Ok Devs, you tell us. What would make a server transfer function difficult?

[Brene]

"It's just data"... Wow.

 

All sorts of potential storage, QoS, continuity, and integrity issues that could be at play.

 

Character transfer systems have been in place in MMOs for years now. I doubt these are road blocks competent developers come across.

[Inarai]

Character transfer systems have been in place in MMOs for years now. I doubt these are road blocks competent developers come across.

 

Should they decide to implement server transfers, it's a question of when, not if. In development, competency doesn't prevent problems from needing to be solved. It doesn't particularly make things faster, either. The best developers tend to take more time, not less.

Edited December 21, 2011 by Inarai

[Scan]

The only answer that makes sense is that they do not have a system in place to moce characters from server to server.

 

They have computer programmers working around the clock to implement this program as we speak.

 

This I can take and will suffer the queues no problem.

[LordJestR]

Character transfer systems have been in place in MMOs for years now. I doubt these are road blocks competent developers come across.

 

Go make your own game then you can come back and tell us how easy it is, mmmmmkay?

[HexCaliber]

Then you must have heard about data integrity and duplicates?

 

All you need is a player with the same name as you. And maybe that isn't even the hardest part. Just add a -1 to the name, I'm sure, the transferring player will be very happy.

But then take into account the 100.000's of other data within a database for one server alone. Have fun.

Migrating databases is not as easy as you make it out to be. This isn't just some little CD Home database.

GM's already have the facility to force name changes so your point was moot, players will happily take a temporary name as happens in all other mmo character transfer systems, assigning a new name should duplicates occur is simple. The size of a database matters not either, it is nothing more than a collection of records, the size of each record is immaterial, as each contains related data, which can be ordered and manipulated on any number of fields.

 

As my old tutor used to say, a record, is a record, is a record, do not lose sight of that simply because the size of the record changes.

 

Even early Pdp systems, incorporated the facility to extract, duplicate, sort, move, and manipulate individual records, using old reel to reel magnetic tape storage, its data management 101. The last such system I worked on was for Ford UK's parts system. Not only did we migrate their entire parts database to a system used internationally, we also designed and wrote an entirely custom dbl to mange it, while completely changing their existing record structure, to facilitate parts tracking across the globe.

 

What we have here is simple by comparison, copying a record from one database to another that uses an identical record structure.

[Jaku]

In a well written system it is JUST that, a series of organised records set out by a pre defined class, each representing a single player character, his stats and inventory.

 

Transferring records between databases is nothing more than a simple script away, their problem it is clear, is how to manage that across separate server blades, However, frankly, that too should have been taken care of during design, its basic and intrinsic database management. I have no idea what sort of programmer you claim to be, but you are clearly not very capable.

 

The structure, organisation and management of records within a database is taught to school children now, as its fundamental to data management, maintaining a healthy database and ensuring data security.

 

Wow, either someone is upset they are in a queue, or is looking for an argument.

 

Way to blow my tongue-in-cheek comment out of proportion.

[Brene]

Go make your own game then you can come back and tell us how easy it is, mmmmmkay?

 

I never said it was easy, Mr. Literate.

 

I said developers whom have done it for some time (CLEARLY MEANING THEY'RE SKILLED AND EXPERIENCED) should have no issue implementing a system which has been in effect for this genre of gaming (thus similar server set ups) for years.

 

I doubt you're a programmer though, generally you need to know your primary language (English here,) before you learn others.

[Inarai]

I doubt you're a programmer though, generally you need to know your primary language (English here,) before you learn others.

 

Making a crack about expecting programmers to use proper language, mechanics, and so forth?

 

Clearly you haven't worked with many programmers.

[Brene]

Making a crack about expecting programmers to use proper language, mechanics, and so forth?

 

Clearly you haven't worked with many programmers.

 

Actually, one of my two jobs (both involve language), revolves around communicating with programmers on a day to day basis. It also entails communicating what the programmers want said to investors and clients.

 

edit: sorry, didn't see who the poster was.

Edited December 21, 2011 by Brene

[Zhazam]

What we have here is simple by comparison, copying a record from one database to another that uses an identical record structure.

 

By hand? Where is the webapplication interface that allows users to select which server they want to move to? Where's the logic that checks if the target server already has that character? Where's the part that makes DAMN sure that if something goes wrong, you can simply rollback the change and don't leave this character hanging in the void between system?

 

No, it's not rocket science, but it's not a matter of two hours of work either. And you, like many non-developer people, underestimate the amount of work involved. And I'm quite sure that the programmers working on both the backend and the site are quite busy with something that's 'important'. What that is we don't know, it might be that they're working in this right now.

[Inarai]

Actually, one of my two jobs (both involve language), revolves around communicating with programmers on a day to day basis. It also entails communicating what the programmers want said to investors and clients.

 

A lot of programmers are pretty terrible in so far as correct grammar, mechanics, spelling, and such are concerned - that's all I'm saying there.

 

Also, if it's a communications role, I'm not surprised you're taking the viewpoint you are. As I said earlier in the thread, skill in development doesn't really tend to make things go faster - often the opposite, in fact. But, of course, without direct interaction with the work most people don't understand this.

[Sarah_Kerrigan]

I just don't see how the server queue problem gets fixed without a server transfer option. If that is too hard or unworkable for some reason, I'm willing to listen to why. Every MMO goes through server transfers, so why not do them now to free up the queues instead of letting people get more and more pissed off.

 

It's just data, so what's the hold up?

 

The game was released YESTERDAY. Would you stop moaning and wait patiently?

 

Damn minors. "Want this, want that, give me this, give us that, now, now, now, I need, we need, everyone needs, bla, bla, bla, can I haz *********** cheezeburger in my CE box so I don't get hungry while UNBOXING YOUR GAME, xxBiowarexx?"

[Hotte]

The game was released YESTERDAY. Would you stop moaning and wait patiently?

 

Damn minors. "Want this, want that, give me this, give us that, now, now, now, I need, we need, everyone needs, bla, bla, bla, can I haz *********** cheezeburger in my CE box so I don't get hungry while UNBOXING YOUR GAME, xxBiowarexx?"

 

I'd have never thought this possible but I'm 100% with Sarah here. ( Yes, we had our non-disagreement-moments )

There's still hope...for all of us.

[Brene]

A lot of programmers are pretty terrible in so far as correct grammar, mechanics, spelling, and such are concerned - that's all I'm saying there.

 

Also, if it's a communications role, I'm not surprised you're taking the viewpoint you are. As I said earlier in the thread, skill in development doesn't really tend to make things go faster - often the opposite, in fact. But, of course, without direct interaction with the work most people don't understand this.

 

I never said anything about them going faster, nor did I imply skill made things go faster. I simply stated, once again, that skill (and therefore, experience) would make it EASIER and perhaps smoother to implement than if this was a revolutionary and/or groundbreaking feat to apply to an MMO. This is a service that Mythic has had in it's repertoire for years, and much of the SWTOR team has come from BioWare-Mythic.

 

I've also studied basic programming languages such as C++, Pearl (very, very, VERY basic PEARL), Python, and several web-based languages (HTML, JScript, and I've worked with AJAX as well). So I'm not simply looking in from the glass plane being an armchair programmer. While my knowledge is nowhere near as comprehensive as of those actually being paid to program and be scriptkiddies for a living, I do understand the qualms, tribulations, and frustration that can occur if you don't take your time.

[Inarai]

I never said anything about them going faster, nor did I imply skill made things go faster. I simply stated, once again, that skill (and therefore, experience) would make it EASIER and perhaps smoother to implement than if this was a revolutionary and/or groundbreaking feat to apply to an MMO. This is a service that Mythic has had in it's repertoire for years, and much of the SWTOR team has come from BioWare-Mythic.

 

I've also studied basic programming languages such as C++, Pearl (very, very, VERY basic PEARL), Python, and several web-based languages (HTML, JScript, and I've worked with AJAX as well). So I'm not simply looking in from the glass plane being an armchair programmer. While my knowledge is nowhere near as comprehensive as of those actually being paid to program and be scriptkiddies for a living, I do understand the qualms, tribulations, and frustration that can occur if you don't take your time.

 

Easier from the standpoint that less goes wrong, sure. That doesn't change the fact that it's an inherently unpredictable exercise, however, and the furthest thing from being simple - there's so many concerns to factor.

 

And Mythic's past work is really only of relevance if they were recycling the database/server stuff. Clearly, they're not.

[Brene]

Easier from the standpoint that less goes wrong, sure. That doesn't change the fact that it's an inherently unpredictable exercise, however, and the furthest thing from being simple - there's so many concerns to factor.

 

And Mythic's past work is really only of relevance if they were recycling the database/server stuff. Clearly, they're not.

 

I'm relatively certain at least 75% of the server 'stuff' is the same. A game engine does not change how client server negotiations, server maintenance, and server setup are proceeded through.

[Hotte]

What we have here is simple by comparison, copying a record from one database to another that uses an identical record structure.

 

I'm glad that you know how their database is structured. So why don't you just create a server transfer tool, which will make sure, that nothing goes wrong. It's comparatively simple, isn't it?

 

And here I wonder why database managers and analysts earn so much money.

I'm afraid you underestimate the amount of things that can be wrong.

 

Yes, there ARE tools for database migration, but usually, every single database that deserves the name is hand tailored to specific needs. So where to begin?

 

Again, it's not that easy.

 

But of course we can have some half-assed front and backend to transfer (or better yet, just copy) data from one database to another. Enjoy the hate when things go wrong like missing items. And don't tell me people will gladly accept the loss of, say, their speeder skill if just one field was selected wrong.

I don't believe that for a milisecond.

[Brene]

I'm glad that you know how their database is structured. So why don't you just create a server transfer tool, which will make sure, that nothing goes wrong. It's comparatively simple, isn't it?

 

And here I wonder why database managers and analysts earn so much money.

I'm afraid you underestimate the amount of things that can be wrong.

 

Yes, there ARE tools for database migration, but usually, every single database that deserves the name is hand tailored to specific needs. So where to begin?

 

Again, it's not that easy.

 

But of course we can have some half-assed front and backend to transfer (or better yet, just copy) data from one database to another. Enjoy the hate when things go wrong like missing items. And don't tell me people will gladly accept the loss of, say, their speeder skill if just one field was selected wrong.

I don't believe that for a milisecond.

 

 

Generally speaking, a system like this should have been in the works side by side with game development anyway.

[Valkaern]

It's just astounding that while designing and launching a modern day MMO with a massive budget, they never once considered incorporating the tools necessary to facilitate server transfers. They must have realised they would need this.

 

I can overlook smaller tools that are neglected such ast target of target, the ability to adjust UI size, a combat log and various other smaller scale tools in gameplay, but something as fundamental as server transfers being left out makes me a bit skeptical about where this game will go from here now that I've had a taste of the decisions made.

 

I love the game itself, great job there without a doubt, despite spending more time queueing than playing at the moment, but there's no denying there's some underlying decisions that are, politely put...interesting.

[Inarai]

I'm relatively certain at least 75% of the server 'stuff' is the same. A game engine does not change how client server negotiations, server maintenance, and server setup are proceeded through.

 

No, but completely different development with different protocols, different needs, different data to handle and so on could. They started working on this long before Mythic had a connection, too.

 

Also, the game engine could well have a server side implementation to handle much of that.

 

I don't expect there's much if any recycling, frankly.

[Yatabaza]

The game was released YESTERDAY. Would you stop moaning and wait patiently?

NO!

My server (EU - Nightmare Lands) suffers from the queues of 45 mins and more before the launch. The argument of the release being recent is a weak one at best. They saw it coming since the pre-order log-ins and they just crossed their fingers that everything's gonna be fine.

 

And still, no decent announcement was made concerning some fixing of this. Our subscription time is still being spent, though...

 

A shame for a product that beautiful.

[Jaku]

The fastest I've seen an MMO developer come out with a character transfer system is Trion, and that still was not until about 1.5 months after Rift released.

[Inarai]

It's just astounding that while designing and launching a modern day MMO with a massive budget, they never once considered incorporating the tools necessary to facilitate server transfers. They must have realised they would need this.

 

I can overlook smaller tools that are neglected such ast target of target, the ability to adjust UI size, a combat log and various other smaller scale tools in gameplay, but something as fundamental as server transfers being left out makes me a bit skeptical about where this game will go from here now that I've had a taste of the decisions made.

 

I love the game itself, great job there without a doubt, despite spending more time queueing than playing at the moment, but there's no denying there's some underlying decisions that are, politely put...interesting.

 

"We might need this"

 

"When?"

 

"Probably not for at least a year after launch"

 

"All right, then that gets to wait"

 

Given the emphasis on server communities in terms of some of the decisions they've made, I'm hardly surprised that this would have been pushed to post launch.

[Naltharial]

What we have here is simple by comparison, copying a record from one database to another that uses an identical record structure.

 

Ho ho ho!

No.

 

I'm willing to bet important parts of my anatomy that a character is nowhere close to 'a record'. It's more like a million records spun together with all the other records in a spider web. Go ahead and try to move a spider web perfectly. I'll be here waiting.

 

You see, I'm a database engineer. As one, I have a viewpoint through all the nice buzzwords like "database normalization" and "transactional integrity". Where most people err is to assume a vacuum environment. That's nice when you're solving theoretical physics, but has no place in database engineering. Every record is connected with the system. You don't just move one record, you move a whole section of the database, while having to preserve all the connections between the player, their items, their mails, their bank accounts, everything. That's the spider web I've mentioned earlier. Same character names are the least of your problem. It's the primary key by which you store it in a DB that's usually the problem. But it's all manageable. It's just not pretty, fast, or efficient.

 

It's not something you'd want to be doing when your server array is already kneeling under load. I hear people are more upset over servers mother****ing exploding then waiting in the bloody queue for half an hour.

[Brene]

"We might need this"

 

"When?"

 

"Probably not for at least a year after launch"

 

"All right, then that gets to wait"

 

Given the emphasis on server communities in terms of some of the decisions they've made, I'm hardly surprised that this would have been pushed to post launch.

 

You need to remember (this is concurrent with your previous post as well) EA is still at the helm of development when it comes down to: What will make money, what will be needed to make money, what will keep money incoming.