Wulfhelm II.'s page

And for that matter, why are weapon proficiencies and saving throws not also rolled into skills? Or a general "proficiency" category, if that fits better?

At first glance, I thought "Hey, Perception has been separated out as its own thing. Good, so it will grow automatically like BAB and saves as it should, because it is so important to every character."
Except that goes for *everything* now. The +1/level system has obviated the reason for which Perception as a mere skill among others was problematic in the first place. Classic "two fixes applied to one problem" situation, there.

Generally, if you are determined to go with the +1/level mechanic, there are ways in which you could and should simplify things:

a.) The mechanics description should be changed to begin with the idea that a check is performed by 1d20 + character level, + modifiers for ability and proficiency (which are both fixed values independent of level.) That is clearer and for lack of a better term more honest than sneaking in the fact that your level is added to any check in the description of proficiency bonuses.

Alternatively, you could simply increase ability modifiers, which also figure into every check, by +1 every level. Whether you do this by increasing all ability *values* by 2 every level and keeping them meaningful or by ditching the 3-18 paradigm altogether and have ability bonuses be the only remaining stat is a matter of preference.
In this case, the description of a check would be: 1d20, +your ability modifier, +your proficiency modifier

b.) There is little need to distinguish between skills, weapon proficiencies, saving throws and perception if they all work according to the same "level check" paradigm. Even if you want to keep saving throws separate for nostalgia reasons, rolling weapons, armor, perception and skills back into the general category of "proficiency" would simplify and clarify things. I cannot see any rationale to keep them as separate things, but if there is one, I'd like to hear it.

Regards

Wulfhelm

Two remarks in advance:

1.) I have only read the system, and not played it so far. My observations are thus based on a theoretical approach.
2.) I have no objections to the consolidation of the skill list, nor to the separation of Perception as its own thing. I applaud both of these decisions.

With that in mind: I cannot discern what the point of the new skill mechanics is or what advantage the new system offers over the old one.
To explain, here are the main points of the new underlying system as I see them:

a.) Level of complexity
In PF1e, your overall expertise in a skill (measured as a chance of success in most cases) was influenced by the following factors:
- Your corresponding stat.
- The number of ranks you put into the skill (limited by level.)
- A bonus for a class skill (and a further bonus for skill focus.)
- (depending on what you wanted to do: If you were trained in the skill at all.)

In PF2e, we have:
- Your corresponding stat.
- Your level.
- A bonus for your proficiency rank. (limited by level and by whether it is a signature skill.)
- (depending on what you want to do: If you are trained in the skill at all.)
- (also depending on what you want to do: Possession of relevant skill feats.)

So, right off the bat I do not see any simplification. You might say that no longer getting to distribute skill ranks streamlines choices, but this is in fact more than made up for by having to make new choices about levels of training and skill feats.

b.) Relative importance of factors
Since all skills improve by the character's level, it is by far the most important factor in your overall proficiency. The effects of this are:
- Flattening of proficiencies across skills as you progress in level.
- Higher level characters outperforming lower level characters even in the latters' given professions. For example, a level 1, Strength 18 Barbarian who is trained in Athletics is worse at any typical feat of strength (grapping, breaking down doors etc.) than a level 8, Strengh 10 untrained Wizard.
It should be readily apparent that this is a step backward as far as verisimilitude is concerned.
(On a personal level I should mention that this is the exact same way D&D 4e changed from 3.x and this was one major reason for me to leave 4e behind.)

c.) Relevance and descriptive accuracy
This is closely related to the preceding point. If a Strength 10 character demonstrably performs better at all feats of strength than a Strength 18 character, then the numerical value for "Strength" has little worth as a descriptor.
Consider this: If we took the proficiency rank as the actual level of training in a given area of expertise and wanted to replace the level bonus by attribute increases, it would mean that every attribute would increase by 2 points every level. A level 5 character who started out with Strength 12 would have Strength 20. At level 10, Strength 30. And at level 20, Strength 50.
For purposes of checks, this is the same as the flat level bonus.

A related problem is the negligible numerical impact of proficiency ranks both between the different ranks and compared to character level. Not only is the difference between a "trained" user and a "master" a mere 10 percentage points (and thus irrelevant in ~90% of all practical cases), it is also much less relevant, again, than the character's level.
The effect is that the mechanical description "She is a master diplomat" simply does not correspond to the connotations such a description would indicate to a layman. It merely means she has a small bonus which may be easily offset by her overall level of experience or her basic attributes.

To summarize: In my opinion, for reasons I have just explained, the new skill system does not improve a.) ease of use (beyond the consolidation of the skill list), b.) verisimilitude or c.) accuracy of description. If any of these were design goals, I would argue that the new system does not meet them. If they were not, or if there were other design goals, I would like to know what they are.

Regards,

Wulfhelm