Pages: 1 2 [3] 4 5 6 7 8 9 10 11 12 13 14 15 16 .. 16 :: one page |
|
Author |
Thread Statistics | Show CCP posts - 3 post(s) |
Keira Fordring
Caldari State War Academy
|
Posted - 2007.07.07 10:31:00 -
[61]
Quote: And of course, if you wanted to, you'd always have the option to "break formation" and start flying on your own again.
At which point your client freezes indefinitely as the server suddenly has to calculate positon and collision information for all ships that were in formation.
As soon as formation breaks, the server HAS to convert a "single object" formation into it's "multiple object" components which means calculating initial data such as individual coordinates and then doing a collision check to see if any ships that broke formation have collided with something.
Unless people stayed in formation for the entire battle, the only thing this would do is delay the inevitible lag.
|
MissileRus
|
Posted - 2007.07.07 11:07:00 -
[62]
we got squads already, so formation flight would only enhance that for larger fights where you need certain players with certain skills at certain positions, if the leader pops next in line takes over and his skills start to apply etc.
i like the idea of formation flight alot, but i dont want to see it everywhere being a lover of small chaotic battles without the "standard" target calling if im going to team(witch i dont do often now), id like formations to mostly affect only larger battles 50vs50+ etc (that would be where it would benefit if my brain works propperly..) small gangs would still have the option but no real benefits over a conventional gang. mostly tweaking issues but it would be nice if bonuses scaled with amounts of ships and/or [insert stuff here] if it was done
im no programmer or expert but sounds easier to juggle 100 balls then thousands.. ..
a thought, how mutch do bonus calculations affect? if it becomes one "entity" concerning movement maybe other things like bonus calculations could be improved too when in formation somehow? even if it doesnt affect alot if would always be beneficial if that could improve too as a good side effect instead of calculating every shot(witch i think the server does?), server keeps bonus givers in formations in "memory" or what to call it, and doesnt "calculate/update pakets/whatever" at every shot, i think some dev said it calculates every single shot.. as said im no expert... dont look at me like that! *hides*
/signed
--------------------------- 4. i like pizza |
Reithan
Caldari LEGI0N SOUL CARTEL
|
Posted - 2007.07.08 00:36:00 -
[63]
Originally by: Keira Fordring
Quote: And of course, if you wanted to, you'd always have the option to "break formation" and start flying on your own again.
At which point your client freezes indefinitely as the server suddenly has to calculate positon and collision information for all ships that were in formation.
As soon as formation breaks, the server HAS to convert a "single object" formation into it's "multiple object" components which means calculating initial data such as individual coordinates and then doing a collision check to see if any ships that broke formation have collided with something.
Unless people stayed in formation for the entire battle, the only thing this would do is delay the inevitible lag.
I never meant the whole formation would break on one person leaving. Though, even if it did, it would just mean those ships would then "enter" the grid. So, even in a WORST case scenario it would cause less lag than the current system, or at VERY worst equal lag. -----------------------------------------------
|
Reithan
Caldari LEGI0N SOUL CARTEL
|
Posted - 2007.07.08 00:41:00 -
[64]
Originally by: Zoi Opia Only the brute force approach to collision detection scales O(n^2). Using optimizations like simplified models, temporal/spatial coherence, and pruning algorithms, it is possible to achieve O(n) scalability. So no, I don't see 'forcing' formations bringing that much improvement to lag.
Yeah, I know that you can exceed O(n^2) efficiency, though I still doubt O(n) is possible without losing important functionality and error-checking.
Even so, formations don't really change the algorithm, they simply reduce n.
So, even if you DID get to O(n)...which is better:
n = 204+204 or n = 34+34
I think the answer is obvious.
And in any case, I think based on server performance it's obvious that their algorithm is not fully optimized to O(n), and probably even closer to O(n^2) than O(n), so reducing n by large amounts (as this would) would be a big boost to performance.
If you coupled this with optimizations to the algorithms used for fleet battles, I'm sure we could cover a lot of ground towards lagless fleet battles...until some ****** goes, "Hey, we can do 200v200 lagless....let's bring OVER NINE THOUSAND!!!!!!" -----------------------------------------------
|
SiJira
|
Posted - 2007.07.08 15:26:00 -
[65]
good idea but needs a lot of work and support from the big alliances - they feel the most lag anyways ____ __ ________ _sig below_ the jet cans are made so that people that dont mine can get free ore
miners ritually donate the ore to anyone wishing to take some |
Jiggs Casey
Minmatar Asguard Security Service Rule of Three
|
Posted - 2007.07.08 15:45:00 -
[66]
I've been in fleet battles where we've tried dividing up who jumps first by name, by ship type, even tried to divide the group into smaller groups... still lagged out nearly every time. I've lost too many ships to lag in large battles. I don't do it anymore..
Unfortunately, the concept of large, long lasting fleet battles is one of the reasons I was drawn to eve. Anybody remember the video of the ccp interview where the developer spoke of '1000 people' in a system, in combat...?
So, until they figure out how to make it a reality, I guess I'll continue forking over money... to play... lag and all...
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.08 16:37:00 -
[67]
Originally by: Reithan
Originally by: Zoi Opia Only the brute force approach to collision detection scales O(n^2). Using optimizations like simplified models, temporal/spatial coherence, and pruning algorithms, it is possible to achieve O(n) scalability. So no, I don't see 'forcing' formations bringing that much improvement to lag.
Yeah, I know that you can exceed O(n^2) efficiency, though I still doubt O(n) is possible without losing important functionality and error-checking.
Even so, formations don't really change the algorithm, they simply reduce n.
So, even if you DID get to O(n)...which is better:
n = 204+204 or n = 34+34
I think the answer is obvious.
And in any case, I think based on server performance it's obvious that their algorithm is not fully optimized to O(n), and probably even closer to O(n^2) than O(n), so reducing n by large amounts (as this would) would be a big boost to performance.
Unfortunately, I highly doubt that it would be possible to reduce positional checking for fleet battles down to O(n), I think the best we'll get is something along the lines of O(nlogn) or perhaps O((n/x)^2)*x with x being the number of subgrids...or something like that. I'm too lazy to run the calculations at the moment.
Quote:
If you coupled this with optimizations to the algorithms used for fleet battles, I'm sure we could cover a lot of ground towards lagless fleet battles...until some ****** goes, "Hey, we can do 200v200 lagless....let's bring OVER NINE THOUSAND!!!!!!"
WHAT NINE THOUSAND?!
LAG IBTL! IBDS! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
Leneerra
Minmatar Trinity Nova KIA Alliance
|
Posted - 2007.07.08 16:41:00 -
[68]
I doubt that collision detection is such a major player in the battles. yes it requires a lot of check in large numbers of ships, but the calculations for tracking and damage escalate much faster, as there are usually mutible guns (and or other modules that affect other ships) for each ship. General status on each and every ship is transmited to everyone in the grid (you still get data to see all kinds of visual effects even if you select not to have them shown on your system) and then there is the more detailed data you get on whomever you have locked and everyone that is in your squad or command tree. Add to this the data about every ship entering the grid that is transmitted to every ship in the grid as well as the data about everone there to the new arrival. Formations would do absolutely 0 for this huge flood in data that has to be shared on large scale battles. the reduction for collision detection is realy negligeble.
However they could try having formations for drones so a player can choose what drones are to work together before launcing and see if it makes such a difference
|
William Young
Caldari
|
Posted - 2007.07.08 16:44:00 -
[69]
Disregarding the whole idea of fixing lag,
I really like the idea of formations with bonuses. The drawback would be that you'd need alot of coordination to get a formation to work effectively (turn MWDs on/off! NOS this dude! Fire on that dude! etc.), but the payoff would be the bonuses, hopefully with the end result being a well-coordinated fleet with extremely good communication.
Might solve lag, might not, but who cares? It sounds like a fun idea.
|
Kartikeya
|
Posted - 2007.07.08 16:57:00 -
[70]
Some one might have already posted this but I dont think it will work. Formations would be nice but you couldnt just mass all the ships into a signle entity on the server because it would mess up game mechanics. What would happen when 2 formations bumped each other each blob bounces in its own direction. So you could then abuse this by having a signle cepter being able to bump your 100 man sniper fleet. Or setting your many smaller gangs up around a gate right at the jump in point causeing people to bounce the sec they decloak. As cool as it would be I think it wouldnt work with game mechanics. At least not in a lag reducing way.
|
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.08 16:58:00 -
[71]
Edited by: Tarminic on 08/07/2007 17:03:00
Originally by: Leneerra I doubt that collision detection is such a major player in the battles. yes it requires a lot of check in large numbers of ships, but the calculations for tracking and damage escalate much faster, as there are usually mutible guns (and or other modules that affect other ships) for each ship. General status on each and every ship is transmited to everyone in the grid (you still get data to see all kinds of visual effects even if you select not to have them shown on your system) and then there is the more detailed data you get on whomever you have locked and everyone that is in your squad or command tree. Add to this the data about every ship entering the grid that is transmitted to every ship in the grid as well as the data about everone there to the new arrival. Formations would do absolutely 0 for this huge flood in data that has to be shared on large scale battles. the reduction for collision detection is realy negligeble.
However they could try having formations for drones so a player can choose what drones are to work together before launcing and see if it makes such a difference
Actually we went over this point before and I believe that collision checking escalates faster than weapons fire. Allow me to explain:
Weapons: Ship A fires on Ship B - this involves about ten calculations assuming that tracking, resistances, and all those things are being dealt with. These interactions only involve the ship being fired on and the ship doing the firing, so this is a linear equation, e.g. O(n).
Status Updates: The health of ships is not immediately known to every other ship, only to that ship's /fleet/wing/squad commander and ships locking it. So this is also a linear equation, albeit one with a few variables. Assuming a ration of 8 to 1 of commanders to non-commanders, and each ship has an average of three other ships locked, the formula is O(3n+ n/8) or O(25n/8)
Graphical Effects: Graphics in eve consist of an animation whereby one ship fires on another (regardless of whether it hits the target or not. So once again this is a linear equation. Assuming every ship is firing on three other ships in the battle, this is O(3n), still a linear equation.
Positional Updates: This is more complicated because assuming that the grid is divided into subgrids to reduce the work, every ship still have to check that it's not colliding with every other ship, and there are also factors of accelleration/decelleration to calculate as well. The accel/decel calculations are linear, but collision detection must be performed on every ship in the (sub) grid. Assuming the grid is divided into X subgrids, the full formula is still O(n) + O((n/x)^2)/X, which still increases exponentially.
Our total formula for fleet battles is O(7n + n/8) for non-movement control and O(n^2) for movement control. Therefore, once a battle reaches 8 or more people, movement checking becomes the largest source of processing resources. Assuming that subgrids can be used like in the example above, movement checking still becomes the major factor.
Phew.
EDIT: Bolded O(n) notation and added some stuff.
LAG IBTL! IBDS! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
DarkFenix
Caldari Pilots Of Honour
|
Posted - 2007.07.08 17:13:00 -
[72]
Edited by: DarkFenix on 08/07/2007 17:13:27 Wow there certainly are a lot of large egos being swung around in here. Well, I know next to nothing about programming, computing and all that crap, but reducing lag by 50% wouldn't help at all. Why? Blob sizes would just increase by 50% and create precisely the same problem as before.
Whatever CCP do to reduce lag, the players' stupidity will always create precisely the same problem all over again.
While formations would certainly look shiny, and would probably bring in numerous more tactical considerations, it sounds like it would stop Eve being what it is in favour of a) making it a meta-gamer's heaven (ie. even more than it is now) and b) removing more player control, bringing in more automation.
I don't want to bring my ship along to a fleet battle so someone else can then control it for me. In fact, I'll stick to the far more reliable way of having non-laggy fights. I won't fly as part of a 500 man blob.
|
Reithan
Caldari LEGI0N SOUL CARTEL
|
Posted - 2007.07.08 23:37:00 -
[73]
Originally by: DarkFenix Edited by: DarkFenix on 08/07/2007 17:13:27 Wow there certainly are a lot of large egos being swung around in here. Well, I know next to nothing about programming, computing and all that crap, but reducing lag by 50% wouldn't help at all. Why? Blob sizes would just increase by 50% and create precisely the same problem as before.
Whatever CCP do to reduce lag, the players' stupidity will always create precisely the same problem all over again.
You're right, we should all just cancel our account now and let CCP close the servers. I mean why bother fixing problems with the game, that's just lunacy!!!
Originally by: DarkFenix While formations would certainly look shiny, and would probably bring in numerous more tactical considerations, it sounds like it would stop Eve being what it is in favour of a) making it a meta-gamer's heaven (ie. even more than it is now) and b) removing more player control, bringing in more automation.
I think most of these concerns have already been covered. Feel free to discuss them again in some meaningful manner...if you can handle that.
Originally by: DarkFenix I don't want to bring my ship along to a fleet battle so someone else can then control it for me. In fact, I'll stick to the far more reliable way of having non-laggy fights. I won't fly as part of a 500 man blob.
Yeah, the best way to fix fleet battles is to not have fleet battles! How did I not miss that? Jeeze, is my face red.
Yeah, and having someone else gangwarp for you, pick your targets for you ("Primary is ____! Secondary is _____!"), etc, is complete free control on your part, right? As stated before your squad/fleet leader can ALREADY control your movement through gang-warps and the "Regroup" command. Formations would change very little...
Unless you forgot to read the thread and are yet ANOTHER person thinking that we meant that the formation command controls your locking and guns? -----------------------------------------------
|
Acacia Everto
State War Academy
|
Posted - 2007.07.09 00:29:00 -
[74]
This does sound like a decent idea...as long as it's implemented after the new graphics engine, that way the overview calculations can be done client-side without slowing down the game.
Also, how would you handle bumping of individual ships? Some explanation that Formation Link Generators (possibly a module only able to be used by BCs/Command Ships, which would be appropriate) keep all ships in place, thus making the formation a single bumpable entity?
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.09 00:36:00 -
[75]
Originally by: Acacia Everto This does sound like a decent idea...as long as it's implemented after the new graphics engine, that way the overview calculations can be done client-side without slowing down the game.
Also, how would you handle bumping of individual ships? Some explanation that Formation Link Generators (possibly a module only able to be used by BCs/Command Ships, which would be appropriate) keep all ships in place, thus making the formation a single bumpable entity?
Yes, for the purposes any location/positional computations would treat the formation as a single entity instead of particular ships, so you couldn't bump a single ship - just the formation itself.
LAG IBTL! IBDS! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
Deja' Vu
Super Secret Hot Chic's Hideout
|
Posted - 2007.07.09 01:21:00 -
[76]
This would work well for large scale fleet warfare. Even if formations were loaded up into groups of two.
Not sure what the big worry about movement would be as ships last about 5 seconds in a fleet scenario once they are primaried.
Here is one problem with the formation, who's mass and agility do all the ships attain and how would that be applied skillwise?
It would have to be class oriented forming only. Using the highest formation skilled person like gangs are atm.
I could see this being buggy if not used as such with BS formations having an Inty formed getting his skills, mass, agility and control.
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.09 01:27:00 -
[77]
Originally by: Deja' Vu This would work well for large scale fleet warfare. Even if formations were loaded up into groups of two.
Not sure what the big worry about movement would be as ships last about 5 seconds in a fleet scenario once they are primaried.
Here is one problem with the formation, who's mass and agility do all the ships attain and how would that be applied skillwise?
It would have to be class oriented forming only. Using the highest formation skilled person like gangs are atm.
I could see this being buggy if not used as such with BS formations having an Inty formed getting his skills, mass, agility and control.
We had decided previously that the mass and agility of the formation would be the mass and agility of the slowest ship x the number of ships - so a fleet with 6 battleships and an interceptor would have the mass of 7 battleships and the agility of a battleship.
LAG IBTL! IBDS! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
Deja' Vu
Super Secret Hot Chic's Hideout
|
Posted - 2007.07.09 01:32:00 -
[78]
Well thinking of my last post I think this could be a partial solution.
1. As stated before Hull Class forming Only.
2. Each formation when formed would be calculated as an average mass so the Formation leaders skills and bonus's can be applied easily.
Now the rest of the tough questions I have no idea how to solve.
1. What would the collision parameters be of a formation?
2. Should Formations be able to fly through each other with a small collision focal point to prevent the inability to warp your formation out in emergency situations.
3. How would the formation be affected if the leader is killed? Would it be assigned to a random person already existing in formation?
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.09 01:40:00 -
[79]
Originally by: Deja' Vu Well thinking of my last post I think this could be a partial solution.
1. As stated before Hull Class forming Only.
2. Each formation when formed would be calculated as an average mass so the Formation leaders skills and bonus's can be applied easily.
Now the rest of the tough questions I have no idea how to solve.
1. What would the collision parameters be of a formation?
2. Should Formations be able to fly through each other with a small collision focal point to prevent the inability to warp your formation out in emergency situations.
3. How would the formation be affected if the leader is killed? Would it be assigned to a random person already existing in formation?
1. Th collision parameters could be a sphere with the radius being the number of ships in the formation...A.K.A. a formation of battleships would have a collision radius of 5 x battleships radiating out from the center of the formation - this would require almost the same processing power as calculations for individual ships
2. Interesting question, but if we keep things simple then no
3. I think formations should operate like gangs...if something happens to the gang leader than someone else has to become the leader, and gangs can be freely joined and left.
LAG IBTL! IBDS! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
The RepoMan
Caldari Red Horizon Inc Red Horizon
|
Posted - 2007.07.09 03:33:00 -
[80]
...... you seem to be addressing this as though the problems with blobs is the lag they cause... the problems with blobs is that they are pointless laggy stupid battles that revolve around who's faster on the draw vs lag. If you remove the lag portion, they're still stupid and pointless. So basically even if your idea was sound, which it really isnt, you havent really thought or at least not written about how these formations would work, it still wouldnt fix anything, just move the problem around a bit.
|
|
The 0verseer
|
Posted - 2007.07.09 04:02:00 -
[81]
Originally by: The RepoMan ...... you seem to be addressing this as though the problems with blobs is the lag they cause... the problems with blobs is that they are pointless laggy stupid battles that revolve around who's faster on the draw vs lag. If you remove the lag portion, they're still stupid and pointless. So basically even if your idea was sound, which it really isnt, you havent really thought or at least not written about how these formations would work, it still wouldnt fix anything, just move the problem around a bit.
your opinion about fleets being stupid has no bearing on the subject so gtfo
imo giving up flight control during fleet would have little impact on the individual as all you do is align and shoot anyways. why not give the task of aligning to the fleet commander and not have to worry about it yourself? i dont see why people would be worried about this.
using a bonus system for formation flying seems like a good idea as you can still break your ship from formation incase you get primaried etc. but you are still usefull and it provides a good enough reason for fleets to form up.
also, stop complaining that this wouldn't cure all lag. any improvement is a good improvement and i think this could possibly help a lot. i dont see ccp doing anything worthwhile currently to reduce lag, so something like this is needed imo.
|
Kransthow
AWE Corporation Intrepid Crossing
|
Posted - 2007.07.09 06:18:00 -
[82]
I have been thinking about the possibility of formation flying for a while as well and have a few thoughts about it. Formation flying could not only be used to decrease lag in fleet battles, but to slow them down as well (Which CCP seems quite keen on).
Instead of being able to target individual ships in a formation, perhaps you could only target the formation as a whole and all your shots would be randomly distrubuted between the formation (Possibly depending on signature radius which ships are more likely to be shot). This would cause it to take alot longer to destroy ships in a formation, and therfore slow combat down by giving the formation commander much greater time to decide if it's time to split.
|
Tarminic
Black Flame Industries
|
Posted - 2007.07.09 06:22:00 -
[83]
Originally by: Kransthow I have been thinking about the possibility of formation flying for a while as well and have a few thoughts about it. Formation flying could not only be used to decrease lag in fleet battles, but to slow them down as well (Which CCP seems quite keen on).
Instead of being able to target individual ships in a formation, perhaps you could only target the formation as a whole and all your shots would be randomly distrubuted between the formation (Possibly depending on signature radius which ships are more likely to be shot). This would cause it to take alot longer to destroy ships in a formation, and therfore slow combat down by giving the formation commander much greater time to decide if it's time to split.
Unfortunately I think that may have the opposite effect indirectly. By dividing fire among the individual ships of the fleet it means that 5 battleships, all tanking, could hold off a much larger attacking force because their fire would be divided amont 5 different ships. It also means that an equally-sized formation on each side might not be able to destroy each other at all. LAG IBTL! IBDS/DC! IBTC! 1st in a BoB post! And other such forum tom-foolery. |
Kransthow
AWE Corporation Intrepid Crossing
|
Posted - 2007.07.09 06:35:00 -
[84]
Edited by: Kransthow on 09/07/2007 06:42:37
Originally by: Tarminic
Originally by: Kransthow I have been thinking about the possibility of formation flying for a while as well and have a few thoughts about it. Formation flying could not only be used to decrease lag in fleet battles, but to slow them down as well (Which CCP seems quite keen on).
Instead of being able to target individual ships in a formation, perhaps you could only target the formation as a whole and all your shots would be randomly distrubuted between the formation (Possibly depending on signature radius which ships are more likely to be shot). This would cause it to take alot longer to destroy ships in a formation, and therfore slow combat down by giving the formation commander much greater time to decide if it's time to split.
Unfortunately I think that may have the opposite effect indirectly. By dividing fire among the individual ships of the fleet it means that 5 battleships, all tanking, could hold off a much larger attacking force because their fire would be divided amont 5 different ships. It also means that an equally-sized formation on each side might not be able to destroy each other at all.
True, I hadn't thought of that, but then the whole enemy would have to be tanking orientated to achieve this. And if this happened it would boost the popularity of smartbombing and stealth bombers, which makes the game a little more exciting. For example if a squad of 5 tanking abbadons jumps into your system you drop 5 smart bombing domi's on their formation and watch the explosions .
P.S. Just though of another way to stop this happening, get your tacklers to warp scram them, (Which would make one break formation) and then proceed as normal
|
Kharadran Sullath
Caldari IntoXication Inc
|
Posted - 2007.07.09 07:46:00 -
[85]
On the topic of formations, not lagg: If bonuses would be given to ships in formation (aforementioned shield harmonics, damage increases and whatnot), a whole new tactical element in fleet battles could be breaking formations. This role could be given to smaller ships, who then would be given some new means to break formations (e.g. specialized weapon/ewar mod). Then all fleet battles would be Star Wars epic and it would be teh awesomes!
Well, one can always dream. ^^ ------ --Don't get saucy with me Bernaise!-- |
BluOrange
Gallente Agony Unleashed Agony Empire
|
Posted - 2007.07.09 08:10:00 -
[86]
Originally by: Reithan
Originally by: Lisento Slaven How would this look on an enemy pilots overview? I'm trying to picture it in my head on how there is no location data for the ship being processed or stored yet...there is still location data present for each individual ship in relation to - at least - the enemy ships not part of the formation. I can see how it would remove location data for friendlies within the formation but not for everyone on grid.
Ok, as far as the overview goes, nothing would change at all.
INTERNALLY, your overview would receive one location for each formation, the shape of that formation a link the each pilot in the formation. It would then calculate, from this data the actual data for each ship and display it on your overview.
So, to the eyes of the enemy, the overview would not change AT ALL.
-----
The way this removes location data for the ships even for enemies is such:
Right now, the server has to check each ship to see if it's touching another ship. It does this by knowing the locations and shapes of the ships and comparing all this data in every way possible.
With formations, it would only be comparing the locations and shapes of the formations - not individual ships.
I would be stunned if the amount of lag generated by that activity on the server was more than infintesimal.
The movement-related lag comes from sending updates of ship locations and other needed information to all the parties in the combat. (That, and the fact that the communication protocol is TCP/IP, which is more reliable than the UDP that game programmers usually use, but can be much slower.) Your proposal doesn't seem to change that at all.
Recruitment FAQ |
BluOrange
Gallente Agony Unleashed Agony Empire
|
Posted - 2007.07.09 08:15:00 -
[87]
Originally by: Kransthow
P.S. Just though of another way to stop this happening, get your tacklers to warp scram them, (Which would make one break formation) and then proceed as normal
So, warp scrambling pops ships out of formation? Warp scramblers and webbers are among the most commonly-used modules in PVP. At which point, there's going to be a bunch of extra calculations as the formations get broken up and all of the stuff that was calculated for the formation has to be recalculated. In other words, it'll reduce lag until the fighting starts, at which point there will be just as much lag as before, plus a spike?
Recruitment FAQ |
Kransthow
AWE Corporation Intrepid Crossing
|
Posted - 2007.07.09 08:49:00 -
[88]
Originally by: BluOrange
Originally by: Kransthow
P.S. Just though of another way to stop this happening, get your tacklers to warp scram them, (Which would make one break formation) and then proceed as normal
So, warp scrambling pops ships out of formation? Warp scramblers and webbers are among the most commonly-used modules in PVP. At which point, there's going to be a bunch of extra calculations as the formations get broken up and all of the stuff that was calculated for the formation has to be recalculated. In other words, it'll reduce lag until the fighting starts, at which point there will be just as much lag as before, plus a spike?
A warp scrambler would only pop one ship out of a formation (A webber will just slow down the entire formation), and unless you have a tackler for every ship in the apposing fleet the lag will be less than without formations. |
Digital Anarchist
|
Posted - 2007.07.09 11:56:00 -
[89]
I kind of like the idea. Lots of info. CCP should definitely do this! Get your fix today |
Terazuk
|
Posted - 2007.07.09 12:20:00 -
[90]
Reithan you mentioned OOP?
I could be wrong but I was under the perhaps misguided impression that Eve is a hodge podge of scripting languages consisting mainly of Python but also a nightmare cavalcade of many different procedural high level languages...
If this is true, then some (more) OOP may well be the cure. ~
"*BANG* you're dead!"
|
|
|
|
|
Pages: 1 2 [3] 4 5 6 7 8 9 10 11 12 13 14 15 16 .. 16 :: one page |
First page | Previous page | Next page | Last page |