Skip to content
This repository has been archived by the owner on Jul 1, 2022. It is now read-only.

Added new param: end_run_in_act #857

Open
wants to merge 12 commits into
base: master
Choose a base branch
from
Open

Conversation

jobithu
Copy link
Contributor

@jobithu jobithu commented Jun 11, 2022

The most efficient way to farm Travincal is to end the run in A4, instead of A3.
A3 town is rather large and traversing to WP & Vendors takes longer than in A4 town.
Thus subsequent runs will be faster when previous runs end in A4.

Setting "end_run_in_town" to 4 will trigger a TP at the end of Travincal and our char traversing to A4 town to end the run there.
Any town can be selected for ending the run (1,2,3,4,5), if param set to 0 the standard town for the corresponding run is used (default behaviour).

Intensive testing on a fully geared Enigma Hammerdin revealed that average run duration decreased from 82 to 71 seconds (12% improvement). For a typical farming day of 12h, instead of completing 525 runs, additional 85 runs can be performed (total 610 runs).

The methodology can be adapted to other runs such as Arcane, the benefit here, will however be much less.

@namdnas
Copy link

namdnas commented Jun 11, 2022

I run pindle eld shenk trav then nihlathak for this reason. Ends back in act 5 so it traverses less distance

@D2RLegit
Copy link
Contributor

2 extra load screens for this even though they may somewhat quick sometimes. Not sure if its really cuts that much time for most people due to the extra loading screens/sell items/stashing/pickit and especially on vm's. Also when joining a new game you typically walk to vendor to resupply or sell items which ormus is roughly similar distance as jamal when starting a5. Just my 2 cents. It better to start in act 5 and end in act 5 aka start with pindle and end with eld/shenk or nihlathak.

@jobithu
Copy link
Contributor Author

jobithu commented Jun 12, 2022

Thanks for your feedback, Legit!

Due the points you mentioned, the default value of the param is 0 - in case multiple targets are farmed, I agree with you: we should end up in A5 instead.

Based on your feedback, I did pressure tests on VM and as Walkadin with / without the param activated, just farming Trav.
Even taking into account the extra loading screens, my average run duration was 21s shorter 85 more games in 12h):
param off = Avg Game Length: 00:01:58 (120 games)
param on = Avg Game Length: 00:01:37 (200 games)

Looking at it: I would definitely merge it :)

@aliig
Copy link
Collaborator

aliig commented Jun 12, 2022

I'd like to incorporate a feature like this, but one that is more generalizable. Like a way to end in whatever act you put in a param.

it doesn't make much sense to include as a param in [char] when it only is implemented on trav.

Copy link
Collaborator

@aliig aliig left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

see prior msg

@jobithu
Copy link
Contributor Author

jobithu commented Jun 12, 2022

So for Arcane, it could also make sense to just pop the tome, when at summoner and leave through the WP of Duriel. This could allow you to end up in A4 town, as well (when powerfarming hate keys).

so in this case: the param would be: preferred_end_town= "A1, A2, A3, A4, A5". Relevant runs: Trav & Arcane.

@jobithu jobithu changed the title Added new param: Prefer_act4_town Added new param: end_run_in_act Jun 13, 2022
@jobithu
Copy link
Contributor Author

jobithu commented Jun 16, 2022

Ready to merge (for Trav - I dont think it makes sense for other currently implemented runs)

@aliig
Copy link
Collaborator

aliig commented Jun 16, 2022

This is not ready to merge. It says "end_run_in_act" which implies you want botty to force to end the run in a certain act regardless of what run you're on, but this is not at all implemented. The only thing it does is if trav is your last run it'll end the run in X act.

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants