Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 30 40 50 .. 51 :: one page |
|
Author |
Thread Statistics | Show CCP posts - 7 post(s) |
Teckos Pech
The Executives Executive Outcomes
4752
|
Posted - 2016.02.23 17:58:59 -
[1111] - Quote
Crackforbreakfast wrote:What seems to be forgotten by the salty gankbro's in this thread pulling up 2014 RFF statistics is that the ganks are mostly happening in a very select few systems, the Niarja and Uedama pipes with the surrounding few systems because these are connecting the major trade hubs to one another.
The statistics of RFF2014 also include ALL of the other contracts not travelling through those gank choke-points (I'm not sure if gankbro's are just playing dumb or really don't understand this). Scouting will only help so much as the pipe consist of multiple jumps judging by the map, this means that a Mach could be waiting 3-4 jumps ahead while you get scanned down in the current system to see if you're worth ganking. This is furthermore hindered with the fact that Freighter warps are slow, multiple minutes on occasion, in that time a lot of relocating can be done by a Mach with it's warp speed increase bonus. Even if you have a webber, the fact you have to go multiple jumps through the pipe and can be caught up to means your webber can also be ganked.
What would be the solution gankbro's? Three scouts? Two webbers? So a total of six accounts to avoid 160mil worth of Catalysts that can kill you if they desire to do so while the risk free Machariel happily bumps you. Or would you need four Tornado accounts to alpha the Mach of the field to keep you safe?
Twisting the facts (or misinterpreting them purposefully would be a nicer wording) in regards to the safety of freighter pilots is not making you look like the sharpest tool in the shed so best to stop that.
My sincere apologies if you're not purposefully misinterpreting them and are actually not understanding how to read into them.
This is a load of errant nonsense that any statistician would laugh at.
Yes, RFF really only went through Uedama 245 in 2014 and every single time they were ganked. [/sarcasm]
The point is, even if we limited ourselves to those instances when RFF went through Uedama and Niajara we would doing the following,
Gank rate = 245/a big number = a small....very small number. Maybe not 0.1% but probably less than 1%.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Ylmar
Spontaneous Massive Existence Failure
47
|
Posted - 2016.02.23 18:15:27 -
[1112] - Quote
Teckos Pech wrote:Red Frog's own annual reports are not good enough? The "creative" interpretation is not good enough, which has been explained several times before. |
baltec1
Bat Country Pandemic Legion
17447
|
Posted - 2016.02.23 18:41:23 -
[1113] - Quote
Ylmar wrote:Teckos Pech wrote:Red Frog's own annual reports are not good enough? The "creative" interpretation is not good enough, which has been explained several times before.
The only people getting creative are the ones who think just two systems should count for the entirety of highsec. |
Lucas Kell
Internet Terrorists SpaceMonkey's Alliance
7233
|
Posted - 2016.02.23 19:50:41 -
[1114] - Quote
baltec1 wrote:Ok I'll do it for you then.
Your ranis fit if you get rid of the DCU II and go with a EANM II you lose several hundred EHP.
Looking at the new Imperium mach you will lose around 10k EHP by doing the same.
Your FYF celestis is 2k ehp worse off if we go with the next best thing which is a 400mm plate.
Your Drake is also worse off without the DCU to the tune of over 10k ehp. Again though, that's all great to EFT warrior, but the reality is that it will change how the average player chooses their fits. They will be less likely to choose the DC simply because it has less of an extreme effect.
The Indecisive Noob - EVE fan blog.
Wholesale Trading - The new bulk trading mailing list.
|
Teckos Pech
The Executives Executive Outcomes
4753
|
Posted - 2016.02.23 20:02:14 -
[1115] - Quote
Crackforbreakfast wrote:baltec1 wrote: Both of those systems can be avoided or the risk in the reduced to virtually zero. All of my ships are bought in jita and pass through those systems and not once have I lost a shipment. Equally my industrial alt has never been ganked in said systems despite traveling through them twice a day, nearly every day. The reason RFF get use is because it is the single largest body of date available on freighter trips and the likelihood of failing a contract. Both systems can not be avoided unless you're going through low-sec with your freighter, to get from Jita to Amarr you either have to go through Niarja, or go 40-ish jumps around and go through Uedama. ; see http://evemaps.dotlan.net/route/2:Jita:Amarr:-Niarja:-Uedama for reference, the same goes for Jita to Dodixie http://evemaps.dotlan.net/route/2:Jita:Dodixie:-Uedama:-Niarja, and thus also for Jita to Hek http://evemaps.dotlan.net/route/2:Jita:Hek:-Uedama:-Niarja.
Also the fact you're "using" RFF data does not mean you're capable of interpreting it right, which leads me to believe you actually don't understand the previously made argument as to how you (and others) are misinterpreting the statistics. How about using the most recent and best updated source you can find on where these freighters actually die, regardless of if they're RFF freighters or not; https://zkillboard.com/kills/freighters/. Hard to deny the trend much? This does not include statistics on how many freighters pass through, but is merely to show that on these popular routes you WILL have to go through one of two systems, Niarja or Uedama, and at those points, have a fair chance of getting picked off. The fact is the more interesting freighters (ISK-wise) during times of activity be ganking groups are being picked off. (logically) However the current EHP of the freighters allows this to be done at such a low cost that the hull resistance buff will even it out a bit. You'll still have your easy life in regards that there are only two pipes for these major hubs, just have to be a bit pickier on what you decide to blap by pressing F1. Instead of 160 mil worth of Catalysts you might need 240mil worth of Catalysts, or just up your game and only go for the juicy targets using Talos if fielding that amount of players seems unmanageable.
First off zkill does not show us a trend. Nor does it show us what the rate of ganking is. Nor do I think it can show us that because it does not log freighters traveling around and not dying.
For somebody who is complaining about the misuse of statistics these posts make you look completely incompetent and incoherent when it comes to statistics.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Crackforbreakfast
Imperial Academy Amarr Empire
7
|
Posted - 2016.02.23 20:13:28 -
[1116] - Quote
Teckos Pech wrote:Crackforbreakfast wrote:baltec1 wrote: Both of those systems can be avoided or the risk in the reduced to virtually zero. All of my ships are bought in jita and pass through those systems and not once have I lost a shipment. Equally my industrial alt has never been ganked in said systems despite traveling through them twice a day, nearly every day. The reason RFF get use is because it is the single largest body of date available on freighter trips and the likelihood of failing a contract. Both systems can not be avoided unless you're going through low-sec with your freighter, to get from Jita to Amarr you either have to go through Niarja, or go 40-ish jumps around and go through Uedama. ; see http://evemaps.dotlan.net/route/2:Jita:Amarr:-Niarja:-Uedama for reference, the same goes for Jita to Dodixie http://evemaps.dotlan.net/route/2:Jita:Dodixie:-Uedama:-Niarja, and thus also for Jita to Hek http://evemaps.dotlan.net/route/2:Jita:Hek:-Uedama:-Niarja.
Also the fact you're "using" RFF data does not mean you're capable of interpreting it right, which leads me to believe you actually don't understand the previously made argument as to how you (and others) are misinterpreting the statistics. How about using the most recent and best updated source you can find on where these freighters actually die, regardless of if they're RFF freighters or not; https://zkillboard.com/kills/freighters/. Hard to deny the trend much? This does not include statistics on how many freighters pass through, but is merely to show that on these popular routes you WILL have to go through one of two systems, Niarja or Uedama, and at those points, have a fair chance of getting picked off. The fact is the more interesting freighters (ISK-wise) during times of activity be ganking groups are being picked off. (logically) However the current EHP of the freighters allows this to be done at such a low cost that the hull resistance buff will even it out a bit. You'll still have your easy life in regards that there are only two pipes for these major hubs, just have to be a bit pickier on what you decide to blap by pressing F1. Instead of 160 mil worth of Catalysts you might need 240mil worth of Catalysts, or just up your game and only go for the juicy targets using Talos if fielding that amount of players seems unmanageable. First off zkill does not show us a trend. Nor does it show us what the rate of ganking is. Nor do I think it can show us that because it does not log freighters traveling around and not dying. For somebody who is complaining about the misuse of statistics these posts make you look completely incompetent and incoherent when it comes to statistics.
Easy to quote, hard to read it seems, I put it in bold for you so you can indeed see I stated exactly what you said. Also, how is it not a trend if the majority of high-sec freighter kills occurs in and around the pipes? To blame someone for misusing statistics, be sure to read properly first.
|
Teckos Pech
The Executives Executive Outcomes
4753
|
Posted - 2016.02.23 20:38:44 -
[1117] - Quote
Crackforbreakfast wrote:
Easy to quote, hard to read it seems, I put it in bold for you so you can indeed see I stated exactly what you said. Also, how is it not a trend if the majority of high-sec freighter kills occurs in and around the pipes? To blame someone for misusing statistics, be sure to read properly first.
That is precisely what makes your posts so ironic and continue to highlight why you are a bumbling and incompetent when it comes to statistics. If you want to show that there is an upward trend in freighter ganking, then you have to look at the rate. Freighters ganked/number of freighters. We could look look at it on a per system basis as well, but again it is the rate we need. If the total number of freighters is increasing and the number of freighters ganked is also increasing one possibility is that there is no significant increase in the rate at which freighters get ganked. There is no trend.
Further, the fact that most freighter ganks (note you need to take out freighters killed for other reasons such a war dec or just doing something damn dumb) take place in 2 systems is not a trend. You are confusing trend with tendency or likelihood. Yet another reason you sound not just incompetent but incoherent.
What we want to know is,
P(G|X).
That is the probability of being ganked given the characteristics X (think of it as a tuple where each element contains things like do you have a scout, does your scout have webs, do you use perches, etc.). We don't and cannot observe this. But we can observe:
P(X|G).
That is what is the probability you had a certain characteristic type given you were ganked. Then we can, if we had the data, use Bayes Theorem to come up with P(G|X), that is,
P(G|X) = P(X|G)*[P(G)/P(X)].
We don't really have all this data.
What we do have is data from RFF. So people have been pointing to it and making the following claim:
RFF is prudent in its use of freighters. RFF does not die very much. They conclude that prudence reduces the risk of flying a freighter in regards to ganking. It is not at all an unreasonable conclusion to make.
Pointing to an incomplete data set and asking why nobody is using that is just stupid beyond belief.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Masao Kurata
Perkone Caldari State
413
|
Posted - 2016.02.23 20:59:28 -
[1118] - Quote
Teckos Pech wrote:That is what is the probability you had a certain characteristic type given you were ganked. Then we can, if we had the data, use Bayes Theorem to come up with P(G|X), that is,
But, like, that's just a theorem, man.
I somehow think that probability theory will be wasted on this audience :) |
Teckos Pech
The Executives Executive Outcomes
4754
|
Posted - 2016.02.23 21:06:05 -
[1119] - Quote
Masao Kurata wrote:Teckos Pech wrote:That is what is the probability you had a certain characteristic type given you were ganked. Then we can, if we had the data, use Bayes Theorem to come up with P(G|X), that is, But, like, that's just a theorem, man. I somehow think that probability theory will be wasted on this audience :)
You wound me sir...you are cutting deep into my Bayesian heart.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Crackforbreakfast
Imperial Academy Amarr Empire
7
|
Posted - 2016.02.23 21:09:06 -
[1120] - Quote
Using a data set which is not valid in regards as to what is in question is beyond stupid, furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral. You can keep on rambling about your RFF data.
Zkill is showing the following: The majority of high-sec freighter kills happen in the pipes, that is what can be concluded from the data at hand. The fact that these systems are camped during high times of activity makes the chance RELATIVELY high to get ganked in a freighter. Especially since the costs of ganking one are so low.
Have fun keyboard warrioring in this thread regardless, and props to CCP for this re-balance. |
|
Lucas Kell
Internet Terrorists SpaceMonkey's Alliance
7233
|
Posted - 2016.02.23 21:47:41 -
[1121] - Quote
Teckos Pech wrote:RFF is prudent in its use of freighters. RFF does not die very much. They conclude that prudence reduces the risk of flying a freighter in regards to ganking. It is not at all an unreasonable conclusion to make. Of course it's an unreasonable conclusion to make. Again, that's because RFF are only able to have high survival rates by being less appealing than other targets, so the stats first off have absolutely no relation to the EHP of freighters and also are not representative of freighting in general, in fact they are quite the opposite.
Here. Imagine you have a bowl of apples and a bowl of pears for feeding the hungry, and you know full well that people like both an equal amount. Every person that arrives takes two pieces of fruit, one from each bowl. As people arrive, the two bowls deplete at the same rate. One day you start putting mould and dirt around the edge of the bowl containing pears. Now some people aren't put off and take their two different fruits as normal, however most of the people now take two apples instead of one of each. At this point, the fruits overall are still depleting at exactly the same rate, but if you were to look only at the pears and form an opinion over how much people like fruit from that single statistic, you'd get the impression they don't like fruit. If however the apples ran out, people would still take the pears as they are hungry and need to eat.
This is exactly the problem with the RFF stats. They are not only skewed because of the additional safety precautions that their pilots take, but because pilots who aren't in RFF which we don't have stats for are likely killed at a much higher rate, because when given the choice between the two, non-RFF pilots are generally more appealing. In order for RFF to achieve their level of safety, other haulers have to be destroyed in their place.
Again though, when discussing this change it's entirely irrelevant, since EHP is not what kept those pilots from being shot down. Reduced cargo and webbing alts were much bigger factors, and so the increase in EHP is not relevant. It's just another way for people who want to refuse to adapt to twist the subject to push their agenda. They'll just have to learn to be better EVE players and get over it. I'm sure they'll manage.
The Indecisive Noob - EVE fan blog.
Wholesale Trading - The new bulk trading mailing list.
|
Daichi Yamato
Xero Security and Technologies
3116
|
Posted - 2016.02.23 21:51:55 -
[1122] - Quote
Crackforbreakfast wrote:Using a data set which is not valid in regards as to what is in question is beyond stupid, furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral. You can keep on rambling about your RFF data.
Zkill is showing the following: The majority of high-sec freighter kills happen in the pipes, that is what can be concluded from the data at hand. The fact that these systems are camped during high times of activity makes the chance RELATIVELY high to get ganked in a freighter. Especially since the costs of ganking one are so low.
Have fun keyboard warrioring in this thread regardless, and props to CCP for this re-balance.
So after all that, you're just saying that you are more likely to be ganked in uedama and niarja at times gankers are active than the rest of hi-sec? Well no ****. Another incredible contribution by an anti-ganker.
If only there was a way to not travel in those systems whilst not afk...
EVE FAQ "7.2 CAN I AVOID PVP COMPLETELY? No; there are no systems or locations in New Eden where PvP may be completely avoided"
Daichi Yamato's version of structure based decs
|
Teckos Pech
The Executives Executive Outcomes
4759
|
Posted - 2016.02.23 22:19:03 -
[1123] - Quote
Crackforbreakfast wrote:Using a data set which is not valid in regards as to what is in question is beyond stupid, furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral. You can keep on rambling about your RFF data.
Zkill is showing the following: The majority of high-sec freighter kills happen in the pipes, that is what can be concluded from the data at hand. The fact that these systems are camped during high times of activity makes the chance RELATIVELY high to get ganked in a freighter. Especially since the costs of ganking one are so low.
Have fun keyboard warrioring in this thread regardless, and props to CCP for this re-balance.
Why is it not valid. You say this as if it is trivially obvious. Are you telling us RFF never go through one of these camped pipes? Ever?
And offering collateral makes the pilots more prudent....which in turn impacts their rate of being ganked. Thank you for proving my point.
Finally, has anyone disputed that ganks usually take place in the pipes through Uedama and Niarja?
And ganking is not low cost. It is relatively low cost. That is, 200 million ISK in catalysts to kill a freighter full of 5 billion ISK of goodies is why ganking is "low cost" the costs for ganking is low relative to the gains. And the reason why the gains are high is because a player put 5 billion ISK in a big fat slow moving ship that has no offensive capabilities without even a single ship for an escort. There is a word for that: idiot.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
baltec1
Bat Country Pandemic Legion
17460
|
Posted - 2016.02.24 08:58:27 -
[1124] - Quote
Lucas Kell wrote:baltec1 wrote:Ok I'll do it for you then.
Your ranis fit if you get rid of the DCU II and go with a EANM II you lose several hundred EHP.
Looking at the new Imperium mach you will lose around 10k EHP by doing the same.
Your FYF celestis is 2k ehp worse off if we go with the next best thing which is a 400mm plate.
Your Drake is also worse off without the DCU to the tune of over 10k ehp. Again though, that's all great to EFT warrior, but the reality is that it will change how the average player chooses their fits. They will be less likely to choose the DC simply because it has less of an extreme effect.
This less of an extreme effect being 0.02% less than today. Feel free to not use the DCU but to say it's going to be less effective than today is just a lie. |
baltec1
Bat Country Pandemic Legion
17460
|
Posted - 2016.02.24 09:00:20 -
[1125] - Quote
Crackforbreakfast wrote:Using a data set which is not valid in regards as to what is in question is beyond stupid, furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral. You can keep on rambling about your RFF data.
Zkill is showing the following: The majority of high-sec freighter kills happen in the pipes, that is what can be concluded from the data at hand. The fact that these systems are camped during high times of activity makes the chance RELATIVELY high to get ganked in a freighter. Especially since the costs of ganking one are so low.
Have fun keyboard warrioring in this thread regardless, and props to CCP for this re-balance.
How many freighters pass through your two sample systems unmolested? |
Lucas Kell
Internet Terrorists SpaceMonkey's Alliance
7253
|
Posted - 2016.02.24 19:27:54 -
[1126] - Quote
baltec1 wrote:This less of an extreme effect being 0.02% less than today. Feel free to not use the DCU but to say it's going to be less effective than today is just a lie. If it's 0.02% less effect then saying it's going to be less effective can't be a lie. Not really sure where you've got the 0.02% from but mate, you're still talking about EFT warrior fits where the only important factor is tank. Someone that removed a DC and decides that with the natural hull resist boost they are tanky enough to instead increase another stat will be the types of players who will drop it.
I'm really not sure what you are hoping to achieve here anyway. Would you say that perhaps CCP should drop of some more of the DCs effects and naturally add those to all ships too? I'm game for that.
The Indecisive Noob - EVE fan blog.
Wholesale Trading - The new bulk trading mailing list.
|
Chip Flux
Center for Advanced Studies Gallente Federation
48
|
Posted - 2016.02.24 19:42:29 -
[1127] - Quote
Why must we look forward to your changes with trepidation? Why must you always upset your customers with these horrible changes?
This is now an attack on energy neutralizers and the core gameplay
Just stop the stupid |
Teckos Pech
The Executives Executive Outcomes
4765
|
Posted - 2016.02.24 21:01:39 -
[1128] - Quote
baltec1 wrote:Crackforbreakfast wrote:Using a data set which is not valid in regards as to what is in question is beyond stupid, furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral. You can keep on rambling about your RFF data.
Zkill is showing the following: The majority of high-sec freighter kills happen in the pipes, that is what can be concluded from the data at hand. The fact that these systems are camped during high times of activity makes the chance RELATIVELY high to get ganked in a freighter. Especially since the costs of ganking one are so low.
Have fun keyboard warrioring in this thread regardless, and props to CCP for this re-balance. How many freighters pass through your two sample systems unmolested?
He took his ball and went home once he realized he could not answer that question and that it was fundamental to the discussion.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Xe'Cara'eos
A Big Enough Lever
408
|
Posted - 2016.02.24 21:27:33 -
[1129] - Quote
Is it me or will this change act as a nerf to indiscriminate freighter ganking, whilst probably allowing the clever/selective gankers to carry on pretty much as normal (with more needed per gank)?
Also - has anyone yet done the maths on how long a gankalyst will now last against concord? Or are the rumours of concord doing a 5th damage type true?
For posting an idea into F&I:
come up with idea, try and think how people could abuse this, try to fix your idea - loop the process until you can't see how it could be abused, then post to the forums to let us figure out how to abuse it.....
If your idea can be abused, it [u]WILL[/u] be.
|
Teckos Pech
The Executives Executive Outcomes
4765
|
Posted - 2016.02.25 05:12:05 -
[1130] - Quote
Xe'Cara'eos wrote:Is it me or will this change act as a nerf to indiscriminate freighter ganking, whilst probably allowing the clever/selective gankers to carry on pretty much as normal (with more needed per gank)?
Also - has anyone yet done the maths on how long a gankalyst will now last against concord? Or are the rumours of concord doing a 5th damage type true?
That will likely be the effect. Ganking empty freighters will likely go the way of the dodo, and ganking groups will look for freighters that meet some minimal requirements.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
|
Black Pedro
Yammerschooner
2222
|
Posted - 2016.02.25 05:51:25 -
[1131] - Quote
Teckos Pech wrote:Xe'Cara'eos wrote:Is it me or will this change act as a nerf to indiscriminate freighter ganking, whilst probably allowing the clever/selective gankers to carry on pretty much as normal (with more needed per gank)?
Also - has anyone yet done the maths on how long a gankalyst will now last against concord? Or are the rumours of concord doing a 5th damage type true? That will likely be the effect. Ganking empty freighters will likely go the way of the dodo, and ganking groups will look for freighters that meet some minimal requirements. I am not sure this will be case. Nobody, even CODE., goes out to hunt empty freighters. A freighter with cargo will always be preferred over an empty one all other things being equal and empty freighters are exploded primarily just because nothing else is available to keep the fleet busy, or because the pilot is red to their coalition in the case of Miniluv. Both reasons are still valid and any increase in cost will be absorbed to keep moral up, or to serve a greater sandbox reason and empty freighters will still explode.
This change will have little effect as long as the large ganking groups still have enough pilots to be over the group size threshold needed to field enough DPS. In fact, with PL now seeming to be experimenting with ganking, and even ganking empty freighters, it appears even more empty freighters might be dying in the near future. All this does is lock out smaller groups even more from choosing to hunt freighters in highsec.
The fundamental problem can't balance ganking with EHP beyond a certain point when players can just bring more pilots. All that does is restrict the ability to attack in highsec to the largest groups in the game, and given the wide disparity in group sizes, you end up eventually locking out most players from using the mechanic. How is it fair that Goonswarm can field sufficient pilots to gank your freighters, but your 30-man corp now no longer has enough people to meet this arbitrary DPS number to attack their freighters back? I mean, if you get in a real fight with such a size difference you should be crushed, but you don't even have the ability to try and force a response because of an absurdly high EHP wall. |
Teckos Pech
The Executives Executive Outcomes
4765
|
Posted - 2016.02.25 05:53:36 -
[1132] - Quote
Black Pedro wrote:Teckos Pech wrote:Xe'Cara'eos wrote:Is it me or will this change act as a nerf to indiscriminate freighter ganking, whilst probably allowing the clever/selective gankers to carry on pretty much as normal (with more needed per gank)?
Also - has anyone yet done the maths on how long a gankalyst will now last against concord? Or are the rumours of concord doing a 5th damage type true? That will likely be the effect. Ganking empty freighters will likely go the way of the dodo, and ganking groups will look for freighters that meet some minimal requirements. I am not sure this will be case. Nobody, even CODE., goes out to hunt empty freighters. A freighter with cargo will always be preferred over an empty one all other things being equal and empty freighters are exploded primarily just because nothing else is available to keep the fleet busy, or because the pilot is red to their coalition in the case of Miniluv. Both reasons are still valid and any increase in cost will be absorbed to keep moral up, or to serve a greater sandbox reason and empty freighters will still explode. This change will have little effect as long as the large ganking groups still have enough pilots to be over the group size threshold needed to field enough DPS. In fact, with PL now seeming to be experimenting with ganking, and even ganking empty freighters, it appears even more empty freighters might be dying in the near future. All this does is lock out smaller groups even more from choosing to hunt freighters in highsec. The fundamental problem can't balance ganking with EHP beyond a certain point when players can just bring more pilots. All that does is restrict the ability to attack in highsec to the largest groups in the game, and given the wide disparity in group sizes, you end up eventually locking out most players from using the mechanic. How is it fair that Goonswarm can field sufficient pilots to gank your freighters, but your 30-man corp now no longer has enough people to meet this arbitrary DPS number to attack their freighters back? I mean, if you get in a real fight with such a size difference you should be crushed, but you don't even have the ability to try and force a response because of an absurdly high EHP wall.
As somebody who found their empty freighter being bumped by CODE. I have to disagree. They go for ransom in those cases. It may not be common, but it does happen.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
baltec1
Bat Country Pandemic Legion
17460
|
Posted - 2016.02.25 06:16:04 -
[1133] - Quote
Lucas Kell wrote:baltec1 wrote:This less of an extreme effect being 0.02% less than today. Feel free to not use the DCU but to say it's going to be less effective than today is just a lie. If it's 0.02% less effect then saying it's going to be less effective can't be a lie. Not really sure where you've got the 0.02% from but mate, you're still talking about EFT warrior fits where the only important factor is tank. Someone that removed a DC and decides that with the natural hull resist boost they are tanky enough to instead increase another stat will be the types of players who will drop it. I'm really not sure what you are hoping to achieve here anyway. Would you say that perhaps CCP should drop of some more of the DCs effects and naturally add those to all ships too? I'm game for that.
These are the ships and fits that YOU currently use. You said you wont be fitting the DCU because they are not as effective, I just showed what you said was wrong. Perhaps if you spent more time looking at this change rather than blindly supporting what you see as a nerf to ganking you would have seen that this change does nothing to lower the importance of the DCU. |
Black Pedro
Yammerschooner
2222
|
Posted - 2016.02.25 06:16:51 -
[1134] - Quote
Teckos Pech wrote:As somebody who found their empty freighter being bumped by CODE. I have to disagree. They go for ransom in those cases. It may not be common, but it does happen. Of course, but there is already a huge financial disincentive when they gank an empty freighter. People get blinded by that billion ISK number on the killmail, but if the freighter was empty, the gankers get nothing and lose hundreds of millions in gank ships. So while they will bump any freighter than can catch, they don't usually shoot the empty ones, and rather will try a ransom or just let them go when something with more cargo comes along.
The only time they do shoot them is when they have a restless gank fleet sitting in station and no other target is available. Or if they are Miniluv and they have intel you are working for a red. In both cases, an extra hundred million will be swallowed as the cost of doing business and that empty freighter will still explode. |
baltec1
Bat Country Pandemic Legion
17460
|
Posted - 2016.02.25 06:44:16 -
[1135] - Quote
Black Pedro wrote:Teckos Pech wrote:As somebody who found their empty freighter being bumped by CODE. I have to disagree. They go for ransom in those cases. It may not be common, but it does happen. Of course, but there is already a huge financial disincentive when they gank an empty freighter. People get blinded by that billion ISK number on the killmail, but if the freighter was empty, the gankers get nothing and lose hundreds of millions in gank ships. So while they will bump any freighter than can catch, they don't usually shoot the empty ones, and rather will try a ransom or just let them go when something with more cargo comes along. The only time they do shoot them is when they have a restless gank fleet sitting in station and no other target is available. Or if they are Miniluv and they have intel you are working for a red. In both cases, an extra hundred million will be swallowed as the cost of doing business and that empty freighter will still explode.
Other way around in my book. The people who gank for profit are hardest hit as this raises the bar for making a profit which means the have to target bigger cargo which means fewer targets for them. CODE operate very differently, the operate on donations and don't pick targets based upon profit. CODE will continue as they do now in the same way they operate in barge ganking which is more like a terrorist organisation. |
Masao Kurata
Perkone Caldari State
415
|
Posted - 2016.02.25 07:06:13 -
[1136] - Quote
Xe'Cara'eos wrote:Also - has anyone yet done the maths on how long a gankalyst will now last against concord? Or are the rumours of concord doing a 5th damage type true?
Er mate, you know that CONCORD has infinite strength ECM, right? As soon as you're locked by concord, you're out.
(Exception: smartbombing battleships with cap boosters.) |
Teckos Pech
The Executives Executive Outcomes
4765
|
Posted - 2016.02.25 07:18:19 -
[1137] - Quote
Black Pedro wrote:Teckos Pech wrote:As somebody who found their empty freighter being bumped by CODE. I have to disagree. They go for ransom in those cases. It may not be common, but it does happen. Of course, but there is already a huge financial disincentive when they gank an empty freighter. People get blinded by that billion ISK number on the killmail, but if the freighter was empty, the gankers get nothing and lose hundreds of millions in gank ships. So while they will bump any freighter than can catch, they don't usually shoot the empty ones, and rather will try a ransom or just let them go when something with more cargo comes along. The only time they do shoot them is when they have a restless gank fleet sitting in station and no other target is available. Or if they are Miniluv and they have intel you are working for a red. In both cases, an extra hundred million will be swallowed as the cost of doing business and that empty freighter will still explode.
Right, ganking an empty freighter does happen. After this change my guess it will become far, far less common.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Violet Crumble
Funtime Factory
790
|
Posted - 2016.02.25 07:40:38 -
[1138] - Quote
Crackforbreakfast wrote:... furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral... Which was also corrected here:
https://forums.eveonline.com/default.aspx?g=posts&m=6340733#post6340733
The maximum collateral only relates to the size of individual contracts accepted. It has no bearing on how those packages are transported and delivered by individual pilots, who are not actually in RFF. Most are in NPC Corps and carry RFF contracted packages alongside other contracts. It's totally down to how individual pilots minimise risk and pilots transporting RFF contracts pass through Uedama and Niarja safely many times every single day.
There is nothing special about a RFF package that protects it from being lost in a gank. The only thing special is that the pilots don't think we need special treatment. We look after our own safety, just as it should be.
Funtime Factory - We put the fun back in funtime
|
Teckos Pech
The Executives Executive Outcomes
4765
|
Posted - 2016.02.25 07:48:42 -
[1139] - Quote
Violet Crumble wrote:Crackforbreakfast wrote:... furthermore it has been explained multiple times why RFF is able to keep their ganks down, for example the 1Bil collateral... Which was also corrected here: https://forums.eveonline.com/default.aspx?g=posts&m=6340733#post6340733
The maximum collateral only relates to the size of individual contracts accepted. It has no bearing on how those packages are transported and delivered by individual pilots, who are not actually in RFF. Most are in NPC Corps and carry RFF contracted packages alongside other contracts. It's totally down to how individual pilots minimise risk and pilots transporting RFF contracts pass through Uedama and Niarja safely many times every single day. There is nothing special about a RFF package that protects it from being lost in a gank. The only thing special is that the pilots don't think we need special treatment. We look after our own safety, just as it should be.
Blah, blah, blah. Please do not distract the dim witted with facts, it merely upsets them and causes them to ask their mom's for more hot pockets.
"The curious task of economics is to demonstrate to men how little they really know about what they imagine they can design."--Friedrich August von Hayek
|
Lucas Kell
Internet Terrorists SpaceMonkey's Alliance
7265
|
Posted - 2016.02.25 08:00:39 -
[1140] - Quote
baltec1 wrote:These are the ships and fits that YOU currently use. You said you wont be fitting the DCU because they are not as effective, I just showed what you said was wrong. Perhaps if you spent more time looking at this change rather than blindly supporting what you see as a nerf to ganking you would have seen that this change does nothing to lower the importance of the DCU. But it will be not as effective. They are reducing all of the DCs stats, therefore it will be less effective than it was. What you are saying is "If your only goal is to maximise defense, then the DC will still be needed" which I'm sure is true, except most people don't build normal ships solely with defense in mind, otherwise noone would ever fit offensive midslots, they'd all add shield tank. With a base increase to hull, many ships will have enough tank that the removal of the DC and replacement with a prop mod or damage/yield mod will be a viable option. Stop looking solely at EHP and perhaps you'd understand this.
The Indecisive Noob - EVE fan blog.
Wholesale Trading - The new bulk trading mailing list.
|
|
|
|
|
Pages: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 30 40 50 .. 51 :: one page |
First page | Previous page | Next page | Last page |