New Content - Implemented Changes in the Rat Quest

[Theraze]

The reason I mention it is, when I did actually make a system that did exactly what you said... followed the right side down, and repeated going down in stripes until it found the faucet... every attempt, it would check one square, but mark the second when the first combat attempt failed to complete.

 

[Rinn]

It's not a very efficient way to solve the quest, since the faucet is typically in one of the 4 tiles in the lower left corner.

 

[Veracity]

The reason I mention it is, when I did actually make a system that did exactly what you said... followed the right side down, and repeated going down in stripes until it found the faucet... every attempt, it would check one square, but mark the second when the first combat attempt failed to complete.

So, we can assume that the old Tavern quest completer would have failed for you - and only you - in precisely the same way. I see no reason to put in special handling for "fights failing to complete", since fights should never "fail to complete".

As I said, your "failure to complete fights" bug has nothing to do with the Tavern Quest solver. It's orthogonal and is addressed in a different bug report, which will be addressed if and when you find a way for me (for example) to reproduce it.

It's not a very efficient way to solve the quest, since the faucet is typically in one of the 4 tiles in the lower left corner.

Where is the spading for that?

If it's true, then my "spiral in" proposal seems as good as any.

 

[StDoodle]

As I mentioned on the previous page, the reason that appears to be so is likely because that's the method most people use.

If you're normally going to get the faucet on your 8th square visited, then if you're going left-then-down, you'll usually see the faucet in the bottom-left square.

But the KoLspading thread suggestions that it's simply a weighted chance on Nth square, with 8 being the avg. value of N.

 

[Theraze]

So, we can assume that the old Tavern quest completer would have failed for you - and only you - in precisely the same way. I see no reason to put in special handling for "fights failing to complete", since fights should never "fail to complete".

As I said, your "failure to complete fights" bug has nothing to do with the Tavern Quest solver. It's orthogonal and is addressed in a different bug report, which will be addressed if and when you find a way for me (for example) to reproduce it.

A problem for me, for StDoodle, for Winterbay, for alfagetti, for NardoLoopa...

Unfortunately, editing the CCS settings to try to test things fixes the issue. But as you say, it's an issue for that thread, not this. It just means I'm less tempted to come up with a solution that I know won't work in my own case, especially when it causes improper setting of the tavern layout to believe that squares are combat when they're still unvisited...

 

[Rinn]

The only evidence I have to support my claim is the 8 or so ascensions where going left then down makes the quest take between 5-8 turns. When I ran Theraze's script the once it uncovered all but 3 squares before finding the faucet. Maybe it's the outlier but I'll stick with following the edge.

 

[StDoodle]

The only evidence I have to support my claim is the 8 or so ascensions where going left then down makes the quest take between 5-8 turns. When I ran Theraze's script the once it uncovered all but 3 squares before finding the faucet. Maybe it's the outlier but I'll stick with following the edge.

Did you do that really early, before Jick changed the revamped tavern?

If not, it's possible that lost fights don't count... which would be a KoL bug, as that would make it possible to fail the quest entirely!

 

[Theraze]

The only evidence I have to support my claim is the 8 or so ascensions where going left then down makes the quest take between 5-8 turns. When I ran Theraze's script the once it uncovered all but 3 squares before finding the faucet. Maybe it's the outlier but I'll stick with following the edge.

5 turns? So that would be 1 step down the left side? That's some awesome luck!

Edit: Will be testing a tweak to the script... for y, then for x. If it works as I hope, should go down, then across, instead of doing stripes... we'll see. To go down the left, would need to have a reversed for (counting backwards), and not certain if that's easily possible in ASH. I could do it by making a while loop that counts backwards, but...

Edit2: Yep, works counting backwards. I'll need to consider how to make it best circuit the system.

Edit3: There we go... for y from 5 to 1, for x from 5 to 1... should follow the top wall left first, then left wall going up, then explore from the bottom left corner.

 

[Darzil]

I always go left all the way, then down all the way, then right all the way. Have sometimes not found Ratsworth, but have found Faucet every time. So, 5-12 turns.

According to the spading here, it has always been found in the following locations : (X marks the spot)

XOOO
XOOOO
XOOOO
XXOOO
XXXXX

You find the Shiny Ring, then Faucet, then Ratsworth, in three of those squares, in that order.

 

[Theraze]

Excellent. New system does 4 possibilities, in order...

y = 1, for x from 5 to 1.
x = 1, for y from 5 to 1.
y = 5, for x from 5 to 1.
for x from 5 to 1, for y from 1 to 5.

So if that works properly, should be top wall, left wall, bottom wall, and then second row to left, second from bottom (being the last available with that match). If it still hasn't found a match, it should continue up that row and look around for more...

 

[Veracity]

Revision 8984 should fix the Tavern Quest solver. Untested.

 

[slyz]

Maybe it's the outlier but I'll stick with following the edge.

Don't, and participate in the spading effort


EDIT: I went through the logs of my last three ascensions, and the upper-left corner looks definitively more likely (from a sample of 3)... That would be the reason for my 20-turn taverns ^^

 

[Terion]

FWIW: I just went through the tavern cellar manually, visiting all squares that weren't on the left or bottom edges, and got none of the special squares. Then I started from the lower right corner and worked left and then up. Got the ring in the lower right corner, then found the faucet on the left edge just above the bottom row (4th row counting down) and finally found the Baron in the upper left corner.

 

[xKiv]

What a coincidence. I just (yesterday) did something similar, except the ring was in lower left and faucet to the lower right. Which is because that's the direction I went.