|
I think you are seeing Perception as a much more important skill than our gaming group. Certainly at least one or two members of the party should have maxed out Perception for scouting, but aside from that, there is no panic amongst our players.
As well, the "skill tax" of which you speak is greatly reduced in Pathfinder -- in 3.5, Spot was one of 45 skills, now Perception is one of 36.
As well, just because you are a high CR creature, doesn't mean you are observant. Take a stone giant skeleton, for example.
While some of the skill merges make me concerned for skill balance, I believe Perception is about equal in value to Acrobatics, Bluff, or Stealth.
This, as I see it, is a problem. When a skill becomes so important that everyone want to take it, it isn't a skill anymore, it's just a skill point tax for something that everyone assumes an adventurer should be able to do anyway.
I respect what you're saying, but can we please have a moratorium on the phrase "skill point tax"?
Maybe I'm just a grungy old grognard, but it used to be that the only characters with real skills were thieves, assassins and monks. That's just the way it was.
Now in Pathfinder, everyone is clamouring that the poor cleric and fighter "only" have two skill points per level (plus whatever they get for Int, for being a human, for having a favored class and/or having a headband of intellect, not to mention the various merged skills compared to 3.5) and that's horribly crippling somehow. Oy vey...
(Sorry to derail the thread, but I've had that on my chest for a while now.)
RPG Superstar 2008 Top 32, 2011 Top 16
|
I don't see this as a must have skill either. In one group I'm in, 2 of the 4 PCs have maxed it out, and they're a rogue and a ranger/rogue. The other group I'm in, I'm not sure if anyone has ranks in it, though it's possible the monk does (and this group has 6 PCs.)
|
helpful hint for those who have no ranks in perception, but the DM ask for it all the time...take 10. If you have a DM that ask for more Perception checks than initiative checks, then hes probably a little novice and it doesnt really matter if you make it or not. why role something that make you feel inept, just take ten.
note: check to make sure that the DM isnt always the supriser (the badguys always seem to know your coming and make you have to make perception rolls to spot their ambush.
also check to make sure that distance is done right (every point you rolled over their stealth is ten more feet distance between you)
as well, make sure your DM isnt always forcing suprise round when there shouldnt be one (the elf Merris notices the sound of orcs charging though the woods, toward the party. unless he made it spot on he should have a free action to yell a warning to unwary allies, and initiative should be rolled normally
perception should be a character role, not a part ability. one character with eagle senses should protect the party from surprise, most of the time.
CR is a game contstruct. It should only be used, ever, to calculate XP. It should not feed back into anything else.
If you want something that scales with level, pick HD, BaB, or Saves.
I'm seeing the same problem in our playtest as the OP remarked on. There are really very few actual skills now, plus a bunch of "leftovers" that people put, at most, 1 point into.
What I notice is that the skills people are putting more than 1 rank into are, almost without exception, the same skills that benefited by the skill combinations. Skills that didn't get combined or bumped are now treated by all players as "leftovers" that don't get points.
Again, I very strongly urge that those other skills be either consolidated or improved so that they merit the status of "full skills" again:
Without doing so, what we have is a de facto skills vs. "fields of experience" system, that we're just pretending is a unified skill system instead.
|
I agree with Hogarth - your characters have possibly more skill points with Pathfinder RPG (favoured class bonus) and those points purchase more than they did under 3.5 (skill consolidation).
If all your players are suddenly maxing Perception in Pathfinder, then they're sacrificing some other skill choice they would have chosen under 3.5.
I suspect this may be a temporary blip, as players are enticed by the perceived benefits of consolidated skills. But once they've played their maxed Perception character, realised it wasn't that big a deal and that there is a whole list of other skills to choose from, particularly if everyone else at the table has maxed the same skill, then they'll drift back to choosing skills to develop their character concept, or to balance the table, or to take that skill they'd sacrificed for Perception with their last character.
If all your players are suddenly maxing Perception in Pathfinder, then they're sacrificing some other skill choice they would have chosen under 3.5.
My point is, those other skills are, in comparison, quite useless -- but they have the exact same cost: 1 point per rank. Things with the same cost should have roughly the same value, or the system is out of whack. Jason made a move in the right direction with the Master Craftsman feat, which just might entice someone into putting more than 1 rank into Craft (weaponsmith).
But Knowledge (history) is still nowhere near as useful as Acrobatics or Perception. And why is Perception all one skill, whereas Perform (piano) and Perform (harp) and Perform (singing) and Profession (songwriter) are "each sold separately"?
Knowledge (history) is still nowhere near as useful as Acrobatics or Perception. And why is Perception all one skill, whereas Perform (piano) and Perform (harp) and Perform (singing) and Profession (songwriter) are "each sold separately"?
Part 1: Knowledge (history) can and does get rolled in my runelords campaign. It has saved the party in its way, and the Wizard loves rolling it. There's a lot of history in that AP that can save a TPK
Part 2: I agree that perform subskills are silly. In "real life" it is quite common to play only one instrument, or to play music but not write it, or to write great music but play poorly. But it doesn't make sense for non-bard characters to have a perform skill unless we can set DCs to do something useful with it, besides busking.
I believe that Performance Subskills should be feats that actually grant feat-like benefits, in addition to a revenue generating benefit (all they offer right now for non-bards) and a "Bard Talent" that beefs up Bardic performance for free. That's my preferred fix for Perform.
If all your players are suddenly maxing Perception in Pathfinder, then they're sacrificing some other skill choice they would have chosen under 3.5.My point is, those other skills are, in comparison, quite useless -- but they have the exact same cost: 1 point per rank. Things with the same cost should have roughly the same value, or the system is out of whack. Jason made a move in the right direction with the Master Craftsman feat, which just might entice someone into putting more than 1 rank into Craft (weaponsmith).
But Knowledge (history) is still nowhere near as useful as Acrobatics or Perception. And why is Perception all one skill, whereas Perform (piano) and Perform (harp) and Perform (singing) and Profession (songwriter) are "each sold separately"?
If you take Perception away then you could make the exact same argument about acrobatics or climb. The problem is that some skills (Acrobatics, Ride, Perception, etc) are have clearly defined mechanical benefits while others (Craft, most knowledge skills, perform, profession, etc) have no clear cut in-game benefit. The more valuable the skills with mechanical benefits become the less likely people are to take the other skills.
Taking Perception away won't solve this issue, it will just shift it to some other skill with a mechanical benefit.
Knowledge (history) can and does get rolled in my runelords campaign. It has saved the party in its way, and the Wizard loves rolling it. There's a lot of history in that AP that can save a TPK ** spoiler omitted **
I was hoping someone would point that out. Runelords is a special case, in which the authors went out of their way to make sure that Knowledge (history) got some use. This was partly done to introduce the new Golarion setting, I think, but it's still a good practice: selectively emphasizing the use of certain skills in ALL prewritten adventures is a nice way to "bump up" the relative value of those skills.
But now I look through my 3.5e adventures from Dungeon and elsewhere, and I see very little use indeed for it. Certainly nothing that a Perception roll wouldn't do better.
I'd kind of like to see a single Knowledge (lore) skill, that covers all of Knowledge (the planes), Knowledge (history), Knowledge (religion), and Knowledge (arcana). Much like there's just a single Perception skill now.
Taking Perception away won't solve this issue, it will just shift it to some other skill with a mechanical benefit.
Exactly right. I like Perception. I'd also like for the other skills to have mechanical benefits, either through consolidation of skills and/or feats (e.g., roll Quick Draw into the Sleight of Hand skill) or addition of other abilities.
If a skill has no mechanical benefit, it shouldn't be priced the same as a skill that provides consistent and significant benefit. That's just basic game design. There are two possible fixes: (1) make the "useless" skills cheaper (which I disagree with, as it makes for a clunky and non-standard system), or (2) give the "useless" skills uses (which I'm all in favor of).
I was hoping someone would point that out. Runelords is a special case, in which the authors went out of their way to make sure that Knowledge (history) got some use. This was partly done to introduce the new Golarion setting, I think, but it's still a good practice: selectively emphasizing the use of certain skills in ALL prewritten adventures is a nice way to "bump up" the relative value of those skills.
But now I look through my 3.5e adventures from Dungeon and elsewhere, and I see very little use indeed for it. Certainly nothing that a Perception roll wouldn't do better.
I'd kind of like to see a single Knowledge (lore) skill, that covers all of Knowledge (the planes), Knowledge (history), Knowledge (religion), and Knowledge (arcana). Much like there's just a single Perception skill now.
It's a good point, and it comes down to a matter of opinion. Specifying skill lists is guaranteed to be maddening.
I've always figured that the #1 duty of Pathfinder RPG was to provide me with stand-alone rules that would let me play the existing Pathfinder APs. Other 3.5 material is a fringe benefit, but should still be playable in a system that can handle the existing Pathfinder APs. If part of a quality campaign is using all the skills at some point, then I feel Paizo doesn't owe me much more than that.
|
I do think certain knowledges can be subsumed into others, I don't think their needs to be a knowledge (the planes) just knowledge (arcana):covers elemental planes and the far realms and knowledge (religion): covers alligned planes. Then knowledge nobility and royalty needs to be eaten by knowledge history.
I do think certain knowledges can be subsumed into others, I don't think their needs to be a knowledge (the planes) just knowledge (arcana):covers elemental planes and the far realms and knowledge (religion): covers alligned planes. Then knowledge nobility and royalty needs to be eaten by knowledge history.
That sounds about right, LKL.
I do think certain knowledges can be subsumed into others, I don't think their needs to be a knowledge (the planes) just knowledge (arcana):covers elemental planes and the far realms and knowledge (religion): covers alligned planes. Then knowledge nobility and royalty needs to be eaten by knowledge history.
Knowledge nobility in knowledge history is a good idea I think.
I'm less inclined on the other, but it's interesting.
My point is, those other skills are, in comparison, quite useless -- but they have the exact same cost: 1 point per rank. Things with the same cost should have roughly the same value, or the system is out of whack. Jason made a move in the right direction with the Master Craftsman feat, which just might entice someone into putting more than 1 rank into Craft (weaponsmith).
But Knowledge (history) is still nowhere near as useful as Acrobatics or Perception. And why is Perception all one skill, whereas Perform (piano) and Perform (harp) and Perform (singing) and Profession (songwriter) are "each sold separately"?
I'm with you on this.
Huh? I really really REALLY doubt every single character is going to max out perception at every level, having seen it not happen in my own games. Stealth has sort of become a skill tax in my games, but I'm ok with that.
The way perception used to be, it acted as a skill tax for rangers and rogues because everyone expected those classes to have those skills maxed (since they were class skills and rangers and rogues had extra skill points). In this new system, the 'perceptive' classes can say 'No, I don't have it maxed. Why don't you?'. And frankly, if I were a fighter with average int, and not a ton of points to throw around, I'd leave perception to others so I could, ya know, climb, swim, jump, or sneak. So no, I don't think it's a skill tax at all.
|
CR is a game contstruct. It should only be used, ever, to calculate XP. It should not feed back into anything else.
If you want something that scales with level, pick HD, BaB, or Saves.
ummmm HD, BAB and Saves are game constructs also...
Just saying...
So, obviously, CR could be used for other uses. There is no reason it cannot.
We have a unique opportunity to create a game that expands upon and become greater than its origins.
Let's think outside the box.
EVERYTHING is a game construct... so let's use everything to advantage.
Just saying...
Huh? I really really REALLY doubt every single character is going to max out perception at every level.
Doubt it all you like, but like I said, in our playtest, that is exactly what's happening -- even the newbies realize that Perception, Stealth, and Acrobatics are simply better than all the other skills (and Spellcraft for casters, of course).
Even if it's not an issue at your table, ask yourself: does it make sense for different skills, which cost the same amount (1 point per rank), to have such wildly different levels of usefulness? Whether or not it's a problem at one particular table, it's just flat-out bad design.
Huh? I really really REALLY doubt every single character is going to max out perception at every level.Doubt it all you like, but like I said, in our playtest, that is exactly what's happening -- even the newbies realize that Perception, Stealth, and Acrobatics are simply better than all the other skills (and Spellcraft for casters, of course).
In my experience, it really makes a difference whether you've had experience with a DM who loves ambushes or not!
I used to play in a 2nd edition game where we would get bushwhacked every time, no matter how many precautions we made. That's why I like to max out Listen (or Spot or Perception) on my characters now, when possible.
But I'm a pretty lenient DM myself, so my players don't go overboard with that sort of thing.
|
|
In my experience, it really makes a difference whether you've had experience with a DM who loves ambushes or not!
True, my first DM always surprised us. Built my character for me and maxed my perception skills, but we never even made a roll!
I felt especially sorry for mybrother, the ranger, with a +36 spot. Never got to roll it once. heh.
Ranks in Perception are like hit points. Everybody of every class wants them, and everybody of every class takes them whenever they can. In my experience, "2+Int skill points per level" now really means "1+Int skill points per level, plus a rank in Perception."
Back when Perception was several skills, there were some adventurers who were perceptive and others who weren't. These days, all adventurers are perceptive; a few don't get a class bonus to Perception, but Perception is otherwise maxed out for every character.
This, as I see it, is a problem. When a skill becomes so important that everyone want to take it, it isn't a skill anymore, it's just a skill point tax for something that everyone assumes an adventurer should be able to do anyway.
So things were better off when characters couldn't see or hear anything. Similarly, by that reasoning, rogues technically have 5 + Int mod skills ranks per level, since they're (probably) automatically putting points into Stealth, Perception, and Disable Device.
Every skill has to be demonstrably equal, whatever that means, when you are discussing something like skills?
And if they aren't equal, then we have to get rid of even more skills to fix the problem?
We've already consolidated a bunch of skills and eliminated cross class penalties, and we are probably going to see spellcasters no longer having to take a skill in order to concentrate . . .
Following this logic, it almost sounds like the conclusion here would be to scrap the skill system entirely and go to something more like what Castles and Crusades does, i.e. having certain "class specific" things that you can make a level check against.
Honestly, I'm not seeing perception being easier for non-rogues to take as a problem, at least not a problem that calls for the system to make it even easier to get more skills by eliminating another skill from the system.
Every skill has to be demonstrably equal, whatever that means, when you are discussing something like skills?
And if they aren't equal, then we have to get rid of even more skills to fix the problem?
Well, no, but the skills could be closer to equal in value. When you have one skill that is near must-have (like Perception to a rogue), and another that is nearly useless (like Craft(Cobbler) or something), you have allegations, like those that appeared near the end of the 3.5 era, that if you aren't using the 'good' skill then you're playing the game wrong.
Now, I'm not saying that the Craft and Profession skills need to be quite as useful to PCs (they're disproportionately useful to NPCs) as other adventuring related skills. However, some effort should be made to make the other skills all vaguely comparable.
I don't know if its broken right now. I haven't had enough experience with the new system. But it's worth thinking about.
Disbalanced or not (and I disagree, as the level of usefulness of a skill varies from game to game), there's no real way to change it without adding the completely retarded search, spot, and listen that used to exist. Why don't we find ways to make the other skills more useful instead?
I understand your suggested fix, meepo, but I don't like the idea of making everything inherently observant. I actually LIKE having characters and monsters without perception. It makes it more realistic.
I'm with epic meepo on this one...
Cross class skills are now too easy to develop. It was too difficult before in 3.5, and now it's the opposite...
I'm with epic meepo on this one...
Cross class skills are now too easy to develop. It was too difficult before in 3.5, and now it's the opposite...
I very much agree.
IMO class roles are getting seriously trampled and it is a stark mar on the system. But that is just my opinion.
Right now a real problem has been solved, but at the expense of a worse problem. And the old problem only impacted prep. The new problem hits against how the game actually plays at the table.
Every skill has to be demonstrably equal, whatever that means, when you are discussing something like skills?
I realize you are being somewhat rhetorical here, but to follow this point...
It is much easier to maintain a degree of equity when the fucntion of the skill within the context of the class is retained. The value of Stealth is different for a rogue than it is for a wizard.
Following this logic, it almost sounds like the conclusion here would be to scrap the skill system entirely and go to something more like what Castles and Crusades does, i.e. having certain "class specific" things that you can make a level check against.
I think this is a good point.
Not that I want to go that way. The system certainly has the capacity to offer much more depth. But as it stands, you are right.
If you don't believe me, use the 3.5 rules to run a high-level dungeoncrawl where every fourth monster substitutes Hide and Move Silently for two of its normal skills. Note how encounters with monsters that have Hide and Move Silently as cross-class skills don't change very much.
That's why I said that a skill being "mandatory" or not is very DM-specific; some DMs love having ambushes, and some don't.
Look at it this way: a 3.5 DM could give every monster an Elixir of Hiding just as easily as a Pathfinder DM could give every monster maxed-out ranks in Stealth. It would have the same effect -- making monsters into super-hiders. Or likewise, a 3.5 DM could give them all a level in rogue and max out Hide. If he did, that would make Spot a "skill tax" in 3.5, as well, but that would be the DM's choice, not the system's choice.
|
Look at it this way: a 3.5 DM could give every monster an Elixir of Hiding just as easily as a Pathfinder DM could give every monster maxed-out ranks in Stealth. If he did, that would make Spot a "skill tax" in 3.5, as well.
Well, unless they have a crafter in their ranks, handing out Elixirs of Hiding would constitute the DM manipulating the (supposedly) random treasure tables,
Whereas having every creature spend skill points on Stealth is justifiable as being both legally possible, and totally in character, for any creature who realistically values its life.
Well, unless they have a crafter in their ranks, handing out Elixirs of Hiding would constitute the DM manipulating the (supposedly) random treasure tables,
*groan*
Let me put it another way. In a core 3.5 game, the DM could have lots of enemies with high spell resistance; then you might argue that [Greater] Spell Penetration is a "feat tax". But in that case your DM would be at least partly to blame; it wouldn't just be an artifact of the rules.
Every skill has to be demonstrably equal, whatever that means, when you are discussing something like skills? And if they aren't equal, then we have to get rid of even more skills to fix the problem?
Why don't we find ways to make the other skills more useful instead?
Velderan is spot on. Not all skills will be "exactly equal," that's a given. But if, in your mind, it makes perfect logical and game-design sense that (a) Listen + Search + Spot + Taste + Smell + Touch = Perception ... but that (b) Perform (piano), Perform (singing), Perform (flute), Perform (drums), Craft (composer), Profession (roadie) are all SEPARATE skills, each of which costs the same as Perception ... then there's really no basis for conversation here. Let's just say I personally find that situation to be absurd.
3.5 had very specific sub-skills, all sold separately. In the Beta, we've taken some of the favorite 3.5 sub-skills and combined them into super-skills, but left the rest alone, which seems odd -- either combine a lot of things, so that there are a number of super-skills, or else don't combine things, and keep the sub-skills. Or, there are two other possibilities as well, which need to be considered also:
1. Make sub-skills cost less than super-skills (clunky and annoying; not my preferred solution); or
2. "Beef up" the sub-skills so that they have more uses.
Personally, my preferred solution would be a mixture of selective additional combining and "beefing" as needed. I know that many people (e.g., DeadDM) would prefer a return to the individual 3.5 sub-skills. I'm guessing that, from your post, you prefer the Beta system, in which some skills are totally awesome and can do lots of things, and other skills are so intensely specific and so mechanically useless that players don't even consider them (which really, to my mind, sort of makes them not skills anymore -- kind of an "alternative 1" situation by default). I wonder if anyone else has another solution that's more palatable to all?
Usefulness of skills is highly campaign-specific and DM-specific. For example, somebody suggested that knowledge skills are not useful and need consolidation. To that I would say "no way"! Knowledge skills tend to be already among the most useful in my campaigns and consolidating them would make them outright uber-skills.
As to Perception, I do think it might be more useful than most skills and thus more powerful. I would probably favor going back partially to he old system and institute "Sight" and "Hearing" skills, which would subsume former Spot, Search and Listen according to the sense used. The disadvantage, of course, would be the loss of the other senses, such as smell, touch and taste (and having those as separate skills would be over-the-top, though it could be applicable to some monsters perhaps). Having those is kind of neat, though in practice they are rarely used, so it might not cause many in-game problems.
That said, if I am not going to be upset if Perception stays as one skill either. It is somewhat more powerful than most skills, but not so much so that all characters want to take it in my campaigns. What others are describing (all characters maxing out perception) has simply not happened in my campaigns.
When we first started playtesting, everyone did max out perception. But it didn't take long for me to catch on that only one of us actually needed to make the check.
So I stopped trying to keep up.
And I've been just fine.
I've had too many characters killed by a surprise-round-partial-charge/first-round-full-attack combo to agree with that conclusion. If even one encounter per character level is an ambush, the ability to act during a surprise round is a significant advantage.
But how was this different in 3.5? Doesn't the fact that your DM did this to you in 3.5 mean that a Spot/Listen "skill tax" existed there, too? To me it sounds like you're complaining about a "tax rebate", not a "tax increase" (since Perception is cheaper than 3.5 Spot/Listen)!
I certainly agree that it's reasonable to ask why certain skills give you so much of a "rebate" compared to others (Perception, Acrobatics versus Perform [percussion] or Knowledge [nobility]), though.
then a high-level wizard who's almost certainly survived several dozen monster attacks should certainly be able to better anticipate and react to monster attacks.
They are. That's why they have more hit points.
I certainly agree that it's reasonable to ask why certain skills give you so much of a "rebate" compared to others (Perception, Acrobatics versus Perform [percussion] or Knowledge [nobility]), though.
Yeah, that's my main issue here. It offends my sense of thrift to spend 2 skill points/level on Perform (sing) and Craft (songwriting), when the same 2 skill points/level will give me Listen, Spot, Search, Taste, Touch, Smell, Balance, Tumble, AND Jump.
Some skills are, and have always been, alot more "fluff" than "crunch".
Most craft and profession skills, for example, are in alot of ways Very non-adventure type and really have most use in the "well i did something before i got my sword and board" type skills.
The problem isn't really with the skills. The problem (and one that isn't going to be fixed) is with Paizo's treatment of skill consolidation and the redo of skill points.
The problem isn't really that everyone is taking perception- its that you get so *few* skill points at first level that each and every one has to be used in a skill you are going to USE. Because dropping 1 skill point into something is actually *giving up the equivalent of 4 points*.
What do I mean?
I mean that in 3.5 you could take a point in Craft: Underwater Basket Weaving or Profession: Rancher- or whatever- and it really didn't deteriment your character much. The wizard could have perform: lute (1-2 ranks) or the fighter could drop a rank into Cooking, thus adding a little spice to the character without terribly screwing up the fellows skills. You do Not have that option anymore.
Now you would Think with skill consolidation it would more than even out since the fighter who takes 1 rank in Cooking actually *gets* 1 rank (rather than half a rank). In reality however, taking 1 rank in cooking *costs him the 2 free points*(two because 1 rank in the old system cost 2 actual points). He doesn't get them back. They are GONE.
What it means is that the characters no longer have the points to drop a dot or two on relatively minor but flavorful skills. Instead, they take the very few points they get, and they absolutely maximize them.
Combine that with skill consolidation and you end up with every physical person having perception and acrobatics, and every mental person taking perception and their relevant magic skills.
It's a function of *both* the changes Paizo has done to the skills. The Old skills themselves are as they always have been- and bunching them up or whatever really isn't going to change the fact that PC's no longer have the points to spend on non-essential skills.
The solution that will never happen? Reverse the skill point change so that PC's once again get 4 times as many as they do now. Reverse the consolidation so that when they do have more skill points, they have enough skills to put them in, while still having enough points to drop a dot or two into the freebies. Again- that isn't going to happen. But it really is the only solution. The only other thing you can do is mandate that some folks aren't allowed to take (insert skill here) while allowing others to do so.
It is an unfortunate, natural evolution given what they have done to the skills.
-S
Because they are opposed rolls, Perception and Stealth create an arms race. In 3.5, there were two different levels of arms race.
This is an assumption on your part. I don't currently see a lot of creatures with ranks in Move Silent or Hide and I don't foresee the Paizo folks adding a bunch of them in some sort of arms race. In order for there to be some sort of 'arms race' you need 2 sides ramping up... and I just don't see that.
Selgard, your argument doesn't apply after first level.
Dennis, DMs can rearrange Monster skill points, but I think the problem Meepo is trying to discuss is that the dynamics of class vs. cross-class skills has really changed with Pathfinder.
At first level (assuming you maxed out the skills you actually use in 3.5) the systems are more or less the same. At 20th level, though, they're really different. A 20th level 3.5 cleric who maxed out their Move Silently still only has 11 ranks. He can certainly out-sneak a level 1 rogue, and is exactly as skilled as a 7th level rogue, but there's no way he can keep up with even the clumsiest 20th level rogue (who has 23 ranks). That same cleric, in Pathfinder, has 20 ranks (in Stealth, now), and can keep up with a 17th level rogue. If he's a cleric19/Rogue1, he can sneak just as well as the 20th level rogue.
That's a heck of a switch. The +3 class skill bonus doesn't scale properly with level the way the old cross-class system did.
That's probably ok for PCs, because if a Wizard has been studying from their Rogue companion for 19 levels they could probably learn to sneak just fine, and that Cleric might be of a trickster god. But if gets weird when Ancient Red Wyrms can sneak up on you too easy.
I think I'd say calling it a tax is not on target.
Rather than being a drain or a cost, the problem is that is to much of a bargain to pass up.