Using 3d6 over 1d20, or Vice Versa

>>49683718
Adjust crit number, as 3 and 18 are much rarer to get than 1 and 20s.

It's much easier to succeed, there is a lot less randomness involved.

>>49683718
Do it like GURPS.
3 is always a critical success, 17 always a failure, 18 always a critical failure.
But the better (or worse) your skill, the more the range moves. So if you're really bad at something, 3 is crit success, but 15 to 18 is crit failure. On the other hand, if you get really good, 3 to 6 can be a crit success, but 17 still a failure and 18 still a critical failure.

>>49683816
>>49683859

Do I need to adjust any DCs from the 1d20, or can I keep them exactly the same? On a 3d6, you don't get a natural roll of 19 or 20 after all.

File: 3d6 v d20 with double-size steps (+1 on 3d6 = +2 on d20).jpg (60KB, 890x603px) Image search: [Google]

60KB, 890x603px

>>49683718
>If I decide to switch to using a 3d6 in place of a 1d20 dice, then what other mechanical adjustments should I make to make the conversion smooth and successful?
Going to 3d6 roughly doubles the effect of every step away from a 50/50 chance (10 or under with a roll-under system, 11 or over with a roll-over system). So if you're doing a roll-under system, a target number of 10 is the same in both systems, while an 11 in a 3d6 system (+1 above 10) is equivalent to a 12 (+2 above 10) in the d20 system. So a +4 bonus on a d20 is equivalent to a +2 bonus on 3d6.

>>49684067
As per >>49684469, look at the actual number people need to roll on the dice (after bonuses and modifiers are applied) to succeed. If the typical person has to roll a raw 16 or over on a d20, that's about like having to roll a 13 or over on 3d6. So generally speaking, if you want to keep probabilities roughly the same, you want to moderate everything, moving DCs halfway between what they are in the book and whatever would require the average person who would be attempting them to roll a natural 10 (actually, it's what would require them to roll a natural 11, but 10 is close enough and it's easier to work with).

But if you just convert the probabilities to keep them the same, what's the point in using 3d6?

>>49684537
Less variance. 3d6 is a bell curve, 1d20 is a straight line. While it doesn't make a difference for the average person, extraordinary individuals will succeed at challenges much more reliably.

>>49684469
>>49684537

So it doesn't sound like it would be as easy a conversion as I thought.

The 3d6 makes a nice bell curve, meaning that its more likely you'll get either a 10 or 11 specifically. It feels like I will need to do more to get my modifiers when it comes to the 3d6 so that I can get above average results.

A 1d20 on the other hand is a very swingy dice, and the point of the modifiers is to make it less likely that the 1d20's randomness will shaft you. The 1d20 also has the advantage of being simple, as you only need to see the number rolled, and then you can quickly add it.

I was hoping to include a variant rule in my homebrew game to use the 3d6 instead of the 1d20. I'm making the 1d20 the main rule because it is less addition than the 3d6 (And I'm familiar with the 1d20 as well).

>>49685101
You could always use a 2d10. It compromises between swinginess and bell curve while not requiring as much adjustment to DC's and requiring less adding.

>>49687032

So basically make the base DC 11 instead of 10 if I adopt this?

>>49685101
>I'm making the 1d20 the main rule because it is less addition than the 3d6

This seems like a terrible reason. 3d6 is adding only three single digit numbers. If you get into double digit numbers it's either 10, 11 or 12 + a single digit number and that's trivial. The hardest is probably 4 to 9 + a single digit of at least 6 to 1 respectively since you have to actually carry a digit. It's something you'll do over and over again and it becomes second nature with either pip dice or numeral dice.

Being unable to do this is either a sign of severe and deplorable laziness, probably encouraged by over reliance of calculators inherent in current education, or mental retardation or dyscalculia or acalculia or some other sort of actual neurological impairment that inhibits number skills. If my seven year olds couldn't do this they would be flagged, lamentable but true, as in need of special attention.

If you think this is hard or slow you severely need to do it and to practise basic numeracy skills in general. If you were rolling a huge pool of dice and adding them many many times I could understand that the delay would start to take time, but only three dice should take no more time than saying aloud the numbers on the dice themselves as you read them.

>>49684913
Is it "much more," though? If someone truly shouldn't fail at something, it's actually easier to roll a 3 with 3d6 than two 1s with a d20 with advantage. You can convert the system, but you could also just lower DCs or be more active with how you rule skill checks as a DM, like giving people advantage if you don't think they should fail, or having them auto-succeed because it's something they've successfully done a million times already.

I say this, but I have been spending my work breaks charting out realistic rules for falling.

>>49683718
Use 4d6 - 4.

File: 1338816177968.jpg (19KB, 328x167px) Image search: [Google]

19KB, 328x167px

>>49685101
3d6 is really more similar to a d8 or d10 than a d20.

>>49689974

Patternwise, yes you are right. But in terms of the numbers themselves, the d20 and the 3d6 can each come up with similar numbers eventually. Although it will likely take more time for a 3d6 to get a natural 18 than for a 1d20 to get a 20.