Deadlines are deadlines. And your honesty is a credit to the whole organization. But please don't put this off after getting 11.5 into the hands of people in Vegas.
^^ This. I'm OK with it not being there on technical grounds, Bort made an excellent case for why there's simply no time and for impossibility of just re-using the old PowerCart (which was quite kludgy to begin with.)
But all those attempts to convince everyone who will miss it that it's just us Voth defending our dinosaur ways, or that it's an MMO standard or even somehow an upgrade (it kind of is for the initial leveling -- but not the respecs...) They're unlikely to convince anyone but the dev team itself. And the only thing worse than believing your own propaganda is to be the only one who believes it.
"made an excellent case" ? Say what? He didn't make a case at all, he just said it can't be done.
Given their sales pitch for 'experimenting with builds' and such Id hoped for something like DCUO's armory system where we could save builds and quick hop to them out of combat.
Loadouts is kinda like the Armory system. The big difference though, is that in DCUO, you can reset your skill points with in game cash rather than having to spend real money.
100,000 EC to reset skills would be decent. Newer players will be cautious of where their skill points go and veteran players will have the funds to experiment/adjust for different scenarios.
In Champs you can redo skills one at a time, but that's because the game keeps track of what you purchased most recently, also the cost is based on how many skills you're changing. A full respec is actually expensive.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Um, yeah.... it only sounds easy until you try to do it.
There are some similarities to how Champions does it, but Champions permanently remembers the order of all power purchases.
Also, filling the new skill tree requires around 120 clicks, filling the old one actually took MORE since you hade to mess with quantities for each skill.
Think what you want about my credentials, I am not all that concered but if you read the post that I was responding to, you would know enough info to grasp Cryptics decision making process for this effort. It doesnt take having access to internal info to figure what was going on
Keep in mind that there are three facets of a project/program:
Scope (quality, performance)
Cost
Schedule
"Pick two"
A project can only prioritize two (max), at the remainders expense.
It does require having access to internal info to figure out why the picks were made as they were.
There is more to project managing then just saying: "TRIBBLE the schedule, we want all these features for that money, that way we maximize customer satisfaction!" It's simply not true. Sometimes features that are small have high cost attached to them.
The skill system isn't an isolated aspect of the game's development either. There are other things in development at all time. Maybe sacrificing skill parts in the skill revamp means having the programming resources available to fix some serious bugs, or implement a different feature in a different part of the game that players want.
The other person stated that the reason why Cryptic decided not to fix the "accept" function, and place it at the end, resulting in making players pay money (or the equivalent) for minor mistakes (clicking the wrong thing and moving on), was they wanted it released for some event.
They knew they had the issue but decided to move anyways, and decided that having the players fit the bill, instead of crashing a revised schedule to fix the problem, in time for the event.
I strongly suspect that you're vastly overestimating the actual likelihood of people making mistakes that force them to respec. It's not like the previous system made it impossible to make errors, either.
Cryptics decision shows that they made the Schedule the top priority for the event, and Cost the other must have. The expense has to be Scope, there is no other area to take from, which is why people will have to pay for mistakes. If you've been reading this forum, you can tell that people are considering it a performance/quality issue. All so they can show a partially improved system that isnt ready for prime time because what should be a simple, basic function got in the way of an event schedule.
Whats worse is that Cryptics mitigation plan is to give us two free respecs, really?
Why not have a reasonable grace period, on Holodeck (not everyone uses Tribble) and take time to fix the issue? The answer is that they arent concerned with the Voice of the Customer on this issue, and are going to say "working as intended" (besides knowing that itsa problem, and just shrugging thier shoulders), and "The new skills system is great, and the players love it!", and Scope was chopped to satisfy schedule and cost, and its not like Cryptic hasnt done this before.
A "grace period" is not free to develop, either.
I will not disagree that the new system will bring quality in some areas but its exremely bad decision making on a rushed product, especially when it comes to money, its even worse to make the comment that had been made, so overtly, that is parallel to "let them eat cake". Never leave the customerholding the bag.While I have no intentions of being personal with you, if you can't/won't see that, I (nor anyone else) can help you.
It's not "extremely bad decision making". It's just making a decision you would have preferred to go another way. I would like the skill cart system and the skill revamp together, too, but having seen the new skill system, I rather want it live soon. If i have to respec an extra time because I couldn't keep track of my points, so be it.
It's definitely not a rushed system, either. It has been on Tribble for a few weeks now even, we rarely see systems so long in public testing!
Star Trek Online Advancement: You start with lowbie gear, you end with Lobi gear.
Seriously taking a step back Cryptic. No undo and no mistakes. Remember folks Cryptic is in the business of making money no matter how underhanded they have to be. Still haven't logged in since last part of Feb and feeling fine.
Deadlines are deadlines. And your honesty is a credit to the whole organization. But please don't put this off after getting 11.5 into the hands of people in Vegas.
^^ This. I'm OK with it not being there on technical grounds, Bort made an excellent case for why there's simply no time and for impossibility of just re-using the old PowerCart (which was quite kludgy to begin with.)
But all those attempts to convince everyone who will miss it that it's just us Voth defending our dinosaur ways, or that it's an MMO standard or even somehow an upgrade (it kind of is for the initial leveling -- but not the respecs...) They're unlikely to convince anyone but the dev team itself. And the only thing worse than believing your own propaganda is to be the only one who believes it.
"made an excellent case" ? Say what? He didn't make a case at all, he just said it can't be done.
He didn't say that. He said it could be done, but not by re-using the old system and not in time for whatever their internal deadline is, with an estimate of exactly how much it might set them back. This is about as good an explanation as it gets without explaining actual code or publishing meeting minutes (which, for obvious and very good reasons, no one would ever do.)
Now you may choose to believe that REAL PROGRAMMERS can do anything over a weekend... And sometimes they can, but if you want your client not crashing and your server not accessed in an unauthorised manner*, they better shouldn't.
* apparently we're not allowed to type the h-word.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
All right, I'm no psychic and never worked for Cryptic or PWE (or any gaming company) but I'll go out on a limb and assume that they aren't the Borg, and also assume that were they truly incompetent, they'd be out of business ages ago.
Your design was literally the very first idea I had when Bort mentioned "technical reasons," back in the old feedback thread. I'm not a particularly good coder, so someone with far more qualification and much more recent and much more relevant experience (like Bort) probably had that exact same idea at some point. (He and the other devs may or may not have thought of things like "grace periods" or other near-replacements, but it's the most straightforward way to make a straight-up copy of the old system on the new UI.) So I assume it's either impossible for some reason (network security, UI widget library not willing to stand up and do this particular trick, or whatever) or the 2 weeks estimate is for this exact design.
Since they are neither the Borg nor the idiots, the deadline decisions are probably made by some person at least a bit higher in the chain of command than the devs that post and read here (there's ALWAYS something that could be improved in a project if you had just one more week... And good and competent devs want things to be better, and they want that extra week. Leaving that kind of judgement call to people actually doing it is basically experimenting with inmate-run asilums.) That said, because they aren't idiots, it's not all one-way, and they listen to what their subordinates are saying.
Again, I'm no psychic, but my general experience of such situations shows that for a customer the best strategy here is to (a)not to act like they know something the devs don't or accuse them of all the deadly sins, it's annoying and demotivating, (b)accept such decisions aren't entirely theirs to make and (c) very firmly insist that you actually WANT the thing in question, not taking any sour grapes excuses, so they would have some ammunition when talking to people who do make those decisions.
This post brought to you by Niccolo Machiavelli via the Ouija Board app. He may not be a psychic, but I am. /kidding./
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Given that there's 1/3rd the things we need to click... I'm cool with that.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
All right, I'm no psychic and never worked for Cryptic or PWE (or any gaming company) but I'll go out on a limb and assume that they aren't the Borg, and also assume that were they truly incompetent, they'd be out of business ages ago.
Your design was literally the very first idea I had when Bort mentioned "technical reasons," back in the old feedback thread. I'm not a particularly good coder, so someone with far more qualification and much more recent and much more relevant experience (like Bort) probably had that exact same idea at some point. (He and the other devs may or may not have thought of things like "grace periods" or other near-replacements, but it's the most straightforward way to make a straight-up copy of the old system on the new UI.) So I assume it's either impossible for some reason (network security, UI widget library not willing to stand up and do this particular trick, or whatever) or the 2 weeks estimate is for this exact design.
Since they are neither the Borg nor the idiots, the deadline decisions are probably made by some person at least a bit higher in the chain of command than the devs that post and read here (there's ALWAYS something that could be improved in a project if you had just one more week... And good and competent devs want things to be better, and they want that extra week. Leaving that kind of judgement call to people actually doing it is basically experimenting with inmate-run asilums.) That said, because they aren't idiots, it's not all one-way, and they listen to what their subordinates are saying.
Again, I'm no psychic, but my general experience of such situations shows that for a customer the best strategy here is to (a)not to act like they know something the devs don't or accuse them of all the deadly sins, it's annoying and demotivating, (b)accept such decisions aren't entirely theirs to make and (c) very firmly insist that you actually WANT the thing in question, not taking any sour grapes excuses, so they would have some ammunition when talking to people who do make those decisions.
This post brought to you by Niccolo Machiavelli via the Ouija Board app. He may not be a psychic, but I am. /kidding./
Let me tl;dr my reply by saying I adhere mostly to ' (c) very firmly insist that you actually WANT the thing in question.' Even in the podcast it came up, that the single 'Commit' button is really an issue amongst players (not just me).
And that yes, I think 'my' solution would be easier than rewriting the whole system from scratch. And I think it's worth doing it. Maybe it can't done for reasons out of the Devs' control (marketing or any other brass department setting a deadline for them), like you say. But I'm also saying that if it isn't done now, it's fair to assume it will never be done. So, I please hope they'll reconsider and add it, after all.
And that is all I'll say on it further. No, not because I'm angry or something, but yeah, you can only lead a horse to water, but you can't make him drink. If they really don't want to implement a single 'Commit' button, then that is ultimately their own call, of course.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Given that there's 1/3rd the things we need to click... I'm cool with that.
Except those 1/3rd of the clicks will now be immediately final. See you at the Respec store!
Let me tl;dr my reply by saying I adhere mostly to ' (c) very firmly insist that you actually WANT the thing in question.' Even in the podcast it came up, that the single 'Commit' button is really an issue amongst players (not just me).
Well, we at least agree on the most important thing, we both want the our PowerCart back and we both worry that "maybe sometime after the release" often means "when pigs fly" in this line of work. We just differ on whether we should attach those pigs to a balloon or go for rocket propulsion...
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Given that there's 1/3rd the things we need to click... I'm cool with that.
Except those 1/3rd of the clicks will now be immediately final. See you at the Respec store!
So? how is that any less final than the current way? The current version doesn't let you test anything before you commit either. So it goes from clicking repeatedly then hitting commit, to clicking a few times.... Both ways have the same finality in the end.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Given that there's 1/3rd the things we need to click... I'm cool with that.
Except those 1/3rd of the clicks will now be immediately final. See you at the Respec store!
So? how is that any less final than the current way? The current version doesn't let you test anything before you commit either. So it goes from clicking repeatedly then hitting commit, to clicking a few times.... Both ways have the same finality in the end.
Yeah, no. In the current system, you can shift around your points to your heart's content, until you're finally okay with it all. Under the new paradigm, every click you make is immediately final, no backsies. The new system requires you to calculate, in full, your entire plan already, upfront. One mistake, and you can respec directly. You can't count down from 46, and then realize you're a point short (at least not without having to respec yet again).
This isn't about 'finality in the end,' but about finality in the moment.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Given that there's 1/3rd the things we need to click... I'm cool with that.
Except those 1/3rd of the clicks will now be immediately final. See you at the Respec store!
So? how is that any less final than the current way? The current version doesn't let you test anything before you commit either. So it goes from clicking repeatedly then hitting commit, to clicking a few times.... Both ways have the same finality in the end.
Yeah, no. In the current system, you can shift around your points to your heart's content, until you're finally okay with it all. Under the new paradigm, every click you make is immediately final, no backsies. The new system requires you to calculate, in full, your entire plan already, upfront. One mistake, and you can respec directly. You can't count down from 46, and then realize you're a point short (at least not without having to respec yet again).
This isn't about 'finality in the end,' but about finality in the moment.
ok... but if you're allocating points as you level you won't be spending points all at once. And if you're doing a serious build you're probably going to use some sort of skill planner and not make it up as you allocate points anyways.
It's only a 'technical' issue in that they simply (lazily?) copied the Specialization tree mechanics, designed to be an 'over-the-course-of-time' deal, in which you purchase points on an individual basis, and which choices become effective immediately, and are now stuck with that, instead of using the logical, and conceptually superior, single 'Commit' button that belonged to the old skill system.
N.B. I can understand how reworking the whole behind-the-scenes 'Specialization tree mechanics,' just for the skills, could be considered cumbersome. But I'd like to point out that such a time-consuming thingy isn't necessary, either. Were I to program this, all I'd do is make a simple, superimposed, 'faux' GUI for it, which does nothing but simply collect the assigned points, and then batch-supply those choices to the underlying 'Specialization tree' interface upon 'Commit.' Maybe not extremely elegant, but extremely easy to make.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
I strongly suspect that "Not missing the deadline for season 11.5" is not just "oops, we're a week late". We know that Cryptic can have things in the works for months before it gets released, and it's also pretty easy to see how some things would have dependencies - if there is a programmer that was expected to be available for developing a critical aspect for another feature is suddenly still working on another thing, that could mean to people twiddling their thumbs as they can't continue on their feature.
Star Trek Online Advancement: You start with lowbie gear, you end with Lobi gear.
Personally without the confirm purchase option or reset/undo option on the new skill tree, which currently exists for the current system, i'd say the new tree is an unfinished product. And i share others scepticism about "if its not added now it won't be added anytime in the near future".
No backsy, no option for a fully levelled character to spend points/create a build, then check its what they want and click confirm.
Sure theres a skill planner offered by 3rd party development. But what about those players that aren't aware of such a tool ?.
There's no margin for error with this new system, in its current format it is setup to punish mistakes and force/encourage respect token purchases. its monetisation of the skill system for the less knowledgable.
I'm in the corner where skill tree tweaking should be free of charge. Why are we being charged 500 zen a pop to create/change/rework a character to play the game. I thought the game was F2P. The skill tree is a core part of the game and core to your character to play the game. Same question for renaming a character.
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
Except that it's wrong.
A 'fake' UI like that would still need to be programmed with all of the same restrictions/allowances of the real Skill UI, in order to track the spending of points, unlocking of tiers, and choosing of unlocks. It would also need to be able to check other character attributes, such as Class/Profession.
Basically, it would be exactly as complicated as the real thing, but with the added complexities of being safeguarded against actually having any effect, saving every click you make so that they can be undone on the fly, calculating not just points-spent but also points-unspent. And then, you propose the ability to be able to apply those saved changes to the actual UI, which would need yet another set of unique translation restrictions to get Layer A to accurately speak to Layer B.
Making the assumption that you know an "easy" way to "just" implement something entirely new into our game - especially one that includes user interface interactions - is only setting yourself up for embarrassment.
Jeremy Randall
Cryptic - Lead Systems Designer
"Play smart!"
The Dev has spoken. No PowerCart and no official skill planner for 11.5. For the short term all we can do is accept or or walk away. Personally I am accepting and adapting and don't consider it an unbearable hardship. It's not that bad, as long as the underlying skill mechanics work. Can we maybe focus on that?
The Dev has spoken. No PowerCart and no official skill planner for 11.5. For the short term all we can do is accept or or walk away. Personally I am accepting and adapting and don't consider it an unbearable hardship. It's not that bad, as long as the underlying skill mechanics work. Can we maybe focus on that?
Yeah, really, people used Skill planners before, and they'll probably continue to do so after.
it would be interesting if cryptic adopted STOAcademy's skill planner as a sort of 'official' unofficial skill planner like they apparently adopted the STOWiki as an 'official' game guide - could even stick a button in the skill section along the bottom where the confirm button currently is leading directly to the planner
not that it'll actually happen, but it's nice to dream...
A normie goes "Oh, what's this?"
An otaku goes "UwU, what's this?"
A furry goes "OwO, what's this?"
A werewolf goes "Awoo, what's this?"
"It's nothing personal, I just don't feel like I've gotten to know a person until I've sniffed their crotch." "We said 'no' to Mr. Curiosity. We're not home. Curiosity is not welcome, it is not to be invited in. Curiosity...is bad. It gets you in trouble, it gets you killed, and more importantly...it makes you poor!"
Passion and Serenity are one.
I gain power by understanding both.
In the chaos of their battle, I bring order.
I am a shadow, darkness born from light.
The Force is united within me.
Seems like all this nonsense could be avoided by leaving the skill tree alone.....JMO
That the Devs haven't stopped and explained the reason they're sinking hundreds (maybe thousands) of man hours into this project doesn't mean there aren't compelling reasons... Reasons that are pretty easy to intuit if you consider the timing... This wasn't done on a whim and it's not gonna stop on a few petulant complaints.
I think the absence of the PowerCart tech is gonna be a lot more noticeable to the target audience then the Devs are trying to reassure themselves because frankly the center of mass of the whole industry has moved on from real-money paid re-spec's towards actual freedom to experiment with these kinds of systems. It's a black eye that undoes a lot of the good being done when you toss the new lighter/cleaner skill system in front of that fresh new audience we all need to hope will be joining us here in STO in the aftermath of the movies and the rise of a new TV show.
Comments
"made an excellent case" ? Say what? He didn't make a case at all, he just said it can't be done.
My character Tsin'xing
You know, meimei, people who rely upon "because it's too hard" excuses just hate it when some smart aleck like you comes along and shows how easy it would actually be
There are some similarities to how Champions does it, but Champions permanently remembers the order of all power purchases.
Also, filling the new skill tree requires around 120 clicks, filling the old one actually took MORE since you hade to mess with quantities for each skill.
My character Tsin'xing
There is more to project managing then just saying: "TRIBBLE the schedule, we want all these features for that money, that way we maximize customer satisfaction!" It's simply not true. Sometimes features that are small have high cost attached to them.
The skill system isn't an isolated aspect of the game's development either. There are other things in development at all time. Maybe sacrificing skill parts in the skill revamp means having the programming resources available to fix some serious bugs, or implement a different feature in a different part of the game that players want. I strongly suspect that you're vastly overestimating the actual likelihood of people making mistakes that force them to respec. It's not like the previous system made it impossible to make errors, either.
A "grace period" is not free to develop, either.
It's not "extremely bad decision making". It's just making a decision you would have preferred to go another way. I would like the skill cart system and the skill revamp together, too, but having seen the new skill system, I rather want it live soon. If i have to respec an extra time because I couldn't keep track of my points, so be it.
It's definitely not a rushed system, either. It has been on Tribble for a few weeks now even, we rarely see systems so long in public testing!
Seriously taking a step back Cryptic. No undo and no mistakes. Remember folks Cryptic is in the business of making money no matter how underhanded they have to be. Still haven't logged in since last part of Feb and feeling fine.
Now you may choose to believe that REAL PROGRAMMERS can do anything over a weekend... And sometimes they can, but if you want your client not crashing and your server not accessed in an unauthorised manner*, they better shouldn't.
* apparently we're not allowed to type the h-word.
Yea, it sounds pretty easy, about two weeks project, with testing and reasonable safety margin... Oh, wait...
Actually, yeah, it really is rather easy. Peeps keep saying "Maybe so, but it can't be done within the deadline fot 11.5, though." To which I retort: "Sure, but said deadline is of your own choosing too." Nobody is holding a gun to their head, saying 11.5 needs to come out within the next 2 weeks. Were it me, yes, I'd postpone a few days (like who cares, really?) and just get the 'Commit' thingy over with, even it it took as much two extra weeks (which assuredly it wouldn't).
The way it's going down now, is that the 'Commit' thing will not currently be done, and maybe later... which means never.
Your design was literally the very first idea I had when Bort mentioned "technical reasons," back in the old feedback thread. I'm not a particularly good coder, so someone with far more qualification and much more recent and much more relevant experience (like Bort) probably had that exact same idea at some point. (He and the other devs may or may not have thought of things like "grace periods" or other near-replacements, but it's the most straightforward way to make a straight-up copy of the old system on the new UI.) So I assume it's either impossible for some reason (network security, UI widget library not willing to stand up and do this particular trick, or whatever) or the 2 weeks estimate is for this exact design.
Since they are neither the Borg nor the idiots, the deadline decisions are probably made by some person at least a bit higher in the chain of command than the devs that post and read here (there's ALWAYS something that could be improved in a project if you had just one more week... And good and competent devs want things to be better, and they want that extra week. Leaving that kind of judgement call to people actually doing it is basically experimenting with inmate-run asilums.) That said, because they aren't idiots, it's not all one-way, and they listen to what their subordinates are saying.
Again, I'm no psychic, but my general experience of such situations shows that for a customer the best strategy here is to (a)not to act like they know something the devs don't or accuse them of all the deadly sins, it's annoying and demotivating, (b)accept such decisions aren't entirely theirs to make and (c) very firmly insist that you actually WANT the thing in question, not taking any sour grapes excuses, so they would have some ammunition when talking to people who do make those decisions.
This post brought to you by Niccolo Machiavelli via the Ouija Board app. He may not be a psychic, but I am. /kidding./
My character Tsin'xing
Let me tl;dr my reply by saying I adhere mostly to ' (c) very firmly insist that you actually WANT the thing in question.' Even in the podcast it came up, that the single 'Commit' button is really an issue amongst players (not just me).
And that yes, I think 'my' solution would be easier than rewriting the whole system from scratch. And I think it's worth doing it. Maybe it can't done for reasons out of the Devs' control (marketing or any other brass department setting a deadline for them), like you say. But I'm also saying that if it isn't done now, it's fair to assume it will never be done. So, I please hope they'll reconsider and add it, after all.
And that is all I'll say on it further. No, not because I'm angry or something, but yeah, you can only lead a horse to water, but you can't make him drink. If they really don't want to implement a single 'Commit' button, then that is ultimately their own call, of course.
Except those 1/3rd of the clicks will now be immediately final. See you at the Respec store!
Well, we at least agree on the most important thing, we both want the our PowerCart back and we both worry that "maybe sometime after the release" often means "when pigs fly" in this line of work. We just differ on whether we should attach those pigs to a balloon or go for rocket propulsion...
My character Tsin'xing
Yeah, no. In the current system, you can shift around your points to your heart's content, until you're finally okay with it all. Under the new paradigm, every click you make is immediately final, no backsies. The new system requires you to calculate, in full, your entire plan already, upfront. One mistake, and you can respec directly. You can't count down from 46, and then realize you're a point short (at least not without having to respec yet again).
This isn't about 'finality in the end,' but about finality in the moment.
My character Tsin'xing
I strongly suspect that "Not missing the deadline for season 11.5" is not just "oops, we're a week late". We know that Cryptic can have things in the works for months before it gets released, and it's also pretty easy to see how some things would have dependencies - if there is a programmer that was expected to be available for developing a critical aspect for another feature is suddenly still working on another thing, that could mean to people twiddling their thumbs as they can't continue on their feature.
No backsy, no option for a fully levelled character to spend points/create a build, then check its what they want and click confirm.
Sure theres a skill planner offered by 3rd party development. But what about those players that aren't aware of such a tool ?.
There's no margin for error with this new system, in its current format it is setup to punish mistakes and force/encourage respect token purchases. its monetisation of the skill system for the less knowledgable.
I'm in the corner where skill tree tweaking should be free of charge. Why are we being charged 500 zen a pop to create/change/rework a character to play the game. I thought the game was F2P. The skill tree is a core part of the game and core to your character to play the game. Same question for renaming a character.
Except that it's wrong.
A 'fake' UI like that would still need to be programmed with all of the same restrictions/allowances of the real Skill UI, in order to track the spending of points, unlocking of tiers, and choosing of unlocks. It would also need to be able to check other character attributes, such as Class/Profession.
Basically, it would be exactly as complicated as the real thing, but with the added complexities of being safeguarded against actually having any effect, saving every click you make so that they can be undone on the fly, calculating not just points-spent but also points-unspent. And then, you propose the ability to be able to apply those saved changes to the actual UI, which would need yet another set of unique translation restrictions to get Layer A to accurately speak to Layer B.
Making the assumption that you know an "easy" way to "just" implement something entirely new into our game - especially one that includes user interface interactions - is only setting yourself up for embarrassment.
Cryptic - Lead Systems Designer
"Play smart!"
My character Tsin'xing
not that it'll actually happen, but it's nice to dream...
#LegalizeAwoo
A normie goes "Oh, what's this?"
An otaku goes "UwU, what's this?"
A furry goes "OwO, what's this?"
A werewolf goes "Awoo, what's this?"
"It's nothing personal, I just don't feel like I've gotten to know a person until I've sniffed their crotch."
"We said 'no' to Mr. Curiosity. We're not home. Curiosity is not welcome, it is not to be invited in. Curiosity...is bad. It gets you in trouble, it gets you killed, and more importantly...it makes you poor!"
http://www.stf-vanguard.com/STO-Skill-Planner/
Reddit discussion thread for above planner:
https://www.reddit.com/r/sto/comments/49a1j0/space_skill_planner_v01/
Cryptic - Lead Systems Designer
"Play smart!"
original join date 2010
Member: Team Trekyards. Visit Trekyards today!
That the Devs haven't stopped and explained the reason they're sinking hundreds (maybe thousands) of man hours into this project doesn't mean there aren't compelling reasons... Reasons that are pretty easy to intuit if you consider the timing... This wasn't done on a whim and it's not gonna stop on a few petulant complaints.
I think the absence of the PowerCart tech is gonna be a lot more noticeable to the target audience then the Devs are trying to reassure themselves because frankly the center of mass of the whole industry has moved on from real-money paid re-spec's towards actual freedom to experiment with these kinds of systems. It's a black eye that undoes a lot of the good being done when you toss the new lighter/cleaner skill system in front of that fresh new audience we all need to hope will be joining us here in STO in the aftermath of the movies and the rise of a new TV show.