GPGSL S11: R14 Flevoland GP [VACANCIES, join now!!! || R13 QUAL online p.157, RACE p.158!!!]

what option determines the length of pit-stop? while one of my tests i just noticed that one of the teams got both their cars refueled in 5 seconds while the rest of the field got 9-10 seconds in average for refueling. i checked it for another races, and it seems like it was a one-time bug, but still... also it seems that wing repairment time is way too short.

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]



Edited 1 time(s). Last edit at 03/21/2017 05:55PM by Soutsen.

pitstop.ini in the tweaker folder, both look fine.



JohnMaverick schrieb:
-------------------------------------------------------
> Exactly that, at least with the old boost system.
> Whether or not this stays like this after the
> boost testing will have to be seen.


I hope that doesn't mean that each team will receive same amout of BHP boost again...

---

Some mods
F1 1996 | F1 2002 | F1 2007 | F1 2011 | F1 2013 | F1 2015 | F1 2018

well, that was the issue:
1 - c.warrington got his rear wing off
2 - he got it repaired in 5 seconds and goes out of the pits
3 - later on his teammate cremasco got refueled in 5.7 seconds
4 - end of the race, they both have 1 pit and it was 5,0 and 5,7 seconds while the rest of the field had 9-10 seconds.
i checked it on melbourne again and got no similar problem, my guess is that the rear wing repairment broke the pit pattern for that team or something, maybe it is a native yet unnoticed gp4 bug
anyway 5 seconds for replacing a rear wing is a bit too fast.

btw i dunno why my cam/helm coordinates are messed up that much, i tried to delete the mod folder and reinstall it, but got absolutely the same result
nvm i reinstalled the whole game completely, and now the mod works fine

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]



Edited 2 time(s). Last edit at 03/21/2017 07:03PM by Soutsen.

In regards to feature races for testers, it's up to the RD if they decide to run a feature race for the test drivers. Otherwise, it ends up being a regular free practice session. Think it's been a tradition to run a feature race for the testers in the season opener, but I could be wrong though.

---

Did some research about how variance works for AI drivers. The track I used was Melbourne from the GPGSL S11 trackpack, with some magic data modifications (AI errors set to 0, Stop 1 on lap 14, Window 0). Did 4 quick race sessions (100% distance), with the same weather seed (0,0,0,0,0,0 - which is the driest set possible, to prevent the different weather conditions distorting the results) and with the one single car on the track being starting from the grid (no other cars to prevent them distorting the results). The results shows the lap times for the first full 13 laps until the pit stop on lap 14:

It seems that the higher variance value produces a better lap times, so the current formula of adding points to the performance and subtracting from the variance now looks questionable.

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]

Obviously higher variation value improves the laptime, since it's a random amount between 0 and the value, which is added to the ability. Higher ability -> better laptimes.

SDI properly explained how it works a while back [www.grandprixgames.org]

---

Some mods
F1 1996 | F1 2002 | F1 2007 | F1 2011 | F1 2013 | F1 2015 | F1 2018



Edited 1 time(s). Last edit at 03/22/2017 10:09AM by kedy89.

With the adding to the perf and subtracting from the variance, it doesn't instantly improve how good a driver is in the next race. It just means they're more likely to perform at the maximum level. As Tobi says variance adds performance on, so if a driver starts at 16000 perf and 600 variance, his perf for the race will be somewhere between 16000 and 16600. With 10 points added this becomes 16010 and 590, giving a range of 16010 to 16600. Both before and after the points a driver can reach the same maximum performance, but after the points he's just more likely to do so. With the added bonus of never being able to perform at 16000.

I think perhaps the biggest question that's been passed around here is how much of a difference the driver and team boost makes. For the driver, 100 seems to be too little, whilst something like 500 (aka GPGNC) is almost certainly too big. You can do some basic probabilities and say if everything's equal and we use the GPGSL variance of 500, Driver A he will beat driver B x amount of times using y amount of boost:

0 boost: 50% of the time
100 boost: 68% of the time
150 boost: 75% of the time
200 boost: 82% of the time

It's worth doing some tests and seeing whether this is close to how it actually is, what the similar numbers are for team boost, and then see what people want these numbers to be. Obviously things like individual variance exist, but if you use a rough estimate of 7% extra chance of beating someone per 50 perf gain, then you'll be able to work out best vs worse drivers and such. Personally boost shouldn't exceed (Maximum cap level for best driver) - (starting performance level), but as that's 320 (I think), I doubt anyone will choose to go that high.

_________________________________________________

For a list of EVERY download for GP4, look here: [docs.google.com]

What about technical failure chance for the trackpack, Connor? Have you edited that at some point? For example Melbourne has only Wheel (744) and Transmission (744) failure chance, while Spa has Suspension (744), Wheel (744), Puncture (1488), Engine (2488), Transmission (1488), Leak (744) Throttle/Brake (744) and Electrics (1488) which looks much more challenging for the cars. In fact while using the starting non-boosted perf file I have 0-1 technical DNFers for both tracks normally :-/

also my first suggestion is this obviously: less consecutive cars on track during qualy to prevent the traffic jam (had 2 live qualy sessions and both times AMR drivers qualified somewhere at the bottom because of traffic)

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]



Edited 1 time(s). Last edit at 03/22/2017 11:45AM by Soutsen.

Soutsen Wrote:
-------------------------------------------------------
> What about technical failure chance for the
> trackpack, Connor? Have you edited that at some
> point? For example Melbourne has only Wheel (744)
> and Transmission (744) failure chance, while Spa
> has Suspension (744), Wheel (744), Puncture
> (1488), Engine (2488), Transmission (1488), Leak
> (744) Throttle/Brake (744) and Electrics (1488)
> which looks much more challenging for the cars. In
> fact while using the starting non-boosted perf
> file I have 0-1 technical DNFers for both tracks
> normally :-/

The technical failure hasn't been edited at all, and I don't think there has ever been rules around this. Each track will be different to reflect its real life characteristics, and I'm quite happy to keep it that way to add some extra variation. Spa should be more technically demanding than Melbourne shouldn't it
No track should have excessive retirements for any reason, but if you notice a massive technical DNF spree do let me know and i'll tweak that track a little.

_________________________________________________

For a list of EVERY download for GP4, look here: [docs.google.com]

truecrysis schrieb:
-------------------------------------------------------
> The technical failure hasn't been edited at all,
> and I don't think there has ever been rules around
> this. Each track will be different to reflect its
> real life characteristics, and I'm quite happy to
> keep it that way to add some extra variation. Spa
> should be more technically demanding than
> Melbourne shouldn't it
> No track should have excessive retirements for any
> reason, but if you notice a massive technical DNF
> spree do let me know and i'll tweak that track a
> little.


Yeah, thanks to that (I guess) 2 teamboosts for Shadow last season ended in 2 double DNFs, while 2 (way bigger) boosts for GGP gave them 2 wins. Utter joke...

---

Some mods
F1 1996 | F1 2002 | F1 2007 | F1 2011 | F1 2013 | F1 2015 | F1 2018

That's good, I just don't think that Melbourne is famous for being demanding to transmission and the wheels. While testing the boost ideas and stuff at Melbourne having to see the one car retiring because of the same wheel problem every time is a bit tiring (I mean, it would be nice to have some variation on these). And I agree that Spa should be more technically demanding, especially to engines (F1 1994 Belgian GP - 6 cars out because of engine problems), but the old Hockenheim should be the same (F1 1995 German GP - 6 cars out because of engine problems), while on the trackpack Hockenheim has only 744 chance of engine failure. And I disagree with the excessive retirements statement, I think we should have the higher probability of engine problems (for example) on tracks like Spa, Hockenheimring and Monza, so for the team bosses it should be risky to use high boost on such tracks. This kind of gambling adds more fun with making it spicier

I can make some quick researches with the F1 stats (Indycar stats for Elkhart) to make a realistic chance of failures if needed (I dont mean a higher chance, but realistic, if community votes for having technical DNFs as little as possible).

Or we can make it standart for every track with the basic 64/128/256 (?) chance to break for everything except engines, which can be different for all the tracks. 3723 (highest) for Monza, Hockenheim and Spa, 2488 (medium) for the medium-speeded tracks, and something like the lowest 744 for the rest.

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]



Edited 4 time(s). Last edit at 03/22/2017 12:21PM by Soutsen.

kedy89 Schreef:
-------------------------------------------------------
> Yeah, thanks to that (I guess) 2 teamboosts for
> Shadow last season ended in 2 double DNFs, while 2
> (way bigger) boosts for GGP gave them 2 wins.
> Utter joke...



Audi power is a joke, Porsche power rules! YEEHAA!!!

---

My Grand Prix 4 Files
I'm a total dick. How many people can say that?

Correct me if I'm wrong, all you track specialists, but so far I understood it that it's the race retirement value (the third in the team perfs) that determines how likely it is for a car to DNF because of technical issues and the track values only determine which part of the car breaks IF the car has a technical issue?! Of course we can go in all realism mode but there's still a lot of work ahead of us and deciding which track causes which technical failures isn't really at the top of my priority list. The goal is to have not more than 1-3 DNFs at a normal race, which has and still does reflect pretty well the fortitude of the GPGSL cars. Whether a car now drops out because of any technical issue or a driver mistake is secondary, imo.


GPGSL : Team Owner of 'Maverick Track Performance' (MTP)

JohnMaverick Wrote:
-------------------------------------------------------
> Correct me if I'm wrong, all you track
> specialists, but so far I understood it that it's
> the race retirement value (the third in the team
> perfs) that determines how likely it is for a car
> to DNF because of technical issues and the track
> values only determine which part of the car breaks
> IF the car has a technical issue?! Of course we
> can go in all realism mode but there's still a lot
> of work ahead of us and deciding which track
> causes which technical failures isn't really at
> the top of my priority list. The goal is to have
> not more than 1-3 DNFs at a normal race, which has
> and still does reflect pretty well the fortitude
> of the GPGSL cars. Whether a car now drops out
> because of any technical issue or a driver mistake
> is secondary, imo.

Whenever I've been playing around with tracks I've assumed it was a combination of both numbers. I could be wrong but if that is the case, the entire trackpack is available for everyone to use for strategy decisions. If someone wants that information but can't get it, then I can provide it. But i'm not overly fussed about changed failure in the trackpack as the variation is interesting, and the tracks I've chosen are all very well made so I assume the failure rates are set in a reasonable way by the authors, even if they're not 100% correct.

_________________________________________________

For a list of EVERY download for GP4, look here: [docs.google.com]



Edited 1 time(s). Last edit at 03/22/2017 02:01PM by truecrysis.

what's the reasoning behind keeping the specific (limited) boosts (=100 points as for now)? why can't we give the drivers for example 1200 points to spend through the season with limiting the lowest (100) and highest (300) (for example) ammount of points they can spend for one race? and why we have no separate boosts for qualy and race as well? what was the problem with implementing these in the past (if it was ever implemented before)?

also spangenberg's variation is 501 for both qualy and race at the first page and at the mod perfs as well.

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]

The main reason is that it makes things a lot harder for the perf creators and currently our main goal is to keep things as fluent as possible to avoid longer than necessary waiting times. So it's highly unlikely that we'll implement free boost values for drivers this season.


GPGSL : Team Owner of 'Maverick Track Performance' (MTP)

one more question: for the R1 at Melbourne no driver boosts allowed, but are the team boosts allowed for this race?

___________________________________________________________________________
For a list of EVERY download for GP4, look here:[docs.google.com]

no, no boosts at all in the first round


GPGSL : Team Owner of 'Maverick Track Performance' (MTP)

JohnMaverick schrieb:
-------------------------------------------------------
> Correct me if I'm wrong, all you track
> specialists, but so far I understood it that it's
> the race retirement value (the third in the team
> perfs) that determines how likely it is for a car
> to DNF because of technical issues and the track
> values only determine which part of the car breaks
> IF the car has a technical issue?!

Imo that is the correct way to think about it. Iirc the maximum value for reliability does not mean 100% chance of failure, though.

---

used to be GPGSL's Nick Heidfeld

Greg sent me a message to be more active here, but with all these small details I got lost hahahaha

Anyway, let's push for the next season