Why would it not? Apart from the DB structure of the character itself you have to take into account that the House changes have to happen first, trade skills have to be split away and replaced with something if needed. All of this is a massive amount of work and only then do you start looking at Multiclass itself.
Sure I think it would be simpler than how it would have been in the days before Dragon became a class on its own, but I do not think Multiclass itself is the holdup, it is all the other work needed for it to happen.
I'm talking about multiclassing itself, none of that other stuff. The current class system would barely need to be tweaked to allow it.
Yeah, it's super helpful to guesstimate the timing on a project which bears no actual resemblance to the one that's actually being done
I said nothing of the sort, was just responding mostly to Jukilian's post regarding multiclassing being a huge feature that would have to be coded, when it's really just a small extension of existing features.
@Hod have you seen any of the code behind Achaea? I know that I haven't seen any of it. But to simply assume that anything in it would be easy to code isn't a good assumption to make.
It could be a small extension, like copypasta the Dragonform code except make it like 'classform Serpent'. Or it could be a big change. It depends on the engine that Achaea is built upon and how much stuff would need to be changed or how deep into the code you need to dig to make the necessary changes that would allow it. Then you'd have to make sure you didn't break anything else (like breaking Dragon, as an example), then there's testing, debugging, etc.
It sounds simple enough, sure. A lot of the ideas my colleague and I have thrown around with each other at work in regards to ways we can upgrade our systems at work are pretty simple. Until you get into the code to actually do it, and realise you need to change this and this and this. This seemingly unrelated feature suddenly gets impacted. What are our size constraints? etc etc.
I don't think it's fair at all for us to say, 'Hey, come on now. It would be a simple and easy-to-do change! It shouldn't take long.' Then the software/computer engineer just smiles and nods, screaming on the inside. Unless you're a coder for any of the IRE games, then sure.
Yeah, it's super helpful to guesstimate the timing on a project which bears no actual resemblance to the one that's actually being done
I said nothing of the sort, was just responding mostly to Jukilian's post regarding multiclassing being a huge feature that would have to be coded, when it's really just a small extension of existing features.
Yes, but obviously Jukilian is talking about multiclassing, as it would be implemented in Achaea, not some hypothetical coding exercise that is never going to exist. Your post just read as a snarky "jeeze admin, coding is easy, hurry up" and a "people don't understand this is really simple, let me explain" kind of obnoxious thing. Whereas actually, you don't have any point at all.
My thought on this is that it's not necessarily the code to make it mechanically possible, it's the balancing that's going to take the time and effort. Either way, griping about how long it's taking isn't helpful or productive.
Tharos, the Announcer of Delos shouts, "It's near the end of the egghunt and I still haven't figured out how to pronounce Clean-dat-hoo."
If it sounded like I was badgering admins to "hurry up" that's not what I meant. And honestly, I was more talking about modifying Certimene for class changes. Literally all you have to do is save the lesson count for the next time you become class. Simple. Just to be clear I'm talking about multiclassing as in switching between classes NOT hybrid classes.
If it sounded like I was badgering admins to "hurry up" that's not what I meant. And honestly, I was more talking about modifying Certimene for class changes. Literally all you have to do is save the lesson count for the next time you become class. Simple. Just to be clear I'm talking about multiclassing as in switching between classes NOT hybrid classes.
Either you don't code yourself or you're just not thinking will all the ramifications of multiclass.
E.g. how often can the person switch between classes? Cooldown? How many classes max? Mechanism to prevent priest/apostate/occultist/druid. Will it work just like dragonforming does now or potentially some other mechanic involved in switching? Need to take care of all class defences, but not those granted by a 3rd party (i.e. runes), what about class specific eq? Fullplate, necklace of purity, etc. Also need to handle class loyals, but not pets. Can the person still teach other class skills or would they have to switch back to do that? What if the person is in another class-specific realm when changing - i.e. chaos-planes?
Plus probably a hundred more bits that I haven't thought of as I don't code for them. Each of those questions need to be answered before you can start or you risk having to refactor everything and doubtless each of them then affect database structures, which then may affect various other service calls, etc., etc.
When you're working with something and complex as Achaea and with so much interaction between modules, any small change can magnify quickly and this is a major change to the structure of the world and how classes are handled in general.
If it sounded like I was badgering admins to "hurry up" that's not what I meant. And honestly, I was more talking about modifying Certimene for class changes. Literally all you have to do is save the lesson count for the next time you become class. Simple. Just to be clear I'm talking about multiclassing as in switching between classes NOT hybrid classes.
Either you don't code yourself or you're just not thinking will all the ramifications of multiclass.
E.g. how often can the person switch between classes? Cooldown? How many classes max? Mechanism to prevent priest/apostate/occultist/druid. Will it work just like dragonforming does now or potentially some other mechanic involved in switching? Need to take care of all class defences, but not those granted by a 3rd party (i.e. runes), what about class specific eq? Fullplate, necklace of purity, etc. Also need to handle class loyals, but not pets. Can the person still teach other class skills or would they have to switch back to do that? What if the person is in another class-specific realm when changing - i.e. chaos-planes?
Plus probably a hundred more bits that I haven't thought of as I don't code for them. Each of those questions need to be answered before you can start or you risk having to refactor everything and doubtless each of them then affect database structures, which then may affect various other service calls, etc., etc.
When you're working with something and complex as Achaea and with so much interaction between modules, any small change can magnify quickly and this is a major change to the structure of the world and how classes are handled in general.
A lot of your points are moot if it is done through Certimene, like I said. Also, some mechanisms are already in place, like Mhaldorians not being able to become forestals (Certimene gives a message that says your city doesn't permit this class or something to that effect.) Honestly, a lot of this stuff IS already in place, that's why I said it's not much more work to expand the system.
If it sounded like I was badgering admins to "hurry up" that's not what I meant. And honestly, I was more talking about modifying Certimene for class changes. Literally all you have to do is save the lesson count for the next time you become class. Simple. Just to be clear I'm talking about multiclassing as in switching between classes NOT hybrid classes.
Either you don't code yourself or you're just not thinking will all the ramifications of multiclass.
E.g. how often can the person switch between classes? Cooldown? How many classes max? Mechanism to prevent priest/apostate/occultist/druid. Will it work just like dragonforming does now or potentially some other mechanic involved in switching? Need to take care of all class defences, but not those granted by a 3rd party (i.e. runes), what about class specific eq? Fullplate, necklace of purity, etc. Also need to handle class loyals, but not pets. Can the person still teach other class skills or would they have to switch back to do that? What if the person is in another class-specific realm when changing - i.e. chaos-planes?
Plus probably a hundred more bits that I haven't thought of as I don't code for them. Each of those questions need to be answered before you can start or you risk having to refactor everything and doubtless each of them then affect database structures, which then may affect various other service calls, etc., etc.
When you're working with something and complex as Achaea and with so much interaction between modules, any small change can magnify quickly and this is a major change to the structure of the world and how classes are handled in general.
A lot of your points are moot if it is done through Certimene, like I said. Also, some mechanisms are already in place, like Mhaldorians not being able to become forestals (Certimene gives a message that says your city doesn't permit this class or something to that effect.) Honestly, a lot of this stuff IS already in place, that's why I said it's not much more work to expand the system.
Got it. So with your negative experience or familiarity with coding a MUD or this MUD in particular, you're convinced that changing one of the basic fundamentals of the game is a simple, fast fix. Tracking. Thank you for your opinion. Can you STFU now?
You can't walk to Carriere every time you want to morph. That shit will be such a buzzkill
Certimene* Mobile posting ftw
"Hey, Certimene. Hit me up with THAT." *Whoooooosh*
edit: On second thought, I think if multiclass allow you to switch over to different class in shor time, i would be first to say - "It's Morphin time!" oh wait, it's dragon already doing that.
2015/01/12
Tecton, the Terraformer has bestowed His divine favour upon you. It will last for approximately 1 Achaean month.
As someone who's had experience working with large, old, complex systems: there is no such thing as a quick, easy change. Believing otherwise is the road to many heartaches, and even more headaches.
And remember folks, Achaea was originally written by someone who thought 'var1' sounded like a good, solid variable name; God only knows what it really looks like inside ( @Sarapis ).
If anything, Certimene's involvement will only be picking what your other classes are. That is it. He'll stop you from picking classes you can't go, that is all. That still doesn't begin to touch on all of the other changes that will need to be done. Yes, adding code to make Certimene add your extra classes might not be hard (once the rest of the system is done), but is a lot more than having "I am also a Shaman" on my SCORE.
The mechanics of switching between classes etc that @Suladan posted (and likely much more) still count. I don't see how Certimene is the quick and easy fix.
I'm about 100% certain that it doesn't work the way that you think it does, no offense. As Suladan said, classes is one of the fundamentals for the game. Adding extra classes isn't that much of an issue (IMO, without thinking of balance) because all of the structure would be present already. I don't think that's the case for multiclass.
Moral of the story: this very likely won't be a quick, simple, easy change to do to implement multiclass. Good things take time. I'd rather a system brought out in a year that is bugless and as close to perfect as possible that works "straight out of the box" than a system that is released in a month that is dead on arrival and riddled with bugs.
I would say that any assumptions that this is easy to do and all the systems are already in place etc would tend towards ignorance. Unless you've seen the code, you're only guessing.
Since I'm honestly not sure if you are simply trolling or not, I won't touch on the subject again.
On the other hand, without access to the code, assuming that it will take a lot of time is also somewhat baseless.
You can be cynical about it and say it probably will, but it could end up taking @Makarios and @Tecton a weekend or so at most to get everything coded up and most of the obvious testing done.
The thing is that coding is only a small part of software development. You'll probably spend much more time designing it or debugging it, your choice.
Ah, @Hod, my naive little friend. You wandered into this thread with a heart pure as driven snow and are now understandably defensive and bewildered at the concentrated invective you have encountered. Know that this hate is not directed at you. It burgeons from a deep wound in the programmer’s soul, a hurt born of crushed dreams and race conditions, and we lash out at you because we envy the innocence that was so cruely torn from us.
You must understand, you are like the bright-eyed Starbucks-imbibing, scarf-accessorizing, Ugg-wearing college freshman who raises their hand in Econ 101 and suggests with perfect sincerity that the government should just print more money when it goes into debt. Or like the first year physics student who still dances, nymph-like, in Newton’s beautiful fantasy and has not yet discovered the Lovecraftian truth of particle-wave duality, the lie of simultaneity, or the uncertainty principle. Your viewpoint is based on the understandable but utterly incorrect premise that fundamentally the world is a simple, sensible thing, and that at the end of the day everyone goes home to their families and enjoys a wholesome turkey dinner. Perhaps you are lucky enough to live a life like that. But you do so because gaunt figures with dead eyes go out into the night and strangle memory corruption bugs with their bare hands, and one cannot come away from such a thing unscarred.
For if there is one truth I must communicate to you, it is that demons lurk in the depth of the codebase. Always, without exception. They are spawned by accident, ignorance, and pride. By sudden deadlines and shifting specification. Also, interns. Humanity’s one saving grace is that the demons rarely rouse themselves while the codebase lies dormant. But when features have to be implemented, someone has to saddle up and go boldly forth armed with a fully loaded debugger and armored in source control. They face horrors the rest of us will never know, such as a fast square-root implementation with i = 0x5f3759df - ( i >> 1 ); embedded boldly in the middle, which is like picking up your morning newspaper in Wyoming and discovering the headline is written in Sanskrit that smokes and smells faintly of sulfur. Strange, otherworldly truths will become manifest in their minds, like that NULL < -1 < 0 < NULL, and they will laugh and cry in turns at the terrible magnificence of this knowledge. And the worst part about this process is that to combat demons one often must resort to similarly unholy practices. I once was forced to call upon the Curiously Recurring Template Pattern and ended up vomiting frogs for a week.
So when you blithely suggest simplicity in a codebase over a decade old, written for a custom engine and under tight performance constraints, know that after we are finished with our cruel laughter we will brood quietly in the shadows of a bar where no one talks to anyone else and Billy Joel plays out of greasy, crackling speakers, nursing a stiff drink and trying to remember what it was like when we were happy.
On the one hand, there is assuming it would be quick and easy to implement and making comments that can come off as ignorant to the actual situation and make it seem as if people don't understand how easy their job supposedly is.
On the other, there is knowing that it could and more likely take a while to do, so you don't make comments saying "Oh come on, it is really easy to do" and you wait patiently.
I don't think Hod has to be trolled any longer. By his lack of replies, I'm going to guess that he's bowing out of the whole "b-but Certimene" nonsense.
This is not funny. This is the sad tale of a man being utterly destroyed by a soulless corporation. The demeanor of the 'expert' at the end is of a guy who's going to stop by his local retail superstore on the way home to pick up a box of utility blades and a Coldplay album.
It's got little to do with coding and the necessary code for multiclassing is already basically done anyway. As someone above pointed out, multiclassing can't happen until House Renaissance is done, and we may also hold off until we've split tradeskills off from the classes we intend to split them off from. And, as we're splitting off tradeskills, we're also taking the opportunity to, in some cases (like Forging), completely redo how they work. And then, of course, replacements for the classes that lost skills have to be put in too. Generally speaking the design time takes considerably longer than the actual coding time.
I was sleeping. And you can throw around all kinds of insults, but I've been coding for over a decade, so I'm not sitting here making armchair hypotheses. My main point was really that it is nowhere even close to the complexity of ships being added, not that you could literally write one line of code and add multiclass. Obviously you're gonna have to work around some things. I haven't seen the code, so certain things may or not be somewhat difficult to do, since I realize some people have said Achaea's code is quite monolithic, but adding multiclass is orders of magnitude easier than writing a completely new feature. Again, this was not a call out of staff for being lazy, I'm perfectly aware there are plenty of things to be done and decisions to be made before they actually implement this.
I do hope Enchanting gets reworked. Its system seems quite old-fashioned and I'm sure it could be spiced up a bit, who knows. And there's some enchantments that are kind of useless (i.e. baubles).
Looking forward to it all!
And @Hod, yes it is nothing compared to ships. That I cannot deny. But to assume that it would be a quick and easy to do change is still a silly assumption to make, in my opinion. Especially if you haven't seen the code.
I don't understand why anyone thinks that adding multiclassing would require an immense coding effort.
This is the comment that bothered me.
In other news, great to hear that most of the work for multiclass is done. Just got to, y'know, keep working at those Renaissances :P (We've not heard anything official happening in ages )
I was sleeping. And you can throw around all kinds of insults, but I've been coding for over a decade, so I'm not sitting here making armchair hypotheses. My main point was really that it is nowhere even close to the complexity of ships being added, not that you could literally write one line of code and add multiclass. Obviously you're gonna have to work around some things. I haven't seen the code, so certain things may or not be somewhat difficult to do, since I realize some people have said Achaea's code is quite monolithic, but adding multiclass is orders of magnitude easier than writing a completely new feature. Again, this was not a call out of staff for being lazy, I'm perfectly aware there are plenty of things to be done and decisions to be made before they actually implement this.
Actually, adding multi-classing is one of the most complex and difficult changes we could make to the game because doing so mandates wide-reaching changes to other game systems in a way that something like ships, which is largely a stand-alone system, does not.
I was sleeping. And you can throw around all kinds of insults, but I've been coding for over a decade, so I'm not sitting here making armchair hypotheses. My main point was really that it is nowhere even close to the complexity of ships being added, not that you could literally write one line of code and add multiclass. Obviously you're gonna have to work around some things. I haven't seen the code, so certain things may or not be somewhat difficult to do, since I realize some people have said Achaea's code is quite monolithic, but adding multiclass is orders of magnitude easier than writing a completely new feature. Again, this was not a call out of staff for being lazy, I'm perfectly aware there are plenty of things to be done and decisions to be made before they actually implement this.
This is a game where the code for the weather system likely involves variables from the denizen respawn, credit transfer, and player namechange systems, with math functions copy-pasted wholesale from Windows 95 and Wolfenstein 3D, all tied together with incense, ominous Gregorian chanting, and fervent prayers to the God-Emperor of Mankind (In nomine Imperator!).
Never make the mistake of thinking that changing old Achaea code is easier than writing new code.
I was sleeping. And you can throw around all kinds of insults, but I've been coding for over a decade, so I'm not sitting here making armchair hypotheses. My main point was really that it is nowhere even close to the complexity of ships being added, not that you could literally write one line of code and add multiclass. Obviously you're gonna have to work around some things. I haven't seen the code, so certain things may or not be somewhat difficult to do, since I realize some people have said Achaea's code is quite monolithic, but adding multiclass is orders of magnitude easier than writing a completely new feature. Again, this was not a call out of staff for being lazy, I'm perfectly aware there are plenty of things to be done and decisions to be made before they actually implement this.
Actually, adding multi-classing is one of the most complex and difficult changes we could make to the game because doing so mandates wide-reaching changes to other game systems in a way that something like ships, which is largely a stand-alone system, does not.
I suppose I'm in the dark on something then. It seems to me that quite of bit of the game is largely independent of what class you are, besides class abilities themselves. I realize that is an assumption that could be wrong. That, and a lot of the work was already done in that regard which I was taking into account. From my (ignorant) standpoint it seemed to me that only these things remaining that needed to be implemented: Additional character information, which I assume if you're using an actual database system (another assumption), would be trivial. Second, a way to change between classes, which Certimene already allows (however I get the feeling another route is being persued, hence the additional complexity.)
I was sleeping. And you can throw around all kinds of insults, but I've been coding for over a decade, so I'm not sitting here making armchair hypotheses. My main point was really that it is nowhere even close to the complexity of ships being added, not that you could literally write one line of code and add multiclass. Obviously you're gonna have to work around some things. I haven't seen the code, so certain things may or not be somewhat difficult to do, since I realize some people have said Achaea's code is quite monolithic, but adding multiclass is orders of magnitude easier than writing a completely new feature. Again, this was not a call out of staff for being lazy, I'm perfectly aware there are plenty of things to be done and decisions to be made before they actually implement this.
This is a game where the code for the weather system likely involves variables from the denizen respawn, credit transfer, and player namechange systems, with math functions copy-pasted wholesale from Windows 95 and Wolfenstein 3D, all tied together with incense, ominous Gregorian chanting, and fervent prayers to the God-Emperor of Mankind (In nomine Imperator!).
Never make the mistake of thinking that changing old Achaea code is easier than writing new code.
Maybe I'm giving it to much credit. But the code can't be that bad...
Comments
It could be a small extension, like copypasta the Dragonform code except make it like 'classform Serpent'. Or it could be a big change. It depends on the engine that Achaea is built upon and how much stuff would need to be changed or how deep into the code you need to dig to make the necessary changes that would allow it. Then you'd have to make sure you didn't break anything else (like breaking Dragon, as an example), then there's testing, debugging, etc.
It sounds simple enough, sure. A lot of the ideas my colleague and I have thrown around with each other at work in regards to ways we can upgrade our systems at work are pretty simple. Until you get into the code to actually do it, and realise you need to change this and this and this. This seemingly unrelated feature suddenly gets impacted. What are our size constraints? etc etc.
I don't think it's fair at all for us to say, 'Hey, come on now. It would be a simple and easy-to-do change! It shouldn't take long.' Then the software/computer engineer just smiles and nods, screaming on the inside. Unless you're a coder for any of the IRE games, then sure.
E.g. how often can the person switch between classes? Cooldown? How many classes max? Mechanism to prevent priest/apostate/occultist/druid. Will it work just like dragonforming does now or potentially some other mechanic involved in switching? Need to take care of all class defences, but not those granted by a 3rd party (i.e. runes), what about class specific eq? Fullplate, necklace of purity, etc. Also need to handle class loyals, but not pets. Can the person still teach other class skills or would they have to switch back to do that? What if the person is in another class-specific realm when changing - i.e. chaos-planes?
Plus probably a hundred more bits that I haven't thought of as I don't code for them. Each of those questions need to be answered before you can start or you risk having to refactor everything and doubtless each of them then affect database structures, which then may affect various other service calls, etc., etc.
When you're working with something and complex as Achaea and with so much interaction between modules, any small change can magnify quickly and this is a major change to the structure of the world and how classes are handled in general.
Penwize has cowardly forfeited the challenge to mortal combat issued by Atalkez.
Mobile posting ftw
Penwize has cowardly forfeited the challenge to mortal combat issued by Atalkez.
edit:
On second thought, I think if multiclass allow you to switch over to different class in shor time, i would be first to say -
"It's Morphin time!" oh wait, it's dragon already doing that.
And remember folks, Achaea was originally written by someone who thought 'var1' sounded like a good, solid variable name; God only knows what it really looks like inside ( @Sarapis ).
The mechanics of switching between classes etc that @Suladan posted (and likely much more) still count. I don't see how Certimene is the quick and easy fix.
I'm about 100% certain that it doesn't work the way that you think it does, no offense. As Suladan said, classes is one of the fundamentals for the game. Adding extra classes isn't that much of an issue (IMO, without thinking of balance) because all of the structure would be present already. I don't think that's the case for multiclass.
Moral of the story: this very likely won't be a quick, simple, easy change to do to implement multiclass. Good things take time. I'd rather a system brought out in a year that is bugless and as close to perfect as possible that works "straight out of the box" than a system that is released in a month that is dead on arrival and riddled with bugs.
I would say that any assumptions that this is easy to do and all the systems are already in place etc would tend towards ignorance. Unless you've seen the code, you're only guessing.
Since I'm honestly not sure if you are simply trolling or not, I won't touch on the subject again.
You can be cynical about it and say it probably will, but it could end up taking @Makarios and @Tecton a weekend or so at most to get everything coded up and most of the obvious testing done.
The thing is that coding is only a small part of software development. You'll probably spend much more time designing it or debugging it, your choice.
On the other, there is knowing that it could and more likely take a while to do, so you don't make comments saying "Oh come on, it is really easy to do" and you wait patiently.
I do my best to stick to the latter when I can.
Still, it was cute while it lasted.
take like five minutes tops
Looking forward to it all!
And @Hod, yes it is nothing compared to ships. That I cannot deny. But to assume that it would be a quick and easy to do change is still a silly assumption to make, in my opinion. Especially if you haven't seen the code.
This is the comment that bothered me.
In other news, great to hear that most of the work for multiclass is done. Just got to, y'know, keep working at those Renaissances :P (We've not heard anything official happening in ages )
Never make the mistake of thinking that changing old Achaea code is easier than writing new code.
Can it?