seconds after the start, or even right before Gate 1. (But you should have lost him by then anyways)

The key to top times (Note plural)

Quote:

No KD during first minute: 26.6%
Flawless starts: 13.9% (Second gate at 30.5 seconds or lower)
Flawless starts with only Echo: 9.3% (Echo from one of the first 3 boxes)
Flawless starts with only SRB: 1.9% (But none from the Ox's Cage)
Flawless starts with Echo SRB: 0.29%
Flawless starts with 2 SRB: 0.39% (Meaning 1 SRB from the 3 start boxes, and 1 from the ox box)
Echo RB: 3.30%

*I occasionally did a "2-box" run as described in the youtube video. To score a top 100 time however, you can afford to loose a second by grabbing the second box aswell, so I did aswell.*

So in order to get a top time, not only do you need to know what to do and get lucky boxes, but you also need a good connection to the server (Which you have no control over) and a good start. And the truth of the matter is that this latter is, by far, the most important factor in Rollerbeetle Racing for all the reasons mentioned before and one more: It is the first obstacle you encounter.

If you trip at the start, your run is over. And going from my numbers, I've tripped 73.4% of the time. This means that 3/4 of the time, I simply had no control over getting a top 100 score, and it simply came down to "dumb luck".

And if I dig deeper, I find that only 13.9% of my runs actually were a possible top 100 scenario (<30 seconds at gate #2). Because alot of the times, I simply could not afford to use Dash and Ram at the perfect possible time. As I've said before, if you Dash too late (but still before the "perfect" time), you'll simply get behind all the rest and get body blocked behind them.
So only 13.9% of the time, I managed to come really close to the ideal Dash and Ram usage trying to avoid external factors.

Of the runs I had, only 9.3% (So about 67% of my flawless starts) had an Echo in one of the 3 boxes (Which still doesn't result in a shot at top 100) and only 2.29% had atleast one SRB. (Which atleast gave me a shot at top 100). After the Ox cage, tough, only 0.39% remained in giving me a shot at top 100. (Most of which resulted in 475-478K times)

0.29%, that is the number that matters most. Out of all the runs I had, only 0.29% gave me a shot at top 100, whereas the number of "Echo SRB" is 3.30%.

If I had control over the other players, not only would I have had an additional 3% chance of getting a top 100 scenario, but I also could have gotten way more runs in as Green, Purple or Teal, resulting in an overal 6.60% increase of chance of getting a good scenario, regardless of color.

Tough note that these percentages pretty much represent the amount of times you will get a top 100 score, as anyone who has seen the youtube video should be able to mimic the controls at 99% similarity. (I actually do it faster than the youtube video)
And I can only blame myself for not getting a top 100 time this year, because I made a mistake. Out of the 3 runs that I got that gave me a fair chance to get a top 100 score, 2 I lagged back at the spiral, resulting in a 470-475K score *One time I reached gate #6 at 1:38 before lagging back*) and 1 time was fully on my end (I pressed the wrong skill).

Obviously these are still just percentages, and in overal still low, but you do see how controlling the other beetles nets you a significant advantage.

Cheating

And that brings me to my final point, which is the syncing and other tactics being used to get more top 100 runs in:Shouting: "Whoever Rams at start gets ganked", aswell as syncing with a couple of guildies will gain you some serious advantage, because it drastically increases your chances of getting top 100 scenario. Reality is that in the 1000+ runs I've did, I've faced numerous of top 100 players playing with atleast 1 more guildie or alliance member, resulting in "whoever gets the best start, gets covered by the rest". But simply knowing what the beetles next to you are going to do will guarantee you a flawless start, let alone them covering you. Then it only comes down to that 3.30% chance you have of getting an Echo SRB combo.

Multiple acounting will also severely increase your chance of getting top 100 scores, because you don't have to wait out the average 2 minutes 25 seconds it takes for each match to finish. Instead, any time it comes to the point where you no longer can get a top time, you simply switch to another acount, waiting in the district to enter again.

This will increase your chances by 200, 300 upwards to 500% of getting a top 100 time. Simply because you don't have to wait out the match anymore.

And it's not like the people that are multi-acounting are hiding it. The top 100 is filled with xxx "Enter their alias" xxx names. Some even using "Nigr" and other offensive words. These same people, I faced in-game, don't show the least bit of respect towards other players (I actually got camped by top 100 player once he saw I got a SRB), and they even type in all-chat to their synced in friends: "Going to different acount".

Conclusion

Because Rollerbeetle Racing follows a set map every time, and every box, hill and obstacle stands on the same place, Rolleebeetle racing is more routine than it is "skill". One can argue that that you have to have skill to know when exactly you use certain skills, and I agree. Understanding the game and using your experience and abilities to improve your gameplay and times will get you an optimal route. But once that route has been layed out (Youtube video), anyone could possibly copy this route and perform it at 100% efficiency.

So the hardest part of getting a top 100 score is not necessarily the skill required to perform the game, but rather the ability to get in as much good runs as possible, be it through cheats or simply playing a lot.

Yet I will never claim an individual person can not get a top 100 score, as there is enough people who proove that statement wrong. *Myself almost included* When certain people claim: "I don't sync this" despite being a guild that I now have personally found out to sync (Up to 3-4 people from their alliance in the same match), I believe them, and I wish them the best of luck next year, because they are good at RBR, and deserve the spot.

But to the people with upwards of 5 acounts in top 100, I can bring up no respect. You're stealing people's spots, aswell as cheating in a game, probably for real life rewards. A quick scan on infamous black market sites history tells us that each year at around this time, people (who don't even bother to use an alias on the site, but rather post from their GW name) sell mini-greased lightnings or the ecto/zkey equivalent of one for real money.

I have come to terms with me not getting a top 100 score, albeit getting in daily top 100 3 times, and I have only myself to blame for the 1 run I, and I myself only am to blame for not getting a top 100 time.

I just wish everyone a good luck next year, I hope you guys learned something, and maybe are now capable of understanding how exactly the top 100 scoring board works, and maybe even steal a spot on it next year.

Good luck!

king swift · Feb 16, 2010, 01:26 AM // 01:26

Truly Brilliant.

My top score out of 250 runs was only 475,934. I got k'ded 3 times that run and got Echo + super roller beetle. I was blue. I agree with everything you said about luck and skill. If it was just me against the clock I would be flashing my new Zippy this year but I guess I will have to wait. I have also noticed people running bots in roller beetle. If they get k'ded they simply just stop and sit there for a few seconds and move in an weird pattern. I reported a few people who will be upset when they get perma banned.

Shayne Hawke · Feb 16, 2010, 01:27 AM // 01:27

Quote:

Originally Posted by Killed u man

First off, I would like to say "Thank You" to Anet for the rather good state of servers this Canthan New Year, so again thanks for that. (I could actually play at an average ping of 100)

This is true, save for the last two hours of the even that I played, during which my average ping never went below 500ms.

Quote:

Rollerbeetle Color

tl;dr version: If you get the wrong color, you're not getting a top 100.

The only reason color really matters is in the start of your game, except it really doesn't matter even then, because what you should probably be doing irregardless of your color is rolling into a wall, hugging it until you make the turn, and then getting out of the mob of beetles as soon as possible.

Yes, some of the colors might have an easier time with this, due to proximity with other racers, with walls, etc. My experience tells me Red is the best, Blue is a close second, and Green is third. Personally, I've never tried to get the superstart, no matter the color, because it's just so unreliable.

Top 100 times are feasible with all of them, and I very much doubt that it's going to ever get to a point where your color determines whether you're eligible for a record or not. That is, there aren't going to be 100 484/485k's clogging up the list any time soon.

Quote:

Map, experience and knowledge

By your definition of skill, I can agree with your argument that you wouldn't need a lot of it to post lots of records, and that you do need to know and understand which speed boosts do what, where to use them, how the map will let you move, and everything else.

What about when you're not going for a record, and you just want to get first place? The powerups that come out of the boxes aren't static. They're random, and the way that racers decide to use them can easily change the fate of the race. Skill is necessary here. You need to be able to assess whether a player is going to Shield here or there, where they're intending to use their knockdown, where and how you're going to pass the racer in front of you, and the list continues. Again, a lot of know-how and understanding goes into this, like records, but adaptation also becomes key.

Quote:

So unlike GvG, Rollerbeetle Racing is about following one playstyle as close as you possibly can, and if you do get kocked out of it, giving up is the only option you have.

If your goal is a Greased, yes. If your goal is first place, you'd better not start welling up because someone dropped your ass just once.

Quote:

Any person who has the ability to use his 2 hands will be able to learn how to Rollerbeetle merely by watching a video of a player playing it.

HAHAHAHAHAHAHAHAHAHAHA

Quote:

Boxes and skill usage

With Ram, you get the undocumented effect of increasing your RRPM to max, which it effectively does, aswell as putting you back to your intital speed once it's over.

I felt I'd quote this whole line here, because I feel it's something extremely important. Why they maintain the description to say that the skill only serves to knock down other players and say nothing about the speed boost is beyond me.

Everything here was pretty much a "no, duh" to everyone who's been racing competitively for probably not even a year.

Quote:

The start

There is only one good start: Dash at ~50% speed, Shield at ~:08/09 seconds so that it lasts the whole duration until you need to Ram, and then Ram up the hill to get away from everyone else. The exception would be if you have eyes and can keep track of who has already used Ram and who hasn't, and then saving your Shield for when you feel too dangerously close to them.

Above all, you need to get a clean Ram off for your start to mean anything at all. If you are forced into using anything else, it wasn't a run to be bothering with getting a record anyways.

Quote:

Cheating

And that brings me to my final point, which is the syncing and other tactics being used to get more top 100 runs in:Shouting: "Whoever Rams at start gets ganked", aswell as syncing with a couple of guildies will gain you some serious advantage, because it drastically increases your chances of getting top 100 scenario.

First off, I would encourage more people to use Ram at the start, because that means they're going to be slower later before checkpoint 1. It usually results in me avoiding their Ram, catched up to them at the hill, mashing my own Ram to set them back, and me putting on my awesome smiley face while I sail through checkpoint 1 at :18.

Second, I don't know anyone good that syncs. I can sometimes find myself entering a race with someone else who I know to be a top 100 player, and because I know that they're trying to get a top 100 score, and they can sometimes realize that I, too, am trying for a top 100 score, we do what we can to not screw each other over, and if one of us should get that change to strike it big, it's respectful for the other player to lineback the rest of the beetles and make sure that the run isn't screwed. I don't see this mentality being held only between guildmates; it's done between all top racers that I've observed.

You can call it cheating, I call it making up for some of the runs that were previously screwed over by one bad apple or another.

Third, drastically is an overstretch. I've had matches between three or four racers this season that were more fierce than some of my races between six. While it's good to know that you won't have as much competition from some racers, the other racers can all be plenty fierce to make up for it. It's all in what powerups people are able to get. If you're doing the start right and everything up to the first water boxes, it shouldn't matter really whether you have two beetles behind you or twenty (assuming the server wouldn't go into an absolute seizure from having twenty racers on the field at once).

Quote:

Conclusion

Because Rollerbeetle Racing follows a set map every time, and every box, hill and obstacle stands on the same place, Rolleebeetle racing is more routine than it is "skill". One can argue that that you have to have skill to know when exactly you use certain skills, and I agree. Understanding the game and using your experience and abilities to improve your gameplay and times will get you an optimal route. But once that route has been layed out (Youtube video), anyone could possibly copy this route and perform it at 100% efficiency.

I'll say again that I'm okay with the idea of records being more of a routine thing which is based on lots of knowledge and experience in the arena. Outside of the records though, it does take skill to be able to adapt to what the other racers are doing and to put yourself in first place consistently. Also, a greenhorn racer is not going to watch the YouTube video and start posting high scores. It takes more than just watching it happen to have the ability to post high scores.

I'm sorry for you that you weren't able to make it to the top 100 this year, whether that be your fault or someone else's. I'll say that I'm also not pleased with players that are taking multiple accounts and sticking numerous scores up on the list to deny others from getting on there, but I'm at a loss for ideas of how ANet could prevent this without doing some kind of manual investigation, which sounds like taking the issue way more seriously than it needs to be.

Good luck to you next year, Borat.

Killed u man · Feb 16, 2010, 02:05 AM // 02:05

Shayne,

Thanks for the good luck, next year I'll definatly be top 100

Everything I've said was with a focus on getting a top 100 time, hence the title. I agree that Rollerbeetle Racing (for first place) does require a decent amount of skill and versatility abilities (Targetting, strafing, shield usage, anticipation, etc) and that also is one of the reasons I can easily get 10-15+ consecutive runs.

I gues I put too much stress on the glitch, but yes, Red, Blue and Yellow aren't the best because of the glitch, but rather because of their position on the hill. (They have a slight slope)

Same for my "any person with 2 hands" statement. Don't get me wrong, I holeheartidly agree that some aspects in Rollerbeetle Racing require skill (shields, etc), but when going for a top time, after the first 40 seconds, you're flying solo. Pretty much after the first 17 seconds, you're going solo. And everything prior to that, as I've stated, comes down to several moments where you have to take a gamble and surrender yourself to variables outside of your control. (Will they ram at start, or rather in the corner? Will I get blocked or not? Will they dash in front of me)

The fact that Ram carries on your current speed (freezes it pretty much, aswell as it's decay) is something really big which only VERY FEW people seem to realize the full effects of. 95% of the people who I see going for a top time use Ram in all the wrong places, because they exactly copy the youtube video. Given, you'll only scrape off half a second a most by using it correctly in each of those spots, it's still half a second which is easily won.

And yes, I could explain the start as: "Dash at 50%" and Ram 8,9 seconds in, but rather than giving people the fish, I figured I'dd better teach them how to fish. I slowly explained every step of the process as to why Dashing and Ramming on those times is ideal.

Also, I WISH everyone would Ram at the start. And that also kinda proves my point as to how important syncing really is. Merely KNOWING what other beetles will do will get you a really far way, and it actually happened on occasion where I was the Red Beetle, and both Yellow and Blue Rammed straight away. "Yes, free flawless run" is what I thought to myself.

Syncing and Multiacounting pretty much guarantee you a spot in top 100, look at the scoreboard... In my eyes, that is "drastically".

Polgara Val · Feb 16, 2010, 02:28 AM // 02:28

Theres no doubt having multiple accounts gives you a vast advantage in terms of of getting the "echo srb" goldmine, however there really isint anything Anet can do about that.

I feel red and blue have the biggest advantage during a race in terms of getting the absolute pinnacle top 100 time, however last year i got my top 100 time with purple and i got another 481 during the dragon festival also with purple. This year I got it with red which was a pretty standard run with no risks involved.

I dont agree though that after the first 17 seconds your going solo, people can easily influence your time by lunging, blasting and all together making a nuisance of themselves, this is where you have to be smart and shield when appropriate.

Another thing, this thing about raming out of gate is wrong. It might get you to the first gate in good time but in the overall run you will lose speed and its one small detail which leads you to get a big time which in turn gives you that extra push into the top 100, assuming you know what your doing towards the back end of the race. When the 5th GW birthday event happens and rbr is back, feel free to experiment and see your rrpm wont match those of the top 100 after the first gate.

Pol

IattackU · Feb 16, 2010, 02:53 AM // 02:53

Borat, great job, I'd agree with just about all of it. Most of all I'm glad you didn't bring up the idea of bots because well that's already been discussed enough. However, would just like to talk about Multiple Accounts a bit.

Quote:

Originally Posted by Killed u man

Cheating
Multiple acounting will also severely increase your chance of getting top 100 scores, because you don't have to wait out the average 2 minutes 25 seconds it takes for each match to finish. Instead, any time it comes to the point where you no longer can get a top time, you simply switch to another acount, waiting in the district to enter again.

And it's not like the people that are multi-acounting are hiding it, they even type in all-chat to their synced in friends: "Going to different acount".

First off, I think you're giving syncing to much credit. Knowing what the enemy is going to do is a great tool yes, but syncing is just giving you less chance at a top time. If you sync with other people going for a top 100 time, you will all be fighting for the same line (assuming they all actually know what they're doing) which overall, like you said, will lag you because you will be coming in contact with other beetles.

Secondly, I agree that maybe using Multiple Accounts might not be completely fair for the people who have a lesser knowledge of how things work and are still going for a top time, or for the people that are using single accounts (because of the reasons you stated). Would Rollerbeetle be better without multiple accounts? Sure I guess. However, I don't agree that it is cheating at all.

Cheating, by the definition that I use, is "To violate rules deliberately, as in a game." As far as I'm aware of (I don't know the Terms of Usage or Rules of Conduct like the back of my hand), there is no rule against having multiple clients running at the same time. Correct me if I'm wrong, cause I'd be interested to know. Is it against the "spirit of the game"? You could argue that, but I wouldn't go as far as cheating.

Also I quoted that last sentence (snipped all the unimportant stuff about names) because I'm not sure I agree with the word syncing in there, and maybe this is me just being anal, but I figured I might as well mention it. Like I said before, syncing actually gives you less chances at a top 100 time. That being said, just because you see two people talking in All Chat does not mean they synced. Personally I met a lot of people this year that I've talked to multiple times in all chat during a race (when I'm multi accounting yes). I raced friends and people I know a lot (relatively speaking) during this weekend. Just because someone says "Oh I'm going to my other account now." Doesn't automatically mean they synced with the person they are talking to. Again, that's probably just me being anal about wording, so no worries there.

All in all, good job. I was really interested to read this when I saw it. Don't take my criticism the wrong way, it is no more then just discussion. I'm not here to start fights.

So good job and good luck next year Borat.

Killed u man · Feb 16, 2010, 03:51 AM // 03:51

Yeah, thanks for reading, I tried my best to make it readable.

And these syncers/multiaccounters I'm talking about, we'll everyone who aimed for a top 100 acount will know them. They have the same narionality, guild/alliance and/or I know them playing together in other formats. (HA, GvG)

I obviously didn't just get an unlucky team-up with syncers, as a matter of fact, the intire sundaynight multi-acounters were syncing it, and I got teamed up with them for about 40-50 times. I would be lying if I said I wasn't multi-acounting, but atleast I'm not clogging up 20+ spots in the top 100 list (And I only use 2 acounts) or having unsportsmanlike behaviour.

I have no problem with you, (Or even Yuris which I saw some times) any other MoO or dth member getting a lucky sync, or even knowingly syncing Rollerbeetle Racing. As long as the goal is to get 1 single acount in top 100, I obviously do not concider this cheating, tough still not the ideal way of gameplay, and maybe even is necessary to guarantee yourself a spot in the top 100.

I'm talking about people mass-acounting and syncing it, while at the same time causing grief by boy-cotting anyone else attempting to go for a top 100 time.

To avoid calling anyone out any further, I will just end this post by saying that they knowingly "ruin" (big word, but you get the point) the game for other people, both in- as outside the game. (This person with 7+ acounts in top 100 actually followed me after he saw me get a SRB, and made sure he dlunged me on a keymoment, only to go switch to another acount afterwards.)

X Ghoul · Feb 16, 2010, 05:27 AM // 05:27

First off, nicely done explanation

Quote:

Originally Posted by Killed u man

"Whoever Rams at start gets ganked"

I saw a synced team tell me something along these lines, then saw the person they were helping get a top 100 score and then smashed my face towards the keyboard and went to afk on rings for the rest of event >___>

Fril Estelin · Feb 16, 2010, 05:58 AM // 05:58

Very nice, GG!

RBR does require some skill, in the sense of consistently winning each match (playing against other players, using shields, observing, etc.). I'm with you on aiming for a top100, but I also got Gamers point (and one or 2 stacks of Ltokens, leading to 2 minis) and it requires a certain amount of skill to end up with these points (rather than Ltokens only). RBR is 2 racing games into 1: Super Mario Kart (get bonuses, multiplayer) and Trackmania Nations (fullspeed, trajectory).

Strangely enough, I feel my 200+ games gave me similar statistics from yours, without top scores (best 476k ... with a kd!).

You don't mention too much rubberbanding, which is related to "body blocks" and server lag. Yuris' guide on GWG indicated very nicely where you can afford to take corners short (the "S" right after the 2nd gate) or shouldn't (the spiral bridge).

Additional comment: almost impossible to get a top100 at peak times (last day in particular), where you're 90% guaranteed to get kd'd, more than once most of the time. Seems like the CNY start is the best window of opportunity for that.

I hope I'll understand one day (before GW2 is released) how top scorers get above 482k. Personally I'd never, ever want to win the mini without achieving the top score purely, by simply playing (no sync, no multiple account). But (contrarily to you it seems) I had loads of fun playing RBR this weekend, sure it's a bit frustrating to be so far from top100 despite winning games often, but the fun is still there and that's what matters.

Artisan Archer · Feb 16, 2010, 09:32 AM // 09:32

@above poster;

Winning requires a different playstyle then making top 100. You just follow that one perfect line till you get a good start without KD's and the perfect box-upgrades and then you get a topscore. If you play to win, every round is different cause you react to your opponents. Winning =/= Top 100

Drake Slasher · Feb 16, 2010, 10:20 AM // 10:20

Nice post, however
" If you trip at the start, your run is over. And going from my numbers, I've tripped 73.4% of the time. This means that 3/4 of the time, I simply had no control over getting a top 100 score, and it simply came down to "dumb luck. "

isn't entirely true. You can reduce the 73,4% by using shell soon after you get out of the cage, while looking at your opponents. I always look at who used ram (skill animation) and then know who still has it. If let's say 2 people have ram left, move to the right side and wait for them to use it, if they're gonna use it up the hill, stay a bit away from them and ram later. You won't get 484k but 484k - 1 to 1,5 secs is still a top score. Sometimes even all of them use ram before the hill and you can just proceed normally. 73,4% -> 20%?
Anyway , i finally got my beetle this year

Gl to you next year

Killed u man · Feb 16, 2010, 02:25 PM // 14:25

Thanks for the replies.

@ Fril,

I indeed didn't mention the corner of lagging back, and comming to think of it, it might be something I will add to my analysis.

I myself ALWAYS cuz every corner besides the spiral, because it crosses itself. The problem here like in the Z-axis.

Not knowing the real root of the problem, because even 2 dimension race tracks shouldn't have problems with pathing, I ASSUME the problem lies in the client not recognizing the boundries in the path that the server does.

For the client, the spiral is nothing more than a mere "X" where you enter from one of it's lines, go straight, make a u turn and then enter on the other line to go straight.
Yet the client doesn't seem to realize that these paths in reality do NOT cross eachother, (But rather go over eachother) and this results in some serious location issues right before the center of the X after your U-turn. To the client, it seems like you teleported from one leg of the X to another one, because it doesn't have the Z-axis to explain that you simply just went down. Hence, it sometimes "ignores the jump to the second line of the X" and will just keep you stuck on the top part.

Tough this is more technical, and I think I mentioned somewhere that Rollerbeetle racing relies on the Z-axis a bit too much for it to be a reliable racers. This probably is the only reason why that is, as the overal downward slope is getting mimic'ed rather nice by individual pieces of code for each part of the map saying: "Speed up here" and "Slow down here" to create the feeling that all the slopes are real.

As for shielding straight out of the cage, it's something I did alot when I felt the Rollerbeetles next to me were rather nervous/noobie. The problem is that most of the time, I always got unlucky that there was someone else on the other side who DIDN'T have to shield up, and could then save it for the corner (Also using the ideal setup), and he would always outplay me due to him having a shield, and me not.
Though on some occasions, I did succesfully use shield at start, and managed to get the perfect Ram and Dash in to reach the gate at the 17.5s mark.
Heck, one time my speedstart on Red was so flawless, I reached the gate at 16.8-9 seconds. (I had my skills at exactly 17 seconds) Obviously, if I had gotten Echo SRB there, I would be lying if I said I didn't get a shot at a 485-487K time. (My technique isn't perfect, but it's pretty darn close. At one point, I could perfectly sync up my route with Yuris, as we both got a SRB)

GL next year to all. But I think Rollerbeetle is comming back somewear around the Summer, no? Atleast without a mini-greased lightning that is, but it has never been around those for me. I truly enjoy playing RBR, despite what I make you believe apperently

, and improving my play. Though I now seem to have reached my cap. I feel I am now capable of reaching 485K times using pretty much every trick in the book (speed start), and I hope I can proove it next RBR event.

Z-axis part added to the analysis, as this is indeed a large part to understand top times. (Mainly due to the spiral)

Polgara Val · Feb 16, 2010, 02:54 PM // 14:54

A 485 score is definitely possible which the top racers will aim for in the rbr event coming in April, a 486 improbable but possible, however a 487 is impossible in my opinion there simply isint a way to have 100% rrpm in the track from start to finish. That is the only way I believe along with all the tips and tricks to succeed in getting a time which is even close to that.

I had next to nothing echo srbs this year I recieved 2 combos. One was clean and I played a safe route and got a 482.750 thats at least 1.4 seconds slower then my best record. The second combo i got was clean but with four rollbacks of which i got a 481.2 in the end. I had no other echo srbs in the whole weekend apart from those two.

My point is if you get an echo srb and you just want to be in the top 100 and win a beetle then just make sure you do the basics to get one solid time in and work on getting a spectacular time when you have nothing to lose.

I would suggest you focus on getting a 481+ before aiming for something as lofty as a 485 since to be frank this has not been achieved by any of the top racers let alone a 486. Despite the youtube video it still makes numerous mistakes to achieve a 483 let alone a 484. Make small steps, it will make you a better racer.

Pol

Reverend Dr · Feb 16, 2010, 04:27 PM // 16:27

Wait, so rollerbeetle racing is buggy, entirely dependent upon luck to the point that you only get 7 chances out of 1000 matches of getting a top score, and causes so much unneeded rage. Awesome game mode!

drkn · Feb 16, 2010, 04:36 PM // 16:36

+1 Pol about getting the mini and focusing on high score later.

i'd just like to note that the above 'essay' (and the topic that follows it) seems to be much more complex, mature and readable than previous issues, mainly the 'rbr botting' topic. i don't know the OP at all (besides reading his posts here, that is) but i can clearly see some sort of evolution in his posts. way to go and great job with collecting all the data and pointing out the obvious, yet usually ignored.

Morphy · Feb 16, 2010, 04:52 PM // 16:52

You guys take this all way too seriously.

Martin Alvito · Feb 16, 2010, 05:18 PM // 17:18

I'm glad you're finally bringing the scientific method to bear on the problem. It's the only way of making sense out of it.

You have a few inaccuracies.

1) Syncing doesn't pay off. Anything that reduces the number of runs you get is not a good thing. When you observe good players that know one another in a run, it happened by chance. Given that players are doing hundreds (or even thousands) of runs, they're bound to bump into one another randomly.

Consider the possibility that RBR traffic was relatively light this year, and that quality players were not evenly distributed across guilds. The result is going to be observations where half the run knows one another. Those observations are not the norm.

Also consider the possibility that traffic was SO light on Sunday night that the server protocols naturally paired these people repeatedly as long as they continued to join immediately after match end.

If the winner was leaving to sync join, then OK. I believe you. But in the absence of such further evidence that you should already have provided if your assertion is correct, there is a legitimate alternative explanation for the observation that you should not discard.

2) Cooperative norms evolve naturally. Go read The Evolution of Cooperation by Robert Axelrod. Influential book. Long story short, if it's in players' best long-term interest to cooperate, and there's a decent enforcement mechanism to deter short-term cheating, then they'll cooperate. In this case, what happened is that the top players figured out who one another were and started networking. Failure to adhere to the cooperative norm is quickly relayed around the network and punished severely.

You can call such logrolling cheating if it makes you happy, but you're stretching the definition of cheating quite a bit.

Making enemies isn't a good idea in RBR. But once the shadow of the future gets short (last couple hours of the event), the griefing gets pretty intense because the costs of making enemies become quite low. Also, it can pay off to disrupt others' runs to protect a marginal time.

3) There are ways to break a good run. I lost two superior times to disruption. One to a Distracting Lunge right before the beach, and one to a Blast while taking the spiral and unable to mash Target Nearest to time Harden Shell. Both turned 483s into 479s. Some people just live to make other people mad.

4) If you're only getting 15% clean runs, one of two things is true. Either you lag badly, or you're not defending yourself properly. The key to self-defense is NOT to mechanistically use Ram at the same spot. You need to evaluate the radar and the opponents in the match. If there's another quality player, you know you have to save Harden Shell for that player's hill Ram. If not, you can burn it earlier...and the answer to when varies with what your opponents do. You can usually predict when a bad player is about to use Ram to grief if you watch radar.

This is where skill comes into the equation. One of the primary differences between you and a top RBR racer is that the top RBR racer understands that there is some strategy to obtaining quality starts reliably. You are correct that otherwise the race is largely an optimization problem, but once again you continue to downplay complex strategy you don't understand as "skilless". Getting out cleanly is an art. Someone that won a ton of beetles probably did so because they understand that art better than other players, and got more opportunities to get a clean run with Echo-SRB as a result.

The flaw in your logic is that you're treating your observed frequency of clean starts as a constant in some underlying distribution, when in truth it is a variable that depends upon your behavior.

5) The spiral is not random. You need to evaluate the mechanics that create rubber banding more carefully. It is true that there is a small degree of randomness, in that if you ping spike while taking it you WILL band. But it can be taken aggressively with about a 90% success rate at full speed if you know what you're doing.

6) You don't need to be < 30 seconds to gate 2. You just need to play better after that point. For top 100 this year, < 32 was the absolute cutoff assuming you were flawless from that moment forwards. It sounds to me like you're giving up some time in the spiral and giving up some time somewhere else.

jazilla · Feb 16, 2010, 05:32 PM // 17:32

I have found this game to be way too buggy for my tastes. It's really frustrating in a race mode to bounce back the equivalent of 20 feet or so more often than not. It's like any multi player game really; if it doesn't work from a technical standpoint, I don't play it.

Phaern Majes · Feb 16, 2010, 10:31 PM // 22:31

Quote:

Originally Posted by Morphy

You guys take this all way too seriously.

Agreed, but makes for entertaining reading.

Artisan Archer · Feb 16, 2010, 10:54 PM // 22:54

Quote:

Originally Posted by Reverend Dr

Wait, so rollerbeetle racing is buggy, entirely dependent upon luck to the point that you only get 7 chances out of 1000 matches of getting a top score, and causes so much unneeded rage. Awesome game mode!

It's called farming.