I am a Noob when it comes to being a DM and barely more than a Noob when it comes to being a PC.  Most of the games my table plays are mystery style. I have a work in progress for anyone who cares posted here Howell of the Moon - Work in Progress. Just so i understand the concept correct lets say a PC wants to investigate something .... here is example from my Howell at the Moon campaign. 

Examine a Corpse

• Default
  • Three parallel claw marks slice deep into the Gnome’s flesh.
• Intelligence Check (Investigation) > #
  • Something doesn’t look right … the claw marks look like they were a little too sharp.

So as the DM i will set the DC # (as represented by the #) and the player then has to roll a d20 to pass the check. So in this case the PC would Roll a D20 and then add/subtract their Intelligence modifier and then add their investigation modifier. Is this process correct? 

overall i guess i'm not very good at setting DC values. In a perfect world it would be unlikely but possible that a player without the Investigation skill would pass the DC and get the extra information. It would also be possible but unlikely that a player who has the Investigation skill would miss the roll and not gain access to the extra information. 

Im not sure if i can use the standard formula to make the system consistently work like this. 

I have considered modifying the formula to weigh the skill points higher. For example, 

Success = DC < D20 + Attribute Modifier + (2 x Skill Modifier)

What are peoples thoughts? 

The offical DM screens from from WotC have a little table that divies it by 5's (DC 5 is easy, DC 30 is near impossible). Normally, though, I set most of my DCs at 15 and then adjust based on what makes sense for my story and (when I feel mean) what my player's average profiency bonuses are. Doing the ladder really makes the players have to play to their strengths and makes sure someone is gonna get snared eventually.

I'm familiar with the chart ... what but do to the style of games that my table tends to enjoy we tend to use the system as more of an investigation buffer. So for instance if a character looks at a book shelf ... someone who has the Arcana Skill will notice particular titles ... and characters who have the History skill will notice other titles ... but it is possible that someone may have come across a book by accident ... 

a lot of our games are played more like murder mysteries than a traditional dungeons and dragons game containing more problem solving and puzzles then combat. I've been debating about just making it a conditional idea .... so when they investigate the bookshelf if they have the skill they automatically succeed the problem is that i want to be have an element of random luck that allows a bonus level investigation for example. 

• Player looks at the knife 
  • Wisdom Check < 10
    • Someone carelessly left a knife sitting out .... you have cut yourself with it ... take 1d4 damage ... the knife is now covered in your blood.  
  • Wisdom Check > 10 
    • This knife seems out of place as if it were thrown here hastily.
  • Wisdom Check (Medicine) > 20
    • This knife has residue of blood on it.
  • Wisdom Check (Medicine) > 25
    • This is not human blood.

In order to have these stacked layers i still have to do a D20 roll style. But I really want to reward the player for having the Medicine skill because it represents something particular they were looking for. Someone who has medical training would be more likely to notice blood stains then someone who was a potter .... or at least recognize the stains as blood. 

At the same time in the unlikely event they fail the the initial roll (in this case a Wisdom check of less than 10) they will end up with an accident that usually destroys the piece of evidence so that no other players can examine it. 

Doe this concept make sense? is there a better way to do it? 

I struggled with setting DCs as well.

Here's the system I'm using:

1. Does it need a check at all? Yes? Base of 5
2. How many of of the following would a PC need to perform this task: Luck, Innate Talent, Training ( sharpens up innate talent ), Practice ( hones training)? Add 5 for each that's required.
3. Is there opposing Magic here? Yes? +5
4. Is there aiding Magic here? Yes? -5
5. Are there minor, non-magical, conditions aiding or hindering? -2,-1,+1, or +2, depending on the influence.
6. Is there some sort of extraordinary influence? +/- 5

#2 is the tricky bit. Essentially I ask "Could an ordinary person do this effortlessly? No?  With just a bit of Luck? No? How about Innate Talent?  ... unlikely ... Innate Talent, Training? .... ". I factor in Luck first, and if it's not sufficient, take it back out, and add it again at the end.

Want to convince an Elite Guard to let you past the gates? Base + Talent ( charismatic ) + Training ( in the Con ) +  Practice ( done this before ) - so ... 20. You know him? Really well? OK - 18. The keep is on alert? Sorry - 23. 

Roll :)

I am afraid you are trying to use d&d (with its 1d20 madness) for recreating linear gameplay like in computer adventure point n click games. You should take a look at some other game mechanics like FATE

P.S.

Do not get me wrong i love d&d but it is not for everything

P.S 2

Why don't you just use rolls with advantage and disadvantage ?

in this example it might not be very D&D like but the feel overall is still very D&D. The problem we were facing is that the feeling at the table is that skills dont really matter that much ... and that a lucky roll mattered more then a +2 skill modifier. The idea of saying if you posses the skill it adds the advantage (+5 i think) and if you dont its a disadvantage (- 5 i think) ... is so simple that it never even crossed my mind. thinking about it  .... its simple and easy ... and would show my players how important skills are ... that having a high dex does not automatically make you an acrobat ... there is a skill for that ... 

Thank you Bird Copp ... i was clearly overthinking this DC thing. 

Also consider that the difference in proficiency bonuses between high and low level characters gets greater and greater. As they gain levels, the bonuses for their given skill set climbs and climbs while their other skills are still sitting at +1's and +2's. It feels minor with lesser levels, but the buffs get better with time.

Quote from Cporter5 >>

in this example it might not be very D&D like but the feel overall is still very D&D. The problem we were facing is that the feeling at the table is that skills dont really matter that much ... and that a lucky roll mattered more then a +2 skill modifier. The idea of saying if you posses the skill it adds the advantage (+5 i think) and if you dont its a disadvantage (- 5 i think) ... is so simple that it never even crossed my mind. thinking about it  .... its simple and easy ... and would show my players how important skills are ... that having a high dex does not automatically make you an acrobat ... there is a skill for that ... 

Thank you Bird Copp ... i was clearly overthinking this DC thing. 

I scale DC based on character proficiency and aptitude. The base DC is set for someone who's capable in the required skill.

With your knife example that would be Medicine and I'd set it at 15 (to notice blood).

Someone who's not skilled in Medicine but can justify or note why they know about the body, blood, etc. (a soldier who's taught to bind wounds perhaps) may get an DC 18 WIS check.

Someone who's not skilled in Medicine and has no character based study of such things at all may get a DC 20 WIS check.

Someone who's not only unskilled, but is focused in a completely different area of life may get a DC 25+ check as it's just about impossible for them to notice.

From there, I do "degrees of success" such as your example indicated. A skilled person who passes the Medicine check would learn more based on how much higher they go ("This is not human blood. It's also pretty fresh. This knife was used to cut today."). 

I take this focus to reward the choices characters make in their skills and progression. 

When two trained characters try a skill I grant advantage. If one skilled character is aided by an unskilled character who has some justified understanding of the concept (like my solider example above), I have them make a skill check first, and grant +1 for every 5 points on their roll (rounded down). A completely unskilled character can try to help granting +1 for every 10 points on their roll.

That helps encourage characters working together and let's any one person contribute at least a +1 without creating a runaway scene.

Also, if the information to be learned is trivial or critical, the DC is always 0. I just let a skilled person know what's up. That prevents a need for a check from gating game advancement (failed rolls that stop the entire game are bad). 

At the end of the day all you need as a DM is consistency in how you choose to set DC's. If it's consistent the players can anticipate what's going to happen and not be shocked or surprised by the challenges you set. Inconsistency is the thing to avoid. 

Degrees of success and failure is my favorite way to run skill checks. I really like borrowing a concept from Dungeon World, where they have total failures, partial successes (or success with a cost), and total success. I tend to add in degrees of total success as well, like if they roll particularly well they notice more than they would have otherwise.

In terms of setting DCs, I usually just wing it. I don't put a lot of thought into it to be honest, and roughly use the chart as a guideline.

I do think that there are times where the system doesn't reward proficiencies enough, so I'll often call for someone with a proficiency in the given skill to be the first to try it, or allow them to roll with advantage depending on the scenario. 

That's a cool idea; how do you scale it? 2 over or so gets extra?

Thanks! To be completely honest, I don't really have a hard-and-fast system and it depends on the check too--for instance, rolling particularly well on a DC 15 Investigation check (say rolling 18 or 20) could result in an extra insight beyond just finding a clue or whatever (e.g. "you discover a series of footprints leading in the opposite direction; from the size of the print and the tread of the boot, you can be reasonably certain the tracks were left by a human--and a large one at that" instead of maybe just finding the foot prints, doing another check to determine the species, and so on) but rolling really well on a lock-picking check doesn't necessarily grant you anything extra--you either pick the lock or you don't. Then again, maybe you do it especially quickly, or quietly, something like that. As I said, I don't have any hard-and-fast rules; I just kind of go with my gut. I'll use this to make nat 20s count for something on skill checks too, since I don't play where you can crit on skill checks. Maybe there's a DC 25 lock; the rogue rolls a natural 20 + 4 -- doesn't quite make it, but thanks to the nat 20, they notice something behind/under the chest, something like that. Or they meet the DC by rolling a natural 20, so they accomplish what they set out to do and gain an extra clue. Not the best examples in the world but hopefully you get the idea.

Where I really have a lot of fun though is really with the partial successes. If you fail a roll by a very slim margin (say 2 or so) I may offer you the option to succeed at a cost. So maybe you're able to pick that lock, but you realize you can't do it without tripping an alarm, possibly alerting guards elsewhere in the dungeon--are the contents of the chest worth it? Or maybe you can manage to push the villager out of the way of the trap they accidentally sprung, but only by exposing yourself to the full brunt of it--how much do you want to save that villager? I find that stuff adds so much to the drama and tends to be even more fun for the players than getting extra stuff for rolling especially well (though that is fun too, of course!).