[0.15.1280] RimWorld Olympics Scandal or Pawns ignoring obstacle move penalties?

[Calahan]

Hello friends! And welcome to the RIMWORLD OLYMPICS!!!

Today we have a very special event for you, the obstacle course! Three top athletic Pawns, who have all set identical times of 4.61 c/s at the 40-yard dash, line up for today's final. The winner will be decided by the first Pawn to reach their designated mining spot. Ahead of them lies an assortment or obstacles, and the Pawn who navigates these obstacles the best will win.

Matt - Has 6 Stone Chunks ahead of him spaced out at 1 tile intervals (he is the clear favourite).
Linnea - Has the even more daunting task of 11 Stone Chunks ahead of her without any spacing (only a fool expects her to win).
Jess - He will have to use his ingenuity, because he has all manner of random obstacles in front of him (the dark horse perhaps?).

So please place your bets now! Who do you think will win? Will it be close? Or will we witness the debut of the next Usain Bolt?

Ready....Set...Go!!!!

And the winner is......
.
.
.

LINNEA !!!!!!!!! She literally walked over the obstacles like they weren't even there!!! But that makes no sense at all, so a request has been made for an official investigation.

Replay this exciting race here.
https://dl.dropboxusercontent.com/u/97382000/RimWorld%20Olympics.rar

And see the attached photograph from our top reporter of the Pawns lined up before the race.


The rest of this news report is intended for official investigators only. So basically boring stuff that our regular readers can safely ignore.

-----------------------------------------------------------

1.What the circumstances were.
There was an obstacle course race at the RimWorld Olympics

2.What happened
The clear favourite, who only had about half as many obstacles to negotiate as his rivals came in a distant last. 

3.What you expected to happen.
I expected the favourite to win in Usain Bolt fashion! Because he only had 6 tiles worth of obstacles to negotiate whereas all his opponents had 11 tiles worth.

All the Pawns have identical running speeds, and were all injury free. So the only variable was the number of obstacles, and 6 is less than 11 (unless the fundamentals of mathematics changed while I wasn't looking). 

4.Steps we can follow to reproduce this issue.
Replay the race from the above link and see for yourself. I'd have thought that by the year 5500 that they'd have sorted all the cheating problems in athletics out, but apparently not


Thank you for your time in reading this (news report)


Edit - UPDATE - With the help of some information from a few forumites (my thanks to them), I think this bug report has turned into a clarification request, which I think can be summed up as...

- Is it intended that <pathCostIgnoreRepeat> should be "True" for most (all?) objects that cause a movement penalty, or should it only be "True" for Sandbags, Chunks, and other "cheap" obstacles etc. To me it seems like a bug that all objects have been tarred with the same brush (to avoid players exploiting cheap movement obstacles).

[attachment deleted by admin - too old]

[Jimyoda]

This was supposed to be one of the those exploits we never tell Tynan about!
Guess we can't trust you with the position of the rebel base. 

Very interesting find tho.

[Ark]

Some things have <pathCostIgnoreRepeat>true</pathCostIgnoreRepeat>.
Set this to false for ChunkBase in /Mods/Core/Defs/ThingDefs_Misc/Various_Stone.xml, and Linnea finishes last.
Presumably it was to fix this 'exploit'. I think the enemy should send sappers or siegers instead of enabling everyone to leap from boulder to boulder like mountain goats.

I am not sure why this exploit is not more common on these forums. Rather than a kill box I use a Gratuitous Ganking Gauntlet!

http://imgur.com/QIRC6Sc

Nearly all baddies will ignore the colonists as they hobble through the sandbag filled hallway while they get stabbed over and over through the corners. This base entrance can kill hordes of attackers using melee attacks with very little danger of retaliation. Even melee raiders/tribals/robots do not fight back even though they could.

It only stops working vs mad animals (because they just target the closest person) and when something in the hallway actually gets line of sight on something it wants to kill. Then it starts looking for targets and fighting back. When this happens I just move the attacked pawn back a square to get rid of agro and the attacker goes back to limping through the gauntlet.

On the one hand, I don't want you to fix this because I love using murder holes like this. On the other hand, I think this is clearly an exploit and it is very powerful.

[arl85]

according to Tynan, it's not a bug, but a feature, indeed

[makkenhoff]

Well, pawns have to carry these chunks, so they can't be ridiculously heavy, and granite typically weights 2,750 kg per cubic meter; hard to be exact with rock weights, since they can vary. So I figure these chunks are smaller than a cubic meter, and more importantly surface rock, so the winds will have eroded the rock to create relatively flat surfaces (when you break mountain to make rock, your just cutting them in a way that they can be used for building later). So think of it as hopping across platforms like a certainly plumber named Mario.

[Calahan]

Thanks for the additional info on this everyone. Much appreciated. I suspected this new bug (as I saw it) had arisen as a result of an attempt to patch some sort of exploit, and thought it had created an unintended bug during the process. But the above Reddit link (kindly provided by arl85) seems to suggest otherwise, and this is indeed intended behaviour (new as of A14).

I would close this thread now as resolved, but Tynan's actual comment makes me think I should leave it open until Tynan (or ison) confirms this is 100% WAD. Because in my OP test, even though the Pawn that had to negotiate a line of stone chunks won the race, the Pawn that had to negotiate an assortment of obstacles was only just behind. So while I can certainly believe... (quoting Tynan from that link)

"It'll slow them down 1 cell earlier, but not more overall. They'll walk overtop of the sandbags."

... that Pawns can walk along the top of a line of sandbags, and if I stop by brain asking to many questions, then I can also believe they can walk along the top of a line of stone/metal chunks. But I find it too implausible that Pawns can simply walk on top of a Hydroponics Basin, then effortlessly transverse to walking on a Turret, then onto a Bed, then onto a Battery, then onto whatever random object is next in line etc etc. As Ark alluded to above, Pawns are meant to be humans, not mountain goats.


My RimWorld modding skills are zero, but if as Ark says this change is because...

Some things have <pathCostIgnoreRepeat>true</pathCostIgnoreRepeat>.
Set this to false for ChunkBase in /Mods/Core/Defs/ThingDefs_Misc/Various_Stone.xml, and Linnea finishes last.

... all obstacle like objects now have <pathCostIgnoreRepeat> set to True rather than False, then to me that seem to be tarring everything with the same brush in order to patch an exploit where players were using corridors, or mass fields, of Sandbags and/or Chunks to slow enemies down. In which case this change is fine for those objects, but I really don't see why it should have been applied to most (all?) objects that cause a movement penalty. As if players want to create corridors/fields of Workbenches, Beds, Hydroponic Basins etc, then that comes with a seriously hefty cost in resources, which would/should more than cancel out any benefit.

So as I said even though this appears to be WAD, I'll keep this bug thread open for now to see if Tynan (or Ison) can confirm WAD relating to...


tl;dr - Is it intended that <pathCostIgnoreRepeat> should be "True" for most (all?) objects that cause a movement penalty, or should it only be "True" for Sandbags, Chunks, and other "cheap" obstacles etc. To me it seems like a bug that all objects have been tarred with the same brush (to avoid players exploiting cheap movement obstacles).