Current master post-4.2.1

[Nick]

The new default points code is fairly simple, and not perfect. What it does is:

1. Buys 7 points of STR at a cost of 8
2. Buys 7 points of DEX, taking note of the last point where blows increased (except casters and classes with the LIKE_SPELLS flag - currently blackguards)
3. Sell back DEX to that breakpoint (same exceptions as 2)
4. Spend half remaining points on spell stat (except warriors)
5. Spend other half on CON
6. Spend any remaining points on DEX, then non-spell stat

The main deficiency is that all this happens before wielding the starting weapon, so it's unarmed blows that are optimized. I think I'm happy to live with this as a simple default for new players; experienced players can work it out for themselves.

[DavidMedley]

That all sounds very reasonable. But it doesn't happen.

Also, armed or unarmed the dex blows breakpoint is the same, so that's good.

[DavidMedley]

One other note: I think Blackguards should also hit the DEX blows breakpoint. They need the blows as much as anyone IMO. Splitting the remainder between INT and CON seems fair.

[Nick]

Originally Posted by DavidMedley

One other note: I think Blackguards should also hit the DEX blows breakpoint. They need the blows as much as anyone IMO. Splitting the remainder between INT and CON seems fair.

OK, wait, so why am I doing this then?

[DavidMedley]

The 4.2.1 and I don't know how many versions prior default buy is bad. It doesn't work like the code says it should. The latest version also doesn't work like the code says it should, but it's coming out a lot better IMO anyway.

The 4.2.1 Dwarf Paladin is a great example of a terrible default buy:
8 pts in STR to bring that up to 18/20 (which is arguably perfect)
12 pts in DEX to bring that up to 15 (this is a huge waste imo)

[DavidMedley]

Current nightly Dwarf Paladins puts 4 pts each in DEX, WIS, and CON, which is far superior. And now I see it's trying to hit the DEX breakpoint but is overshooting by 1*. So you're actually pretty close to a damn good system.

*EDIT: I was wrong

[DavidMedley]

No sorry I'm wrong looks like only an odd point after splitting between WIS and CON is going to DEX so you are hitting that breakpoint. Seems like a very good system. Certainly there is room to quibble, but it's such a huge improvement over the prior versions!

Please do exactly this for BGs.

[Nick]

Originally Posted by DavidMedley

No sorry I'm wrong looks like only an odd point after splitting between WIS and CON is going to DEX so you are hitting that breakpoint. Seems like a very good system. Certainly there is room to quibble, but it's such a huge improvement over the prior versions!

Please do exactly this for BGs.

That is easily doable. However, I thought there was some discussion (which I can no longer find) where we were thinking that blackguards actually wanted more INT for their spells (whirlwind, etc), and didn't care so much about blows because with big weapons they were not taking full advantage anyway.

But tell me I'm wrong and I'll just remove the LIKE_SPELLS flag again and everything will be good

[DavidMedley]

tl;dr yes remove the flag

12 points in Dex is wrong for BGs, Paladins, amd IMO everyone, but less wrong for Rangers, Rogues, and Warriors, and casters weren't suffering this fate. BGs are getting a lot more chatter than Paladins so they were more of the focus. And, yeah, BGs probably benefit a little more from INT early on than Paladins benefit from early WIS.

But the real issue was spending 12 on DEX whether it hit breakpoints or not. Putting some of that into CON and the spell stat is much better. Thanks!!

[Nick]

Update to blackguard birth points pushed to github build page; also includes a fix to flicker on latest macOS by backwardsEric.