Re: Weapon skills improvement

[Synthesis]

I branched 4 of the 5 advanced skills with an old PC, but that was before Morgenes fixed the ride code such that it no longer gave a large penalty to offense while mounted.  He actually changed the code right when I was in the middle of skilling up the last primary skill, and after the code change I couldn't get fails anymore and had to abandon the effort.  Even with the skills I had branched, I couldn't advance the obligatory two-handed weapon skills at all, because a) I rarely missed, even at baseline level and b) it seems like useful weapons for 2 of the 3 single-hand advanced skills don't actually exist (and I was in Salarr at the time, so believe me, I asked).

And as Yam mentioned previously, practically none of this was done while sparring clanmates, because mundane humanoid PCs just can't achieve the kind of defense that is required to max a weapon skill, especially if you're using two-handed.

[Bacon]

I think the other weapon skills shouldn't branch. I think they should just be included in the list to begin with so people can have more variety in weapon use for characters. It would also make it so if you wanted to use that sort of weapon, that's what you can use and train with. The way it works currently, I never bother with using those branched weapons because my character already has a style by then and they're too hard to train. Branching them isn't something I work toward or even look forward to. As someone else mentioned other combat skills don't become reliable at all unless your weapon skill is up there as well making those branched ones nearly useless branching so late on anyway.

I think they could improve slightly faster as well. It seems really odd that things I would consider "more advanced" combat skills can be mastered before a weapon skill can even be bumped up from apprentice to journeyman and then they don't really act as if "master" level because some of them are dependant upon your weapon skill. The skill bump option will help with this but grinding isn't entertainment to me, it's something I have to do to get to a certain point where it is in line with what my character should be able to do at that point of their lives.

[AmandaGreathouse]

I can't begin to fathom how long it must take. Then again, I can't stand combat. It takes way too much effort to be decently good at it, imo. Then again, I don't enjoy playing warriors. I played one, once. She lasted about 12 hours. And with my rangers, I tend to go melee, rather than ranged, so it's probably not surprising that I wind up at oh, about 30 days played, with one, who dies to a carru. Of course. Weapons skills being so slow to get to a point where you can do interesting stuff with them (my opinion, of course), is a lot of the reason I tend to play magickers and merchants, or grebbing rangers who run from like everything. I tried. My god, I tried. There's got to be something I'm doing wrong about it, because I have no problem with noncombat skills. Perhaps it's not for everyone, but if it were changed, I might be more interested in trying.

[brytta.leofa]

I know, I'm being That Guy,  but...

Skill Weapons    (Combat)

Weapon skills represent specialised knowledge in the use of a particular class of weapons. All weapons fall into one of four general categories: bludgeoning, chopping, piercing, and slashing. Stabbing weapons are merely a subset of piercing ones.

If one possesses, for example, the 'chopping weapons' skill, then one's usage of an axe is improved both in ability to land blows and to parry those of others, the degree of improvement depending on one's level of accomplishment in that skill.

Each category of weapon has its own characteristics, which should become evident as your character becomes more familiar with its usage.

See also:

    combat, defense, offense, skill_parry

...the things in this thread that are nowhere mentioned in that helpfile should probably not be in this thread.

[Nyr]

I can't begin to fathom how long it must take. Then again, I can't stand combat. It takes way too much effort to be decently good at it, imo. Then again, I don't enjoy playing warriors. I played one, once. She lasted about 12 hours. And with my rangers, I tend to go melee, rather than ranged, so it's probably not surprising that I wind up at oh, about 30 days played, with one, who dies to a carru. Of course. Weapons skills being so slow to get to a point where you can do interesting stuff with them (my opinion, of course), is a lot of the reason I tend to play magickers and merchants, or grebbing rangers who run from like everything. I tried. My god, I tried. There's got to be something I'm doing wrong about it, because I have no problem with noncombat skills.  Perhaps it's not for everyone, but if it were changed, I might be more interested in trying.

If you don't like combat, you don't like playing warriors, you play rangers like warriors, and you've only played one warrior for a few hours..maybe that's the problem?  You haven't experienced the guild or given it a fair shake.  No, it is not for everyone.  Everyone has a niche that they enjoy, and while it is good to break out of the box, most tend to return to the things that appeal to them.  However, if you haven't given it a shot and don't really do combat anyway, why would you be interested in trying it if minute changes were made?

[AmandaGreathouse]

I don't enjoy it because it is so time consuming (time vs reward ratio), and that alone makes the rest of it like... meh. Then again, I like being able to do other stuff and warriors seem very focused on just combat. I was speaking as someone who enjoys playing rangers, who might enjoy playing rangers who were more inclined to actually be 'hunters' of some sort, if the change was made. As it is, a carru taking out a 30ish days played ranger, who trains regularly and has good wisdom, was enough to put me off it. That's a lot of my life to put in to be taken out on a supposed hunter by something which is... from what I understand, not typically so difficult for people in a comparable position. Eh. I was merely offering my thoughts on the thread and proposals made, with a bit of anecdotal info to give people maybe an idea where I'm coming from to have those positions.

[Delirium]

Rangers are very good hunters... you just have to play them like a hunter.

To each their own though. Crafting bores me out of my skull, I find it nearly impossible to focus on the game.

[Synthesis]

I don't enjoy it because it is so time consuming (time vs reward ratio), and that alone makes the rest of it like... meh. Then again, I like being able to do other stuff and warriors seem very focused on just combat. I was speaking as someone who enjoys playing rangers, who might enjoy playing rangers who were more inclined to actually be 'hunters' of some sort, if the change was made. As it is, a carru taking out a 30ish days played ranger, who trains regularly and has good wisdom, was enough to put me off it. That's a lot of my life to put in to be taken out on a supposed hunter by something which is... from what I understand, not typically so difficult for people in a comparable position. Eh. I was merely offering my thoughts on the thread and proposals made, with a bit of anecdotal info to give people maybe an idea where I'm coming from to have those positions.

Meh, if you can't hunt carru solo with an 8-10 day ranger, you're doing it wrong somehow, or your stats are terrible (possibly doesn't apply to elves).  This isn't to say that you'll be 100% successful and emerge unscathed every time, but by 10 days, you should be able to maul the majority of them up pretty well without even resorting to archery.

Of course, this is assuming you've actually been hunting for that 8-10 days, not sitting around at the bar or mudsexing.

[Sunburned]

If you don't like combat, you don't like playing warriors, you play rangers like warriors, and you've only played one warrior for a few hours..maybe that's the problem?  You haven't experienced the guild or given it a fair shake.  No, it is not for everyone.  Everyone has a niche that they enjoy, and while it is good to break out of the box, most tend to return to the things that appeal to them.  However, if you haven't given it a shot and don't really do combat anyway, why would you be interested in trying it if minute changes were made?

I'm a chronic warrior-player who would  be more interested in playing my favorite guild again if alterations were made to how weapon skills improve.  Not that code changes should/or ever be done to cater to an individual.

I've played one warrior (40+ days) that, prior to certain code changes, managed to get an advanced weapon skill to a skilled level.  This is back when it was easier.  It was so goddamn grindy, so ridicolously hard, and twinkish for that matter, I will never make that effort again.  Yet, for some people, that advanced weapon skill is a golden-ring that needs to be constantly sought after, and regardless of the faults of the player, it seems detrimental to have a large, untapped quantity of coded proficiency that is only rewarded to players that are willing to take extreme, twinky measures.

In my ideal Arm, weapon skills -still- take a ridiculously long time to improve, but warriors who actually remain within realistic limits of their roles (ie, don't go hunting with the sole OOC intention of improving a specific skill) could still improve them through long-term dedicated play, regardless of whether they're in the city or wilderness.    

[Nyr]

I was speaking as someone who enjoys playing rangers, who might enjoy playing rangers who were more inclined to actually be 'hunters' of some sort, if the change was made.

A ranger's strong point is "pew pew pew" with arrows, not "smack smack smack" in melee combat.  Sneaking, hiding, hunting, and shooting have nothing to do with weapons skills.  I'd suggest reading over the helpfiles for rangers, it really helps get more into the mindset of their strengths, weaknesses, and overall playstyle.

[Down Under]

If you don't like combat, you don't like playing warriors, you play rangers like warriors, and you've only played one warrior for a few hours..maybe that's the problem?  You haven't experienced the guild or given it a fair shake.  No, it is not for everyone.  Everyone has a niche that they enjoy, and while it is good to break out of the box, most tend to return to the things that appeal to them.  However, if you haven't given it a shot and don't really do combat anyway, why would you be interested in trying it if minute changes were made?

I'm a chronic warrior-player who would  be more interested in playing my favorite guild again if alterations were made to how weapon skills improve.  Not that code changes should/or ever be done to cater to an individual.

I've played one warrior (40+ days) that, prior to certain code changes, managed to get an advanced weapon skill to a skilled level.  This is back when it was easier.  It was so goddamn grindy, so ridicolously hard, and twinkish for that matter, I will never make that effort again.  Yet, for some people, that advanced weapon skill is a golden-ring that needs to be constantly sought after, and regardless of the faults of the player, it seems detrimental to have a large, untapped quantity of coded proficiency that is only rewarded to players that are willing to take extreme, twinky measures.

In my ideal Arm, weapon skills -still- take a ridiculously long time to improve, but warriors who actually remain within realistic limits of their roles (ie, don't go hunting with the sole OOC intention of improving a specific skill) could still improve them through long-term dedicated play, regardless of whether they're in the city or wilderness.    

I guess I hadn't really thought about this until Sunburned brought it up, and Synthesis RE: advanced weapon skills.

It is pretty unfortunate that by the time a Warrior branches any advanced weapons skills, their base offense is SO good that it's almost impossible to skill up the advanced weapons skill unless you are doing massively twinky stuff to your character that is clearly OOC.

It sounds stupid, but the only thing I can think of is 'nosave training' that allows for more misses, softer blows, and the opportunity for advanced PC's to tone it down a notch and try to actually train instead of just beat ass on anyone that steps into a training ring with them.

Beyond that, it's silly that Warriors have to seek out almost instant death in order to get fails on weapon skills.

I do agree with Jim that it's cool that advanced weapon Warriors are about as rare as Defilers, but the system is actually more grindy than any Magicker system. Not sure what the fix would be, but the problem is definitely there.

[Synthesis]

You don't have to seek out instant death to train up a weapon skill.  However, if 3 or 4 particular creatures were removed from the game, it would be very difficult to get a weapon skill beyond advanced.  It does get somewhat ridiculous fighting/hunting/killing those things over and over and over, especially since they have relatively little economic value.

The simple solution is just to put something like a 5-10% random failure baseline, no matter what the skill level differential is between the attacker and the defender.  So no matter how skilled you are, you'll still get a failure every once in a while.  To a) aid clanfolk and b) prevent it from being an offense nerf, you could apply this random failure baseline only when using items intended to be sparring weapons.  (You could also significantly increase the failure rate, if you limited it to sparring weapons.)

On the other hand, you could leave everything the same, and re-code the way weapon skills increase, removing them from the failure-based system and changing it to a use-based system (i.e. every time you use it, it improves a little bit, regardless of whether you fail or not).  They improve so slowly anyway, I don't see how that would be much of a problem, beyond the coding aspect.

[AmandaGreathouse]

I was speaking as someone who enjoys playing rangers, who might enjoy playing rangers who were more inclined to actually be 'hunters' of some sort, if the change was made.

A ranger's strong point is "pew pew pew" with arrows, not "smack smack smack" in melee combat.  Sneaking, hiding, hunting, and shooting have nothing to do with weapons skills.  I'd suggest reading over the helpfiles for rangers, it really helps get more into the mindset of their strengths, weaknesses, and overall playstyle.

I'm thinking of the utility skill of rangers, that's the appeal more so than the combat. Yes, the combat -can- be done. If it improved a little more easily, it would be appealing. But the cost of arrows makes it really impractical, imo, when the salary of the average clanned pc is somewhere around 150 sid/rl week, and the cost of arrows would afford 3-6 of them at that rate, for the cheaper ones I've seen. Spears and other thrown type things aren't so bad, but slings are godawful (personal opinion, again), so I wind up going with melee more by default. I mean, it's hard to launch a couple small to a half a large worth of arrows at something for a piece of meat and maybe a few sid for the skin, and feel like you're being realistic in the framework of the world (economywise, which I admit is skewed). It is the fact that warriors are nearly -all- combat that makes them unappealing. I really like the idea that was posted elsewhere of merging warrior and ranger, but that's not something to get into here, I suppose. I understand the difference in a warrior and a ranger.

Edit to add: Just caught Synthesis' post because chrome shutdown on me. While I don't like the idea of the sparring weapons being given a higher failure rate, I do like the idea of having the skills slowly improve with use regardless of failure. That seems like a realistic and not drastically different way of handling things which could/might improve them if implemented.

[Synthesis]

Slings don't suck, unless you suck at using them.

[Nyr]

I was speaking as someone who enjoys playing rangers, who might enjoy playing rangers who were more inclined to actually be 'hunters' of some sort, if the change was made.

A ranger's strong point is "pew pew pew" with arrows, not "smack smack smack" in melee combat.  Sneaking, hiding, hunting, and shooting have nothing to do with weapons skills.  I'd suggest reading over the helpfiles for rangers, it really helps get more into the mindset of their strengths, weaknesses, and overall playstyle.

I'm thinking of the utility skill of rangers, that's the appeal more so than the combat. Yes, the combat -can- be done. If it improved a little more easily, it would be appealing. But the cost of arrows makes it really impractical, imo, when the salary of the average clanned pc is somewhere around 150 sid/rl week, and the cost of arrows would afford 3-6 of them at that rate, for the cheaper ones I've seen. Spears and other thrown type things aren't so bad, but slings are godawful (personal opinion, again), so I wind up going with melee more by default. I mean, it's hard to launch a couple small to a half a large worth of arrows at something for a piece of meat and maybe a few sid for the skin, and feel like you're being realistic in the framework of the world (economywise, which I admit is skewed). It is the fact that warriors are nearly -all- combat that makes them unappealing. I really like the idea that was posted elsewhere of merging warrior and ranger, but that's not something to get into here, I suppose. I understand the difference in a warrior and a ranger.

Help ranger = rangers are far poorer at combat than warriors; increasing (or decreasing) the rate at which weapon skills improve would not affect rangers as they are not intended to be combat powerhouses (and as previously mentioned by both you and me, their other skills are what they are intended to be used for and are in fact their strengths).  It would only affect them if they were being played as melee combatants, which isn't their strong suit anyway, so it wouldn't have more than a minute effect.  The average clanned (and unclanned) PC can get ahold of arrows (and 'sid) easier than you might think, but even so, archery is a powerful skill and the cost of supplies comes into play for purposes of balance.

But back to the main point in response to your concern, I do not think that altering weapons skills in any way (greatly or minutely) will affect your play of rangers.  If you are playing them like warriors when the helpfile and the guild disagree with that, you are going to be constantly disappointed unless you like a major challenge.  If you want a ranger that can fight like a warrior, use one of the extended subguilds.

[AmandaGreathouse]

It's not about my play of rangers. It's about weapon skills and whether or not I like the idea of the change proposed (my posting is), I do like the idea of a change. Yes, I like a major challenge. That's why I love playing magickers or horribly disturbed characters. I might well play one with one of the extended subguilds at some point. But regarding the proposal, I still think it would be a positive thing. Completely outside my play of rangers, or lack thereof, or my dislike of warriors. My limited use of them is far from all encompassing, I admit, but when I do use them... I would like the idea of the change that was brought up.

[Yam]

On the other hand, you could leave everything the same, and re-code the way weapon skills increase, removing them from the failure-based system and changing it to a use-based system (i.e. every time you use it, it improves a little bit, regardless of whether you fail or not).  They improve so slowly anyway, I don't see how that would be much of a problem, beyond the coding aspect.

This would fix the problems I have with the way I see and am pretty sure the system of weapon skill improvement works. Weapon skills would be improvable by all long lived, dedicated warriors and not just those who have accrued some obscure and, in my opinion, unrealistic game knowledge.

[Spoon]

I'd much rather go for the increased chance of failure than 'use-based' skill increases. It's works well now because there's a point where you have to fight things that are a 'challenge'. Problem is, 'challenge' doesn't mean challenge. In Arm, you learn more trying to hit a mouse that fighting off a dragon.

As for increased failure rate, it can be tiny, but just so there is some chance. I'm thinking a level where it would only really affect PCs that have been around a great deal of time, just to give them some chance of improving over a long period.

[titansfan]

I like synthesis' idea of use based learning...if any change were to be made imo that fits best with the rate of increase.

[catchall]

I'd much rather go for the increased chance of failure than 'use-based' skill increases. It's works well now because there's a point where you have to fight things that are a 'challenge'. Problem is, 'challenge' doesn't mean challenge. In Arm, you learn more trying to hit a mouse that fighting off a dragon.

As for increased failure rate, it can be tiny, but just so there is some chance. I'm thinking a level where it would only really affect PCs that have been around a great deal of time, just to give them some chance of improving over a long period.

The problem is that simply establishing a fixed minimum miss rate doesn't change anything substantial. You'll still learn more fighting the mouse. This will always be the case as long as a "learn by failure" system is in use.  "Learn by failure" is a fairly straightforward and simple way to make challenging tasks more useful for skill improvement, as they should be, and are in real life.  Though this is a commonplace and generally effective system, at it's core it's something of a kludge.  In real life, you don't get better at something by screwing up. You get better by succeeding at tasks that are just at the boundary of your ability and correcting past mistakes.  

That is, in real life, failing at something and getting better at something are two independent consequences of engaging in challenging tasks.  A learn by failure skill system in a computer game simulates that effect by having skill failure mediate between a challenging task and skill gain. This has an appealing self-balancing mechanism built in, where no one can get better at a skill after a point where they are always successful.  It's also easier from a code perspective to look at one binary data point (success or failure) rather than design a complicated improvement algorithm that calculates the difficulty of the task and weighs your skill roll and current skill level in order to determine if improvement is in order, which would be more "realistic" but much more complicated to code and balance, with little pizzazz visible from player-side. After all, if it were well-implemented, it would feel like the current system most of the time (the current system is perfectly adequate most of the time), and this is a roleplaying game, not a machine model of learning. Too much detail is undesireable.  In the marginal cases discussed in this thread, though, the unrealistic elements in the system compound to a point where a playability problem is experienced by some of the players here.

However practical learn-by-failure may be, it's important to remember that it is a conflation. Learning and failure aren't the same thing, and as long as they are treated the same or close to the same by the game, we will see unrealistic methods of advancement, where the best way to advance a skill is generally to do something FAR too hard for your current aptitude (which in real life would be of virtually no benefit), and/or orchestrate circumstances that are conducive to failing at the skill (but which realistically speaking, are not conducive to learning the fundamentals of the skill). I'm being intentionally vague about the ways someone might go about this and the unrealistic nature of these tasks, for obvious reasons.

So, simply adding a base chance of failure is, in effect, a kludge that attempts to mitigate some marginal shortcomings of a game mechanic--learn by failure--that is already a bit kludgy.   (You could argue that a base failure chance is reasonable in and of itself, and I'd be inclined to agree, but as a solution for the present issue, it's less than a band-aid, since it doesn't even begin to address unrealistic advancement methods being more effective than realistic ones.)  I don't have a solution of my own, but I just hope that learn-only-by-failure is not retained as a fundamental part of the skill system in Arm 2.  Armageddon's skill system is too sophisticated and the interactions between skills too subtle and complex for a simple learn-by-failure system to suffice, as witnessed by the issues reported here. Learn-by-failure is best suited to simple skill systems with few interactions between skills.

[Sunburned]

The problem is that simply establishing a fixed minimum miss rate doesn't change anything substantial. You'll still learn more fighting the mouse. This will always be the case as long as a "learn by failure" system is in use.  "Learn by failure" is a fairly straightforward and simple way to make challenging tasks more useful for skill improvement, as they should be, and are in real life.  Though this is a commonplace and generally effective system, at it's core it's something of a kludge.  In real life, you don't get better at something by screwing up. You get better by succeeding at tasks that are just at the boundary of your ability and correcting past mistakes.  

That is, in real life, failing at something and getting better at something are two independent consequences of engaging in challenging tasks.  A learn by failure skill system in a computer game simulates that effect by having skill failure mediate between a challenging task and skill gain. This has an appealing self-balancing mechanism built in, where no one can get better at a skill after a point where they are always successful.  It's also easier from a code perspective to look at one binary data point (success or failure) rather than design a complicated improvement algorithm that calculates the difficulty of the task and weighs your skill roll and current skill level in order to determine if improvement is in order, which would be more "realistic" but much more complicated to code and balance, with little pizzazz visible from player-side. After all, if it were well-implemented, it would feel like the current system most of the time (the current system is perfectly adequate most of the time), and this is a roleplaying game, not a machine model of learning. Too much detail is undesireable.  In the marginal cases discussed in this thread, though, the unrealistic elements in the system compound to a point where a playability problem is experienced by some of the players here.

However practical learn-by-failure may be, it's important to remember that it is a conflation. Learning and failure aren't the same thing, and as long as they are treated the same or close to the same by the game, we will see unrealistic methods of advancement, where the best way to advance a skill is generally to do something FAR too hard for your current aptitude (which in real life would be of virtually no benefit), and/or orchestrate circumstances that are conducive to failing at the skill (but which realistically speaking, are not conducive to learning the fundamentals of the skill). I'm being intentionally vague about the ways someone might go about this and the unrealistic nature of these tasks, for obvious reasons.

So, simply adding a base chance of failure is, in effect, a kludge that attempts to mitigate some marginal shortcomings of a game mechanic--learn by failure--that is already a bit kludgy.   (You could argue that a base failure chance is reasonable in and of itself, and I'd be inclined to agree, but as a solution for the present issue, it's less than a band-aid, since it doesn't even begin to address unrealistic advancement methods being more effective than realistic ones.)  I don't have a solution of my own, but I just hope that learn-only-by-failure is not retained as a fundamental part of the skill system in Arm 2.  Armageddon's skill system is too sophisticated and the interactions between skills too subtle and complex for a simple learn-by-failure system to suffice, as witnessed by the issues reported here. Learn-by-failure is best suited to simple skill systems with few interactions between skills.

I think you're vastly overcomplicating the problem.  Its clear, from the variety of progression rates and modes of failure by which skills improve, that staff have many means by which to alter what, at face value, seems like a simple, ham-fisted method of progression.  I'm confident there's an elegant code solution, if the staff wanted to implement it.

[Bacon]

I like synthesis' idea of use based learning...if any change were to be made imo that fits best with the rate of increase.

I do too. Perhaps learn from both failures and successes. You'd just learn a bit more from a failure each time vs. what you would learn from each success.

[roughneck]

I like that it's very, very difficult to reach the ceiling with weapon skills and base offense/defense.  I think it reflects the fact that nobody knows -everything- about -anything-.

If a change was made to make weapon skills easier to train or simply or changed to simply require 'putting time in', then it becomes a stat battle when people face off, because they all trained to roughly the same level.

The current system doesn't make perfect sense, realistically, but at least it makes training the skills very difficult.  There should always be room for improvement, that is more realistic than a system that allows you to improve unreasonably fast.  It also makes perfect sense to improve at a slower rate at high levels.

Besides, I feel like the difficulty to train and effectiveness of weapon skill is being misrepresented here.  It's not supremely difficult to reach journeyman level with a weapon skill, at which level a PC is very competent.  Even with a weapon skill at the apprentice level, coupled with journeyman dual-wield or two-handed, a ranger is effective enough in combat, prior to acquiring the defensive skills that class branches to.

Either way, if a change was made, there would be a percentage of players that figured out how to maximize their skilling more efficiently than others and this discussion would pop up again.

[titansfan]

I don't think it would make it easier by any drastic amount...I just feel it would allow advanced players to still advance their skills without running into boredom due to you being too badass with your skills and never learning anything you don't even know anything about.

[Sunburned]

I like that it's very, very difficult to reach the ceiling with weapon skills and base offense/defense.  I think it reflects the fact that nobody knows -everything- about -anything-.

If a change was made to make weapon skills easier to train or simply or changed to simply require 'putting time in', then it becomes a stat battle when people face off, because they all trained to roughly the same level.

The current system doesn't make perfect sense, realistically, but at least it makes training the skills very difficult.  There should always be room for improvement, that is more realistic than a system that allows you to improve unreasonably fast.  It also makes perfect sense to improve at a slower rate at high levels.

Besides, I feel like the difficulty to train and effectiveness of weapon skill is being misrepresented here.  It's not supremely difficult to reach journeyman level with a weapon skill, at which level a PC is very competent.  Even with a weapon skill at the apprentice level, coupled with journeyman dual-wield or two-handed, a ranger is effective enough in combat, prior to acquiring the defensive skills that class branches to.

Either way, if a change was made, there would be a percentage of players that figured out how to maximize their skilling more efficiently than others and this discussion would pop up again.

For warriors, its not just very, very difficult to reach the skill ceiling for a weapon skill... for most - that choose to stay within the realistic constraints of their roles - its impossible (with the exception of a few roles where warriors are under-represented).

Your argument that it would become a "stat battle" doesn't seem valid to me, because it assumes that 1) all players who play warriors are capable of running a long-lived warrior, 2) weapon skills are the most critical skill in determining a warrior's success, and 3) there is no strategy in melee combat.  All of these assumptions are, in my humble opinion, false.  

I believe its valid to say that this has less impact on rangers - if the code was changed, it would be more to the benefit of warriors, because their particularly narrow skill tree is balanced in consideration of their higher potential with weapon skills... but currently, for most warrior players, the level of expertise which grants highest returns is unattainable, no matter if they're 20 days played or 200 days played... or it is only attainable to players who are expert twinks.