GPL 1967 v3 Dev blog

[mcmirande]

Bottom post of the previous page:

Well, I think the "engine blowing pattern" is fairly well done. I don't know its programming details but I like the effect of some random blows mixtured with "well deserved" ones, hehe

Oh for sure, GPL is head and shoulders above many modern sims when it comes to engine damage modelling

The issue I mentioned is with mod EXEs only - something went wrong when Nigel Pattinson converted the code to XML. Try shifting into 1st gear at 300kph in the original GPL and the engine will blow immediately. Do it in any of the mods and you'll get lots of revs, maybe some locked rear wheels, but typically not much will happen to the engine. That's what I have basically fixed.

Ahhh, ok, great!!

[ReneLotus]

And here I was thinking I had become a very car-friendly driver! So basically in the mods you need to try really hard to blow an engine.
I'm interested to see how this change affects my driving in GPL. Great work in figuring it out.

[PTRACER]

And here I was thinking I had become a very car-friendly driver! So basically in the mods you need to try really hard to blow an engine.
I'm interested to see how this change affects my driving in GPL. Great work in figuring it out.

Even without the patch, you still have 6 other ways to damage your engine so drive carefully

[LasseA]

Even without the patch, you still have 6 other ways to damage your engine so drive carefully

How many ways are there overall?

[PTRACER]

Even without the patch, you still have 6 other ways to damage your engine so drive carefully

How many ways are there overall?

Besides damage to the suspension when having an accident, these are the primary damage types listed in the physics editor:

Overrev damage
Oil Temp Damage
Water Temp Damage
Oil Pressure Damage
Fire Damage
Propagation Damage
Random Damage

I have never investigated any of these in depth, but this is how I guess it would work based on various things I've seen:

Overrevving, oil temps and water temps - Below the specified limits there is no damage. Above that, the amount of damage it does probably depends on how far over the limit you are. i.e. Your engine will fail sooner at 20C above peak oil temp than at 10C above.

Oil pressure - Max pressure in game is 79 PSI, so maybe this is damage from low oil pressure. Which would explain why blow ups can happen when warming up on the grid.

Propagation damage - I never knew what it is, but the word "propagation" suggests that damage in one area can affect another. So perhaps this is an overall counter of some kind. Meaning all the damage you have done is combined into one figure. A combination of overrevving, overheating oil and water could cause a failure sooner than any one of those things could on their own.

Random damage - Again, no idea how it's calculated because I haven't looked. If this is a probability based system and damage physics are calculated once per second, then based on the parameter, it looks like there is a 100% chance of a random failure occurring within 20 hours of driving. Someone once said on SRMZ that physics are calculated 4 times a second, in which case a failure is guaranteed after 5 hours of driving. Or 50% chance after 2.5 hours. Might have implications in the 1937 mod, if they have 3-4 hour races.

Besides the strength of the suspension which can be specified per chassis, the values above are applied equally to all cars. So, when we talk about Hondas blowing up before Brabhams, it's only because they can overrev further and run at higher temps than the Brabham can.

[LasseA]

Oh nice, even when there is some room for speculation about how every parameter really works, this really brings some light into the dark. So glad the public gets to see this info... I mean GPL veterans are still guessing how damages exactly work after all these years. Thanks so much for all of this, it's really appreciated.

[PTRACER]

I put another 12 hours into this at the weekend. I made some progress.

I found when you extend the nose collision box beyond the default value of 2.2 metres, the car can get stuck in the ground/launched into the shadow realm, especially when landing on big jumps. Luckily I found the cause (altitude difference between the nose box and the main body collision box which is present in the original game).

I also found some interesting parts in the code related to the suspension and the resistance of different racing surfaces. So, I can do things like making gravel harder to escape from (or easier), or make the car skim across the top of it. I find the way the car sinks into the gravel in GPL a little weird.

The one thing I really wanted to adjust though was the artificial bumps. The racing surface has them but you barely feel them. On grass, it is obviously more noticeable. If I could increase the bumpiness of tarmac it would make driving a lot more interesting, since so many tracks are billiard table smooth and that takes away from the experience somewhat. Sadly I failed to find the code for this. I will keep looking.

[PTRACER]

Another update - I have partially decoded the brake fade patch from the 1955 mod and have tested it in 67v3. Basically, it works!

However, my idea is to turn this patch around. Rather than having brake fade, I want braking ability to be reduced when the brakes are cold. The patch currently allows full braking ability from totally cold up until their peak temperature, then drop off above it to produce the fade effect. If I can then add a slight element of brake fade on top (i.e. after say, 10-15 laps of hard driving), it would be even better, but maybe beyond my ability. A fresh set of eyes would probably help. @Michkov Would you be interested in looking over it with me?

[mcmirande]

Wow, these are GREAT additions Paul!!

Thanks for all your effort!

Marcos.

[andygpl]

I follow this thread with great excitement. These are wonderful findings Paul and I am really looking forward for the new mod.

[PTRACER]

Glad to see you guys are still following this. A few weeks of silence doesn't mean I have given up, I guarantee I will see this through to the end

[Michkov]

Another update - I have partially decoded the brake fade patch from the 1955 mod and have tested it in 67v3. Basically, it works!

However, my idea is to turn this patch around. Rather than having brake fade, I want braking ability to be reduced when the brakes are cold. The patch currently allows full braking ability from totally cold up until their peak temperature, then drop off above it to produce the fade effect. If I can then add a slight element of brake fade on top (i.e. after say, 10-15 laps of hard driving), it would be even better, but maybe beyond my ability. A fresh set of eyes would probably help. @Michkov Would you be interested in looking over it with me?

Can do, send me a PM, or post here what you had in mind.

[PTRACER]

Another update - I have partially decoded the brake fade patch from the 1955 mod and have tested it in 67v3. Basically, it works!

However, my idea is to turn this patch around. Rather than having brake fade, I want braking ability to be reduced when the brakes are cold. The patch currently allows full braking ability from totally cold up until their peak temperature, then drop off above it to produce the fade effect. If I can then add a slight element of brake fade on top (i.e. after say, 10-15 laps of hard driving), it would be even better, but maybe beyond my ability. A fresh set of eyes would probably help. @Michkov Would you be interested in looking over it with me?

Can do, send me a PM, or post here what you had in mind.

Thanks. Give me a few days to convert it to something readable by humans and I'll share with you so we can discuss some changes.

[Michkov]

@PTRACER not sure if that is what you are asking for, but are you looking for the maximum brake force (y-axis) that looks like this for a given temperature (x-axis)? It's totally empirical, but it gives a low force at low temps, followed by a peak and finally a fade towards even higher temps. We'd need to adjust the numbers to get the behaviour we want, but there is only one input variable.

[PTRACER]

@PTRACER not sure if that is what you are asking for, but are you looking for the maximum brake force (y-axis) that looks like this for a given temperature (x-axis)? It's totally empirical, but it gives a low force at low temps, followed by a peak and finally a fade towards even higher temps. We'd need to adjust the numbers to get the behaviour we want, but there is only one input variable.

Something like this yes.

The best we can do is straight lines though. So, let's say when the brakes are at 0C, braking ability is 80%. At 250C, it's 90% and at 500C, 100%. Something like that. The current brake fade code is really strange and has some weird calculations that I need to work out the purpose of first, that is what I will need your help with in the initial stages.

[andygpl]

The best we can do is straight lines though. So, let's say when the brakes are at 0C, braking ability is 80%. At 250C, it's 90% and at 500C, 100%. Something like that. The current brake fade code is really strange and has some weird calculations that I need to work out the purpose of first, that is what I will need your help with in the initial stages.

I wonder what will be the effect of such mod for offline driving with the AI as we are probably not able to simulate such behaviour of the brakes for AI drivers. It might give the human driver a hard time for the first few laps. But I guess similar thing is already happening with tyre heating so maybe it will not be such a big deal.