Non-championship races vs. championship difference

Here you can post questions about GP2, your favourite GP2 things and errorreports/problems!
User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

In other words, if you follow the exact same circumstances after booting the game, you get the exact same results every time. If you run the same quick race, quit, then start a championship, then quit the game, start the same quickrace, start a championship with the same settings, you now have 2 identical championships, the results of each race will be identical. So the quickrace thing is not actually randomising the seed, its just changing it but you can replicate exactly what the results will be

We really need to find a solution because it's not practical entering into 10 different sessions every time you play the game just to ensure the seed has sufficiently changed.

I have a feeling that's why Coulthard tends to be DOMINANT on most of the gameplay videos on YouTube. Their championships are probably all from the same seed. I never remember Coulthard being dominant when I used to play on DOS

User avatar
rremedio
GP2 Star
Posts: 1424
Joined: 14.01.2007, 15:05

Re: Non-championship races vs. championship difference

Post by rremedio »

If you join a non-championship race and go directly to the race, the trick won't work. You need to go through qualifying and then go to the race once you have a grid (after both qualys being finished). If you go directly to the race, the game will generate a grid and it will always be the same, as well as the race result (assuming you don't race and cause the result to be slightly different).

If you just want to go to the race, just join both qualifyings first and use the "accelerated time" feature. That way you have different random generation each time (just don't forget that the seed for the first race is still fixed, so you should join a race and then leave before joining your actual race).

Also, I have not made tests targeting the random seeds on this, but there is a difference in behaviour when you chose "Race again on this circuit" when leaving a non-championship race. Not sure if it affects random seeds, but it doesn't reload the track, meaning it is a simpler loading process and maybe it doesn't change the seed.
"The Fox provides for himself. but God provides for the Lion" - The Marriage of Heaven and Hell by William Blake

Image

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

rremedio wrote:
01.10.2019, 14:12
If you join a non-championship race and go directly to the race, the trick won't work. You need to go through qualifying and then go to the race once you have a grid (after both qualys being finished). If you go directly to the race, the game will generate a grid and it will always be the same, as well as the race result (assuming you don't race and cause the result to be slightly different).

If you just want to go to the race, just join both qualifyings first and use the "accelerated time" feature. That way you have different random generation each time (just don't forget that the seed for the first race is still fixed, so you should join a race and then leave before joining your actual race).

Also, I have not made tests targeting the random seeds on this, but there is a difference in behaviour when you chose "Race again on this circuit" when leaving a non-championship race. Not sure if it affects random seeds, but it doesn't reload the track, meaning it is a simpler loading process and maybe it doesn't change the seed.
Ok I was going straight to the race without simulating qualifying.

The bad news, I've now run the same test again, I will explain step by step so it's clear.

- Booted the game
- Started quickrace at catalunya
- Accelerated time
- Started non championship race Brazil
- Accelerated time through all quali sessions and race

I now have different results to the first lot of tests. BUT, if I quit the game, reboot, and then follow exactly the same procedure, I get the exact same results. Again I tried it across 3 races, exactly the same results one after the other.

Surely this didn't happen in the original game? If the starting seed was actually random like gp3 and gp4 appears to be then I'm sure it would be impossible to replicate the same results on different boots.

It seems that if you follow the same pattern every time, the exact same thing will happen. Any more information will be appreciated.

User avatar
rremedio
GP2 Star
Posts: 1424
Joined: 14.01.2007, 15:05

Re: Non-championship races vs. championship difference

Post by rremedio »

I did the same test today and the results were different for each time. I used the "accelerated time" thing from the cockpit (not from the ESC menu), but it shouldn't affect it since it should give the same results for both accelerated time options.

I'm gonna run more tests later because I'm very intrigued now.

EDIT: ran more tests (under 0.74-3 and it is working fine for me, different results every time). If I was coding something like that, I would use the current date and time to calculate a seed (that way it would rarely be repeated). If GP2 does that, maybe DosBOX is failing to provide date and time under certain circumstances (maybe related to Windows version or DosBOX version idk). DosBox doesn't have the old DATE and TIME dos commands, this may indicate their handling of system dates and times may not be the best. But it is all speculation from me. I don't even have that issue.
"The Fox provides for himself. but God provides for the Lion" - The Marriage of Heaven and Hell by William Blake

Image

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

rremedio wrote:
01.10.2019, 15:14
I did the same test today and the results were different for each time. I used the "accelerated time" thing from the cockpit (not from the ESC menu), but it shouldn't affect it since it should give the same results for both accelerated time options.

I'm gonna run more tests later because I'm very intrigued now.

EDIT: ran more tests (under 0.74-3 and it is working fine for me, different results every time). If I was coding something like that, I would use the current date and time to calculate a seed (that way it would rarely be repeated). If GP2 does that, maybe DosBOX is failing to provide date and time under certain circumstances (maybe related to Windows version or DosBOX version idk). DosBox doesn't have the old DATE and TIME dos commands, this may indicate their handling of system dates and times may not be the best. But it is all speculation from me. I don't even have that issue.
Can I ask your exact testing method?

Are you still getting different results after re booting the game and following the exact same method? Because if I follow the exact same method, I will get the exact same results in the exact same sequence. If I change the method, the results change. But if I then repeat THAT process multiple times (by rebooting the game and replicating the same conditions) I get the same results every time

In other words, I do a test, quit the game, repeat the test identically, and get an identical result every time.

Even when I run 2 or 3 races, I record the results, quit, reboot, repeat the exact same method with all the same settings and parameters, and the results will be identical to the first test. Then if I change the method, the results change, but if I record the results and repeat the new method again, I get the same new results every single time. Even the session length and difficulty level affects it. But game mode (championship, non championship) doesn't affect it. I ran 2 championship races and 2 non championship races under the exact same conditions and got the same results.

Can you confirm if this is the case?

I'm going to do more testing, but selecting a driver and fast forwarding the time from the cockpit. Hopefully that solves the issue, I'll report back the results

User avatar
rremedio
GP2 Star
Posts: 1424
Joined: 14.01.2007, 15:05

Re: Non-championship races vs. championship difference

Post by rremedio »

Yes, I open the game, go to a quickrace, leave the quickrace, go to non-championship race and then go though qualy and race (all accelerated). Then I usually close dosbox, open it again and do it all again, but if I don't close dosbox, just leave the game and start it again the results are the same. I never get repeated results, including the results from the qualifying sessions.

I'm using GP2Lap, are you?
"The Fox provides for himself. but God provides for the Lion" - The Marriage of Heaven and Hell by William Blake

Image

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

Are you doing the same quickrace and non championship race each time?

I'm not using gp2lap, I'll re test with that as well.

EDIT

I can confirm that gp2 is picking up the date somehow. You can see the date in the advanced options menu
Last edited by GP2tifoso28 on 01.10.2019, 21:28, edited 1 time in total.

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

Ok, new test results.

This time, I selected a driver, Gachot, because he doesn't normally qualify for the races and is unlikely to affect the virtual reality.

Next, I set his grid position to 26, so that the AI don't crash into him when I accelerate the quickrace. Any change in the sequence of events seems to produce a different result.

Now I run the quickrace, accelerate the race, and go back to the main menu. I ran at catalunya.

Now, I run a non championship race, 50%, ace. I accelerate the qualifying from the cockpit both times. The track was interlagos.

Now I set the fuel stops to zero, start the race, exit, and accelerate. Save the results. Now I started a new identical race at the same track (NOT by hitting the button to race again on the same track), I go back into the menu and start the race manually. I repeat the same steps in the next 50% race (without running a quick race first)

Now I have 2 lots of 50% race data from a single game boot, after running 1 quickrace at catalunya with gachot from p26. I quit the game, and close dosbox.

Now I reopen dosbox, make sure Gachot is selected, start a quick race (from the same track as last time) in 26th. Accelerate the time. Now I start an identical 50% race (again at interlagos) with all the same settings, accelerate the qualifying in the cockpit, set the fuel stops to zero, accelerate the race, save the data. Now I do the second 50% race back to back, follow the same steps, save the data.

Now I have 4 lots of race data, 2 races on the first boot, and 2 races on the second boot, both after running a single quick race at catalunya at the start of each boot.

The results:
Race 1 boot 1
Race 1 boot 2
THE EXACT SAME RESULT

Race 2 boot 1
Race 2 boot 2
THE EXACT SAME RESULT

It's worth mentioning that these results are completely different from my first lot of tests with no driver selected. All other conditions of the first lot of tests were identical.

Any changes to the settings, proceeding results, difficulty, race length, quick race location, circumstances, seems to change the results. But if you repeat the identical circumstances, the same result will happen.

So based off this data, the seed is still not random. It's producing repeatable results every single time provided the exact same conditions are replicated from the time the game has been booted.

This test was conducted with gp2lap

Any more feedback, data, opinion, criticism, suggestion etc is much appreciated. Seriously I've wasted a full day of my life running these tests over and over, I'm this close to contacting Rene Schmidt because this is a serious problem

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

It seems my gp2 is always running on the same seed, which means every randomly generated number from the seed, is identical and in the same sequence every time if the conditions are replicated.

That is consistent with the data that I have collected. Any more feedback is welcomed. Unfortunately Rene Schmidt's forum seems to be down. I'm not sure how I can contact him

I even tried going into advanced options, saving the game state, then restoring the game state hoping that the game would then load on a different seed, but it's still the same

User avatar
samsepi0l
GP2 Starter
Posts: 120
Joined: 01.03.2019, 01:39

Re: Non-championship races vs. championship difference

Post by samsepi0l »

I woke up this morning and I was blown away by the attention this got.

I wonder if there is a way to alter the "seed" in memory so we can force more variation?

I personally would never use championship seasons, I do all my races one at a time and keep track of the results by spreadsheet. I don't even use the real F1 point-system (I invented my own which is different). I play 100% distance races but I still want to be able to have a realistic variation in the results so the same driver isn't winning every race.

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

samsepi0l wrote:
01.10.2019, 20:10
I woke up this morning and I was blown away by the attention this got.

I wonder if there is a way to alter the "seed" in memory so we can force more variation?

I personally would never use championship seasons, I do all my races one at a time and keep track of the results by spreadsheet. I don't even use the real F1 point-system (I invented my own which is different). I play 100% distance races but I still want to be able to have a realistic variation in the results so the same driver isn't winning every race.
Believe me, I've wasted a whole day running tests trying to find a solution. I'm going to try and private message Rene Schmidt on the grandprixgames forum. He's the only person who has an advanced enough knowledge of the sim to come up with an explanation or solution. Unfortunately, he's no longer active in gp2. If I have any luck I'll post it here.

In the meantime I encourage anyone else to conduct their own tests and report the results here.

Based off my current data, the same results are being produced every time the game is booted, provided the exact same identical conditions and sequence of events are performed each time. Any changes to the conditions or sequence of events will then change the reality. But it always resets back to the same start point when you reboot the game. Which means you can always replicate results if you replicate the process.

That is based off my data. Rrmedio is reporting different results so I'm going to try and re install the game and re run the same tests again. I have the results sheets saved so I can run the exact same test again and see if the result will be truly random.

Feel free to come up with any suggestions

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

I tried to PM Rene on gpgames but his inbox is full and his forum appears to be down. I don't know of any other way to contact him.

User avatar
rremedio
GP2 Star
Posts: 1424
Joined: 14.01.2007, 15:05

Re: Non-championship races vs. championship difference

Post by rremedio »

GP2tifoso28 wrote:
01.10.2019, 18:55
Are you doing the same quickrace and non championship race each time?

I'm not using gp2lap, I'll re test with that as well.

EDIT

I can confirm that gp2 is picking up the date somehow. You can see the date in the advanced options menu
Yes, always the same. Tested under Pro and Semi-Pro levels and same results: the second seed is always different.
"The Fox provides for himself. but God provides for the Lion" - The Marriage of Heaven and Hell by William Blake

Image

User avatar
rremedio
GP2 Star
Posts: 1424
Joined: 14.01.2007, 15:05

Re: Non-championship races vs. championship difference

Post by rremedio »

samsepi0l wrote:
01.10.2019, 20:10
I woke up this morning and I was blown away by the attention this got.

I wonder if there is a way to alter the "seed" in memory so we can force more variation?

I personally would never use championship seasons, I do all my races one at a time and keep track of the results by spreadsheet. I don't even use the real F1 point-system (I invented my own which is different). I play 100% distance races but I still want to be able to have a realistic variation in the results so the same driver isn't winning every race.
Maybe. Some things we can just change in memory and put new values while others need to have the program flow changed and for that, the best option would be to build this feature into GP2Lap. But that's way out of my league. I'm gonna try to find where the seed it to check what we can do when I have time and will lol. Not sure when it will be.
"The Fox provides for himself. but God provides for the Lion" - The Marriage of Heaven and Hell by William Blake

Image

User avatar
GP2tifoso28
GP2 Starter
Posts: 84
Joined: 29.09.2015, 14:21

Re: Non-championship races vs. championship difference

Post by GP2tifoso28 »

rremedio wrote:
01.10.2019, 21:32
GP2tifoso28 wrote:
01.10.2019, 18:55
Are you doing the same quickrace and non championship race each time?

I'm not using gp2lap, I'll re test with that as well.

EDIT

I can confirm that gp2 is picking up the date somehow. You can see the date in the advanced options menu
Yes, always the same. Tested under Pro and Semi-Pro levels and same results: the second seed is always different.
Ok, so you're getting different results every time.

Can you run 2 quickraces and record the results, then tell me the session settings and I'll see if I get the same results

Post Reply