open All Channels
seplocked EVE Information Portal
blankseplocked New Dev Blog: Iterative development and what's happening in 2011
 
This thread is older than 90 days and has been locked due to inactivity.


 
Pages: first : previous : ... 2 3 4 5 [6] 7 8 9 10 ... : last (66)

Author Topic

Mira O'karr
Posted - 2010.07.16 20:45:00 - [151]
 

Originally by: Bartholomeus Crane
Edited by: Bartholomeus Crane on 16/07/2010 20:40:09
Originally by: ihcn
Making AI to help test fleet fights sounds like just about the most interesting thing ever. Things like that make me wish I was graduating sooner.


Knowing a bit about Artificial Intelligence I can almost assure you that that "borderline sentient AI-testing automation feature" will be neither borderline sentient nor have proper AI in it. As for automation feature for testing: welcome to the '90ies, glad you could make it.

If CCP wants to disagree, fine, prove me wrong. I'd be happy to have a look at it. Almost free of charge.


hopeing for an AI that can not tell "hold" from "jump" and "align" from "warp"

should at least be entertaining to watch

Bartholomeus Crane
Gallente
The Crane Family
Posted - 2010.07.16 20:46:00 - [152]
 

Originally by: BeanBagKing
However, when that 18 months is up I hope you will remember this post and return the devs working on those areas so they can make some major improvements in Eve that players want to see.


You took three poast to come up with this? 18 months? Are you bloody insane? You think the players out in 0.0 will still be around in 18 months with Sov the way it is now?

This isn't just about lag, it is about all the bloody expansions that have gone before! Designed to be iterated upon, and now destined to stay on the bloody backlog for two years!

Seriously! 18 months? That's a bloody lifetime in game development! No one's going to wait that long!

What are you, a fanboy? Looking for a job?

URSODMB
Posted - 2010.07.16 20:47:00 - [153]
 

Originally by: Mira O'karr
Originally by: Bartholomeus Crane
Edited by: Bartholomeus Crane on 16/07/2010 20:40:09
Originally by: ihcn
Making AI to help test fleet fights sounds like just about the most interesting thing ever. Things like that make me wish I was graduating sooner.


Knowing a bit about Artificial Intelligence I can almost assure you that that "borderline sentient AI-testing automation feature" will be neither borderline sentient nor have proper AI in it. As for automation feature for testing: welcome to the '90ies, glad you could make it.

If CCP wants to disagree, fine, prove me wrong. I'd be happy to have a look at it. Almost free of charge.


hopeing for an AI that can not tell "hold" from "jump" and "align" from "warp"

should at least be entertaining to watch


If it wants to mimick human behaviour it will simply have these behaviours in them !!!

Listen to Barth he knows what he is on about unlike you dev muppets.

CCP Explorer

Posted - 2010.07.16 20:47:00 - [154]
 

Originally by: Malcanis
CCP assigning just one out of their 17 dev teams to working on the issues and bugs backlog, publicising the existence of this team, and maintainging a list - just a simple list - of what has been fixed by this team, what is currently being fixed, and what will be fixed next would go a long long long way to placating a currently very disillusioned and disappointed playerbase.
Slight correction here, all the non-Incarna teams allocate 20% of their time to bug fixes. The other 80% is spent on feature development, which may include an overhaul of an existing feature if that overhaul is more than a bug fix. Then there is a team working on server performance and another team on network performance and scalability.

Bartholomeus Crane
Gallente
The Crane Family
Posted - 2010.07.16 20:48:00 - [155]
 

Originally by: Mira O'karr
Originally by: Bartholomeus Crane
Edited by: Bartholomeus Crane on 16/07/2010 20:40:09
Originally by: ihcn
Making AI to help test fleet fights sounds like just about the most interesting thing ever. Things like that make me wish I was graduating sooner.


Knowing a bit about Artificial Intelligence I can almost assure you that that "borderline sentient AI-testing automation feature" will be neither borderline sentient nor have proper AI in it. As for automation feature for testing: welcome to the '90ies, glad you could make it.

If CCP wants to disagree, fine, prove me wrong. I'd be happy to have a look at it. Almost free of charge.


hopeing for an AI that can not tell "hold" from "jump" and "align" from "warp"

should at least be entertaining to watch


If that's all you're looking for, I'm sure CCP will be able to pull it off ...

Shinma Apollo
BURN EDEN
Posted - 2010.07.16 20:49:00 - [156]
 

Originally by: CCP Explorer
I deeply apologise, but your assumptions and your conclusions are all wrong.


What about your numbers that you released is even remotely indicating that your company has prioritized a side project when your entire customer base is indicating to you your core product is broken. There are several serious defects, most notably fleet fights, that you simply have failed to address. Simply because one portion of your customers doesn't experience it, or choose to experience it, should not dictate such inadequate resource distribution.

CCP you are selling a false product at the moment. Fix it, or find yourself a new playerbase.

Motriek
Selective Pressure
Rote Kapelle
Posted - 2010.07.16 20:51:00 - [157]
 

I think the disconnect is simple. A conscious decision has been made to ignore CSM feature/fix requests, and there is no way to continue communication with the CSM, except in a 'ears-plugged' fashion. If you want to close the disconnect, address some of the many issues raised by players, and voted on by the CSM. Allow the CSM to pick 3 features, or 3 man-months of work, or whatever, and just go do them. As it stands, you're leaving the CSM (and players) twisting in the wind with this idea that you know what's best for them. As it stands, we are only hearing that you know what's best for expanding the subscriber base.

The only two things raised by the CSM, that has teams focused on it, is the lag issue and PvE content (in a sense). On the issue of lag, I really do agree with CCP's handling. It makes complete sense. We all wish it wasn't a problem, but it is, and you're dedicating significant resources to it. Regarding PvE, it's already lovely, but aside from sleepers, is predictable, and unengaging. I hope the 'sekrit' PvE feature addresses that. However, neither of these really count as payting attention. Lag is a defect, and the PvE content is not anything specifically asked for.

While I recognize that all CSM submissions are not possible, the reason players are justifiably up-in-arms right now is that you are not implementing anything *new* being asked for, aside from some PvE content. Go read this:
http://wiki.eveonline.com/en/wiki/Category:CSM_Submission

And be accountable to which ones have been addressed, and which ones WILL be addressed in the next six months. Either that, or skip all upcoming CSM elections until you plan on paying attention.

Gripen
Posted - 2010.07.16 20:52:00 - [158]
 

Out of around 300 people in CCP only 14 working on the stuff I care about... Great news...

CCP Explorer

Posted - 2010.07.16 20:52:00 - [159]
 

Originally by: Gerard Deneth
No Offence, but one of the first things I learned in the single HR class I took on the way to getting my BBA:Accounting degree was that "claiming tens of defect fixes without being able to cite a single one" is something that quite handily erodes support and belief.
Sorry, I can't link you the defects nor the changelists since they are all in internal systems.

Mira O'karr
Posted - 2010.07.16 20:52:00 - [160]
 

Originally by: Bartholomeus Crane
Originally by: Mira O'karr
Originally by: Bartholomeus Crane
Edited by: Bartholomeus Crane on 16/07/2010 20:40:09
Originally by: ihcn
Making AI to help test fleet fights sounds like just about the most interesting thing ever. Things like that make me wish I was graduating sooner.


Knowing a bit about Artificial Intelligence I can almost assure you that that "borderline sentient AI-testing automation feature" will be neither borderline sentient nor have proper AI in it. As for automation feature for testing: welcome to the '90ies, glad you could make it.

If CCP wants to disagree, fine, prove me wrong. I'd be happy to have a look at it. Almost free of charge.


hopeing for an AI that can not tell "hold" from "jump" and "align" from "warp"

should at least be entertaining to watch


If that's all you're looking for, I'm sure CCP will be able to pull it off ...


you seem awfully sure of that, good sir.

Batolemaeus
Caldari
Free-Space-Ranger
Morsus Mihi
Posted - 2010.07.16 20:53:00 - [161]
 

Originally by: CCP Explorer
We are actually actively working on the "blackscreening client" issue (which is most likely a server code service locking issue as far as we can tell) and those code changes are being tested on Singularity.


No offense, but for a game breaking regression, that's a pretty bad response time. Especially coupled with the GM's refusal to offer any support in that area.

Originally by: CCP Explorer
The team that worked on EVE Gate had to build a the foundation for a web site cluster that connects to a real-time simulation cluster and provide a two-way real-time communication between the two clusters. The web site had to be architectured to scale to 100's of thousands of users. A lot of time was spent on testing and performance analysis, as well as defensive coding and exploit testing. On top of that the team had to provide the features for the first iteration of the web site.

May I ask why the groundwork wasn't used to provide a fully functional API access to these features, with Evegate as a frontend?
You do know that you willingly created a situation like in pre-api days, where tools like Evemon required users to use their login details to access web pages and parse them? This time it's Evegate and the exclusive features it uses.
Also, why on earth did you create an attempt at a social networking site without the social networking features? Facebook is so successful partly because of the interfacing it provides, not because it's locked down.

Lolion Reglo
Caldari Provisions
Posted - 2010.07.16 20:53:00 - [162]
 

figures... dust 514 AND incarna come out the year im deployed...lol. i had better get a decent internet connection out there to be able to play these on my time off...lol.

Mors Magne
Astral Adventure
Posted - 2010.07.16 20:54:00 - [163]
 

I like the direction Eve is heading in, and it's well oranised.

Here's to an infinate number of years playing Eve Online.

Dzajic
Gallente
Posted - 2010.07.16 20:55:00 - [164]
 

I dont know, is "White Wolf" still publishing rulebooks and stuff, and is CCP cashing in on it? As they like own them... But even if so, its a tiny amount of cash.

All work and Incarna, Dust and still "secret and unannounced" WOD MMO is payed by our, EVE subscriber money. OK, CCP is a free company, they can do whatever they think is best for them. They should really pray that Blizzards next MMO isn't fantasy setting (or god forbid, "urban dark fantasy").

Now, 70 people walking on walking in stations is a bit extremely too much. I only hope that Incarna is planned to be the basis of WOD engine so its shared work. Otherwise it is obscenely too much resources devoted to something that can be categorized as glorified station UI redesign.

And I agree that CCP Warlcok's paper sound delicious to my CS&IT eyes and ears, is it public? Please CCP, pretty please?

Ix Forres
Caldari
Righteous Chaps
Posted - 2010.07.16 20:55:00 - [165]
 

Edited by: Ix Forres on 16/07/2010 21:00:15
Originally by: iP0D
Also, please, no selective quoting :-)



Nail on the head there, excuse the selective quote for brevity. To make my own response...

Originally by: CCP Explorer
The team that worked on EVE Gate had to build a the foundation for a web site cluster that connects to a real-time simulation cluster and provide a two-way real-time communication between the two clusters. The web site had to be architectured to scale to 100's of thousands of users. A lot of time was spent on testing and performance analysis, as well as defensive coding and exploit testing. On top of that the team had to provide the features for the first iteration of the web site.


Yes, and the technical challenges were many, nobody doubts that. However, we're still talking a year to accomplish what amounts to an incredibly basic website. Yes, you have to make sure that it can scale - but scalable websites are not new, and making it work perfectly for 300,000 users is not a challenge in modern terms. Testing and performance analysis is just part of making the website, not an excuse. If you weren't doing that I'd be shocked.

Connecting to a realtime simulation cluster- this is actually fairly trivial if you didn't suffer total not-invented-here; what you need is a messaging framework between the two, there are many readily available tools to do just this on much larger workloads than EVE Gate would ever demand, and they're used by hundreds of companies for huge installs that dwarf EVE Gate; automated low-latency trading in stock exchanges and the like. If you did the sensible thing, set up some sort of messaging broker (say AMQP, since it's a well accepted open standard used heavily already), wrote the appropriate code to handle the endpoints on EVE Gate and the cluster's side, I would estimate you could do that, done and tested, in 2 months with one or two people. Maybe 3 months if the server code is really nasty stuff to hook into.

But as iP0D said - what has EVE Gate gotten us? It's gotten us entirely the wrong product. It may be a good product, it may have been developed as fast as humanly possible, but it is the wrong product for the marketplace, for the community and for CCP as a company. I have asked many times to see statistics showing how many people actually use EVE Gate. So far, we've heard nothing. If CCP were confident that EVE Gate was living up to it's own internal expectations they'd be happy to share those with us, no?

And the point made above is also extremely sound given that the original COSMOS devblog promised it; why aren't EVE Gate features being made available in APIs instead of encouraging people to revert to the dark ages of fake HTTP requests and putting your login details into third party apps?

Edit: Also, to note, EVE Metrics also communicates in near-realtime with background processing tools, and we also maintain public APIs and interfaces to our data streams; we essentially operate a cluster of our own. Not as large in scale as EVE, but the principles are the same and they've been tested to scale to a fairly massive capacity (tens of thousands of messages a second).

Liang Nuren
Posted - 2010.07.16 20:56:00 - [166]
 

Originally by: CCP Explorer
Originally by: Malcanis
CCP assigning just one out of their 17 dev teams to working on the issues and bugs backlog, publicising the existence of this team, and maintainging a list - just a simple list - of what has been fixed by this team, what is currently being fixed, and what will be fixed next would go a long long long way to placating a currently very disillusioned and disappointed playerbase.
Slight correction here, all the non-Incarna teams allocate 20% of their time to bug fixes. The other 80% is spent on feature development, which may include an overhaul of an existing feature if that overhaul is more than a bug fix. Then there is a team working on server performance and another team on network performance and scalability.


That's a hell of a fact to leave out of the dev blog.... ugh

-Liang

BeanBagKing
Terra Incognita
Intrepid Crossing
Posted - 2010.07.16 20:56:00 - [167]
 

Originally by: Bartholomeus Crane
Originally by: BeanBagKing
However, when that 18 months is up I hope you will remember this post and return the devs working on those areas so they can make some major improvements in Eve that players want to see.


You took three poast to come up with this? 18 months? Are you bloody insane? You think the players out in 0.0 will still be around in 18 months with Sov the way it is now?

This isn't just about lag, it is about all the bloody expansions that have gone before! Designed to be iterated upon, and now destined to stay on the bloody backlog for two years!

Seriously! 18 months? That's a bloody lifetime in game development! No one's going to wait that long!

What are you, a fanboy? Looking for a job?


No, I simply recognize the fact that nothing will change in 18 months. CCP is a HUGE company with profits to look after and some major products being developed. As much as people here may complain about 70 devs on Incarna and a max of 37 on the rest of eve, nothing there will change. Yes, this sucks, but it's the truth, nothing you or anyone else can say will change that 18 month timeline. You have a large corporation locked into it with a TON of assets invested, the second coming won't change that timeline and I've simply accepted that.

Given that it won't change now, my only hope is that CCP -will- change once the new timeline comes out. Some of the current players will be gone, but there's hope for the next batch that things will be better for them, and maybe some of the old ones will come back. You note (if you read my 3 pages) that I feel the same way you do, past expansions that were meant to be iterated upon, but are now backlogged with less than 22 devs working to catch up.

So no, I'm not insane, just realistic.

Virtuozzo
The Collective
Against ALL Authorities
Posted - 2010.07.16 20:57:00 - [168]
 

Originally by: CCP Explorer
Originally by: Malcanis
CCP assigning just one out of their 17 dev teams to working on the issues and bugs backlog, publicising the existence of this team, and maintainging a list - just a simple list - of what has been fixed by this team, what is currently being fixed, and what will be fixed next would go a long long long way to placating a currently very disillusioned and disappointed playerbase.
Slight correction here, all the non-Incarna teams allocate 20% of their time to bug fixes. The other 80% is spent on feature development, which may include an overhaul of an existing feature if that overhaul is more than a bug fix. Then there is a team working on server performance and another team on network performance and scalability.


Speaking percentages, has CCP ever at an organisational level considered the merit in both managerial, development and creative processes of adopting a format that builds upon the synergy of smart people having fun, similar to Google's 20% Rule? EVE's IP is in terms of enterprise development doctrines beyond ideal for an implementation of that concept.


Rhaegor Stormborn
BURN EDEN
Northern Coalition.
Posted - 2010.07.16 20:59:00 - [169]
 

How can CCP write this nonsense and keep a straight face? How can you tell us you care about Eve Online the spaceship game which pays your salaries and write the nonsense in this blog? Get a grip. Quit being ******s.

ihcn
Posted - 2010.07.16 20:59:00 - [170]
 

Really though I want to hear more about that AI stuff, if there is enough at this point to write about.

As for the 'bulk' of this thread, I can certainly appreciate ccp's desire to push dust and incarna out the door. Especially considering that incarna has been a punchline for a long time regarding how long it takes ccp to accomplish things, you're in kind of a catch 22. So making one big push to get incarna and dust out the door is fine.

I think the major frustration is the perception that there is nobody allocated to fixing current problems with eve, aside from the lag of course. This dev blog somewhat confirms that.

When you consider that up until now one of the primary jobs of the csm has been to identify problems within eve and communicate those problems with ccp, the issue people are having with the csm process should be obvious. Even smaller problems have nobody to fix them, so all ccp can tell the csm is we'll put it on our list. And i'm sure there will be lots of happy customers in winter 2011/spring 2012 when all those problems get fixed, but what about the year and a half from now and then? Are we going to have to live with all the problems have been identified and just deal with them?

This is what frustrates me most. We shouldn't have to 'put up' with a lot of the problems that exist, and i don't even mean the biggest stuff, i just mean UI stuff like modify order and cancel order being right next to each other. How much 'dev time' would it take to add a separator to that menu? And despite the ease of resolution of many of these problems, *as far as we know* none of them will be resolved, despite the fact that resolving those small issues would go a long way towards making the game easier to play for the year and a half while we wait for the big-boy updates.

Again I understand CCP's big picture plans of getting dust and incarna out the door, but it's hard not to be frustrated when you're told that truly easy-to-solve problems will go completely ignored for a year and a half.

sinfulangel
Body Count Inc.
Pandemic Legion
Posted - 2010.07.16 21:00:00 - [171]
 

Nice attempt at damage control only to confirm what everyone has been saying, if you were really committed to fixing the issues with EvE the number of dev teams committed to EvE would be the other way around.

Like Nathan said though, introducing new (ahbit broken/half finished content) that you can market to the masses will make you more money then retaining players by fixing things.

At least we know where managements priorities are.

Ban Doga
Posted - 2010.07.16 21:02:00 - [172]
 

I don't understand what you are trying to tell us.

The players are upset because their elected representatives (CSM) go to the summit to present issues selected from those brought forward by the playerbase.
The meeting minutes - approved by both CCP and the CSM - indicate that CCP already made a release plan that covers 2-3 releases and basically leaves no room for the suggestions of the CSM.
Instead there is a vague indication that resources available for bug fixing might increase after that release plan.

Now you make a new blog, telling everyone how you think what happened was rather different and much more constructive and then go on to draw that same picture:
fixed release plan until Winter expansion 2011, a large part of the resources will not be spent on improving existing content and telling us how awesome that all is.

Do you think the message will get better if you just continue to repeat it with slightly modified wording?
All you said was "this is how it will be until winter 2011" we already got that from the summit meeting minutes...

iP0D
Posted - 2010.07.16 21:03:00 - [173]
 

Originally by: CCP Explorer
Originally by: Gerard Deneth
No Offence, but one of the first things I learned in the single HR class I took on the way to getting my BBA:Accounting degree was that "claiming tens of defect fixes without being able to cite a single one" is something that quite handily erodes support and belief.
Sorry, I can't link you the defects nor the changelists since they are all in internal systems.


Um. Did you just dodge an honest comment by giving an answer to a question not raised? Or did I get confused.


Jason Edwards
Internet Tough Guy
Spreadsheets Online
Posted - 2010.07.16 21:03:00 - [174]
 

congrats on promotions

CCP Explorer

Posted - 2010.07.16 21:04:00 - [175]
 

Originally by: iP0D
Your argument is somehow, correct me if I am wrong, coming across as if the unilateral bulk of the workload did not relate to specification, research, integration prospects or the project remaining in touch with common principles of the product.
My reply was based on the technical design, programming, analysis and testing that had to be done to deliver what the design called for. I understand your question and I'm not dismissing it, but your question is more about strategy than technology and technology is my area of expertise.

ihcn
Posted - 2010.07.16 21:04:00 - [176]
 

Originally by: Bartholomeus Crane
Edited by: Bartholomeus Crane on 16/07/2010 20:40:09
Originally by: ihcn
Making AI to help test fleet fights sounds like just about the most interesting thing ever. Things like that make me wish I was graduating sooner.


Knowing a bit about Artificial Intelligence I can almost assure you that that "borderline sentient AI-testing automation feature" will be neither borderline sentient nor have proper AI in it. As for automation feature for testing: welcome to the '90ies, glad you could make it.

If CCP wants to disagree, fine, prove me wrong. I'd be happy to have a look at it. Almost free of charge.


Trust me i'm a CS major, i know exactly what AI really means in the field, that doesn't make it not incredibly interesting

Jack Dant
Minmatar
Posted - 2010.07.16 21:05:00 - [177]
 

Originally by: CCP Explorer
Originally by: Gerard Deneth
No Offence, but one of the first things I learned in the single HR class I took on the way to getting my BBA:Accounting degree was that "claiming tens of defect fixes without being able to cite a single one" is something that quite handily erodes support and belief.
Sorry, I can't link you the defects nor the changelists since they are all in internal systems.

Maybe that's one of the things that could be fixed?

CeneUJiti
Posted - 2010.07.16 21:06:00 - [178]
 

Edited by: CeneUJiti on 16/07/2010 21:08:19
Having just a dozen more devs actually working on EVE would be a huge increase looking at current (sad) figures.

I'm amazed that after all the player complaints, after all that CSM presented to you, CCP still wants to have next two eve expansions be "features", not fixes, in a situation where EVE is so starved of developers. Low sec and faction warfare were in need of tweaks and fixes for years, two years more is a bit too much. Only major thing in EVE other than making isk in highsec was 0.0 and not only is Dominion sov mechanic barely worth a meh, something in Dominion broke server performance and 0.0 sov wars have become even more painful fight with game and server than they were long before.

CCP, please. You don't need dozen or more developers working for a year to fix stuck buttons in FW plexes or to maybe balance lowsec rewards. And those need fixing, and are much much easier to achieve than to fix whatever is that you broke with Dominion server side.

Once again, CCP please take note, Jumpgate Evo and Black Prophecy are coming, and with such long timescales even Infinity might start getting closer to release. STO failing isn't a get out of jail for 18 months card.

Galdornae
Caldari
Fat Ugly Guys Security
Posted - 2010.07.16 21:08:00 - [179]
 

Sounds good except for one thing.
Have one team take 30 minutes to fix rockets.

Seriously.

P.S: Really.

Flettch
Posted - 2010.07.16 21:09:00 - [180]
 

I don't think there's anything wrong with the "feedback loop" between CCP and the CSM. It's quite obvious that they have their own ideas about the future of "EvE".

Do yourself a favor, CCP, and abolish the CSM now.

IF you want feedback from players in the future, then adopt a "Referendum" system where all subscribed players get to vote on any issues that CCP feels player feedback would be useful. One vote per account, and people have to login to vote.



Pages: first : previous : ... 2 3 4 5 [6] 7 8 9 10 ... : last (66)

This thread is older than 90 days and has been locked due to inactivity.


 


The new forums are live

Please adjust your bookmarks to https://forums.eveonline.com

These forums are archived and read-only