what's the maximum COF (chance of failure) possible in GP4?

hello again
I was wondering what is the COF value that will 100% result in a retirement?

also while I'm on with COF, how does it work?
I assume the game runs random failure numbers for each car (so between 0 and x), and if the number is lower than the COF value it's a DNF, but I might be wrong.


(ps: used the search bar, found nothing)

---

GPGSL - Yakuza driver

GPGSL activity check app: Direct link - Source code - Have you posted?

I think its between 0 - 32000 maybe

Latest Themes / Mods > [www.mediafire.com]
You Tube Video's [www.youtube.com]
________________________________________±

I can tell you that a failure chance of 1000 is somewhere between 5% and 15%, 10000 is between 35% and 45% I think (own experiences).

According to GP3 Edit: 0 - 32767 = 0% - 80%.
32787 is the maximum number that GP3 Edit allow me to put. The system is the same as the GP4 I think. Test if with a greater number than 32787 works.

I tested this quite some time ago and 32787 is indeed the maximum.



5 x WCC - S5, S6, S7, S8, S10 | 4 x WDC - S5 Nick van der Voort ; S6, S7, S8 Ed Greenhalgh | 2 x WTC - S6 Stuart Ingers; S8 Andrei Sevastian

so 32767 or 32787?
Also does that mean you can't put a 100% chance of failure?

---

GPGSL - Yakuza driver

GPGSL activity check app: Direct link - Source code - Have you posted?

In GP4 it's:

if race:
  FP = teamFailureProbability * max(20, distance%) / 100
else if qualify:
  FP = teamFailureProbability / 2
else:
  FP = teamFailureProbability
willGetFailure = randomNumberBetween(0, 16383) < FP

So with 16384 it should be 100% already, for a 100% distance race (haven't tested this, I have only read the code in the executable).

René Smit, Independent Software Developer.
Download my GPx tools here.



Edited 3 time(s). Last edit at 08/23/2015 01:29PM by SDI.

thanks Rene!

---

GPGSL - Yakuza driver

GPGSL activity check app: Direct link - Source code - Have you posted?

By the way, it's the same in GP3.

And if in your experience 32768 is 100% failure, then you're probably doing 50% distance races.

René Smit, Independent Software Developer.
Download my GPx tools here.

Well, is this proportional then? 16384 means exactly 25% failure chance for 100% races and 8192 12.5% failure chance for the same distance? And I thought the last four years that about 12000 are 50% failure chance for 100% races. So this is definitely wrong, isn't it?

Someone quoted on RaceDepartment many years ago that 5000 was around 10% chance, that sounds about right-ish?

well I did want to know exactly, but I know about that too.

---

GPGSL - Yakuza driver

GPGSL activity check app: Direct link - Source code - Have you posted?

It is proportional, but 16384 means 100% failure chance for 100% races and 8192 means 50% failure chance for the same distance.
The failure chance is propertionally reduced with the race distance until 20% distance.

René Smit, Independent Software Developer.
Download my GPx tools here.