Pages: 1 [2] 3 4 :: one page |
|
Author |
Thread Statistics | Show CCP posts - 12 post(s) |
TornSoul
BIG Libertas Fidelitas
|
Posted - 2008.10.16 19:44:00 -
[31]
Quote:
We want to make it handle things in batches rather than as single items. Basically if you move items from one place to another, trade items, etc. the system moves these objects one at a time and makes a call to the server, then back to the client for each item. It is, of course, more efficient to move those items in batches so that the number of calls between the server and client is reduced
Identifying those areas of the game and getting it batched up, will easily be the one largest improvement area you can find.
We had similar issues with the application I work with at work.
Going from single item to batched calls, gave us a ~50 times (5000%) improvement on those particular (DB) operations (also using MS SQL).
So start digging for those where ever you can find them!
BIG Lottery |
IR Scoutar
Caldari State War Academy
|
Posted - 2008.10.16 20:38:00 -
[32]
hmmm my view
interesting blog
improvement to code etc is an ongoing thing and a must anyways so no special praise just a good job IF it works on tq
|
Franga
NQX Innovations
|
Posted - 2008.10.16 20:53:00 -
[33]
This is some tasty crap.
I don't understand all of it (/me would love to have more of a technical understanding of things), but nevertheless, always interesting stuff. ----------
|
Trebor Daehdoow
Gallente LEAP Corp Ursa Stellar Initiative
|
Posted - 2008.10.16 21:27:00 -
[34]
If I might make a humble suggestion, why not post a list of the top 10 or 20 areas where you are looking for better algorithms to improve performance. Given the number of programmers who play EVE, the chances are good that someone's run into a similar problem and might have an insight that would be really helpful.
Sure, 99% of the time, the suggestions will be things you've thought of, or that aren't applicable to your particular environment, but every so often, you'll find a diamond in the turds. There's a ton of bizarre algorithms out there, most of them invented in the '60s and now buried somewhere deep in one of the early volumes of Knuth, that you'll never know about unless someone points you at them.
If you've got an order n**2 bottleneck and someone knows an obscure tape-sorting algorithm that can be adapted to make it order n log n, or even linear, that's a huge win. World Domination - It's fun for the entire family! EViE - The iPhone / iPod Touch Skill Training Monitor
|
Dmian
Gallente The Scope
|
Posted - 2008.10.16 21:28:00 -
[35]
Great blog and incredible work by you guys!
----
Eve Alpha - The font of Eve - Get it here |
Haradgrim
Tyrell Corp INTERDICTION
|
Posted - 2008.10.16 21:34:00 -
[36]
So....does this have anything to do with Tanis. dropping sov in the Jove region? --
Originally by: CCP Oveur ...every forum whine feels like a baby pony is getting killed
|
WachinDaGame DrinkinABud
|
Posted - 2008.10.17 02:34:00 -
[37]
If we could sort out loot by META LEVEL then we could reprocess all our loot in one operation rather than picking them out individually. Combine this with batch operations and you have efficiency.
|
|
CCP Tanis.
|
Posted - 2008.10.17 09:01:00 -
[38]
Originally by: Andy Rogerson
Originally by: CCP Tanis. [snip] Rest assured that characters with neutral (zero) standings will NOT factor into the mix. The formula is still counting the same members it was counting before, we've just cut out a lot of fluff. If you were to think of your corporation standings like a bucket of water the old system filled that bucket one cup at a time where as the new system fills it all in one go. [snip] I hope this alleviates any worries you guys may have had.
Yep -- it'll take a bit longer for a contractor to do their thing if you're anchoring in 0.5, but perhaps a bit less in 0.7. Regardless, it's definately an easier system to understand than the old way (too much math). I guess I was just trying to use this as an example to suggest that changes in game mechanics (even if it seems like a minor thing and a big performance boost) should at least be mentioned in patch notes or the like. Or maybe I'm just feeling that way because I got chewed out by a user this morning over a bug that we fixed on our website that he was relying on :)
Changes to mechanics generally are in teh patch notes; however the patch notes for the next expansion haven't even been released yet. ____________________________ I break things.
Tanis. Testing Lead EVE Online CCP Games
"GM Voodoo > That plan really straddles the fine line between genius and idiocy." "CCP Tanis. > And that differs from everything else I do how?" |
|
Eldar Boon
|
Posted - 2008.10.17 09:05:00 -
[39]
The Dev Blog mentions that NPC will have a greater alpha strike while maintaining the DPS. This, in theory, sounds fine, but often in missions large numbers of NPCs will spawn and open fire similtainiously. This could result in a ship, that could normally absorb the DPS, unable to deal with the alpha strike of a spawning mob.
I can understand that CCP don't want to change the balance of missions, but a rebalancing of missions seems long over due? By their very nature, missions create lag, as players flock to the most popular systems. Would a simple change, like increasing the number of systems a player has to travel for a mission eleviate the problem?
If batching of calls to the DB has a dramatic effect on lag, will CCP be looking at altering game mechanics to enable batching? One example could be the jumping of fleets through a jumpgate.
|
|
CCP Tanis.
|
Posted - 2008.10.17 09:49:00 -
[40]
Originally by: Eldar Boon Edited by: Eldar Boon on 17/10/2008 09:09:59 The Dev Blog mentions that NPC will have a greater alpha strike while maintaining the DPS. This, in theory, sounds fine, but often in missions large numbers of NPCs will spawn and open fire similtainiously. This could result in a ship, that could normally absorb the DPS, unable to deal with the alpha strike of a spawning mob.
I can understand that CCP don't want to change the balance of missions, but a rebalancing of missions does seem long overdue. By their very nature, missions create lag, as players flock to the most popular systems. Would a simple change, like increasing the number of systems a player has to travel for a mission eleviate the problem?
If batching of calls to the DB has a dramatic effect on lag, will CCP be looking at altering game mechanics to enable batching? One example could be the jumping of fleets through a jumpgate. Perhaps even a jump queue where a busy jumpgate batches players jumping through.
As I'd mentioned in my blog, for the NPC ROF changes we're carefully testing the changes to ensure that the missions and encounter sites will still not suddenly be claiming more ships than they should. We've pulled in our mission testing expert just for this because we wanted to be sure we're not throwing off the balance.
We would not likely be changing the number of jumps to get to the missions as that would have a lot of negative effects; missions would take longer, mission running would become less profitable, bonus times would go out of whack, etc. This is a perfect example for why we spend so much time researching and investigating a variety of possible solutions then test them to be sure that what we end up putting out will do what we want it to without any really nasty side-effects. ____________________________ I break things.
Tanis. Testing Lead EVE Online CCP Games
"GM Voodoo > That plan really straddles the fine line between genius and idiocy." "CCP Tanis. > And that differs from everything else I do how?" |
|
|
Daan Sai
Polytrope
|
Posted - 2008.10.17 10:35:00 -
[41]
Awesome work folks!
Maybe when you get some DB access bandwidth freed up we might be able to look at the whole BPO/BPC facorting and normalisation, so we can oneday telll them apart in the client UI.
|
Eldar Boon
|
Posted - 2008.10.17 11:07:00 -
[42]
Originally by: CCP Tanis.
As I'd mentioned in my blog, for the NPC ROF changes we're carefully testing the changes to ensure that the missions and encounter sites will still not suddenly be claiming more ships than they should. We've pulled in our mission testing expert just for this because we wanted to be sure we're not throwing off the balance.
We would not likely be changing the number of jumps to get to the missions as that would have a lot of negative effects; missions would take longer, mission running would become less profitable, bonus times would go out of whack, etc. This is a perfect example for why we spend so much time researching and investigating a variety of possible solutions then test them to be sure that what we end up putting out will do what we want it to without any really nasty side-effects.
Thanks for the response. Small changes can have dramatic effect as the infamous "butterfly effect" would suggest; so it's great that so much time is invested in testing to making sure these performance improvements have no nasty side-effects.
I was wondering about performance improvements that would have a beneficial side-effect to the way the game is played, like weapon grouping for example. Surely a more holistic approach where changes could be made to the way the game is played, in tandem with performance improvements, would see far greater benefits.
I used mission running as an example because I see any game mechanic that encourages players to flock to a single point (where the network effect is not beneficial, due to missioning being a largely solo occupation) as a bad thing. In this case, performance improvements may alleviate the lag, but they don't resolve the underlying problem. Is this something that can be discussed with your mission testing experts?
|
Jufi Wekior
Arachnid Logistics Fate Weavers
|
Posted - 2008.10.17 11:16:00 -
[43]
Originally by: CCP Tanis.
Originally by: Eldar Boon ...
As I'd mentioned in my blog, for the NPC ROF changes we're carefully testing the changes to ensure that the missions and encounter sites will still not suddenly be claiming more ships than they should. We've pulled in our mission testing expert just for this because we wanted to be sure we're not throwing off the balance.
We would not likely be changing the number of jumps to get to the missions as that would have a lot of negative effects; missions would take longer, mission running would become less profitable, bonus times would go out of whack, etc. This is a perfect example for why we spend so much time researching and investigating a variety of possible solutions then test them to be sure that what we end up putting out will do what we want it to without any really nasty side-effects.
Has there been any consideration on making NPC's stronger but less numerous in missions, such as halving the number of NPC's but each NPC being worth twice they were previously? It has always irked me that a capsuleer can take on half a dozen NPC BS's(and support) at once and not break a sweat. Capsuleers are powerfull but such a humongous difference is unrealistic to me. After all why would the NPC's still bother fighting with capsuleers lined up to take potshots at them for monies if they lose bs by the dozen per mission? Not to mention that supposedly not all systems on a capsule-fitted bs are directly controlled by the pilot but still have crew operating them.
|
Kody Apollo
Eve University Ivy League
|
Posted - 2008.10.17 11:39:00 -
[44]
Originally by: CCP Tanis.
New system: * ALL members who have non-neutral standings are counted After the new member has been in corp for 7 days or more your corporation standings would now be +9 for Caldari state.
When you say non-neutral do you mean that anyone with corp standing over 0.5 standing?
I've noticed there's quite a few people that have less than 0.1 standing with various corps. I'm curious whether you're planning on doing it by true neutral standing or just standings listed as "neutral". This will effect whether it's possible to get jump clones for some of the larger alliances with many members often with 0.01 standing with various corps.
I imagine the calculations may be a little bit less if you ignore standings that are fairly close to 0, since many people tend to run a single mission with a corp then never return. |
|
CCP Ytterbium
|
Posted - 2008.10.17 11:48:00 -
[45]
Originally by: Jufi Wekior Has there been any consideration on making NPC's stronger but less numerous in missions, such as halving the number of NPC's but each NPC being worth twice they were previously? It has always irked me that a capsuleer can take on half a dozen NPC BS's(and support) at once and not break a sweat. Capsuleers are powerfull but such a humongous difference is unrealistic to me. After all why would the NPC's still bother fighting with capsuleers lined up to take potshots at them for monies if they lose bs by the dozen per mission? Not to mention that supposedly not all systems on a capsule-fitted bs are directly controlled by the pilot but still have crew operating them.
Of couse there has been, but as I said in my previous reply on that thread, such a change is not that easy to come up with.
We know quite well that NPCs in general are too predictable in their behavior and would like to change that in the future. Just reducing their number and increasing their flat DPS is not the way to go since it does not solve the core problem.
Examples on making NPCs less predictable would be to have them focus on player logistic ships first, sensor boost/repair each other, call for reinforcements, try to escape if outnumbered or change ammunition type to exploit a player tank weak resist hole.
But again, this is a very delicate mechanism to tamper with since a NPC which may be entertaining for you may prove to be impossible to come up with for other pilots with different skills or experience. It all matters on where and how you meet the NPC in question (mission, asteroid encounter, exploration complex...) and what is the overall difficulty supposed to be for that encounter. |
|
Viilaa
Caldari OH Corp
|
Posted - 2008.10.17 11:50:00 -
[46]
Originally by: Kody Apollo
When you say non-neutral do you mean that anyone with corp standing over 0.5 standing?
Non-neutral... anyone with a standing that is not 0 - whether positive or negative.
And to reply to someone who posted earlier - I believe Corp Standings will be calculated daily at DT just like they are now, but they will only count characters with non-0 standings and that have been in the corp 7 days. So standing will still change daily, but new members joining the corp will not effect the standings for a week.
Viilaa
|
|
CCP Tanis.
|
Posted - 2008.10.17 11:59:00 -
[47]
I'll answer your questions in turn...
Originally by: Jufi Wekior
Has there been any consideration on making NPC's stronger but less numerous in missions, such as halving the number of NPC's but each NPC being worth twice they were previously? It has always irked me that a capsuleer can take on half a dozen NPC BS's(and support) at once and not break a sweat. Capsuleers are powerfull but such a humongous difference is unrealistic to me. After all why would the NPC's still bother fighting with capsuleers lined up to take potshots at them for monies if they lose bs by the dozen per mission? Not to mention that supposedly not all systems on a capsule-fitted bs are directly controlled by the pilot but still have crew operating them.
We've looked into it, yes; however, in order to reduce the number of NPCs we would have to re-balance every mission in game for damage, alpha strike, bounties, payout, time to complete the mission, etc. which is an obscenely massive undertaking. We know that we can get "more bang for our buck", eg. more performance boost, if we focus our efforts on other types of changes. we would have to re-balance every mission in game for damage, alpha strike, bounties, payout, time to complete the mission, etc. which is an obscenely massive undertaking. We know that we can get "more bang for our buck", eg. more performance boost, if we focus our efforts on other types of changes.
Originally by: Eldar Boon
Thanks for the response. Small changes can have dramatic effect as the infamous "butterfly effect" would suggest; so it's great that so much time is invested in testing to making sure these performance improvements have no nasty side-effects.
I was wondering about performance improvements that would have a beneficial side-effect to the way the game is played, like weapon grouping for example. Surely a more holistic approach where changes could be made to the way the game is played, in tandem with performance improvements, would see far greater benefits.
I used mission running as an example because I see any game mechanic that encourages players to flock to a single point (where the network effect is not beneficial, due to missioning being a largely solo occupation) as a bad thing. In this case, performance improvements may alleviate the lag, but they don't resolve the underlying problem. Is this something that can be discussed with your mission testing experts?
That is exactly what we are already doing with regard to issues like with mission-hubs. We know that there are a number of factors that cause people to "herd" into these areas and as a result we must find a solution that addresses many of these factors well enough to achieve our desired results. Unfortunately identifying and understanding all of those factors takes time and a lot of investigating so it can be very time consuming to find a solution, much less a good solution.
In order to best attack these kinds of problems; we are making changes in iterations, or stages. This approach allows us to make more frequent but smaller changes which would have less chances of negative side-effects and then continue building upon things until we've gotten to where we are happy with things.
____________________________ I break things.
Tanis. Testing Lead EVE Online CCP Games
"GM Voodoo > That plan really straddles the fine line between genius and idiocy." "CCP Tanis. > And that differs from everything else I do how?" |
|
|
CCP Tanis.
|
Posted - 2008.10.17 12:01:00 -
[48]
Bah, silly character limit has pwned my reply :(
Originally by: Kody Apollo
When you say non-neutral do you mean that anyone with corp standing over 0.5 standing?
I've noticed there's quite a few people that have less than 0.1 standing with various corps. I'm curious whether you're planning on doing it by true neutral standing or just standings listed as "neutral". This will effect whether it's possible to get jump clones for some of the larger alliances with many members often with 0.01 standing with various corps.
I imagine the calculations may be a little bit less if you ignore standings that are fairly close to 0, since many people tend to run a single mission with a corp then never return.
By "non-neutral" I mean anyone who has a standing that NOT equal to zero. Basically 0.1 -> 10.0 as well as -10.0 -> -0.1. So any standing that has changed from the default value. ____________________________ I break things.
Tanis. Testing Lead EVE Online CCP Games
"GM Voodoo > That plan really straddles the fine line between genius and idiocy." "CCP Tanis. > And that differs from everything else I do how?" |
|
Dee Carson
Caldari Eve University Ivy League
|
Posted - 2008.10.17 12:52:00 -
[49]
Originally by: CCP Lingorm The calculation goes like this ... get me all the none 0 standings of all members of the corp that have been in the corp 7 days or more. Now calculate the average and set the corp standing to this.
Here's our specific situation that I'd like to use as a frame for clarification:
Some time in the past, a student of Eve University ran a lot of missions for an NPC corp in a sparsely populated area. He was the only person running missions for this NPC corp. His standings crossed over the magic number to allow him jump clone access. Eve University's corp standing toward this NPC corp followed his personal standing rise until the entire student body had access to these jump clones. The student in question subsequently left Eve University. We put the NPC corp on a "do not mission" list, preserving access to these jump clones (no one running missions = no dilution of the NPC corp standings).
In this context then:
- Under the proposed changes, do existing Corp standings figure into the calculation at all? - If so, please indicate how. - If not, confirm that in the instance described above, our Corp standing to the NPC corp will go to zero. - Please confirm that your scope for the query to calculate standings is all current members of a corp who have been a member for more than 7 days with non-neutral standings.
Regards, Dee Carson Director of Operations Eve University
Join channel: "Eve University" for more info. |
Hon Kovell
Gallente Neh'bu Kau Beh'Hude Ushra'Khan
|
Posted - 2008.10.17 12:53:00 -
[50]
What is the problem caused by NPC load? It seems that it would be spread evenly throughout the day rather than having any big spikes. Are you aiming to reduce that constant load or are there situations where the load peaks and causes a problem itself?
More generally, do you have a focus on constant or peak load and tools to see the difference (other than servers dying)? Oh, and do you have tools to find related processes causing problems? i.e. no single process is a problem but there's a group that will happen at the same time and together cause problems.
You would have to be careful with that last one. It might lead to reducing load by automatically combining processes like activating stargate, destroying ship, moving player to clone bay and activating new pod.
|
|
M1AU
Gallente
|
Posted - 2008.10.17 14:36:00 -
[51]
Another technical dev-blog and AGAIN nothing new about the long time promised Linux and MacOSX client dev-blog. >:(
|
Earthican
|
Posted - 2008.10.17 16:22:00 -
[52]
Originally by: WachinDaGame DrinkinABud If we could sort out loot by META LEVEL then we could reprocess all our loot in one operation rather than picking them out individually. Combine this with batch operations and you have efficiency.
Devs please respond to this awesome suggestion.
It would also improve gameplay when it comes to sorting through tonnes of crap items while fitting ships. |
Dakisha
D00M. Triumvirate.
|
Posted - 2008.10.17 17:18:00 -
[53]
Originally by: Jaala Creed Is the Eve Performance Group also working on InfiniBand? Or is that a different developer group?
To continue on that, how is InfiniBand developing?
We haven't heard from that for a while and its a VERY promising technology.. Please toss us some bones from that?
Thanks, Jaala
|
|
CCP Tanis.
|
Posted - 2008.10.17 19:21:00 -
[54]
Originally by: Dee Carson
... Here's our specific situation that I'd like to use as a frame for clarification:
Some time in the past, a student of Eve University ran a lot of missions for an NPC corp in a sparsely populated area. He was the only person running missions for this NPC corp. His standings crossed over the magic number to allow him jump clone access. Eve University's corp standing toward this NPC corp followed his personal standing rise until the entire student body had access to these jump clones. The student in question subsequently left Eve University. We put the NPC corp on a "do not mission" list, preserving access to these jump clones (no one running missions = no dilution of the NPC corp standings).
In this context then:
- Under the proposed changes, do existing Corp standings figure into the calculation at all? - If so, please indicate how. - If not, confirm that in the instance described above, our Corp standing to the NPC corp will go to zero. - Please confirm that your scope for the query to calculate standings is all current members of a corp who have been a member for more than 7 days with non-neutral standings.
Because the corporation standings would always follow the average of it's current members' standings; once a pilot leaves the corporation, the corporation's standings would be lowered if his standings were higher than the average.
This may prove an annoyance for some of the smaller corporations; however it gives a pretty significant edge to folks who sell their services to boost a corp's standings making a persons standings a more valuable commodity both to the pilots and the corporations who could benefit from their services. In the end, we feel it is a much better solution. ____________________________ I break things.
Tanis. Testing Lead EVE Online CCP Games
"GM Voodoo > That plan really straddles the fine line between genius and idiocy." "CCP Tanis. > And that differs from everything else I do how?" |
|
Dee Carson
Caldari Eve University Ivy League
|
Posted - 2008.10.17 19:47:00 -
[55]
Originally by: CCP Tanis.
Because the corporation standings would always follow the average of it's current members' standings; once a pilot leaves the corporation, the corporation's standings would be lowered if his standings were higher than the average.
Just so I'm 100% clear. You are saying that existing Corp standings are no longer included in the calculation at all?
THX DC
Join channel: "Eve University" for more info. |
Hon Kovell
Gallente Neh'bu Kau Beh'Hude Ushra'Khan
|
Posted - 2008.10.18 02:26:00 -
[56]
Originally by: CCP Tanis. Because the bulk of the issue was when NPCs were engaged in combat and the fact that NPCs only spawn when players are in a system, to save on load; this issue would become more severe as more players log in to the server by simple virtue of more people being around to kill them or cause them to spawn in the first place.
I did assume that the load would change like that. I was thinking that it was spread across Eve in thousands of locations spread through the day (peaking as players cause them to spawn) as opposed to all spiking in the one spot or at the same time. You answered me anyway with the rest of your response, though, thank you.
Originally by: CCP Tanis. We use a wide variety of tools in our investigation. Because the problems we are tackling almost always have several contributing factors; we must look at a wide variety of information to get the best picture we can about the problem. Our tools cover the entire field from hardware monitors to specialized pieces of code injected into the server or client code, for testing, to trace an issue.
We may or may not focus on a particular system or peak/off-peak behavior depending on the situation. We create a list of "top issues" based on petitions, bug reports, player feedback, monitoring data, server trends, and a myriad of other sources. We then begin prioritizing and researching the issues to determine what is most critical and also which issues are affecting the most people.
Thank you for your reply.
|
Chainsaw Plankton
IDLE GUNS IDLE EMPIRE
|
Posted - 2008.10.18 08:10:00 -
[57]
Originally by: CCP Ytterbium
Originally by: Batolemaeus On the Npc changes:
Of course, reducing rof for npcs is easier for you guys, just run a script that reduces rof and increases damage. <insert one-line-of-code joke here>
But please also poke the mission design team about it. I'm pretty sure reducing the numbers of those npc, and increasing their damage potential, will free up a lot more ressources.
We are aware of this situation and you bring up a good point here.
The problem remains it is far less easy than it sounds to change, since reducing NPC numbers means balancing bounties/loot to keep rewards on the same level. Also, plain increase in damage is not the best way to go since a mission that may be easy for a veteran pilot with skills and experience may turn horribly difficult for a relatively new character.
5 rats doing 100 dps each, total 500 dps 2 rats doing 250 dps each, total 500 dps
less server calculations, less red blinky craping out older computers. I don't see how this could bone noobs.
npc loot tables are messed up anyways, Might as well make a broad sweeping change to npc mechanics (it seems to work for the nano nerf) miner I's from battleship rats, and 1mn microwarp drives. and at the same time 5 or so 1.2mil bs rats can be tanked no problem with 1 rep and 4 hardeners. (empire versions the 0.0 version would be the 1.8 mil bs)
of course you can't just up damage if you want to keep missions of the same difficulty. (well probably you should as level 4 missions are ludicrously easy (if nano speeds are ludicrously fast missions are like 10x ludicrous easy)) people saying you only need like 300 dps tanked in eft to do level 4s? seriously?
|
Chainsaw Plankton
IDLE GUNS IDLE EMPIRE
|
Posted - 2008.10.18 08:29:00 -
[58]
Originally by: CCP Tanis.
Originally by: Eldar Boon
Thanks for the response. Small changes can have dramatic effect as the infamous "butterfly effect" would suggest; so it's great that so much time is invested in testing to making sure these performance improvements have no nasty side-effects.
I was wondering about performance improvements that would have a beneficial side-effect to the way the game is played, like weapon grouping for example. Surely a more holistic approach where changes could be made to the way the game is played, in tandem with performance improvements, would see far greater benefits.
I used mission running as an example because I see any game mechanic that encourages players to flock to a single point (where the network effect is not beneficial, due to missioning being a largely solo occupation) as a bad thing. In this case, performance improvements may alleviate the lag, but they don't resolve the underlying problem. Is this something that can be discussed with your mission testing experts?
That is exactly what we are already doing with regard to issues like with mission-hubs. We know that there are a number of factors that cause people to "herd" into these areas and as a result we must find a solution that addresses many of these factors well enough to achieve our desired results. Unfortunately identifying and understanding all of those factors takes time and a lot of investigating so it can be very time consuming to find a solution, much less a good solution.
In order to best attack these kinds of problems; we are making changes in iterations, or stages. This approach allows us to make more frequent but smaller changes which would have less chances of negative side-effects and then continue building upon things until we've gotten to where we are happy with things.
quality 18 level 4 agent + sec 0.5 -> mission hub
given that 1.0-0.5 all has concord intervention that risk does not scale anywhere near the way rewards scale for the change in sec status. that and the agents are supposed to give out missions from the same pool. I did a lower quality agent in a higher sec system and the lp drop was just not worth the decrease in lag (which wasn't that noticeable, occasionally yes I have some problems like it takes 2 seconds to change mods rather than 1)
if the agent is near lowsec forget it. If I have to turn down a mission every few missions that makes the agent worthless.
and if it is with a desirable corp oh god the people there caldari navy anyone? (why so many people choose cn I have no idea)
normalizing agent rewards and boosting rewards based on sec range (sec >= .5, .4>=sec>=.1, sec<= 0.0) rather than sec (1.0 , .9, ... , .1, 0.0)
agent quality is just confusing and I see many noobs going well I hvae a q20 agent I should be getting nothing but worlds collide/blockade right? but I keep getting crappy courier missions and 100k bounty missions, wtf?
|
mr dragothur
|
Posted - 2008.10.18 10:18:00 -
[59]
i wonder if you guys switch to newly advanced 64bit stuf you know about the direct x 10 problems and all the older harwareissues it might give? old stuf dont work well with new stuf allso shaders will cause allot of problems i think and 1 more thing is will someone with a slow connection slow the whole deal on this cluster or doesnt it affexct the overall preformance? i got a rly fast computer and i notice that the market must have allot of issues in it cus when i start market my whole screeen freeze about 3 sec to fetch info that cant be right well i mean i can understand why a node would crash if allot would do that sugest to remove market when you fly and only usable when docked naa thats an excuse i think the market is rly a uge problem well at least to me.
duno why i write this but i wasa reading the logs hehe just expressing my thoughts
good luck on youre amazing works
|
Clansworth
Burning Sky Labs Libertas Fidelitas
|
Posted - 2008.10.18 16:50:00 -
[60]
Regarding the Faction -> Corp standings changes:
I understand that a new member will have to wait 7 days before his standings are counted towards the average, but what about existing member's leaving? As i understand it, that would take effect immediately (well, next DT)?
For instance, A corp has a couple guys with +7 or greater faction standing, they want to anchor a POS in 0.6. All corp members who DON'T have the base standings >6 leave the corp. After Downtime, the corp standings would already be high enough to anchor. This would actually allow corps to anchor high-sec POS's in 2 days in most cases (24hr role removal, then wait or next DT for new calculation), instead of the 10-11 days as before.
POS Personal Storage |
|
|
|
|
Pages: 1 [2] 3 4 :: one page |
First page | Previous page | Next page | Last page |