Grip: two issues

Smalltalk
Here you can post everything which doesn't fit into the other forums!
User avatar
samsepi0l
GP2 Racer
Posts: 301
Joined: 01.03.2019, 01:39

Re: Grip: two issues

Post by samsepi0l »

Guys- I am not as skilled as you, but I modify the carsets that I download- so I want to ask you this.

*are you referring to the "range" when you go to the little "helmet" icon in GP2edit?

When I play my game- I use a much tighter window of values so that the front of the field and the back of the field are closer together. I know it's not historically realistic, but it make the game more interesting because the slower cars are more competitive. I use the same value for the "range" for everyone.

For example, I am racing a 2001 season now. Here are the values I have. Can you please comment?
(I am only using 26 of the cars, 2 cars are disabled)

Image
Attachments
2001_range.jpg
User avatar
MaxX
GP2 Star
Posts: 1073
Joined: 28.02.2007, 11:41
Location: Salzburg, Austria

Re: Grip: two issues

Post by MaxX »

rremedio wrote:
29.05.2021, 22:38
Wow! That's very nice!

So this confirms it works like in the later games where the RG is a range that goes both ways.

I'm a little surprised tho since I've never found actual grip values that were higher than the driver's grip when supervising the memory during the races. That value is probably stored in one of the many unknown sections.

Also very surprised nobody ever noticed that pattern with the original grips and grip ranges before. Well done! All of this makes total sense if you think about it since the grip range becomes more important than the driver's grip when there are such big grip range values like in the original game and the top drivers are on the top most of the time anyway.
Seconded, awesome findings. tbh, I haven`t thought in that way so far :D
User avatar
samsepi0l
GP2 Racer
Posts: 301
Joined: 01.03.2019, 01:39

Re: Grip: two issues

Post by samsepi0l »

sorry to interrupt - but do you think that I've accomplished my goals with the numbers I use?
User avatar
rremedio
GP2 Legend
Posts: 1559
Joined: 14.01.2007, 15:05

Re: Grip: two issues

Post by rremedio »

samsepi0l wrote:
30.05.2021, 05:18
Guys- I am not as skilled as you, but I modify the carsets that I download- so I want to ask you this.

*are you referring to the "range" when you go to the little "helmet" icon in GP2edit?

When I play my game- I use a much tighter window of values so that the front of the field and the back of the field are closer together. I know it's not historically realistic, but it make the game more interesting because the slower cars are more competitive. I use the same value for the "range" for everyone.

For example, I am racing a 2001 season now. Here are the values I have. Can you please comment?
(I am only using 26 of the cars, 2 cars are disabled)

Image
The range values are the ones in the screenshot to the right (1500 for most drivers in the picture).

In the case of your carset, when starting a session the game will generate a random number for each driver between -750 and 750 (1500/2) and add that generated number to the appropriate grip (R-Grip or Q-Grip, depending on the session). The result will be the actual grip value during the session.

For a modern F1 season, I think 1500 may be too big of a value, but not sure about 2001 without looking back at the results. With all cars having 1500 range you are basically giving 60-70% of them a chance of winning a race or being last. But since your grip values are all somewhat close together, it may help spread the cars a little bit so they are not all racing a few tenths from each other all the time (which is not as fun as it sounds in road racing).
"The Fox provides for himself. but God provides for the Lion"
Image
Early 90s F1 forever!
roland ratzenberger
GP2 Racer
Posts: 356
Joined: 18.03.2010, 23:51

Re: Grip: two issues

Post by roland ratzenberger »

TdK wrote:
29.05.2021, 19:15
Well in the past 10 years in my absence of Gp2 i discovered something about the random grip:

It works both ways. If the Random Grip is 0, the driver will always produce about the same lap time.
If you give a driver Random Grip range of 938 (Schumacher '94 Microprose), it's grip will be between 15915 and 16853. His standard grip is 16384.
So he can perform better, and worse. Previously i thought it only worked negative. But as it turns out, it works positive as well. So the 938 value is a value effecting the 16384 level. Positive and negative.

Also i discovered that only Schumacher's Random Grip matters: The Random Grip (RG) of the other drivers are calculated from his Grip values.
An Example 1994 Microprose values:

GRIP: RG:
Schumacher: 16384 938
Hill: 15980 1342
Alesi: 15468 1854
Belmondo: 13846 3738

Maybe if you are very very sharp you will notice it already. Hill his RG is calculated from the difference in grip between Schumacher and him. The calculation goes like this: 16384 - 15980 = 404. 404 + 938 = 1342. Alesi the same: 16384 - 15468 + 938 = 1854. Belmondo: 16384 - 13846 + 938 = 3738.

I discoverd this only just last years summer. And i implemented it within my 2020 carset, with a few deviations, and it seems to work perfect. In the meantime i tested it on other seasons\years and it works fine. It maybe even works better than what i first used as RG values, each driver it's own random grip, instead of reflecting/feeding it back to the fastest driver. You can always higher or lower the RG value, like i did in my 2020 carset as well, to get a more realistic result with some drivers.

If you want i can upload some carsets with the new data regarding to Random Grip. I personally believe this is how it should be, and it works better.

Commenting on that 16384 is the highest grip: as it turns out now, some drivers will exceed that grip number. And if you give a driver a higher than 16384 level, they are hard to beat, you need to have more horsepower than them to beat them. But that can also be solved, or 'tuned' by adjusting the "Grip Factor" to a lower number in the "Car Settings" menu in Gp2Edit. With this only the player gets more grip. A little fine-tune value within Gp2.
if you dont mind mate , please if you can upload some. the last 4 years i always manipulate carsets values etc, and what a better <<lesson>> than to have some of yours carset which i believe are the cream of the crop as guidance :D :D
ismael
GP2 Racer
Posts: 395
Joined: 02.05.2016, 14:58

Re: Grip: two issues

Post by ismael »

I'm so amazed by this TdK post that I'll issue a question only to a part of this, leaving this rest for later (if ever).
TdK wrote:
29.05.2021, 19:15
Well in the past 10 years in my absence of Gp2 i discovered something about the random grip:

It works both ways. If the Random Grip is 0, the driver will always produce about the same lap time.
If you give a driver Random Grip range of 938 (Schumacher '94 Microprose), it's grip will be between 15915 and 16853. His standard grip is 16384.
So he can perform better, and worse. Previously i thought it only worked negative. But as it turns out, it works positive as well. So the 938 value is a value effecting the 16384 level. Positive and negative.
This is surely the biggest breakthrough in "GP2 Science" for some 20 years! Are you sure about this? How did you find out?
How do we calculate what 'grip' a human driver can reach? Are we to drop the 16100? Is 16100 valid only for a decent race trim for 50% or 100% Race Distance, without using ultra-low ride height which would render a permanent penalty for the human driver after a few laps - in which case the grip would be something like 17000?
User avatar
TdK
GP2 Pro
Posts: 900
Joined: 11.12.2006, 20:16
Location: Netherlands

Re: Grip: two issues

Post by TdK »

ismael wrote:
16.06.2021, 22:26
Are you sure about this? How did you find out?
This one i can answer: Make a carset where every driver has the same amount of grip, power, and 0 Random Grip. In the car settings menu turn on the 'Random Power Variation' box. Than do a Non-Championship Race-->Qualifying and you will see that all the drivers will be very, very close to one another. Sure there are differences, it will be about a second between 1st and 26th, depending on the track.

Next step: Give one driver a quite high Randon Grip Range (1024 for example), and you will see that one race he is tenths (0.100, 0.200, ....) of a second in front, at another event/race he's at the back. So this for me means that the Randon Grip works both ways. :-s
TdK - (re)tired GrandPrix 2 manic!
ismael
GP2 Racer
Posts: 395
Joined: 02.05.2016, 14:58

Re: Grip: two issues

Post by ismael »

TdK wrote:
02.09.2021, 09:44
ismael wrote:
16.06.2021, 22:26
Are you sure about this? How did you find out?
This one i can answer: Make a carset where every driver has the same amount of grip, power, and 0 Random Grip. In the car settings menu turn on the 'Random Power Variation' box. Than do a Non-Championship Race-->Qualifying and you will see that all the drivers will be very, very close to one another. Sure there are differences, it will be about a second between 1st and 26th, depending on the track.

Next step: Give one driver a quite high Randon Grip Range (1024 for example), and you will see that one race he is tenths (0.100, 0.200, ....) of a second in front, at another event/race he's at the back. So this for me means that the Randon Grip works both ways. :-s
I see. It's puzzling, but amazing and I'll take your word =D> . Yet I have another doubt; the grip means its number either way, or does it mean the total number as represented? For example, let's take the inevitable vanilla GP2 Schumacher benchmark, with a grip range of 938.
Does it mean 938 either way, which means that Schumacher can go as fast as 17322 (=16384+938) and as slow as 15446 (16384-938? Or,
Does it mean a total 938, which means that Schumacher can go as fast as 16853 (=16384+469) and as slow as 15915 (=16384-469)?

Maybe the later, as there is no 17322, right?
So that Hill can go as fast as 16651 (15980+671). Curiously, if you take 15980+1342 you get also 17322. The same happens with Coulthard.
But if you take Gachot (14055+3476) the number is higher than 17322. What does all of this mean?

(Why doesn't Geoff Crammond heavenly appear in this forum to clear all this matter :shock: :o :D )

I'm not dropping a previous question of mine 8) You once wrote that "The maximum grip a player can reach is about 16100", you having been quoted here viewtopic.php?p=5884#p5884. Do you still think this way?

"Is 16100 valid only for a decent race trim for 50% or 100% Race Distance, without using ultra-low ride height which would render a permanent penalty for the human driver after a few laps - in which case the grip would be something like 17000?"

Thanks and keep up the amazing work!...
ismael
GP2 Racer
Posts: 395
Joined: 02.05.2016, 14:58

Re: Grip: two issues

Post by ismael »

I'm sorry. You had already answered the essential of my questions: "If you give a driver Random Grip range of 938 (Schumacher '94 Microprose), it's grip will be between 15915 and 16853." And I forgot you had replied, on May 29th, regarding that calculation:

"The calculation goes like this: 16384 - 15980 = 404. 404 + 938 = 1342. Alesi the same: 16384 - 15468 + 938 = 1854. Belmondo: 16384 - 13846 + 938 = 3738."

I suppose that the best way to create a Senna Challenge in an otherwise vanilla GP2 it to set Senna in Williams number 2 with 16384 and a Random Grip Range (RGR) of 938, same as Schumacher, ensuring that over a season Senna would be ahead of Schumacher given the 30 HP advantage of Renault over Ford.

Betting back to TdK:
"Commenting on that 16384 is the highest grip: as it turns out now, some drivers will exceed that grip number. And if you give a driver a higher than 16384 level, they are hard to beat, you need to have more horsepower than them to beat them."

This means it would be, theoretically, impossible to beat a CC-car with 16384 grip and 938 RGR while carrying the same amount of horsepower than him/it (I don't know if, in English, a CC-car is him or it). At least over a season and with races of 100%, 50% or 25% Race Distance, something that would make hotlapping impossible or useless.

Now that would be the Ultimate Challenge. Thanks!
User avatar
samsepi0l
GP2 Racer
Posts: 301
Joined: 01.03.2019, 01:39

Re: Grip: two issues

Post by samsepi0l »

Ismael- I feel kind of dumb asking this, but if I wanted to have the same horsepower for all the cars, and just change the drivers so some are faster/slower than others- can't I just use the grip values?

If I give everyone the same "range" of grip values, it will give everyone equal "difference" in how much they have a fast day, or a slow day right?
ismael
GP2 Racer
Posts: 395
Joined: 02.05.2016, 14:58

Re: Grip: two issues

Post by ismael »

Samsepiol - please don't feel. The matter is complex, and the text development is so long here that I had already forgotten previous topics and responses. But thankfully we got TdK onboard.
samsepi0l wrote:
10.09.2021, 00:40
if I wanted to have the same horsepower for all the cars, and just change the drivers so some are faster/slower than others- can't I just use the grip values?
Yes, at least I think that yes. It would be something like F3000 or GP2 (as far as I now), everyone has the same power. Their "grip" is what differentiated them.

samsepi0l wrote:
10.09.2021, 00:40
If I give everyone the same "range" of grip values, it will give everyone equal "difference" in how much they have a fast day, or a slow day right?
Yes, keeping in mind that the game will randomly give some performance level for each driver inside his grip range, so they won't have all the same level in a given weekend. Unless you give a grip range of 0 to everyone - that's the experiment that TdK has done and he described it before, in this same thread. (I strongly recommend reading him)
And there is also the "Random Power Variation", which, as far as I know, also changes a little the grip of each driver from weekend to weekend.
Claudio
GP2 Newbie
Posts: 11
Joined: 10.09.2020, 21:02

Re: Grip: two issues

Post by Claudio »

TdK wrote:
29.05.2021, 19:15
Well in the past 10 years in my absence of Gp2 i discovered something about the random grip:

It works both ways. If the Random Grip is 0, the driver will always produce about the same lap time.
If you give a driver Random Grip range of 938 (Schumacher '94 Microprose), it's grip will be between 15915 and 16853. His standard grip is 16384.
So he can perform better, and worse. Previously i thought it only worked negative. But as it turns out, it works positive as well. So the 938 value is a value effecting the 16384 level. Positive and negative.

Also i discovered that only Schumacher's Random Grip matters: The Random Grip (RG) of the other drivers are calculated from his Grip values.
An Example 1994 Microprose values:

GRIP: RG:
Schumacher: 16384 938
Hill: 15980 1342
Alesi: 15468 1854
Belmondo: 13846 3738

Maybe if you are very very sharp you will notice it already. Hill his RG is calculated from the difference in grip between Schumacher and him. The calculation goes like this: 16384 - 15980 = 404. 404 + 938 = 1342. Alesi the same: 16384 - 15468 + 938 = 1854. Belmondo: 16384 - 13846 + 938 = 3738.

I discoverd this only just last years summer. And i implemented it within my 2020 carset, with a few deviations, and it seems to work perfect. In the meantime i tested it on other seasons\years and it works fine. It maybe even works better than what i first used as RG values, each driver it's own random grip, instead of reflecting/feeding it back to the fastest driver. You can always higher or lower the RG value, like i did in my 2020 carset as well, to get a more realistic result with some drivers.

If you want i can upload some carsets with the new data regarding to Random Grip. I personally believe this is how it should be, and it works better.

Commenting on that 16384 is the highest grip: as it turns out now, some drivers will exceed that grip number. And if you give a driver a higher than 16384 level, they are hard to beat, you need to have more horsepower than them to beat them. But that can also be solved, or 'tuned' by adjusting the "Grip Factor" to a lower number in the "Car Settings" menu in Gp2Edit. With this only the player gets more grip. A little fine-tune value within Gp2.
Belmondo: 16384 - 13846 + 938 = 3738. the calculations are wrong, should give 3476. This is because Belmondo did not race every gp of the season.The simplest way to calculate would be 17322-grip=range. Drivers Gounon, Gachot and Belmondo are the only ones who won't crash with this rule. Because they didn't qualify for all the GPs.Now I would like to know how to work the parameters in gp 3 (grip and range).
Post Reply