These forums have been archived and are now read-only.

The new forums are live and can be found at https://forums.eveonline.com/

EVE General Discussion

 
  • Topic is locked indefinitely.
123Next pageLast page
 

Asakai fight round up.

Author
Alizabeth Vea
Doomheim
#1 - 2013-01-28 01:56:33 UTC
http://themittani.com/news/asakai-aftermath-all-over-cobalt-moon

I wrote an article, see? You should go read it.

One thing that didn't make it into the article as it was out of scope:

CCP, there really needs to be a way to reinforce a node on the fly. I was told by one employee that node reinforcement can only be done at downtime. This needs to be fixed. The best part of this whole fight was the unplanned "oh holy poo-poo." Then Everyone-like 10% of logged in players-came. No timer, no objective just a giant fight. Best fun that I have ever had in Eve.

Alizabeth, staff writer themittani.com

Retainer of Lady Newelle and House Sarum.

"Those who step into the light shall be redeemed, the sins of their past cleansed, so that they may know salvation." -Empress Jamyl Sarum I

Virtue. Valor. Victory.

Rain6639
GoonWaffe
Goonswarm Federation
#2 - 2013-01-28 02:00:47 UTC
first! in a thread about an article about... tidi? node reinforce? ... did I mention "first!"
Bane Necran
Appono Astos
#3 - 2013-01-28 02:11:48 UTC  |  Edited by: Bane Necran
Alizabeth Vea wrote:
Then Everyone-like 10% of logged in players-came.


I think every single person in 0.0 space might be 10% of the total logged-in players at any given time. Lets say 1000 people showed up, or even 2000, with anywhere from 40-50k players logged in.

But more on topic, it's mainly a hardware problem, not software. In order to do what you want, they'd have to 'reinforce' every single system in game in case someone decided to fight in one. Not only would they need to buy new hardware, but there would be so much needed that they'd need to construct a new building to keep it all in.

"In the void is virtue, and no evil. Wisdom has existence, principle has existence, the Way has existence, spirit is nothingness." ~Miyamoto Musashi

Xercodo
Cruor Angelicus
#4 - 2013-01-28 02:18:11 UTC
I'm pretty sure TiDi was their solution to their inability to reinforce a node on the fly. TiDi allows any unreinforced node to survive an event like this while still not dropping packets and such that would result in modules not turning on/off and those moments when no one can load grid.

The Drake is a Lie

Frostys Virpio
State War Academy
Caldari State
#5 - 2013-01-28 02:35:54 UTC
Alizabeth Vea wrote:
http://themittani.com/news/asakai-aftermath-all-over-cobalt-moon

I wrote an article, see? You should go read it.

One thing that didn't make it into the article as it was out of scope:

CCP, there really needs to be a way to reinforce a node on the fly. I was told by one employee that node reinforcement can only be done at downtime. This needs to be fixed. The best part of this whole fight was the unplanned "oh holy poo-poo." Then Everyone-like 10% of logged in players-came. No timer, no objective just a giant fight. Best fun that I have ever had in Eve.

Alizabeth, staff writer themittani.com


As much as the fight was probably the best in quite some time, you are basicly asking them to rewrite probably most of the code to the game engine because once in a blue moon, some one make a misstake which cause an escalation of engagement followed by a counter escalation followed by the beggest dogpiling ever because everyone and thier mother want to be on a KM involving killing goons.

It probably involve additionnal blade buffer to re-route the systems on the node where the battle is taking palce so they can be sent on another unit. This would most likely require setting on "pause" everything in the affected system (either the one where the fight is happening or the others on teh same node) to create an exact snapshot of everything in there then applying it to a fresh instance on a free blade before resuming everything. This resume only lasting a fraction of second and stoping again to fix any problem caused by the first stop because of people chaging session during that time to and from any ststem that was on that node previously. Then we can actaully start back the action.
RubyPorto
RubysRhymes
#6 - 2013-01-28 02:36:39 UTC  |  Edited by: RubyPorto
I have to say, for the number of people fighting on one grid, I was amazed at how playable it was.

And on an Unreinforced node to boot. I think this is more proof that TiDi is an unmitigated success.

Maybe create a process to semi-reinforce a node? Shift all the other systems on the node to a new node so that you don't disrupt the affected node. Who knows if that's feasible, ofc.

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths." -Abrazzar "the risk of having your day ruined by other people is the cornerstone with which EVE was built" -CCP Solomon

Alizabeth Vea
Doomheim
#7 - 2013-01-28 02:59:33 UTC
I know jack all about hardware and programming. I would hope, though, there could be something to allow unplanned emergent gameplay to happen without it taking thirty minutes for me to refit for triage.

Retainer of Lady Newelle and House Sarum.

"Those who step into the light shall be redeemed, the sins of their past cleansed, so that they may know salvation." -Empress Jamyl Sarum I

Virtue. Valor. Victory.

Caviar Liberta
The Scope
Gallente Federation
#8 - 2013-01-28 03:01:08 UTC  |  Edited by: Caviar Liberta
Was there in a smallish fleet, nemesis myself. Warped in 100 off someone and managed to land 15km off a group of CONDI Scimitar. Decided me being there was pointless and warped off to a safe I had in system then spent the next 30 minutes trying to leave the system.

As its been pointed out TIDI did keep the node and or server from tipping over but its a bandaid to a bigger problem. If CCP wants to keep growing the game population then things like this will become more common place. So a long term solution to 2700+ in local needs to be addressed.
SuicidalPancake
Deep Core Mining Inc.
Caldari State
#9 - 2013-01-28 03:02:11 UTC
Unlikely they will upgrade the hardware after a single fight like that. Have a few more and they might consider it.

_ _

Frostys Virpio
State War Academy
Caldari State
#10 - 2013-01-28 03:11:56 UTC
Alizabeth Vea wrote:
I know jack all about hardware and programming. I would hope, though, there could be something to allow unplanned emergent gameplay to happen without it taking thirty minutes for me to refit for triage.


The problem is we are talking about moving tons of independant data actively changing all the time from the system handling it to another and making sure nothing is lost in the process.

Cluster flip on file server for example can be handled because most of the stuff is static. At most you only lose the transfer happening while cluster 1 was crashing till cluster woke up and took charge.

MAYBE (thats a big ass maybe) if the code can be handled with load balancing they could assign all the systems where no reinforcement was scheduled to be handled by more than one of the blades but that require more CPU to be sure all unit resolve stuff exactly the same. If this was possible and done, then maybe (another one) they could disable the extra ssytem being ressolved on the blade where the battle in happening. It would remove the load from those ssytem to help a bit while the other system would fall in a "no backup blade" status untill the load can be restored.

Major problem : EvE contain some OLD code. Old stuff is hard to modify without destroying it.
Beekeeper Bob
Beekeepers Anonymous
#11 - 2013-01-28 04:53:50 UTC
Alizabeth Vea wrote:
http://themittani.com/news/asakai-aftermath-all-over-cobalt-moon

I wrote an article, see? You should go read it.

One thing that didn't make it into the article as it was out of scope:

CCP, there really needs to be a way to reinforce a node on the fly. I was told by one employee that node reinforcement can only be done at downtime. This needs to be fixed. The best part of this whole fight was the unplanned "oh holy poo-poo." Then Everyone-like 10% of logged in players-came. No timer, no objective just a giant fight. Best fun that I have ever had in Eve.

Alizabeth, staff writer themittani.com



Ohh, I'm excited! I'm gonna run right over there.....Oh wait.

Sorry, reading propaganda from the hivemind gives me gas...I'll have to pass......Cool


Signature removed - CCP Eterne

March rabbit
Aliastra
Gallente Federation
#12 - 2013-01-28 04:55:57 UTC
Alizabeth Vea wrote:
http://themittani.com/...

Side note:
i never visit such sites....

you know: viruses, spam and all this stuff.... Cool

The Mittani: "the inappropriate drunked joke"

Andski
Science and Trade Institute
Caldari State
#13 - 2013-01-28 05:04:41 UTC
Xercodo wrote:
I'm pretty sure TiDi was their solution to their inability to reinforce a node on the fly. TiDi allows any unreinforced node to survive an event like this while still not dropping packets and such that would result in modules not turning on/off and those moments when no one can load grid.


that is precisely what happened in asakai

Twitter: @EVEAndski

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths."    - Abrazzar

Andski
Science and Trade Institute
Caldari State
#14 - 2013-01-28 05:05:40 UTC
SuicidalPancake wrote:
Unlikely they will upgrade the hardware after a single fight like that. Have a few more and they might consider it.


yeah it seems lowsec runs just fine on 486 boxes 99.99% of the time

Twitter: @EVEAndski

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths."    - Abrazzar

Alizabeth Vea
Doomheim
#15 - 2013-01-28 09:14:55 UTC
In the future, an FC is going to see an opportunity to drop on someone and they are going to get denied by the Skyteam since the node is unreinforced. That decision would be completely correct on their part. No one is going to risk supercapitals on a regular node again.

What that means is, effectively, meeting engagements will not happen in Eve, again-at least on this scale. Subcapital fleets will continue to do their own thing, but those almost never make the news. Meeting engagements, for those that are not up with military terminology, is a battle which neither side prepared for; it just kind of happens. A good example would be the battle of Gettysburg in the American Civil War. That battle essentially broke the back of General Lee's Army of Northern Virginia. They are interesting for a few reasons: no battlefield prep, often intelligence is incomplete or nonexistent, they force the commander to think on his feat. All of these can contribute to spectacular losses or victories. Without meeting engagements in Eve, every article from here on out is going to be X was coming out of reinforce. Side A formed up at whatever time. Side B at whatever time. They jumped in system and had a fight. That, I find boring.

From a player's perspective, this fight was amazing (and I say that having lost my carrier). I got the jabber ping and jumped right in, started to heal broadcasts, neut primaries, cycle drones as they took damage and so on. The events were unfolding in front of me, rather than have been posted in the war room a few days in advance. I didn't know how many ships we had, unlike in Tribute where ops would be canceled if Intel reported that we didn't have enough. (Oddly enough, that really didn't happen.) There is a saying that the USMC used at Wake Island (which we lost) and Tarawa (which we won): "Issue in doubt." For a long while there, the issue was in doubt and that made for good, fun, exciting game play. That's why this fight is going to be remembered and not some timer in Tribute.

Retainer of Lady Newelle and House Sarum.

"Those who step into the light shall be redeemed, the sins of their past cleansed, so that they may know salvation." -Empress Jamyl Sarum I

Virtue. Valor. Victory.

Sergeant Acht Scultz
School of Applied Knowledge
Caldari State
#16 - 2013-01-28 09:40:07 UTC
Nice reading.

And once again space Jesus and space Moses are right, gudfights can happen everywhere at any moment, I just can't hold my self say'in "yeah but that lag there...."

Lol

removed inappropriate ASCII art signature - CCP Eterne

Kalle Demos
Ironic Corp Name
#17 - 2013-01-28 10:03:35 UTC  |  Edited by: Kalle Demos
I like reading themittani.com and no im not a fanboi sheep.

Its like reading those stories on Intergalactic Summit, and having people on theMittani.com try far too hard to copy the style thats been old for almost 3 years now.

Sometimes for lolz I even compare articles, just to amuse myself of how obsessed some people are and what they would do just to post something 'readable'. Ofc I am not speaking about the OP here, he much like every other 'features editor' on theMittani likes to mimic the the guy at the top, which in this case would be Mittani.

I think anytime anyone gives any compliments to an article on theMittani.com for the writing styles, they should immediately give credit to the old writers in IGS, its a shame Mittani himself never did this, he out of all people should know why plagiarism is bad, being an ex-paralegal and all
Lord Zim
Gallente Federation
#18 - 2013-01-28 10:08:09 UTC
Bane Necran wrote:
But more on topic, it's mainly a hardware problem, not software.

Nope, it's a software problem.

The problem is that the finest granularity you can get on the cluster is one node for one solar system. This means that when there's 2500+ shooting, popping drones, navigating, deploying fighterbombers etc, that one node has to deal with all of that. This leads to one node having f.ex 50-100 systems, some hisec, some lowsec, and probably some nullsec/WHs too, because there's nothing really happening in them. This would mean that putting each of them on their own server (we're talking about ...what is it again, 5k systems?) would financially bankrupt CCP, and you would still not have gotten rid of the upper limit of how many people can be in a single system, all you would've done is raise that roof slightly.

The best thing we can probably hope for is that CCP were to redo the whole structure, so instead of having the processing structured around what goes on in a single solar system, the processing is structured around what one person does. That way they can distribute the load much more evenly across all the hardware, and they can develop usage-based heuristics to see which users habitually need a lot of processing power, and which just sit around in jita fapping their market orders.

The problem with this solution, however, is making sure all the required information is syncronized between all participants, and this is A Hard Task to solve. It's basically a can of worms, waiting to be opened. Anyone who's worked with multithreading in any way, shape or form knows why this is.

Cyno's lit, bridge is up, but one pilot won't be jumping home.

RIP Vile Rat

Andski
Science and Trade Institute
Caldari State
#19 - 2013-01-28 10:19:08 UTC
Kalle Demos wrote:
I like reading themittani.com and no im not a fanboi sheep.

Its like reading those stories on Intergalactic Summit, and having people on theMittani.com try far too hard to copy the style thats been old for almost 3 years now.

Sometimes for lolz I even compare articles, just to amuse myself of how obsessed some people are and what they would do just to post something 'readable'. Ofc I am not speaking about the OP here, he much like every other 'features editor' on theMittani likes to mimic the the guy at the top, which in this case would be Mittani.

I think anytime anyone gives any compliments to an article on theMittani.com for the writing styles, they should immediately give credit to the old writers in IGS, its a shame Mittani himself never did this, he out of all people should know why plagiarism is bad, being an ex-paralegal and all


i doubt any of our writers read IGS and we don't have internal style guidelines

nice trolling attempt though

Twitter: @EVEAndski

"It's easy to speak for the silent majority. They rarely object to what you put into their mouths."    - Abrazzar

Caviar Liberta
The Scope
Gallente Federation
#20 - 2013-01-28 17:17:33 UTC  |  Edited by: Caviar Liberta
Lord Zim wrote:
Bane Necran wrote:
But more on topic, it's mainly a hardware problem, not software.

Nope, it's a software problem.

The problem is that the finest granularity you can get on the cluster is one node for one solar system. This means that when there's 2500+ shooting, popping drones, navigating, deploying fighterbombers etc, that one node has to deal with all of that. This leads to one node having f.ex 50-100 systems, some hisec, some lowsec, and probably some nullsec/WHs too, because there's nothing really happening in them. This would mean that putting each of them on their own server (we're talking about ...what is it again, 5k systems?) would financially bankrupt CCP, and you would still not have gotten rid of the upper limit of how many people can be in a single system, all you would've done is raise that roof slightly.

The best thing we can probably hope for is that CCP were to redo the whole structure, so instead of having the processing structured around what goes on in a single solar system, the processing is structured around what one person does. That way they can distribute the load much more evenly across all the hardware, and they can develop usage-based heuristics to see which users habitually need a lot of processing power, and which just sit around in jita fapping their market orders.

The problem with this solution, however, is making sure all the required information is syncronized between all participants, and this is A Hard Task to solve. It's basically a can of worms, waiting to be opened. Anyone who's worked with multithreading in any way, shape or form knows why this is.


Perhaps there could be a way to prioritize grids in a system. The larger the number of people on a grid with lots of state changes gets higher priority vs the solo or small group ratting in a belt or docking in a station. Crunch down everything else in the system to help keep the grid with the most activity going smoothly.
123Next pageLast page