Showing Posts For Ayusaki.6739:
Don’t you love bugs that popup out of nowhere… (or seem to, only to be some very silly mistake that triggers that bug once every blue moon…)
When I was in Middle school, a friend of mine wrote a script that would backup a DB for a service once every month, on the 30th…. and then came February… took him a moment.
Is it possible to see what traits or utility skills are being used on each character using API Keys ?
Scroll up 7 posts…
https://forum-en.gw2archive.eu/forum/community/api/Launching-inventory-endpoints/first#post5215567
this is super useful!
GW2 armory, here we go!
I think the last bit we’re missing for the MVP is traits/skills
does that mean that these will be added soon…? : >
https://github.com/arenanet/api-cdi/pull/42
https://github.com/arenanet/api-cdi/pull/5
Its on the todo list
That’s something we can add (using the backend functionality that powers the party/chat system), but I’m a bit concerned about providing that with the current scopes — personally I’d rather not implicitly allow third parties to track the hours I’m online. It’d be better to put that (and some other bits that escape me) behind a separate scope, I think.
Agreed, having that information behind a scope would be safest. The option to turn that off would be very helpful for people who want to use the API but not be tracked in such a manner.
That’s what I was thinking before, just in case you don’t want someone or a service being able to tell if you’re online.
Keep in mind that due to technical constraints and the implications of shared state in distributed systems, the value returned by the API may be stale up to 5 minutes. So it’s not that viable to determine which character is currently online.
That will be just fine, because we’re using this for WvW, so by the time most people are ready to roll out, the information would have repopulated and be relevant again.
Thanks for the reply! I appreciate it.
(edited by Ayusaki.6739)
For the program we’re working on, determining which character the API key holder is currently playing would be helpful.
The team and I were hoping that when “age:” was introduced it would be in minutes, that way we could use that information to check up on their current character by checking if that number was updated since last check, thus letting us make the inference that they are currently active on that character.
Since it seems like it’s in days, it won’t really accomplish what we’re looking to determine. Would it be within the realm of possibility – not necessarily now – to add in an “active:” endpoint as well to the character output to accomplish what we’re looking to do? If the api holder is not logged in, or logged in as invisible, it could have the “active:” value set to 0 (or blank), and if they’re logged in under a character, it should set the flag for that character to 1. I believe, if anything, that having an “active:” endpoint would be a welcome addition to “age:” because of these reasons.
Thanks for all the support and interactions with the community on these discussions.
I know ^^ (ich bin auf Elona). Wouldn’t be the first achievement that’s bugged. Officially the event failed and we were kind of disappointed, because we first thought that we had killed the kitten with 0:00 left.
Latency might have played a huge role regarding this.
And they did it. Sure it would be the one I dc on after 6 HOURS of trying with them…. fml
I’ll be home from my classes in about 3 hours to start organizing and preparing for reset tonight. Gonna be commanding until we get the kill again. Should be much faster this time
I’ll go prep the Teamspeak!
Why did you take a pic of my bad side.
Because your face isn’t much better Derv!
(edited by Ayusaki.6739)
Congratulations, well done.
Do you guys have your victory recorded? I’d really like to see how you guys took him down and how similar our methods were.
http://www.twitch.tv/theonlynemesis/c/2954458
Sorry for quality, crap internet.
Lol @ 15:45… everyone goes ballistic…
Congratulations, well done.
Do you guys have your victory recorded? I’d really like to see how you guys took him down and how similar our methods were.
as far as i know PoserGate got help from devs, at least the GW2 fb said they helped them…which isnt legit in my book, they had people that worked on the event there helping.
If bans were handed out for false and baseless posts, you would be so banned, your mother would feel it.
I’m noticing a trend. People are forgetting the importance of a victory. People seem to be UPSET that we were able to take down Tequatl, and not them. A victory shows that a method worked, and that luck is not a factor involved.
Blackgate did everything that we could to set ourselves up for success. We used teamspeak, we had a dedicated commander calling most of the shots, we had supporting commanders calling out shockwaves, calling out timers, calling out turret health statuses. On top of all that, we had people attempting the event since the patch hit (9 hours of playing, and planning), learning what worked and what didn’t work.
This is in no way a “Blackgate is the best” post, this is a “Learn from someone who has done it before under the best case scenario” post. Blackgate laid the groundwork.
We linked a video earlier, uploaded by our resident Kiwi, Flickky/Landsplash taken from her Twitch account which includes Teamspeak sound source whereas we have all of our chatter and commands, calls, ideas, and executions detailed. Please view it, and use whatever information you think will help you to defeat Tequatl.
This being said, in no way did Blackgate do everything right. I started a small discussion on Teamspeak after the fight stating that I believe we could have reduced our deaths significantly by moving out of the AOE circles as a group once there are 4 or more stacked in one zone. It seems as though they follow large groups, one circle at a time, giving you a chance to get your health back up, without the constant harass of them. Besides my suggestion, we had a lot more suggestions come up as well to improve the run.
In time, if enough people attempt it and understand the mechanics, eventually the information will spread to new people each and every attempt. Soon, using Teamspeak might not even be needed because the mechanics will be second nature to everyone.
For those wondering about our numbers, 110-120 of the people who were on Teamspeak are from Blackgate. We know this because we had verified our users ahead of time and those who joined Teamspeak for Tequatl (around 10) were unregistered guests, who might have been from Blackgate originally, but never verified. If the 150 persons per map capacity is true, that means that only 20 people were not on Teamspeak for the fight.
I hope that people stop saying “Bandwagon” and “Native”, like any of that stuff matters. What matters is that the people who were dedicated enough to stay on the entire time it took leading up to the victory were the ones who showed it is possible, and gave you a good idea of how to achieve victory on your own server.
…I’m disappointed that they had to use outside resources to do it…
Not only that, someone is paying cash for that resource. Are they going to come buy every other server a massive TS host? Is ANet going to be paying for it?
The resource is available to everyone, free. You can host it on your home computer if you so choose to. We decided it’d be best to host ours on a VPS, and yes, that does mean we had to pay for it. This was one of the options open to us. The fact remains, the option is there for someone willing to set it up for others to use, free of cost. I do not think this is a valid excuse at all.
http://npl.teamspeakusa.com/ts3npl.php
“Non-Profit License registration is for NON-PROFIT entities who wish to increase their TeamSpeak 3 Server’s capacity to allow a maximum of 10 virtual servers and 512 slots. Non-profit entities include but are not limited to organizations such as online gaming clans, guilds, or friends and family who utilize TeamSpeak software in a manner in which profit or gain of any kind is NOT intended. "
What do I have to do in order to get my question concerning obvious transfer abuse answered?
Will there be anything implemented (such as weighted league points, server (il-)loyalty debuffs, intraleague server locks or similar) that prevents people flooding the probable winner of the league during the first week, then transfering back to the home server (within the same league) and benefitting from abuse?
I think the ~100G in total transfer fees might be a deterrent to many…. This might not be as large of an issue as you make it out to be.
Saw this on reddit, took me a moment to realize this changed too.
I’ve added few more on top of the one i posted on reddit and you saw :
So have I. Mine is a “What you might have missed-type”, not a complete overview.
Hope this doesn’t get Hugh into trouble!
Saw this on reddit, took me a moment to realize this changed too.
A list of added content:
5 WvW ruins in the borderlands, when you hold 3 you get a buff.
This is all that really matters. Did they go into details? Can you elaborate please? I can’t watch it at work, its blocked, those kittens!
Basically there is five points around the lake very much like a SPvP capture point (don’t go all ape kitten I’m only using SPvP as and example to describe the point) if you cap and hold 3 you get the bloodlust buff. Its looks as if they will flip every few seconds though, realistically you won’t leave a zerg to guard it so the enemy zerg can storm in, cap, and move on. They didn’t go into detail so maybe there will be something put in place to prevent this but, as it is, it looks like a speed bump for a zerg.
It DOES seem like a speed bump for zergs because there are 5 points which need to be capped, of which 3 need to be capped, but the details on how the meta event appears and disappears is still very much lacking. Maybe when the map gets full (or partly capped) it appears. Maybe it’s persistent throughout the entire playing of the matchup, can be flipped at any time. Maybe killing the Grub in EBG makes them spawn on a random map. Maybe they disappear when you cap at least 3, with no chance of flipping them back until X Y Z happens.
I hope they release more information soon, because I’m speculating way too much.
My honest thoughts are theat when you capture these points, a bridge or a gate appears and you are able to traverse it and grab the orb, or fight a boss or something which gives you the buff. If you capture the points, you get guards which watch the point for you while you charge or teleport into the center. The more points you control, the more buffer you have to defeat or grab whatever it is that is lying in waits in the center.
(edited by Ayusaki.6739)
So….
No new map.
Did you even watch the stream?
EDIT: in the screenshot above he is currently STANDING WHERE WATER SHOULD BE (via the mini-map) but is on a new “Colosseum” of sort which works towards Bloodlust in the Borderlands (new name for Orb?)
Oh my gosh..they changed 1 small area on the map.. You call THAT a new map?
Are you complaining about progress? If so, explain to me how a brand new map would work in your eyes. How would you go about balancing it? Did you measure all of the distances between towers and keeps to make sure that trebs aren’t too close between two or more objectives? The more details the better, you seem to have designed a few of your own for a several million-player-game judging by your reply.
(edited by Ayusaki.6739)
Where does that tunnel lead?
it points northwest
perhaps a redesigned bay “watergate” instead of those 2 breakable walls + ally entrance?
That’s an awfully small tunnel, camera angles would be hard pressed to find a good viewable angle. I personally don’t know if it goes ANYWHERE… yet…
Oh, I can see it now…. Necros would have a field day with epidemic and AOE boon stripping.
But on the upside… the bags..
I think it might be THIS GUY:
https://www.youtube.com/watch?v=8pabPl4SYtU
And here I thought the portal would accept enemy projectiles.
Hah, sorry to disappoint. Maybe you could start that thread and then place a link to it in this thread, I won’t mind.
(edited by Ayusaki.6739)
Shameless linking to my suggestion forum thread for this topic.
I hope that this can cause some exposure on this topic as suggestions are one of the most active subforum out there, things get buried.
Thanks for the reply. The timer portion, without a doubt, would be an easier implementation I’d think. I don’t think I fully understand what you mean by range indicator. Do you mean more of a “Too far to drop” or a “this is how far between existing portals”?
The problem that I see with this is that the market is cross-server. All of the people who are on the server would be receiving something that might imbalance the market, seeing as you’re including a massive group of people who might not even have participated in WvW in the first place (lets say 50K+ people per server).
I think your idea has some momentum, but needs some work. How about the winning servers get to participate in an event that week which then whittle down the number of people who have a chance of a precursor? This will avoid a tip in the market.
Quaternions and Euler angles use rotations in order to determine the heading, positioning and starting orientation in a 3d/4d floating point plane of existence. Think, aircraft flight, or underwater submersibles. All of this data is processed through several types of formulas, all with their own way of outputting the data, but the principles remain similar.
To determine the rotations, the formula R = X(a)Y(b)Z(y) is usually used.
To find the space between them is easy, but to find the heading is much more complicated because you just added a bunch of new variables.
(Again, I may or may not be able to add more details later, elaborating on further functions and utilization of this information, my schedule is super SUPER crazy. I appreciate any and all input or comments about this, ranging from “You’re wrong, here’s what you do” to “I like where your head is at, but here is a suggestion”. No negativity without any just reasoning, please. I am posting this because I have been sitting on this for a month now, and it’ll just keep sitting indefinitely otherwise, as is.)
“Oh, so you’ve got it all figured out, huh?” No, I don’t. I do, however, believe that this is a good starting point. As I’ve mentioned before, this is in no way the only solution. This will need to be fine-tuned and carved down for it to be implemented.
(More to come?)
Thanks for reading, I hope someone who knows more on the subject could expand on my post and provide some interesting ideas. A developer response would be fun to read as well, as I am seriously curious about the feasibility on this subject!
I’ve mentioned quaternions and Euler angles, and chances are that if you needed to google what they are, then it only reinforces what I am about to say. Quaternions are… complicated. If this is the system currently used by the movement/positioning systems, then my proposal might fall flat based on complexity alone. I also mentioned a simpler XYZ axis positioning system might be used. If this is the system used, it’s easier.
For the sake of keeping what little readers of this post I still have left, I will go into the math of quaternions and Euler angles later on. I’ll be sure to clearly label it so you can skip over it if you so choose. I might use spoilers; this is yet to be determined. So, in the meantime, let’s talk XYZ!
XYZ, (properly named the Cartesian coordinate system) gives you values to represent a location on a 3 dimensional plane. First, we need to determine the location of where a portal is on a map. Later we’ll talk about the connecting portal. For arguments sake, let’s say we’re only interested in X and the Y, because having a scenario where an arrow pointing straight up might be intrusive (no matter the size) and having it pointing straight down might…umm… not work out too well. Inherently, having the lack of the Z coordinate adds a bit of mystery….ghostly “oOoOOoooOoOOoO” sounds. Not to mention adding a Z coordinate changes the formula entirely, complicating things.
How do we determine the heading between two separate X and Y coordinates? The day after I left college (last year), I’ve lost a lot of my math skills (those useless, useless, math skills), but I believe to get the heading from point A to point B (and back again) you would apply this formula:
Inverse Tan((y2 – y1)/(x2 -x1))
This formula will output the angle which the two points bear.
(I may or may not be able to add more details later, elaborating on further functions and utilization of this information, my schedule is super SUPER crazy. I appreciate any and all input or comments about this, ranging from “You’re wrong, here’s what you do” to “I like where your head is at, but here is a suggestion”. No negativity without any just reasoning, please. I am posting this because I have been sitting on this for a month now, and it’ll just keep sitting indefinitely otherwise, as is.)
Being a closed source game, it is near impossible for those not directly involved with its development to know what type of beast we would be dealing with. To say “All you have to do is X, and you will see Y as a result” would be ill-advised, and I will not be doing that. I will, however, be speculating and offering general ideas about potential implementation/integration.
Let’s begin some speculation!
I believe that when a portal is placed down, it copies the coordinates of the player who dropped it at the time the skill is activated. The portal location is stored on the server for a limited time using an instruction set of either a quaternion/Euler equation (a mechanic used to represent coordinates on a 3D/4D plane, rotation, angle, in a free-floating space) or a much simpler XYZ coordinate type system. (I used to believe the latter was the choice of implementation, but over time, I found more evidence supporting the prior). During this time, this information is passed to all of the clients that meet the criteria for being able to view it.
Once the other side of the portal is placed, similar information is stored, and then sent, and a passage is created letting players traverse through these two portals. “But wait, what happens behind the scenes?” I believe it can be one of two things:
1. Point A (creation point) and point B (exit point) are linked and flagged in the system with a label. The information of their locations are stored in memory and sent to eligible clients. In this case, the information is broadcasted to qualified clients before activation.
2. Point A (creation point) and point B (exit point) are linked and flagged in the system with a label. The information of their locations are stored in the server’s memory and wait for a player to press F to use them. In this case, after they are used, their new location becomes known and the user is teleported.
In scenario 1, overlaying arrows onto portals would most likely be easiest. Scenario 2, it might not ever happen at all. The client must know the information before it can draw the arrow in the proper heading.
I’d like to talk about implementation. “Oh yeah, point A, point B, this shouldn’t be too hard to do” you might say. Although it doesn’t sound like much to implement, there are changes that would need to be made to existing systems before it could work. Referring back to my scenarios above, number 1 makes it easier to implement, but easier doesn’t necessarily mean easy. The existing method of determining placement of the portals greatly governs what math needs to be used to calculate the heading of portals.
In WvW, following a commander is a huge part of a tactical advance, and in some instances, portals will be utilized to help those in back of the line catch up to the commander in front. Unfortunately, this can also have the opposite effect if those in the middle or near-front take those portals and are sent backwards. It’s a problem for those situations where you have an uncontrollable lack of awareness to your surroundings. This is not to say that people have terrible ADHD, or are ignorant to these things and should be paying more attention; rather that you can’t always see all of the portals due to the limitations of our screen not being able to see 360 degrees, and that some things are just out of our control, like obstructions, or large battles blocking vision.
The trouble with portals is this is that no matter the intention of the person dropping them, portals are linear. By their very nature, there is a chance that upon entering them, you might be sending yourself in an unintended direction. This is a chance you take every time you enter the A to B game. Portals go from point A, to point B and do not pass along information on where B’s heading is. Unless you are certain of its endpoint, either visually, or by text/voice chat instructions, you are taking a chance.
The solution to this is to pass along this information to the client in the form of an easy to understand medium. I recommend an arrow.
An arrow overlaying the portal would take up very little space and provide simple, at-a-glance information letting you know which direction the exit portal is. This information helps making the decision to enter the portal much simpler for the player. No longer would a player jump into a portal that is facing backwards (unintentionally). This will also help people distinguish multiple portals from one another, by granting them all the information that they need to know and letting them do an “on-the-fly process of elimination-type judgment call”.
Both DB and JQ kept faling for the same EXACT Portal bombs over and over again.
FTFY. Still waiting for those videos. Make it so!
(edited by Ayusaki.6739)
Hahaha. That’s not a common practice. Probably the people realized there was an abundance of golems and decided to have a party.
^ This… or maybe there was just an overwhelming abundance of supply to burn.
Lets sit here quietly for a moment and think of what the more likely explanation is….
(edited by Ayusaki.6739)
Hahaha. That’s not a common practice. Probably the people realized there was an abundance of golems and decided to have a party.
^ This… or maybe there was just an overwhelming abundance of supply to burn.
(edited by Ayusaki.6739)
Okay, so after much hard work and no small amount of wizardry we are going to get this in for today. We were unable to get the leaderboards fully set up to indicate the matchups, so there will be a disconnect there, but you will see your matchups once you log in.
Awesome news! Now if only matches could reset on time…. They tend to deviate 5-10 minutes until the scores actually reset.
(edited by Ayusaki.6739)
I fully support this type of collaboration. When I get home, I’ll see what I can contribute. Collaboration in this capacity is the best way to learn how to code better, as well, so watch this space guys. I’m excited to see what the community comes up with.
(edited by Ayusaki.6739)
Please consider starting a GitHub or a GoogleCode page so that the community can contribute pull requests in order to help further the development of this project you are working on.
I am a full supporter of open source, and with a community this interested to dive in and utilize this new feature (API) I’m sure they would be too.
Something like…. http://gw2cartographers.com/ ?
YAY! GitHub source! If time permits, I might contribute some code (if allowed)
I fully support a GitHub type of environment for this type of project. This will help those with questions and those with ideas meet and learn from one another.
Perhaps we could come up with a few GitHubs that surround the API in different programming languages to support all iterations of development. A master list of sorts.
You can get many, many ranks past 150.
At what rank do you get the “Evil Overlord” title?
Can you say “Incentive”?
I think that having a “Cash Out” type NPC at the spawn point would be a possible solution to this matter.
How it will work would be that a player racks up points that equal their kills. Badges would have their own currency in which they would function as they do now; a random chance to be looted, but this time automatically. Upon visiting this NPC, he would say “Holy crap! You’ve killed 240 guys this round, a new record!” At this point the server would roll it’s dice of random, and play the RNG game, and spit out the loot you would normally have gathered from bags. He could also double as a merchant, which you could sell it right back to if you so choose.
Hopefully this is a viable type of system which could be implemented.
Just noticed this proposition was already posted. Quoting in hopes that our similar ideas get noticed.
One of the best solutions you could do is rather than having bags drop – have them moved to a Cabinet Container (e.g. Heart of the Mists style) which a player can then grab at a later point. You could even have the inventory in that cabinet decay after timer period; say this period is anywhere from 4 hours to 6 hours. So a player can simply grab their cabinet rewards after heading back to spawn for repair or moving to another BL..
(edited by Ayusaki.6739)
2. To take down a server or infrastructure of any size, you need a few people, or at the very least a few computers on separate connections.
Not with Slowloris or DNS amplification. Single box on a residential connection can do it, but then it’s a DoS not a DDoS.
Slowloris is for web hosting services such as apache, NetScaler, Cisco CSS, and doesn’t out of the box apply to teamspeak servers without heavy modification, unless there’s already a pre-built solution.
DNS Amplification is another slight maybe. If the VPS router has a scanning tool to deny from known DNS routes, then this falls flat.
This thread is fun! It feels good to know we’re educating the thread owner that not everything is about him!
We’re changing lives here people… lives…
(edited by Ayusaki.6739)
The TS isn’t ran by a private company – more like a public nonprofit, if anything.
And there are 300+ members online in the TS at a time … there is a lot of incentive to DDOS a TS server.
And you make it seem like its hard to shut down a small server with a DDOS. Its not, really.
So take your buzzwords and get outta’ here.
Several things I’d like to ask you.
1. You’re not addressing the biggest problem here, rather, you avoid it. Is this company hosting multiple services? Servers? Teamspeaks? You only mention your server and it’s maximum number of participants. Who cares if it’s on a NFP server… This doesn’t answer the question. How many other sites/services are hosted on their network?
2. To take down a server or infrastructure of any size, you need a few people, or at the very least a few computers on separate connections. Unless of course your hosting platform doesn’t meet any of the agreed upon standards for hosting companies. ISP monitoring systems are getting smarter. They are now detecting certain types of patterns, targets, length of time the socket requests remain active, etc, etc. This type of activity doesn’t last long these days. One (or two..or more) person(s) doing this multiple times would also give your hosting company logs to go over; lots and lots of logs! Why haven’t they approached you, told you “We can’t host you any further, you’re being targeted and are crippling our network.”? This happens more than you think… They haven’t approached you because you’re probably not the one being targeted on the rack.
3. Do you think you I’m just using buzzwords now?
Educate yourself (and think) before contesting someone who knows what they are talking about. (I doubt I’ll be hearing the last of you once you’ve learned how to use Google to find retorts.)
Before I forget, fat (smart) German guy speaking about ISP monitoring DDoS traffic preventions. http://www.secdocs.org/docs/detecting-ddos-attacks-and-countermeasures-at-isps-video/
Our teamspeak server was recently being DDOS’ed, but no one knows who we are… it was actually another server on the same box that was being DDOS’ed. If your host has more than one teamspeak, it might not be yours that is the target.
^
THIS!
https://forum-en.gw2archive.eu/forum/pvp/wuvwuv/Really-uncool-JQ-BG/first#post1782204
Lets stop and consider the following. If your Teamspeak is hosted by a company, and not someone’s personal computer (in their house), then chances are it’s on a VPS, or run as a service. If it is on a VPS, then the company houses many different websites and services on the same rack. If one or some of those VPS’s are under attack, chances are the router/switch and the rack are under a load of stress, which in turn affects network traffic to your Teamspeak server.
Before you say that someone is deliberately trying to cause you (and only you) harm, think about how small and insignificant one teamspeak server is compared to what else could be going on in the world. It’s not always someone out to get YOU.
PS. The chances of your rack being under attack by other servers or individuals is much higher than 5 or 6 guys targeting your VPS Teamspeak with a strategic DDoS.
For all the conspiracy theorists out there, this is completely unrelated to any news elsewhere. I’m still at Arena.net, but the next time you see my name it will be for something other than world versus world. Something quite different indeed.
Guild vs Guild / Guild Halls!?
The limit is the same for every team on a map. So if the limit for any team is X then the map limit is 3X. Make sense?
There have been whispers that state that in some cases, players remain active in WvW, taking up space even after logging out or being kicked; “residually” if you would. Is there any truth to this as far as you can disclose?
This could explain why queues seem to pop in waves, after a cleanup is performed of these players.
seriusly,this is crasy -.-
But undeniably fun!
one last question is limit per team same or teams can have differen map limits ?
Hey, woah, no. You can NOT put an end to the potential questions. I call foul!
that statement is completely unreasonable, your patch releases contain hundreds of bugs and errors which you still have to be fixed later on, so releasing the information does not affect in any different way as the bugs and errors you are already including, if you dont want to release the information right now, then atleast come up with better excuses.
If you don’t like how Arenanet operates then play a different game. Simple as that.
operates the game? do you even read what you type, is not about how they operate the game, is about how they handle the information, like the 40 megabyte patch they did a couple of days ago in which they never revealed what it contained.
As we stated at the time, that was a maintenance build. Nothing was changed. The 30-40MB you downloaded was the update process downloading a new EXE file, which happens whenever a build is released, regardless of if any content has changed.
People, the EXE alone is around 22MB, lets not forget that. Even changing the spelling of something in the client means a full redeploy of the EXE.
I like Jeffrey Vaughn more then the moderators, he didnt lock the topic after his remark :P
Suddenly; “Badmouthing moderators, thread closed”