Snip..
I wouldn't mind if they used PvP as an excuse to actually adress some of the inbalances of the game.
But they have used PvP feedback to repair/fix/adjust a bunch of abilities and weapons over the years.
(Have the Devs done a thorough job all this time ? No, they've let things slide, but they have made some effort over the years)
Ill point to the Elachi Disruptors as a prime example.
At one point, those Elachi disruptors could proc unimpeded, Sheild bypass after shield bypass, after shiled bypass. (100% shield bypass damage in case people don't know what I'm talking about)
It was the PvP community that got those weapons adjusted, as their performance was far surpassing any other weapon type at the time.
It became a thing in PvP for while to use Elachi Disruptors, as there are those who will always seek out the most OP fighting techniques.
People complained their proc was not in line with any other weapons in game at that point.
People were getting smoked in one or two passes with the DHC, or 1 shotted with Elachi Double BO.
So their proc was changed, so they could not proc over and over and over again. They added in a cooldown on the proc. Think it was 1 proc per 30secs.
What happened you ask ? Well, it forced all those Elachi Exploiters in PvP to drop them like dead weight. And they all moved on to the next OP weapon trend.
Which I think is healthy for game balance.
If there's a clambering towards a new weapon type in PvP you can bet its performing far better then anything else atm.
And is a rightful candidate for adjustment.
If all you did was PvE and used Elachis, sure, one would be upset that they changed it (like many forum users were). But on the other hand, they were "by far" outperforming any other weapon type, and they needed to be adjusted or everyone would have adopted the Elachi's as their weapon type.
You can't have a huge imbalance between weapons.
You can't have a weapon type that is far an away better then anything else at the time.
It's not good for the game. There should be equal benefit between all weapons types.
Do we have equal balance today ? No I wouldn't say we do, but its getting better.
The next trend was the AP craze and plasma doping.
So we still have issues but, some adjustments are better then none.
Otherwise it would really be the wild west of Power Creep, and there would be a lot less flexibility between builds.
If the Elachis were never fixed, you can bet, something else more powerful would have came out, as that is the nature of power creep.
Then things would have gotten out of control even more quickly then they are now.
And currently, look at all the people using Res Beam in PvP these days... Why do people use it ? Because it has a 100% shield bypass, and applies Phys damage directly to hull, for which there is little to no counter.
There is only one measly space trait that gives you 10% resistance to phys damage. Hence many PvP'ers are abusing this skill in PvP.
Id point to it for the next nerf/adjustment.
And it would be perfectly justified imo.
Since DR, there have been several adjustments, but I have noticed that since STO has gone full bore into selling power creep (since DR), they haven't been as diligent in making fixes for over performing weapons.
Probably fearful of the backlash it creates.
But they have indeed used pvp feedback (many times) in the past as a basis for adjustment.
There is an important relationship between PvE and PvP that people don't fully understand.
PvE side just get's mad that their new toy gets nerfed.
They don't care that the weapon or ability in question was way out of line performance wise.
They only care about their final DPS number.
Which is only looking at things from one side of the issue, and is a very biased position.
Not too mention, very selfish.
And look at the game today, there is "already" a large enough gap between High DPS'ers and casual players.
With no adjustments pointed out by PvP'ers, that gap would be even more cavernous.
Anyway, getting rid of PvP would solve nothing, except to anger those who still enjoy it, small as that community has become.
So...in all of what you posted...you don't see the issue that PvE players would have with PvP players? You don't think somebody who spent the hundreds of millions of EC or hundreds of even thousands of real world money to get a set of the elachi weapons for the PvE fun...won't be miffed that it was nerfed into uselessness because of the PvP crowd? And don't get me started on what you all did to science because you tact jockies couldn't handle it. Hell there is STILL a cry about FBP because you idiots can't learn to STOP SHOOTING if somebody pops one of these up (unless you think that FBP should be nerfed in PvE because it is so OP...I don't mean some of the traits are actually broken...but FBP itself...which is pretty meh in PvE alone). So until I stop seeing a call for a nerf to FBP, I am gonna go with you all do more harm then good and wish you all to go away.
You say PvE people are selfish because they get mad that the stuff they worked and in many cases PAID REAL MONEY for gets nerfed because a part of the game THEY DON'T EVEN PLAY decided it was OP for JUST THEIR SEGMENT of the game. So...who is the selfish one here again? Yes there is a problem with the disparity between the high and low end of this playerbase...but not all of that has to do with gear. Some of it is the PLAYERS themselves. You could take a 250k DPS ship and hand it to some of the lower end players and you'd be lucky to get 50k out of a run. If that high DPS ship is a cannon or sci boat...yeah lucky to get half that. This is because a lot of doing crazy DPS has to do with timing stacking of buffs and debuffs. Now the system COULD be set in a way that skills don't affect the results as much (mostly by taking stacking away)...but seriously, a game is suppose to have an element of skill. The fact that this game has a range of players from people who are hardcore gamers with 1337 skills to people who are so bad at games that they tremble at the idea that they may have to swap out items that orginally came with the ship...yeah...that ain't gonna get solved by adjusted item stats.
Clearly you don't read so well.
I did say the nerf to Elachis were heavy handed and out of step with what was needed. I'm not unreasonable.
My point was, if nothing was done, they would have been without a doubt, far and away, the best weapon type in the game, and that in itself isn't right.
Unless you want everyone using Elachi disruptors and cookie cutter builds because that is what would have happened.
That said, the Elachi weapons are far from useless, they ARE STILL an excellent Disruptor weapon.
And they still sell for VERY LARGE sums of EC on the exchange.
Only now they are more in line with other types of energy weapon.
Also, if you knew me, you'd know I run a SCI for pvp, using FBP+Kinetic FBP build in my T5U Vesta.
LOL
I'm one of the first to speak out when someone complains about FBP builds.
My main reason for running this build, is primarily to punish the players like you, who would run to the latest trends in OP'ness, and power creep.
That kind of behavior is what encourages game developers to push more power creep.
You are the problem.
I also realize the early nerfs to Science truly put those builds at a disadvantage for a long, long time.
But I had also said earlier, these nerfs are sometimes done with a heavy handedness that is not appropriate.
And that Devs need to truly understand the PvE/PvP relationship before making these changes, and this is one of the problems here in STO, where Devs have gone too far in one direction or the other, by not understanding that relationship well enough.
And yes, you are selfish, very selfish, if you thought Elachis were "just fine" on the day they were released.
It's either that or you've deluded yourself into thinking they were somehow balanced in comparison to other weapon types. Or you're just not that bright, and don't do much testing yourself.
The fact you paid large sums of real life money for an item, and are now displeased with how they were changed, is another topic all together. That's more of an ethical issue between the Company and the player.
But for you to say Elachi weapons are useless, even post-nerf is a bald face lie. One thing I can't stand is a liar.
If you want to tell yourself you were cheated out of RL money, by all means, but please don't spread your falsehoods. It just makes you look like you don't know what you're talking about even more.
Those of us who have done extensive testing know what the real truth is.
And no, these changes are not just for pvp, they are used to ensure there are no "extreme cases" of one weapon or ability severely out performing everything else in the PvE setting.
Learn a little, and think about what you say, before you pop off more stupidity.
Ill add something though in your defense, maybe you bought those elachi weapons, with the expectation, they would be balanced on their release. Which I wouldn't blame anyone for. And I wish this were the case for every new item released.
But if you did do the testing, you would've seen yourself that they were indeed vastly superior to anything else at the time.
So, I have some sympathy, for those who put faith in the developers to get things right, the first time.
But that's rarely been the case here in STO. But again, that's an ethics issue between the Developer and Player.
Look at the Kemo controversy for more evidence of this if you need more proof.
Plenty of people invested many millions to acquire this ability.
Granted the Kemo issue is a little different, as it was causing massive lags spikes, while at the same time being OP.
But its just another example of new items/abilities being released without the proper QA.
So...you accuse me of not reading and not know about you while doing the EXACT SAME THING. Where did I EVER say that I got elachi weapons? Or that I play with the most OP weapons?
Well, you seemed pretty upset about them, so I just assumed.
If you bothered to read any of my posts, I repeatedly say to build for fun. Personally I don't give a rat's TRIBBLE if they adjust items for the sake of balance...what I DO care about is they do so for the sake of PvP balance when it make things worse for the PvE people. And quite frankly, this is gonna be true for most PvE exclusive players (which is the vast majority of this playerbase). That means your whole idea of have somebody who know to balance between the two WILL NOT WORK. Even if they are GREAT at their job...it's just a bad idea from the start.
Yep, it's not easy for them to strike the right balance when making changes on the fly. Sometimes they make terrible mistakes, and are too eager to get a fix in, and end up botching the job.
You know what though ? Player feedback made science pretty powerful once again. So it's not always a losing situation.
What really needs to happen is one of two things. EVERYTHING gets two stats. One for PvP and one for PvE so they can adjust each without it affecting the other...or we make PvP a seperate environment. That is to say, you buy PvP ships that comes pre-loaded to be balanced with each other and that is all you are allowed in PvP.
I've been hoping for something like this myself. It's probably the easiest solution.
No Traits Mode. And Stock Ship Mode
Even so, there's no reason they can't also leave the current version in game.
Let the whales beat up on each other if they so choose.
The separate PvP server or two versions of each ability, as you suggest would be nice in theory, but even I don't think it would be a good investment of resources.
I suppose there is the completely unrealistic solution of have NPCs that are specced like players, use player builds and have playerlike AI...but consider that AI does not exist even as an experimental piece of tech...that is a completely unrealistic solution at this time. So since the first solution requires WAY more dev time then they have, the second one has been rejected by the majority of the PvP playerbase and the third one is not even possible with all the money and tech in the world...yeah...it just needs to go away.
I would rather they left PvP in. Even if they let it rot, and never funded any future development ever again.
Like @davefenestrator said; "Keep PVP but let it continue to rot, and only make game balance changes / nerfs for PVE issues not PVP.
Anyone wanting to PVP can still do it, but it will be up to them to agree on the rules for the duel --"
Which is kinda the case now anyway, right ?
Why take it away completely from the people who like it ?
I don't see the logic.
Personally, I don't like when they take away missions and queues, (Ie; NWS/Mine Trap/Mirror Invasion XP/Breach/some of the old story arcs etc), and I wouldn't like losing a whole game mode.
Especially if it's for no logical reason for it.
Oh and we already do have the one true build. BFAW, AP beams FTW. So it's not like the elachi weapons would have made a difference there. The only reason to NOT do BFAW, AP beams right now is because you have more EC/dil/zen to spend just because.
No, Elachi weapons were outperforming even APs at that time. and by a lot, like 20-25% DPS (all things being equal at the time, Elachi CrtDx3 vs AP CrtDx3).
But Yeah, currently, there is no equal to APs, a true min/maxer will go for that combo on the PvE side, if they know what they are doing.
But......
The reason we have build diversity isn't a matter of weapons being similar in power. They ain't. AP beams kick the living snot out of everything else until you get into EXTREMELY costly builds. Hell just BEAMS kick the living snout of other weapons. And don't even get me started on the issues with BFAW. Mines are nigh useless compared to other weapons. But you know what...we can make a viable mine ship...why? Because the content is just that stupid easy.
Now, while I can agree the AP/Faw builds are at the top of the food chain, I don't agree that they;
"Kick the living snot out of everything else until you get into extremely costly builds." That is just not true.
Did you know the difference between an AP build and almost any other beam build is about 5-10% DPS ?
Yes, IT IS better, I won't argue that, but let's not over exaggerate just to make a point.
It is concerning though. I don't disagree.
[...]I've played enough MMOs over the years with some really cutting edge (for their time) PVE encounter design. And it's a shortcoming of the overall system that NPCs can't be trusted to be as effective as human opponents. AI doesn't really have an advantage here. [...]
Curent AI technology, even established game AI technology, can be made to fly in circles around us in a single ship and blow us up with constraints that are equal to players. It is not a technological limitation, but a design decision to have them be this dumb and inflexible.
No...no it can't. If you think this is true, then you are a terrible pilot. Yes, the current AI can defeat TERRIBLE players...but I am not talking about that now am I.
I think what the poster is saying is that the current tech in the game (and maybe some improvements to that technology here and there) which are all reasonable and achievable by Cryptic's development team could create NPC AI that is far more challenging.
And if that is the case, I agree. I think it can be improved. This is where the poster's suggestion wins my heart. The game's PVE does set the bar low. As you point out, the current AI is just too easy.
The Poster is suggesting that can be changed, to create a new AI that is more challenging.
I agree.
What I'm saying is that my mind just realizes that after that happens, and we all adjust ... the new challenging content will become less challenging. And then we're right back here, having the same discussion.
The goal posts will have moved. That's it.
Yeah eventually the players would crack the code.
Either by finding the "path of least resistance" or buying power creep. The cycle would continue a new.
[b]This is not a place to continue arguing about one point, or talking mean about someone (like in the above post). [color="purple"] Do not continually argue every post put in the discussion, or create a long chain of arguments off of one post!!! Also, BE RESPECTFUL TO OTHERS AND DO NOT TALK TO THEM IN OFFENSIVE TERMS! DOING SO WILL RESULT IN REPORTING YOUR POST!!![/color][/b]
ALSO: This is a discussion about PvP, not other aspects of the game like ai. Stay on topic or your post WILL BE REPORTED!
honestly pvp is the most educational and largest revenue for this game I personally learned more from pvp about the galme and am willing to spend money on the game for pvp. The pvp community wants more then anything balanced game play they have honestly as much knowledge of the game as the developers do and that's because they actually care about the game.
I pvp regularly the biggest killer of pvp right now is the constant exotic damage buffs. The pvp queues pop until exotic spam builds star queing. 30k+ feedback pulse builds are low balling what is happening. 50 to 60k feed back pulse is the minimal and that literally one shots people. When surgical strikes did damage like that it got nerfed feed back pulse should get the same simply because it excides that damage threshold...
Cryptic has hurt itself by alienating the pvp community. they spend the most money and care the most about the game.
The Pvp'ers want whats best for the longest running game killing pvp WILL KILL STO!!!!
Why would you want to.remove my favorite pass-time? There is nothing more fun then setting up a handicap match for 1v5 (or more) if the team fighting you are not the typical I hate you kind of player they will see you as a boss fight. (I get this quite a bit actually and have requests to bring in friends which sometimes gets me into the 1v8 area)
Love the comment "to counter FBP just stop firing!" Yes! That is the way to counter it, but when FBP deals 150% of damage dealt that's a little difficult. It's also difficult to spot the FBP icon out of the string of 30 buffs and traits.
Seriously, do some actual PvP isntead of commenting in ignorance.
And yet another post/thread that devolves into PvP is all about the FBP nerf it to death or kill.it outright! *sigh* main issue is: 1.) if you weren't rocking 80k damage per beam.shot on a FAW build you wouldn't get vaped by a counter beam. 2.) With the exception of all the tactical buffs to even get a high return on FBP you need A LOT of Particle Generators!
I don't use a a FAW build and my DPS probably peaks at 30k in my t5u defiant. my DBB is acc3, my 2 DHC are acc2/dmg2 and 1 torp. With a bo2 I've peaked between 70-80k but that's 1 hit followed by crf2. But I've been hit with 140k FBP. Nearly twice the damage I've dealt.
actually - try to use polarize hull when enemy activates FBP - I think it negates the damage.
Also: don'T take a fragile high DPS machine into PvP. my PvP builds do a bitt less damage but survive longer. Balance is the key - unless you go for a sniper, but then you have to know, when to attack (have one, it'S fun, but can go horribly wrong)
18 characters
KDF: 2 tacs, 2 engs, 3 scis
KDF Roms: 3 tacs, 1 eng, 1 scis
FED: 2 tacs, 1 eng, 2 scis
TOS: 1 tac
all on T5 rep (up to temporal)
all have mastered Intel tree (and some more specs Points)
highest DPS: 60.982
i can't do less damage without removing weapons. My ship does resistance and speed tanking, I have science team 1 and hazards to counter sci disables and heal.
You could lower your power setting....swap to a torp (can't FBP kinetic/exotic/special damage types)...watch for cycling of FBP (the effect looks like lines of light moving into the ships hull)
So if I understand you correctly, you are so much ahead of me (and anybody in AI research that I have ever read or heard) that you can't even explain how much more you know.
Well... "If you can't explain it simply, you don't understand it well enough." (Albert Einstein)
So I'll try to explain it simply. You are aware that navigation in video games is basically a super-accurate version of using a GPS device, yes? You get coordinates from the game engine, so the agent knows, without a shade of doubt, where it actually is (which is not what real-world navigation is like, not even for humans, not even with GPS). Also, it gets (or can get, if the game is set up this way) all relevant map data handed over - in a real world situation, of course, it would require lots of computational resources just to aquire that information.
So, all your agent needs to do now for STO space combat is
Set up the right build (abilties, consoles, power levels, etc.). That is, optimize the damage output, resistances, and heals of the ship. This can be even done by a different AI once in a while (for example, the devs could have an AI change the Borg Queen's diamond ship build every week, based on all available data).
In actual combat, put itself in a relative position that optimizes its own damage potential vs the opponent's damage potential. Simple example: "Get behind that all-cannon escort and stay there."
Use the available abilties for maximum effect, such as "Use tractor beam on that all-cannon escort right after all its counters to that have expired, continue firing everything you got until it explodes."
Of these three things, only the second point might require something that even qualifies as AI, because the other two are more like rather simple math problems: You try out all possible combinations and compute the results, then compare them.
Have you followed the whole "Alpha Go" thing? Or the Deep Blue chess duel?
One component of all game AI is basically what is also called a "Monte Carlo" simulation. It means basically considering all the possible moves and evaluate the outcome. And not just the next move - but follow the possible moves down to what is possible then. All of this requires a considerably amount of computing power. The possible moves lead to a tree - each path through the tree representing a sequence of moves, and the entire tree representing all possible moves. Since there is only limited computation time allowed, a chess computer needs a strategy on how to traverse the tree, e.g. which moves to try first, and at what point it should return to a previous move and consider alternate moves. And of course, you also need an evalulation algorithm - to determine whether a position is good (ideally a win) or bad (worst case: a loss).
We're at a point where Monte Carlo simulations on a dedicated computer can beat any grand chess master.
But we couldn't do the same for Go. It was more complex, monte carlo simulation alone couldn't go deep enough to predict an experienced player strategy.
So we went to another area of AI research - arguably where the "real" AI lies. Neuronal Networks. Neuronal Networks seem to be much closer to how we as humans learn. A neuronal network must be trained with data. They basically get an input, and if they pick the right choices (be it a winning move, or identify an object correctly, or whatever) the decision making process that made this decision is reinforced, otherwise it's weakened. One has to be careful though - depending on the type of input data you feed it, and the complexity of the network (e.g. the amount of neurons), you might actually not teach it what you wanted to teach it. An old example of the beginnings of neuronal network research is that of training a network to identify tanks on a air surveillance photo. They seemed to get good results, until they realized that the thing the neuronal network actually identified was night and day photos - the input data happened to be images where tanks were only on the day maps (or vice versa).
Alpha Go was a way to combine both the Neuronal Network approach and the Monte Carlo approach. If I understand things correctly, the neuronal network's job was to pick how to traverse the monte carlo tree and how to evaluate the position.
Now, here are some challenges that I see when applying "real" AI technology to a game:
Chess and Go are turn based games. You get some time to think about your approach, and then pick one move.
STO is real time. You don't get to think 10 seconds before you decide whether to activate EPtS or Evasive Maneuvers or Subnucleonic Beam. You must do it _now_ or miss your opportunity. (Or you must not do it now, because it would be wasted).
Chess and Go have a very limited number of options and pieces. Go has only one type of piece, but a lot of options how to place them. Chess has several pieces, but each piece has only a small number of possible moves in any given turn.
Star Trek Online has a lot of options. A vast amount of powers.
In Chess and Go players have a known option of moves, e.g. all pieces available to the player and their positions are visible.
In STO, Player abilities are not known beforehand since everyone can have his own build. (unless the computer gets to cheat and know your build.)
Go and Chess are two player games, not 5 vs 5 player games.
Alpha Go most simple hardware platform was a 48 processor with a single GPU. Even if for some reason assume that the "Alpha Star Trek Online" AI needs less of all that, all things point to needing roughly one server per simulated NPC.
Star Trek Online Advancement: You start with lowbie gear, you end with Lobi gear.
This is not a place to continue arguing about one point, or talking mean about someone (like in the above post). Do not continually argue every post put in the discussion, or create a long chain of arguments off of one post!!! Also, BE RESPECTFUL TO OTHERS AND DO NOT TALK TO THEM IN OFFENSIVE TERMS! DOING SO WILL RESULT IN REPORTING YOUR POST!!!
ALSO: This is a discussion about PvP, not other aspects of the game like ai. Stay on topic or your post WILL BE REPORTED!
Comments
Well, you seemed pretty upset about them, so I just assumed.
If you bothered to read any of my posts, I repeatedly say to build for fun. Personally I don't give a rat's TRIBBLE if they adjust items for the sake of balance...what I DO care about is they do so for the sake of PvP balance when it make things worse for the PvE people. And quite frankly, this is gonna be true for most PvE exclusive players (which is the vast majority of this playerbase). That means your whole idea of have somebody who know to balance between the two WILL NOT WORK. Even if they are GREAT at their job...it's just a bad idea from the start.
Yep, it's not easy for them to strike the right balance when making changes on the fly. Sometimes they make terrible mistakes, and are too eager to get a fix in, and end up botching the job.
You know what though ? Player feedback made science pretty powerful once again. So it's not always a losing situation.
What really needs to happen is one of two things. EVERYTHING gets two stats. One for PvP and one for PvE so they can adjust each without it affecting the other...or we make PvP a seperate environment. That is to say, you buy PvP ships that comes pre-loaded to be balanced with each other and that is all you are allowed in PvP.
I've been hoping for something like this myself. It's probably the easiest solution.
No Traits Mode. And Stock Ship Mode
Even so, there's no reason they can't also leave the current version in game.
Let the whales beat up on each other if they so choose.
The separate PvP server or two versions of each ability, as you suggest would be nice in theory, but even I don't think it would be a good investment of resources.
I suppose there is the completely unrealistic solution of have NPCs that are specced like players, use player builds and have playerlike AI...but consider that AI does not exist even as an experimental piece of tech...that is a completely unrealistic solution at this time. So since the first solution requires WAY more dev time then they have, the second one has been rejected by the majority of the PvP playerbase and the third one is not even possible with all the money and tech in the world...yeah...it just needs to go away.
I would rather they left PvP in. Even if they let it rot, and never funded any future development ever again.
Like @davefenestrator said;
"Keep PVP but let it continue to rot, and only make game balance changes / nerfs for PVE issues not PVP.
Anyone wanting to PVP can still do it, but it will be up to them to agree on the rules for the duel --"
Which is kinda the case now anyway, right ?
Why take it away completely from the people who like it ?
I don't see the logic.
Personally, I don't like when they take away missions and queues, (Ie; NWS/Mine Trap/Mirror Invasion XP/Breach/some of the old story arcs etc), and I wouldn't like losing a whole game mode.
Especially if it's for no logical reason for it.
Oh and we already do have the one true build. BFAW, AP beams FTW. So it's not like the elachi weapons would have made a difference there. The only reason to NOT do BFAW, AP beams right now is because you have more EC/dil/zen to spend just because.
No, Elachi weapons were outperforming even APs at that time. and by a lot, like 20-25% DPS (all things being equal at the time, Elachi CrtDx3 vs AP CrtDx3).
But Yeah, currently, there is no equal to APs, a true min/maxer will go for that combo on the PvE side, if they know what they are doing.
But......
The reason we have build diversity isn't a matter of weapons being similar in power. They ain't. AP beams kick the living snot out of everything else until you get into EXTREMELY costly builds. Hell just BEAMS kick the living snout of other weapons. And don't even get me started on the issues with BFAW. Mines are nigh useless compared to other weapons. But you know what...we can make a viable mine ship...why? Because the content is just that stupid easy.
Now, while I can agree the AP/Faw builds are at the top of the food chain, I don't agree that they;
"Kick the living snot out of everything else until you get into extremely costly builds." That is just not true.
Did you know the difference between an AP build and almost any other beam build is about 5-10% DPS ?
Yes, IT IS better, I won't argue that, but let's not over exaggerate just to make a point.
It is concerning though. I don't disagree.
Yeah eventually the players would crack the code.
Either by finding the "path of least resistance" or buying power creep. The cycle would continue a new.
[b]This is not a place to continue arguing about one point, or talking mean about someone (like in the above post). [color="purple"] Do not continually argue every post put in the discussion, or create a long chain of arguments off of one post!!! Also, BE RESPECTFUL TO OTHERS AND DO NOT TALK TO THEM IN OFFENSIVE TERMS! DOING SO WILL RESULT IN REPORTING YOUR POST!!![/color][/b]
ALSO: This is a discussion about PvP, not other aspects of the game like ai. Stay on topic or your post WILL BE REPORTED!
Thanks for your respectful opinions about PvP!!!
-Q
(hint: click the image to go to my youtube channel)
http://twitch.tv/profile/quantum_quantonium
http://plus.google.com/+quantumquantonium
http://quantumquantonium.blogspot.com
http://quantonium.tk
EPIC SCREENSHOTS
I pvp regularly the biggest killer of pvp right now is the constant exotic damage buffs. The pvp queues pop until exotic spam builds star queing. 30k+ feedback pulse builds are low balling what is happening. 50 to 60k feed back pulse is the minimal and that literally one shots people. When surgical strikes did damage like that it got nerfed feed back pulse should get the same simply because it excides that damage threshold...
Cryptic has hurt itself by alienating the pvp community. they spend the most money and care the most about the game.
The Pvp'ers want whats best for the longest running game killing pvp WILL KILL STO!!!!
Seriously, do some actual PvP isntead of commenting in ignorance.
1.) if you weren't rocking 80k damage per beam.shot on a FAW build you wouldn't get vaped by a counter beam.
2.) With the exception of all the tactical buffs to even get a high return on FBP you need A LOT of Particle Generators!
Also: don'T take a fragile high DPS machine into PvP. my PvP builds do a bitt less damage but survive longer. Balance is the key - unless you go for a sniper, but then you have to know, when to attack (have one, it'S fun, but can go horribly wrong)
KDF: 2 tacs, 2 engs, 3 scis
KDF Roms: 3 tacs, 1 eng, 1 scis
FED: 2 tacs, 1 eng, 2 scis
TOS: 1 tac
all on T5 rep (up to temporal)
all have mastered Intel tree (and some more specs Points)
highest DPS: 60.982
One component of all game AI is basically what is also called a "Monte Carlo" simulation. It means basically considering all the possible moves and evaluate the outcome. And not just the next move - but follow the possible moves down to what is possible then. All of this requires a considerably amount of computing power. The possible moves lead to a tree - each path through the tree representing a sequence of moves, and the entire tree representing all possible moves. Since there is only limited computation time allowed, a chess computer needs a strategy on how to traverse the tree, e.g. which moves to try first, and at what point it should return to a previous move and consider alternate moves. And of course, you also need an evalulation algorithm - to determine whether a position is good (ideally a win) or bad (worst case: a loss).
We're at a point where Monte Carlo simulations on a dedicated computer can beat any grand chess master.
But we couldn't do the same for Go. It was more complex, monte carlo simulation alone couldn't go deep enough to predict an experienced player strategy.
So we went to another area of AI research - arguably where the "real" AI lies. Neuronal Networks. Neuronal Networks seem to be much closer to how we as humans learn. A neuronal network must be trained with data. They basically get an input, and if they pick the right choices (be it a winning move, or identify an object correctly, or whatever) the decision making process that made this decision is reinforced, otherwise it's weakened. One has to be careful though - depending on the type of input data you feed it, and the complexity of the network (e.g. the amount of neurons), you might actually not teach it what you wanted to teach it. An old example of the beginnings of neuronal network research is that of training a network to identify tanks on a air surveillance photo. They seemed to get good results, until they realized that the thing the neuronal network actually identified was night and day photos - the input data happened to be images where tanks were only on the day maps (or vice versa).
Alpha Go was a way to combine both the Neuronal Network approach and the Monte Carlo approach. If I understand things correctly, the neuronal network's job was to pick how to traverse the monte carlo tree and how to evaluate the position.
Now, here are some challenges that I see when applying "real" AI technology to a game:
STO is real time. You don't get to think 10 seconds before you decide whether to activate EPtS or Evasive Maneuvers or Subnucleonic Beam. You must do it _now_ or miss your opportunity. (Or you must not do it now, because it would be wasted).
Star Trek Online has a lot of options. A vast amount of powers.
In STO, Player abilities are not known beforehand since everyone can have his own build. (unless the computer gets to cheat and know your build.)
Learning is hard, but that shouldn't discourage them from continuing to try.
This is not a place to continue arguing about one point, or talking mean about someone (like in the above post). Do not continually argue every post put in the discussion, or create a long chain of arguments off of one post!!! Also, BE RESPECTFUL TO OTHERS AND DO NOT TALK TO THEM IN OFFENSIVE TERMS! DOING SO WILL RESULT IN REPORTING YOUR POST!!!
ALSO: This is a discussion about PvP, not other aspects of the game like ai. Stay on topic or your post WILL BE REPORTED!
Thanks for your respectful opinions about PvP!!!
-Q
(hint: click the image to go to my youtube channel)
http://twitch.tv/profile/quantum_quantonium
http://plus.google.com/+quantumquantonium
http://quantumquantonium.blogspot.com
http://quantonium.tk
EPIC SCREENSHOTS