Can't those failures done by experts just be waved off by environmental effects?

Like the Guardsman failing to kick open the door because it appears to be too thick or barred at the other side.

As for the Techmarine, maybe there are some environmental disturbances that's causing interference or something.

So they're good at what they do, but sometimes the world is against them.

That requires more on the fly thinking than you'll see from most GMs. We generally waive rolls in such cases where they should be a reasonable success. Kicking in a normal, metal (!) imperial door, though, is probably anything but easy...

So they're good at what they do, but sometimes the world is against them.

That's how I'd try to resolve it, too. But it still happens an awful lot. Maybe it's just my luck, though.

And you need to have a GM willing to work this sort of stuff in favour of your character, to "protect their professional image", so to speak. Not everyone does that. Perhaps this is something the rulebook could talk about...?

Mostly because people don't write it in the rulebook, so DMs don't think of doing it. That's something usually, and unfortunately, lacking, or if it's there, it's not emphasised enough or, the other extreme, emphasised too much. It may also do well to note that some tests are less a matter of "yes/no" and more a matter of time. Picking a lock, for example, is not impossible, with the right tools. A failure may simply mean you take a while. I'm not sure if this is in the rules text of DH2. I don't have a copy myself and I'm not interested in purchasing one given the book binding issues and that I own Only War already. However, it should be. It's definitely a helpful mention, especially for a DM who's not used to improvising.

Perhaps it just depends on the style of the GM in question.

Like if he/she loves to improvise on the spot, rolling those tests, even in case of experts might be oppertunities for him/her to add more content and feeling to the game.

It might be different for other GMs though who like to limit the amount of improvisation to a minimum, like to avoid getting the story and other aspects derailed somehow. In that case I'd just autopass certain tests like DeathByGrotz mentioned.

Mostly because people don't write it in the rulebook, so DMs don't think of doing it. That's something usually, and unfortunately, lacking, or if it's there, it's not emphasised enough or, the other extreme, emphasised too much. It may also do well to note that some tests are less a matter of "yes/no" and more a matter of time. Picking a lock, for example, is not impossible, with the right tools. A failure may simply mean you take a while. I'm not sure if this is in the rules text of DH2. I don't have a copy myself and I'm not interested in purchasing one given the book binding issues and that I own Only War already. However, it should be. It's definitely a helpful mention, especially for a DM who's not used to improvising.

This made me curious, so I looked it up. The book doesn't define failure for individual skills, but at the beginning of the chapter explains how to use skills. It says that skill checks follow the normal rules for tests, defined in chapter 1. In that section, it simply says that any result greater than the target number means the check is a failure.

So, RAW, if you fail at picking a lock, the lock remains locked. If your GM decides it means something else happens, this is not supported by the rules of the game . Which is a **** shame in my mind. Failure is codified to be boring.

The closest the book comes to making failure interesting is this section, buried in the GM chapter:

"The results of a test are simple to determine in some cases,

but there is often room for additional nuance based on degrees of

success or failure. Some tests have obvious possibilities for how

an overwhelmingly positive or negative result might play out—for

example, failing the repairs on a volatile plasma reactor could cause

a dangerous energy build-up and subsequent explosion."

The door remains locked until you try again (which AFAIK no rule prevents you from doing). There's no consequence for failure apart from 'doir stays locked' so all failure adds is time.

Sounds fun.

Depends on your definition of boring.

Come now, simply rerolling until you get the result you want is not a good way to go, because it makes the individual rolls meaningless. (In chargen it's not accepted either, is it? Either you want randomness and you don't reroll, or you just pick what you want.)

D&D as of 3.0 had a better mechanism for dealing with this, for heaven's sake! Take 10 or Take 20, where you could assume an average/best roll if you weren't under pressure and had enough time. Which allowed for special abilities that allowed you to Take 10 even when under pressure, allowing you to auto-succeed at stuff you were supposed to be good in and wasn't extra-challenging.

I wonder if this mechanism could be ported over, but DoS probably makes it a bit complicated to just allow taking 50 or taking 5...

Don't forget that if you don't know how to rangle the rules into a fun game and story, it's your fault for being a bad GM rather than the systems for not giving you any guidance. Good GMs are born of clamshells, too beautiful for this earth, their flowing red neckbeards covering all their unmentionables.

And I'll try to make my point as concise as possible. Chance of success in the game relies on the use of modifiers which must either be provided by the GM or memorized by players/GM, although these modifiers only exist mostly for combat. For everything else, it's up to the GM to make up enough things for players to get a decent chance of success, which isn't actually helping to tell a story. The system is expecting the GM to juggle all of these things, when it could instead give guidance or rules that make for a more consistent experience. In other words, the system is not designed at its core to be enjoyable, and instead relies on the GM to add that. In turn, the system has little guidance for the GM on how to do this job.

As far as swinginess of the d100 goes, it refers to the fact that every result on the die is equally likely. Given how the d100 system works, you have a huge amount of character effectiveness due to random chance rather than actual skill. A character who is trained at something is still looking random chance being the main determining factor, as he would be equally likely to completely screw up as he would be to succeed. Lots of professionals have very difficult jobs that they're also very naturally good at, like surgery. Let's say they have 45 intelligence and +30 to Medicae. That's a 1 in 4 chance of failure every time they do surgery. But then if we look at how difficult surgery is to an average person that's probably -20 difficulty. So now it's a 45% chance of failure. And maybe good tools bump that up. The end result is still a surgeon who would be fired for awful performance. Does this show more how the system breaks down when just run RAW?

Also, a quick note on granularity. With the new DoS mechanic, the 1s digit of a roll will only come into account 10% of the time (you only need to look at the ones digit if the 10s digit of the target number and roll are equal, which will happen 10% of the time. Does that granularity seem that important to anyone? Does anyone notice that point buy chaacters hae everything in multiples of 5s because players can easily grasp that that the 1s digit doesn't matter? As a rule, if your players will take the first opportunity they can to circumvent using a rule, it's probably not a very good rule.

Come now, simply rerolling until you get the result you want is not a good way to go, because it makes the individual rolls meaningless. icated to just allow taking 50 or taking 5...

When there's no time pressure and no consequence for failure (like picking a lock for example) indivudual rolls ARE meaningless. Why even bother rolling instead of just saying 'I do this' ? You will inevitably succeed if you keep at it. Does really 'take 20' need to be written explicitly in the book?

Given there's an allotted time connected to the difficulty of the task?

Yes. You need to know how much time is spent doing it slow and right.

Perhaps some things are left vague on purpose?

Well, it actually is written in the book (on page 96):

Many tasks can be attempted repeatedly with few consequences
until a character finally succeeds. A character using Inquiry to
research vital information might decide, for example, he will
keep examining data until he finally finds the vital clue necessary
to answer a crucial question. In such a situation the GM could
determine how long this might take based on a single roll, with the
character’s only cost the narrative time and effort involved, rather
than have the player make a series of rolls until he is successful.

Also, from the same page (related to the time it takes to do something):

Keep in mind
that these values serve as guidelines but are not always fixed. Game
Masters must alter the time required for a skill use based upon the
overall complexity of the task involved

Huh, whaddaya know, they have it in the book. Making it more visible would've been better (Gridash, I'm fine with stories being vague, rules have to be as explicit as possible) but they actually put it in.

So the game is not doing this bit wrong, which is good

I just meant that certain things are left open for interpretation on purpose so the GM can come up with something. Putting in rules for every explicit situation is just not a feasible thing to do.

The same thing happens with RL law and judges.

Does this show more how the system breaks down when just run RAW?

Depends on how you apply said RAW, especially when it comes to adding modifiers (for example I remember further bonuses being possible for Assisted Tests).

I would agree that / prefer if a lot of tests would be more about the quality of your success, though, rather than determining success or failure. For the surgery, for example, rolling just how much time it takes and thus how long the patient's recovery will take. A critical failure could still mean serious complications up to death! Much more reasonable, no?

But this has nothing to do with the type of dice we roll, just with how we use the basic mechanics. I'm frequently trying to push for minor adjustments in games as well, and not just in 40k. I'm sure most groups could think of something they'd do differently in any game, simply because of PoV's and personal preferences.

Does anyone notice that point buy chaacters hae everything in multiples of 5s because players can easily grasp that that the 1s digit doesn't matter?

Coincidentally, that is a secondary, minor reason for why I suggested characteristics advances in the form of +2s and +3s rather than +5s. It "normalises" people's caracteristics.

However ... now I wonder if it could be fun to make characteristics advances in the form of +1d5 ...

Probably too much potential for being bent over by the dice gods, tho. Imagine rolling a bunch of 1s as opposed to someone rolling a bunch of 5s!

Not around here, it doesn't. We make our laws as clear as possible, being a Civil Law country, and not Common Law. They still need interpretation by the judge, and the rulings of our highest courts are important documents, but we still strive for maximum clarity and least necessary interpretation. Cuts down on arbitrary rulings and difference in treatment. Of course, we still have things like maximum and minimum sentences for certain crimes, because not all cases are entirely equal, and having the human oversight of a judge is valuable, but that's not vague, that's a clearly bounded leeway.

Which I think is a good way to write any rules, really.

Does anyone notice that point buy chaacters hae everything in multiples of 5s because players can easily grasp that that the 1s digit doesn't matter?

Coincidentally, that is a secondary, minor reason for why I suggested characteristics advances in the form of +2s and +3s rather than +5s. It "normalises" people's caracteristics.

However ... now I wonder if it could be fun to make characteristics advances in the form of +1d5 ...

Probably too much potential for being bent over by the dice gods, tho. Imagine rolling a bunch of 1s as opposed to someone rolling a bunch of 5s!

Back in the pre-1.0Beta days (hell, it was pre-BC... good ol' times), we had this idea to replace different characteristic advancement costs with one cost but different characteristic improvement. For example, if you are good with Strength, then you will get 5+1D5 for one characteristic advancement, while another with bad Strength will get 1+1D5 - and you both get it for 250xp (first advancement).

Back in the pre-1.0Beta days (hell, it was pre-BC... good ol' times), we had this idea to replace different characteristic advancement costs with one cost but different characteristic improvement. For example, if you are good with Strength, then you will get 5+1D5 for one characteristic advancement, while another with bad Strength will get 1+1D5 - and you both get it for 250xp (first advancement).

Hmmh, like a "mirrored approach" to Aptitudes, where instead of having to pay more or less XP, yet still arriving at ultimately the same cap, you'd invest the same amount of experience but get a gap in the results...

I like that idea!

Back in the pre-1.0Beta days (hell, it was pre-BC... good ol' times), we had this idea to replace different characteristic advancement costs with one cost but different characteristic improvement. For example, if you are good with Strength, then you will get 5+1D5 for one characteristic advancement, while another with bad Strength will get 1+1D5 - and you both get it for 250xp (first advancement).

Hmmh, like a "mirrored approach" to Aptitudes, where instead of having to pay more or less XP, yet still arriving at ultimately the same cap, you'd invest the same amount of experience but get a gap in the results...

I like that idea!

It's only downside is that it is a pain in the *ss to balance out with Skill advancements .

Worse than the 'low' chance of success (-and I agree with an earlier comment that WH40KRP should have some kind of 'take 10' mechanic to mitigate some of these circumstances) is the boring nature of the 'all or nothing' results on Skill Tests. For example, my players frequently call upon the party's Tech Priest to 'slice' (as 'hacking' is called in the 40Kverse) into computer networks to find information. Rule As Written, this is either a single Tech Use or Security Test (depending on the circumstances)- which is boring even if he succeeds . I'm tinkering on a system to represent 'slicing' that would involve navagating a flow chart (like in the early edititons of Cyberpunk - I'm not sure if that's how they still do it, I don't own anything after 2nd Ed), with different types of tests (Security, Tech Use, Logic, certain Lore skills, plus Agility Tests to react fast enough) required to advance past different junctures.

I think that's what DH2 needed, more that noodling with success rates: some kind of mechanic that would make using Skills more comparable to physical combat, with all the decision-making complexities that involves.

Back in the pre-1.0Beta days (hell, it was pre-BC... good ol' times), we had this idea to replace different characteristic advancement costs with one cost but different characteristic improvement. For example, if you are good with Strength, then you will get 5+1D5 for one characteristic advancement, while another with bad Strength will get 1+1D5 - and you both get it for 250xp (first advancement).

Hmmh, like a "mirrored approach" to Aptitudes, where instead of having to pay more or less XP, yet still arriving at ultimately the same cap, you'd invest the same amount of experience but get a gap in the results... I like that idea!

Personally I think it's not a good idea as it encourages stereotyping. With the current system you can go against type if you so desire, it just costs more. With this system there's no point. You will never be as strong as any stereotypical strong 'race' like a Feral Worlder, as the maximum on 1+1d5 is 6 ( likely to be way less over multiple advancements) while the minimum for 5+1d5 is also 6 (likely to be way more over multiple advances).

It also forces randomness on people. Some of us don't like permanent character effects to be decided by dice and as such use point buy when possible. Having to still roll 1d5 for advancements kinda defeats the point of point buy.

Back in the pre-1.0Beta days (hell, it was pre-BC... good ol' times), we had this idea to replace different characteristic advancement costs with one cost but different characteristic improvement. For example, if you are good with Strength, then you will get 5+1D5 for one characteristic advancement, while another with bad Strength will get 1+1D5 - and you both get it for 250xp (first advancement).

Hmmh, like a "mirrored approach" to Aptitudes, where instead of having to pay more or less XP, yet still arriving at ultimately the same cap, you'd invest the same amount of experience but get a gap in the results... I like that idea!

Personally I think it's not a good idea as it encourages stereotyping. With the current system you can go against type if you so desire, it just costs more. With this system there's no point. You will never be as strong as any stereotypical strong 'race' like a Feral Worlder, as the maximum on 1+1d5 is 6 ( likely to be way less over multiple advancements) while the minimum for 5+1d5 is also 6 (likely to be way more over multiple advances).

It is worth noting that in the system, characteristic advancements are capped at a certain characteristic level rather than number of advancement levels. IIRC, 60 was the maximum characteristic for everyone, and you could take advancements until you reached this level.

Hacking as a mini-game is a bad idea, because it is very niche and time consuming. Condemning the whole party to boredom because their single hacker has to tunnel his way through five-bazillion different tests to open a door is generally not a very good gaming experience.