Starting performance of Human Player

Hey all,

one thing I always considdered angrily was the player's / my performance at the start of a race. For me it is nearly impossible to stay in first position at the start. The drivers in 2nd and sometimes even in 3rd position nearly always get the jump on me.

I made some tests with it an found out that it is no question of my reaction time, launch control, gearbox settings, rear wing setup, horse power or slipstream effect. When my car arrives around 40 km/h (~25 mph) the ai cars just beginn to be a few km/h faster than me. Their acceleration is just better.

Has anyone else detected this issue? What can we do about this performance gap?

Kind regards
Klaus Feldmann

Interesting, as for me it's completely opposite and even from the back end of the grid I can be in top 5 easily on the first corner.

vesuvius schrieb:
-------------------------------------------------------
> Interesting, as for me it's completely opposite
> and even from the back end of the grid I can be in
> top 5 easily on the first corner.

GP4 is a crazy game .

Did you or did anyone else ever found a parameter to influence this behaviour?

klausfeldmann Schreef:
-------------------------------------------------------
> Did you or did anyone else ever found a parameter
> to influence this behaviour?

Yes, throttle handling skill

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

Even when I try to do a race with out Traction Control, I always have to use it off of the start.
(I'm pretty sure the AI use it as well.)

I just hold it in neutral - actually hold it when using Auto-gears - with about 1/4 or less throttle and, when the lights go out, chuck into first, slam the throttle on full and let the TC and Launch Control take over;
Even on Ace difficulty with default car setup this usually works, allowing me to go from 17th at Interlagos to 5th-6th by T1.....with a risky dive up the inside

---

F1 Graphics Swapper --- Lensdump - Renders and Screenshots --- GP4 Performance Generator

Okay, just to make it clear:
I'm not talking about the trick to break on the inside for the first corner . With that trick even I am able to overtake half of the field . I'm just talking about the time between 40 km/h and ~120 km/h at the start. Before that and after it everything is allright with my and the AI cars's speed.

This is how I start:

With Launch Control:
- Switch into 1st gear, keep pushing the button.
- Throttle at 100%.
- When the lights go out: Pull off the finger of the upshift-button.
- At ~40 km/h: disable Launch Control.
- At ~100 km/h disable Traction Control.

Without Launch Control: (most the time I start without Launch Control)
- Switch into 1st gear, keep pushing the button.
- Throttle at ~20%.
- When the lights go out: Pull off the finger of the upshift-button.
- At ~40 km/h: increase Throttle continiously to 100 %.
- At ~100 km/h disable Traction Control.

What are the experts's opinions about it? Am I making something wrong or is this behaviour related to game data?

I find it depends on the mod as to how well the start goes, might have something to do with the torque in the physics file. If I'm struggling to get off the line compared to the AI then generally I find I make better starts by switching off the launch control after the initial wheelspin phase is over, so like when it starts to accelerate rather than hold constant revs.

I start without launch control and it is definitely depending on the performance file. With the original performance file from 2001 (car power about 800) I never lost a position to the first corner when I was on the pole and from 22th starting place I was at least 3rd after the first corner of an average track. Now I'm using performance files with more than 1000 car power and from 22th I never get farer than 14th after the first corner, and starting from pole I'm loosing at least 2 or 3 postions, but often I get the lead back in the first corner by breaking much harder then everyone else. Are other people as well my opinion that the computer drivers are very carefully by breaking for the first corner?

TGF1DRIVER schrieb:
-------------------------------------------------------
> I start without launch control and it is
> definitely depending on the performance file. With
> the original performance file from 2001 (car power
> about 800) I never lost a position to the first
> corner when I was on the pole and from 22th
> starting place I was at least 3rd after the first
> corner of an average track. Now I'm using
> performance files with more than 1000 car power
> and from 22th I never get farer than 14th after
> the first corner, and starting from pole I'm
> loosing at least 2 or 3 postions, but often I get
> the lead back in the first corner by breaking much
> harder then everyone else. Are other people as
> well my opinion that the computer drivers are very
> carefully by breaking for the first corner?

1. This is a good note! Yes, I remember when I drove performance files with about 800 HP I was able to compete with the AI cars at the starts for sure. Now I use a performance file with more than 1,000 HP and have no chance to keep my position. This could be the reason. So, it might be related to game data. Thanks for this note .

2. Yes, support your impression that the AI cars are breaking very carefully in the first corner(s). Sometimes they even cause a traffic jam. I think we are able to see this phenomenon at suzuka in special. There is no 1st lap without a traffic jam in the hairpin. I think Geoff Crammond simply was too conservative at this point. Maybe there is a way to tweak the AI behaviour in traffic situation...?!

klausfeldmann schrieb:
-------------------------------------------------------
> TGF1DRIVER schrieb:
> --------------------------------------------------
> -----
> > I start without launch control and it is
> > definitely depending on the performance file.
> With
> > the original performance file from 2001 (car
> power
> > about 800) I never lost a position to the first
> > corner when I was on the pole and from 22th
> > starting place I was at least 3rd after the
> first
> > corner of an average track. Now I'm using
> > performance files with more than 1000 car power
> > and from 22th I never get farer than 14th after
> > the first corner, and starting from pole I'm
> > loosing at least 2 or 3 postions, but often I
> get
> > the lead back in the first corner by breaking
> much
> > harder then everyone else. Are other people
> as
> > well my opinion that the computer drivers are
> very
> > carefully by breaking for the first corner?
>
> 1. This is a good note! Yes, I remember when I
> drove performance files with about 800 HP I was
> able to compete with the AI cars at the starts for
> sure. Now I use a performance file with more than
> 1,000 HP and have no chance to keep my position.
> This could be the reason. So, it might be related
> to game data. Thanks for this note .
>
> 2. Yes, support your impression that the AI cars
> are breaking very carefully in the first
> corner(s). Sometimes they even cause a traffic
> jam. I think we are able to see this phenomenon at
> suzuka in special. There is no 1st lap without a
> traffic jam in the hairpin. I think Geoff Crammond
> simply was too conservative at this point. Maybe
> there is a way to tweak the AI behaviour in
> traffic situation...?!

1.You are using a performance file with more than 1000 car power as well? Great! Then please, can you try out high speed tracks like Hockenheim 1968 or Monza 1971 with this performance file? Then tell me about the top speed of the AI cars compared to the top speed of you as a human player. Is then the top speed of the AI cars nearly 400 kph as well and your own top speed only about 360 kph? That's a big problem I have since years!

2.Yes, the traffic jam at the Suzuka hairpin, I know it as well, I remember once I was passing about ten cars there in the first lap when breaking down the corner on the inside!

@ TGF1DRIVER:
I just checked the two tracks you told me. Here are the results:

Hockenheim 1968:

CCcars:
Topspeed: 363 Km/h
Topspeed in slipstream: 369 Km/h

Player:
Topspeed: 361 Km/h (compared to CCcars: -2 Km/h)
Topspeed in slipstream: 363 Km/h (-6 Km/h)

Monza 1971:

CCcars:
Topspeed: 384 Km/h
Topspeed in slipstream: 395 Km/h

Player:
Topspeed: 361 Km/h (-23 Km/h)
Topspeed in slipstream: 363 Km/h (-32 Km/h)

So as it seems, the problem isn't that big at Hockenheim, but at Monza it's really silly. Exchanging the magic data of the two tracks does not solve the problem, so it might not be magic data related. It seems as if the rev-counter is different for the CCcars, although the gearbox settings are similar to the player's car.

What else data are track-specific next to the magic data?



Edited 1 time(s). Last edit at 10/12/2014 11:58AM by klausfeldmann.

Have you also tried Monza in a different slot?

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

Trying the track in a different slot leads most time to lower top speeds around 315 kph for AI cars and player. That's not the sense of these tracks as well.

SDI schrieb:
-------------------------------------------------------
> Have you also tried Monza in a different slot?

I didn't try Monza in another slot so far.
Are there any further investigations you recommend?

No, but me and SDI are working on a solution.

TGF1DRIVER schrieb:
-------------------------------------------------------
> No, but me and SDI are working on a solution.

All right, I hope you'll get it !

I just want to confirm the speculation of TGF1DRIVER regarding the original topic of this thread:

When I start with a performance file of about 1,000 HP, I have no chance to follow the CCcars at the starts. With 800 HP it is alright. With 600 HP, however, it is no problem to gain up to 5 positions until the first braking zone.

Thanks at TGF1DRIVER for solving this riddle .

Yes, but now I have to solve another and much bigger riddle!

TGF1DRIVER schrieb:
-------------------------------------------------------
> Yes, but now I have to solve another and much
> bigger riddle!


You mean the top-speed differences at Hockenheim and Monza? I hope you'll get the solution soon .

When somebody will get the solution then it's SDI. I don't know much about how GP4 is programmed. But you can be sure that I won't give up on searching and testing because I am upset with that problem as well!