D&D General What's wrong with Perception?

[Maxperson]

Perception I agree with, but Athletics? Maybe if it replaced Strength. Otherwise the last thing Strength needs is more things taken away from it!

You seem to have attributed that quote to me, rather than the person who wrote it.

 

[glass]

You seem to have attributed that quote to me, rather than the person who wrote it.

Well, ENworld did. I just clicked multiquote and then added my replies. Anyway, fixed now.

_
glass.

 

[amethal]

A good example of overuse is listening at the door or the like - if you listen at a door and some orcs are on the other side, having a conversation, you should hear them. No roll. No perception check. Your ears function. And RAW/RAI I think that's fine, but a lot of DMs, and a lot of adventures (including WotC ones) don't do that - they just have endless Perception DCs and Investigate DCs, and want you to roll them (or passively pass them).

I appreciate it's just an example, but in my case I'd say they automatically hear humanoids talking in rough voices from the other side of the door, but a decent Perception check would enable them to identify the language as Orcish, allow them to understand what was being said (if they speak Orcish and the conversation was interesting, which it probably isn't) and, with a good roll they'd identify that there are at least 4 creatures taking part in the conversation.

I suppose you could say straight away "You can hear voices" and wait for someone to ask for more details before you call for a Perception check but since, in my games, someone always does I find it saves time to ask for the check anyway.

(Also, the request to roll the dice encourages my players to stop chatting amongst themselves and actually pay attention to the ostensible reason we have gathered together, but YMMV.)

 

[Maxperson]

For example, DC 20 is 5%, so lets say your 1st level party has the following modifiers for a Wisdom (Perception) check: +3, +0, +1, +4

That means you need to roll a 17 or a 20 or a 19 or a 16. The odds of them all failing is just over 50%.

In my experience, they aren't going to be rolling individually. The +0 and +1 will be helping the +3 and +4, resulting in 4 rolls still, but 2 at +3 and 2 and +4 due to advantage. What are the odds of failure then?

 

[Ruin Explorer]

I appreciate it's just an example, but in my case I'd say they automatically hear humanoids talking in rough voices from the other side of the door, but a decent Perception check would enable them to identify the language as Orcish, allow them to understand what was being said (if they speak Orcish and the conversation was interesting, which it probably isn't) and, with a good roll they'd identify that there are at least 4 creatures taking part in the conversation.

I think this kind of furthers my point though.

Either your ears work or they don't, IRL (at least in my pretty long lived experience - and I have serious ADHD which means I have worse sensory issues than most people), in a static situation like listening at a door. They don't vary in success in a way that RNG reflects. If I am familiar with orcish, it shouldn't be a perception check to identify it, if I can hear the words, I know what language it is, period. If I can't hear the words, or I'm not familiar with orcish, I don't. External factors (like if people are making noise on my side of the door) may impact it, but it shouldn't really be a roll.

Now let's be clear, part of the problem is the skill design itself, that says: "It measures your general awareness of your surroundings and the keenness of your senses.". That's bad design, frankly. That incredibly overbroad description which encompasses both like, how good hearing and eyesight is, how aware you are generally, and a million other factors is why so many DMs (and adventure designers) call for constant Perception checks. The problem isn't with the DMs or adventure writers - it's with the design of the skill. I'm not saying you're doing it wrong (though I would rephrase the "identify it as orcish" to "hear the words" and then go from the character's backstory - if they have a background where they'll never have heard orcish then no amount of high rolls will let them identify it, though they might be able to repeat the phonemes and have another PC identify it), just that the skill is badly designed.

Even with that, it doesn't seem like this should be rolled - it seems like it should be a passive Perception comparison. I mean, what even is "rolling low on Perception"? Getting distracted? That's simply completely implausible in a lot of scenarios. Equally "rolling high", what exactly is that representing? We can come up with examples but a lot of them are implausible in a lot of scenarios, maybe all of them are in some scenarios - or they just reflect how over-broadly designed it is.

It's also worth noting Take 10 and Take 20 would eliminate some of the weirder issues with Perception (and a bunch of other 5E skills), but 5E doesn't allow for those, and I don't think even suggests them as optional rules (though correct me if I'm wrong).

 

[amethal]

I think this kind of furthers my point though.

Either your ears work or they don't, IRL (at least in my pretty long lived experience - and I have serious ADHD which means I have worse sensory issues than most people), in a static situation like listening at a door. They don't vary in success in a way that RNG reflects. If I am familiar with orcish, it shouldn't be a perception check to identify it, if I can hear the words, I know what language it is, period. If I can't hear the words, or I'm not familiar with orcish, I don't. External factors (like if people are making noise on my side of the door) may impact it, but it shouldn't really be a roll.

Thanks, I finally understand your point - sorry it has taken me so long!

However, given that Perception is a skill in the game, it's understandable that people are going to want to roll for it.

Having purely passive perception could be seen as less fun - either you automatically find it, or you automatically miss it. and since the DM setting the difficulty also knows what the passive scores are, the DM is basically deciding in advance if the party succeeds or fails.

There are plenty of game mechanics that don't make any sense from a real world perspective - e.g. the Fortitude saves in D&D 3.x / Pathfinder to "shrug off" the consequences of drinking poison - but are needed in the game ("you unknowingly drunk poisoned wine, and now you are dead, no save" being very much not-fun.)

 

[billd91]

I appreciate it's just an example, but in my case I'd say they automatically hear humanoids talking in rough voices from the other side of the door, but a decent Perception check would enable them to identify the language as Orcish, allow them to understand what was being said (if they speak Orcish and the conversation was interesting, which it probably isn't) and, with a good roll they'd identify that there are at least 4 creatures taking part in the conversation.

I suppose you could say straight away "You can hear voices" and wait for someone to ask for more details before you call for a Perception check but since, in my games, someone always does I find it saves time to ask for the check anyway.

(Also, the request to roll the dice encourages my players to stop chatting amongst themselves and actually pay attention to the ostensible reason we have gathered together, but YMMV.)

There's nothing at all wrong with considering a base-level DC imparting some information with more information being accessible with higher DCs. For example: Hearing voices at conversation level through a door may be pretty easy. Add 2 to the DC to hear well enough to identify the language as orcish. Add 2 more to pick out 4 distinct voices in the room.
A lot of people just look at attribute/skill checks as being binary and for any particular DC that may be true. But they don't have to be just looked at with a single possible result when you can easily treat it as a series of results that can define multiple degrees of success.

 

[James Gasik]

You can treat it that way, yes, but by default, that's not how the game presents it's checks. If degrees of success was a default, I thin it would be immensely helpful.

 

[DND_Reborn]

In my experience, they aren't going to be rolling individually. The +0 and +1 will be helping the +3 and +4, resulting in 4 rolls still, but 2 at +3 and 2 and +4 due to advantage. What are the odds of failure then?

Even less.

About 31% if my down-and-dirty math is correct. I'll verify after supper when I have more time.

Double-checked: 36% actually.

 

[Ruin Explorer]

You can treat it that way, yes, but by default, that's not how the game presents it's checks. If degrees of success was a default, I thin it would be immensely helpful.

What's really sad is the DMG has a whole pile of vague thoughts about how to run checks as less binary, but they're all presented as "at the DM's whim they might", and all in a few paragraphs of poorly-organised text deep in the DMG. They should have been at least conceptually laid out in the PHB.