These forums have been archived and are now read-only.

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

Player Features and Ideas Discussion

 
  • Topic is locked indefinitely.
 

[Suggestion] Splitting away parts of the game client

Author
L4ST
Sebiestor Tribe
Minmatar Republic
#1 - 2012-09-14 22:23:42 UTC  |  Edited by: L4ST
forget


about


it
Loius Woo
Sebiestor Tribe
Minmatar Republic
#2 - 2012-09-14 22:28:07 UTC
No.

Just no.

Not even a little bit.

Troll?

Also, NO.
L4ST
Sebiestor Tribe
Minmatar Republic
#3 - 2012-09-14 22:31:06 UTC
Bla.

Just blah.

Not even a single argument.

Troll.

You, DEFINITELY.





Get out of this thread.
Spr09
Abyssal Echoes
Invidia Gloriae Comes
#4 - 2012-09-14 22:53:56 UTC
L4ST wrote:
Bla.

Just blah.

Not even a single argument.

Troll.

You, DEFINITELY.





Get out of this thread.


You think he's wrong? Well get your ears ready: HE'S NOT.

Dividing into separate clients means a couple of things; one, they can manage their PI and buy/sell from the market without actually logging in. Two, it just makes the game more complicated rather than just having one client for eve and another for D514. Three, an absolute **** ton of work CCP would need to do to make separate clients. and four, the client is fine the way it is.
L4ST
Sebiestor Tribe
Minmatar Republic
#5 - 2012-09-14 23:13:59 UTC
Just as impolite as the first poster, but wow - an argument. Rare appearance in this forum.
"Without actually logging in" is completely wrong, as they do have to log in into whatever client would be used for PI. Whoever planted the idea in your head that a "PI Client" would be the newest text based EVE Trainer +12 trolled you hard. This sure wouldn't be some sort of lighweight non-graphical thing. It should be an actual game client.

And whoever sais the client would be fine the way it is: even CCP wouldn't believe those lies. Huge performance issues for large fleets, mechanics split into several different places and related bugs, lagging whereever you look even on good machines, etc. etc. etc.
Nestara Aldent
Citimatics
#6 - 2012-09-14 23:35:27 UTC
It could be done, with one complete and several restricted clients, so, for example, trader can log in the game with market client and trade. In theory.

In practice, that means developer team will have to build and maintain several clients instead of one. So its a terrible idea, which would just increase a lot time and manpower needed for client development.
L4ST
Sebiestor Tribe
Minmatar Republic
#7 - 2012-09-14 23:45:30 UTC  |  Edited by: L4ST
I don't know how much you know about developement, but from my experience I'd rather maintain seperated systems than the big lot. The building is the problem, but I think I covered that in the original post, and it still remains my only concern. However f**k any effort for CCP, if its too much they wont do it. Nobody can tell if sth is really doable or not, the developers will have to judge that. We just make suggestions I guess :)

*edit: and bitter/spiteful/contentless/trolling posts as it seems looking at some miserable beings in this forum
Omnathious Deninard
University of Caille
Gallente Federation
#8 - 2012-09-14 23:55:29 UTC
It would seem that the only way to tell you no would be to not post to this thread. The reasons why it is not a good idea have been said above.

If you don't follow the rules, neither will I.

Jett0
Ministry of War
Amarr Empire
#9 - 2012-09-15 01:20:46 UTC
Have you read up on CREST yet?

Also -> In the Dust 514: Seeding the Universe presentation, there's a very quick mention of the idea that building up your Dust districts would be a separate game.

Splitting the current game? Waste of time and resources for too little impact.
New sections, in their own fully-developed games? That, I can get behind.

Occasionally plays sober

Loius Woo
Sebiestor Tribe
Minmatar Republic
#10 - 2012-09-15 03:56:49 UTC
L4ST wrote:
Just as impolite as the first poster, but wow - an argument. Rare appearance in this forum.
"Without actually logging in" is completely wrong, as they do have to log in into whatever client would be used for PI. Whoever planted the idea in your head that a "PI Client" would be the newest text based EVE Trainer +12 trolled you hard. This sure wouldn't be some sort of lighweight non-graphical thing. It should be an actual game client.

And whoever sais the client would be fine the way it is: even CCP wouldn't believe those lies. Huge performance issues for large fleets, mechanics split into several different places and related bugs, lagging whereever you look even on good machines, etc. etc. etc.


First, try looking at my record of posts in this forum, you will see that I usually present a well established logical argument.

Second, realize that when you post in F&I, you have to say "why" your idea is a good one, not just what it is.

Third, lose the attitude.

Now, I usually don't post logical arguments in threads where the OP has completely failed to outline why their idea is worth a shite anyway, why should I when you haven't tried in the first place? but I will here just because you provoked me.

Are you familiar with network theory? If you have a network of a certain number of nodes that all require connection to every other node, then the number of connections required is equal to n(n - 1) / 2 where n is the number of nodes. As you grow, the number of connections grows exponentially such that a network of 3 nodes requires 3 connections, but a network of 10 nodes requires 45 connections.

In the terms of your first post, you want a new client for each style of gameplay, but that means that each client has to be able to talk to one another, thats a connection, a connection that must be coded and tested and that takes time (look at how much time it takes to connect Eve and Dust).

Now you want to do that all over again several times, the effort required grows exponentially.

Now consider that each client has to be able to interact in various ways (corp management is different from market transactions).

now you have a complex exponential mesh network of heterogenous data transfers (thats information that is not all the same).

So take that number for ten nodes and square it, so 45^45 power. Thats, 2.48x10^74 connections (2 with 74 zeros) thats more atoms that there are in the planet Earth.

So you see it is rather unwieldy to accomplish.

That is not even mentioning the sociological complications to the eve political system of breaking everything up into seperate clients. Nor the impact on people who still want to mix and match different play styles.

Now, all of that would be simply termed "an engineering problem" if, IF, you had taken two seconds to explain WHY it should be done in the first place, but you didn't, so why should I take the time to educate you on why your idea is garbage when you can't even take the time to explain why it should be good?

Now, please go away and let this thread die like it deserves.
Omnathious Deninard
University of Caille
Gallente Federation
#11 - 2012-09-15 04:14:43 UTC
Loius Woo wrote:
L4ST wrote:
Just as impolite as the first poster, but wow - an argument. Rare appearance in this forum.
"Without actually logging in" is completely wrong, as they do have to log in into whatever client would be used for PI. Whoever planted the idea in your head that a "PI Client" would be the newest text based EVE Trainer +12 trolled you hard. This sure wouldn't be some sort of lighweight non-graphical thing. It should be an actual game client.

And whoever sais the client would be fine the way it is: even CCP wouldn't believe those lies. Huge performance issues for large fleets, mechanics split into several different places and related bugs, lagging whereever you look even on good machines, etc. etc. etc.


First, try looking at my record of posts in this forum, you will see that I usually present a well established logical argument.

Second, realize that when you post in F&I, you have to say "why" your idea is a good one, not just what it is.

Third, lose the attitude.

Now, I usually don't post logical arguments in threads where the OP has completely failed to outline why their idea is worth a shite anyway, why should I when you haven't tried in the first place? but I will here just because you provoked me.

Are you familiar with network theory? If you have a network of a certain number of nodes that all require connection to every other node, then the number of connections required is equal to n(n - 1) / 2 where n is the number of nodes. As you grow, the number of connections grows exponentially such that a network of 3 nodes requires 3 connections, but a network of 10 nodes requires 45 connections.

In the terms of your first post, you want a new client for each style of gameplay, but that means that each client has to be able to talk to one another, thats a connection, a connection that must be coded and tested and that takes time (look at how much time it takes to connect Eve and Dust).

Now you want to do that all over again several times, the effort required grows exponentially.

Now consider that each client has to be able to interact in various ways (corp management is different from market transactions).

now you have a complex exponential mesh network of heterogenous data transfers (thats information that is not all the same).

So take that number for ten nodes and square it, so 45^45 power. Thats, 2.48x10^74 connections (2 with 74 zeros) thats more atoms that there are in the planet Earth.

So you see it is rather unwieldy to accomplish.

That is not even mentioning the sociological complications to the eve political system of breaking everything up into seperate clients. Nor the impact on people who still want to mix and match different play styles.

Now, all of that would be simply termed "an engineering problem" if, IF, you had taken two seconds to explain WHY it should be done in the first place, but you didn't, so why should I take the time to educate you on why your idea is garbage when you can't even take the time to explain why it should be good?

Now, please go away and let this thread die like it deserves.

I have only one word for this post Loius Woo, Epic!
to the OP everyone posts bad ideas and or comments every now and then, it is best when you realize you did that to just drop it rather than dig a bigger hole and try to defend it.

If you don't follow the rules, neither will I.

L4ST
Sebiestor Tribe
Minmatar Republic
#12 - 2012-09-15 10:58:51 UTC  |  Edited by: L4ST
Are you saying that a client acts as a "node" and thus has to be connected to any other one?

I can see a lot of theory here, bot don't get the point. If I have a network of dozens of points (which the EVE servers are) why would I create more nodes by splitting the client? They would be sending/receiving the same informations as they are now in my eyes - with the difference that "who" sends the data from one connection is splittet from one client into several.


Obviously I didn't outline the advantages I see enough:

User:
a) Performance - Clients become "lighter": more performance for what I am doing right now
b) Accessability - If I want to do PI or market stuff only, why wouldn't I waste time loading other resources
c) Low-End-Performance machines can still "play" - whereas "play" is connected to some sort of enjoyment. If I take my crap laptop there's no way I can play EVE, even doing PI makes it struggle. I know there are a lot of players with crappy PCs that are at the same level as my laptop, which would be able to handle PI or "doing market stuff" on its own, but with the client as a whole I wouldn't bother logging in just because of the endless pain I feel when I hear my laptop almost collapsing under the weight of the client.
d) Play "your" game - so if I'm the industrial and market guy I might not care for what the combat pilots do out there, if I am an alliance leader and have to hold my people together I might not care to fit ships, etc.. With seperate client's everyone just starts "their" EVE, connected by the comms.

Dev (speculation, and no matter how much anyone here considers himself an expert, it's always speculation):
a) Expanding - small clients should be easier to expand, as the only thing to worry about is game balance, and not algorithms influencing each other.
b) Connection - There are tons of people sitting on station 24/7 not caring for 90% of the data transmitted. One could break away a part of that transmitions this way.
c) Patches - usually would become smaller, as you only update partially
d) Focussing on aspects - this is almost (a) but I think it really needs to be stressed that the fact of having for example a PI client would open a wide range of possibilities as it would have it's own engine and resources. Small circles connected with dotted lines on an ugly planet which is so ugly because maybe I'm near Jita with my poor to average PC and have to take low quality settings so I don't waste 3 Minutes undocking and getting out of there isn't really what could be expected of PI. Bittervets only seeing mechanics and nothing else, not caring about look & feel will think different here, but tbh: why would they go to the F&I, its online hell for those people.

The advantages aren't too clearly visible if you only take the current client and break it into 2-4 pieces. They become visible when you think what you can do with these pieces afterwards.

And as I said the downside would be of course setting this up, but stop saying "its too much work" because you don't know it. I don't claim it to be possible either, this is the dev's decision and the reason why I put this here is to get some resonance.

Do you really think an idea in F&I must be thought through to the last detail, filled with technical informations and background knowledge about the client, etc.? Which would mean that CCP just copy & pastes ideas from here if they are "good" and implement them? Sorry but I don't think so. I imagine some CCP forum dude seeing a thread thats discussed a lot, thinking "this might indeed be worth thinking about" and gets it to the DEVs which would not give a s**t about what anybody thought about algorithms, networks, and memory issues, but think it through on their own and maybe adapt 10% of the idea somewhere in some future feature. I imagine. Go on hating >.<


Quote:
to the OP everyone posts bad ideas and or comments every now and then, it is best when you realize you did that to just drop it rather than dig a bigger hole and try to defend it.

I see concerns:
"getting too easy"; answer: this is not some sort of EVEMon, these are actual GAME clients.
"too much work"; answer: let the DEVs decide.

The only thing I am seriously interested in is the enthousiasm anyone would show.
So far not too much to see, idea seems not to be the big one everyone waited for.
And to anyone it applies to: F**k you for critizising my enthousiasm for my idea, are you serious? You want me to STOP advocating MY idea and go outside shouting "im an IDIOT for having an idea because somebody told me so!"? lol :D going to stop responding to those guys.
L4ST
Sebiestor Tribe
Minmatar Republic
#13 - 2012-09-15 11:00:05 UTC
,
Roime
Mea Culpa.
Shadow Cartel
#14 - 2012-09-15 11:25:24 UTC
Do you actually think the PI planet view is running in the background when you are browsing the map? Or that you are loading the Jita undock grid when you update your market orders?

HINT: No, you are not using all the resources at the same time, only when you actually launch them.



.

L4ST
Sebiestor Tribe
Minmatar Republic
#15 - 2012-09-15 11:46:04 UTC
Roime wrote:
Do you actually think the PI planet view is running in the background when you are browsing the map? Or that you are loading the Jita undock grid when you update your market orders?
HINT: No, you are not using all the resources at the same time, only when you actually launch them.


No I don't. However you don't "launch" what you need, you often "launch" what you need to go through to get where you actually want to be. Speaking of the PI example, with my char docked, I wont want to load local & the WOS feature, but (assuming its activated) I have to load it after starting the game.
Considering what you said you could actually preload parts of the planetary interface when starting the PI client, making it faster AND shinier.
Roime
Mea Culpa.
Shadow Cartel
#16 - 2012-09-15 11:50:28 UTC
Others need the local, overview, dscan and inventory when they do PI.

.

L4ST
Sebiestor Tribe
Minmatar Republic
#17 - 2012-09-15 14:05:23 UTC
Roime wrote:
Others need the local, overview, dscan and inventory when they do PI.


The idea is a "modular" use of client parts - thus you would then start your "spaceships yay" client too.
Loius Woo
Sebiestor Tribe
Minmatar Republic
#18 - 2012-09-15 14:29:41 UTC
Im sorry, but you are beyond reason or help.

And no, your idea does not result in LESS server load it results in MORE.
Tippia
Sunshine and Lollipops
#19 - 2012-09-15 14:34:58 UTC  |  Edited by: Tippia
Why on earth would you want to reduce the complexity and interconnectedness, when that's a huge selling-point of the game?

It's a bad idea that doesn't particularly solve anything or provide any added value and (as Loius Woo points out) would only ever increase the server load.