Jump to content
DIRT 5 - Chat about the game and get support here. Read more... ×
Christmas Period - Codemasters Staff and Support Read more... ×

Archived

This topic is now archived and is closed to further replies.

[DiRT 4] Constructive feedback on the subject of car-feel and physics: The Ultimate Thread

Recommended Posts

SamRWD said:
versedi said:
https://www.youtube.com/watch?v=YLtEgWh_2mk

Source: https://www.reddit.com/r/dirtgame/comments/6l2pdt/dirt_4_is_one_of_the_most_technologically/
Every time I see this kind of reassurance of what's going on under the hood in D4's physics modeling dept I feel really sad how that potential got wasted.

IMHO Dirtfish could be used as a test bench, but as soon as all of that got applied to "random" stage generator someone thought that it needs to be simplified for average gamer to enjoy. Assists off or on. One way of doing it is by making cars just stick to the road more.

Unless it is surface parameters indeed, and engine just overemphasizes tyre bite on loose. 

Really hope to get devs involved in our discussion.
They are likely figuring out what can be done and what needs to be done before they make any official statements. As a company you don't want to go saying things in a rush that in the long run end up not being possible.

Share this post


Link to post
Share on other sites
SamRWD said:
versedi said:
https://www.youtube.com/watch?v=YLtEgWh_2mk

Source: https://www.reddit.com/r/dirtgame/comments/6l2pdt/dirt_4_is_one_of_the_most_technologically/
Every time I see this kind of reassurance of what's going on under the hood in D4's physics modeling dept I feel really sad how that potential got wasted.

IMHO Dirtfish could be used as a test bench, but as soon as all of that got applied to "random" stage generator someone thought that it needs to be simplified for average gamer to enjoy. Assists off or on. One way of doing it is by making cars just stick to the road more.

Unless it is surface parameters indeed, and engine just overemphasizes tyre bite on loose. 

I must admit I thought so too at first but now im not too sure. with telemetry I cant get the car to go over 1.5g on gravel, but in real life rally cars exceed 2.5 -3.0 gforces when cornering on gravel.

Dis anyone manage to exceed 2.0 -2.5g on gravel in Dirt 4? and what are the usual g forces people are getting in Dirt4?

Share this post


Link to post
Share on other sites

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


Share this post


Link to post
Share on other sites
Gordouxk said:

https://www.youtube.com/watch?v=5hs-oPzj_9M


For 1:43 and 2:28 it looks more a like a G-Meter glitch than anything else, you can see the point coming back very quickly to the center while the car is still under the same forces. Otherwise the car barely gets to 1.5g the whole stage even on huge breaking zone and tight corners.

Here is some more material, fiesta r5 tarmac at Ypres, high grip, high speed breaking zones, and the car doesn't exceed 1.5g :
https://www.youtube.com/watch?v=0RUIcESAMI8


Share this post


Link to post
Share on other sites
Accro2008 said:
Gordouxk said:

https://www.youtube.com/watch?v=5hs-oPzj_9M


For 1:43 and 2:28 it looks more a like a G-Meter glitch than anything else, you can see the point coming back very quickly to the center while the car is still under the same forces. Otherwise the car barely gets to 1.5g the whole stage even on huge breaking zone and tight corners.

Here is some more material, fiesta r5 tarmac at Ypres, high grip, high speed breaking zones, and the car doesn't exceed 1.5g :
https://www.youtube.com/watch?v=0RUIcESAMI8


No I think the telemetry is working good. I got a book about rally from the library once and it it said that rally cars get nearly as much  grip and as much GeForce as f1 cars on some gravels. When the car turns all the weight goes to the side and the wheels go deep into the gravel and the grip is very high . It's like the car is a train on rails and it becomes hard to turn as well because the tyres are kept in a groove by the forward movement of the car.

I can't remember the name of the book but when I go to the library I will check if it's there and I'll borrow it again.

When I get time I will also post more videos with telemetry.

Also I can't watch your video because it has been disabled but it looks like it's on tarmac, which is different because the tyres can't sink into it.

Share this post


Link to post
Share on other sites
Gordouxk said:


No I think the telemetry is working good. I got a book about rally from the library once and it it said that rally cars get nearly as much  grip and as much GeForce as f1 cars on some gravels. When the car turns all the weight goes to the side and the wheels go deep into the gravel and the grip is very high . It's like the car is a train on rails and it becomes hard to turn as well because the tyres are kept in a groove by the forward movement of the car.

I can't remember the name of the book but when I go to the library I will check if it's there and I'll borrow it again.

When I get time I will also post more videos with telemetry.

Also I can't watch your video because it has been disabled but it looks like it's on tarmac, which is different because the tyres can't sink into it.
Then how do you explain this, the car is still turning right but the g-meter shows forces in the opposite direction. 

Also there is no trace what so ever of ruts and very little gravel in this onboard, it's just a typical finland gravel road, with a very compact and smooth surface.
And thinking that G-Forces of rally cars on gravel are on the same level as f1 cars is just ridiculous, even 2017 wrc (with high downforce aero package) can't pull 4g in corners on gravel... The only time it could come close is at the landing of huge jumps or coming to a stop after hitting a tree.

The video is not disabled, it is just watchable on youtube, here is the link : 
www.youtube.com/watch?v=0RUIcESAMI8

Share this post


Link to post
Share on other sites
"rally cars get nearly as much  grip and as much GeForce as f1 cars on some gravels. When the car turns all the weight goes to the side and the wheels go deep into the gravel and the grip is very high . It's like the car is a train on rails and it becomes hard to turn as well because the tyres are kept in a groove by the forward movement of the car."

Even if that was true it still doesn't apply to any location we have in D4. Even Sweden is way too grippy and I compared to RL on boards a lot. 

Share this post


Link to post
Share on other sites
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?

Share this post


Link to post
Share on other sites
fab1701 said:
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?



















@Fab, At 1:43 it could just be the driver correcting the car by  steering into the opposite direction which caused the g force to go up at that moment.

At 2:28 it does look like a different surface, but maybe its a softer gravel made of smaller stones, but I don't think its tarmac. You can see that the driver turns very hard left at this point, which explains the rise in g force.

I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.

I must go and check what the book is called but it think its something like Rallye competition cars.

Share this post


Link to post
Share on other sites
fab1701 said:
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?


@Fab, At 1:43 it could just be the driver correcting the car by  steering into the opposite direction which caused the g force to go up at that moment.

At 2:28 it does look like a different surface, but maybe its a softer gravel made of smaller stones, but I don't think its tarmac. You can see that the driver turns very hard left at this point, which explains the rise in g force.

I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.

I must go and check what the book is called but it think its something like Rallye competition cars.

Share this post


Link to post
Share on other sites
SamRWD said:
"rally cars get nearly as much  grip and as much GeForce as f1 cars on some gravels. When the car turns all the weight goes to the side and the wheels go deep into the gravel and the grip is very high . It's like the car is a train on rails and it becomes hard to turn as well because the tyres are kept in a groove by the forward movement of the car."

Even if that was true it still doesn't apply to any location we have in D4. Even Sweden is way too grippy and I compared to RL on boards a lot. 
But isn't it true that modern rally tires can give even moree grip on snow than some tires on gravel and Tarmac?

Share this post


Link to post
Share on other sites
Gordouxk said:
fab1701 said:
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?


@Fab, At 1:43 it could just be the driver correcting the car by  steering into the opposite direction which caused the g force to go up at that moment.

At 2:28 it does look like a different surface, but maybe its a softer gravel made of smaller stones, but I don't think its tarmac. You can see that the driver turns very hard left at this point, which explains the rise in g force.

I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.

I must go and check what the book is called but it think its something like Rallye competition cars.
Yeah but it still looks a bit odd at 1:43. Maybe the instrument wasn't well centered? Also at 2:28 you can hear the tires squeak like on tarmac when the car enters the junction, so I'm pretty sure its tarmac. The thing is, even when its true, that one can experience very high g-forces on gravel, the average forces may be well below that. Still, i think that the grip levels in dirt 4 seem to be on the realistic side, because the braking is. Would be helpful to find that book again ;)

Share this post


Link to post
Share on other sites
fab1701 said:
Gordouxk said:
fab1701 said:
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?


@Fab, At 1:43 it could just be the driver correcting the car by  steering into the opposite direction which caused the g force to go up at that moment.

At 2:28 it does look like a different surface, but maybe its a softer gravel made of smaller stones, but I don't think its tarmac. You can see that the driver turns very hard left at this point, which explains the rise in g force.

I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.

I must go and check what the book is called but it think its something like Rallye competition cars.
Yeah but it still looks a bit odd at 1:43. Maybe the instrument wasn't well centered? Also at 2:28 you can hear the tires squeak like on tarmac when the car enters the junction, so I'm pretty sure its tarmac. The thing is, even when its true, that one can experience very high g-forces on gravel, the average forces may be well below that. Still, i think that the grip levels in dirt 4 seem to be on the realistic side, because the braking is. Would be helpful to find that book again ;)

I think maybe also the g force meter might not be perfectly matched with the video like it could be 0.1 or 0.2 of a second out. Also the g force meter in the video looks to be working properly and is centred. Like when the car is stationary the dot is in the centre and then as the car takes off it goes back to 1.0 g which is realistic. Also the dot seems to be moving properly based in what the car is doing.  

And I agree that that the grip levels in dirt4 are on the realistic side and I'm very pleased with my purchase. When I first read this forum I was not sure but now that I have the game I can see that the grip levels and the behaviour of the cars is very realistic. Much better than Dirt Rally in every way.

I think I will go the library on Saturday to get the book.

Share this post


Link to post
Share on other sites
Gordouxk said:
fab1701 said:
Gordouxk said:
fab1701 said:
Gordouxk said:

I didn't find any videos of rally cars getting past 3g but there are some that show rally cars getting close to 2.5g on gravel.

this video show an old Audi Quatro being driven at Rally Finland on gravel and you can see g force getting close to 2g at 1:43 and at 2:28 it goes to almost 2.5g. This is more than on tarmac so it shows that gravel gives very high grip.

https://www.youtube.com/watch?v=5hs-oPzj_9M


I totally agree that in special situations the grip on gravel is higher than one might think. At 1:43 you are right, as the car addionally goes a bit uphill, the g-forces go up too, like it leans into the slope gets slowed down additionally. At 2:28 though, doesn't the surface actually change in the junction?


@Fab, At 1:43 it could just be the driver correcting the car by  steering into the opposite direction which caused the g force to go up at that moment.

At 2:28 it does look like a different surface, but maybe its a softer gravel made of smaller stones, but I don't think its tarmac. You can see that the driver turns very hard left at this point, which explains the rise in g force.

I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.

I must go and check what the book is called but it think its something like Rallye competition cars.
Yeah but it still looks a bit odd at 1:43. Maybe the instrument wasn't well centered? Also at 2:28 you can hear the tires squeak like on tarmac when the car enters the junction, so I'm pretty sure its tarmac. The thing is, even when its true, that one can experience very high g-forces on gravel, the average forces may be well below that. Still, i think that the grip levels in dirt 4 seem to be on the realistic side, because the braking is. Would be helpful to find that book again ;)

I think maybe also the g force meter might not be perfectly matched with the video like it could be 0.1 or 0.2 of a second out. Also the g force meter in the video looks to be working properly and is centred. Like when the car is stationary the dot is in the centre and then as the car takes off it goes back to 1.0 g which is realistic. Also the dot seems to be moving properly based in what the car is doing.  

And I agree that that the grip levels in dirt4 are on the realistic side and I'm very pleased with my purchase. When I first read this forum I was not sure but now that I have the game I can see that the grip levels and the behaviour of the cars is very realistic. Much better than Dirt Rally in every way.

I think I will go the library on Saturday to get the book.
Yes you're right, i just noticed the same thing! The video and the g-forces aren't well synchronized. About the centering: I didn't mean the calibration, but the location of the sensor in the car. But with the delay in mind it makes totally sense.

Share this post


Link to post
Share on other sites
Accro2008 said:
Gordouxk said:


No I think the telemetry is working good. I got a book about rally from the library once and it it said that rally cars get nearly as much  grip and as much GeForce as f1 cars on some gravels. When the car turns all the weight goes to the side and the wheels go deep into the gravel and the grip is very high . It's like the car is a train on rails and it becomes hard to turn as well because the tyres are kept in a groove by the forward movement of the car.

I can't remember the name of the book but when I go to the library I will check if it's there and I'll borrow it again.

When I get time I will also post more videos with telemetry.

Also I can't watch your video because it has been disabled but it looks like it's on tarmac, which is different because the tyres can't sink into it.
Then how do you explain this, the car is still turning right but the g-meter shows forces in the opposite direction. 

Also there is no trace what so ever of ruts and very little gravel in this onboard, it's just a typical finland gravel road, with a very compact and smooth surface.
And thinking that G-Forces of rally cars on gravel are on the same level as f1 cars is just ridiculous, even 2017 wrc (with high downforce aero package) can't pull 4g in corners on gravel... The only time it could come close is at the landing of huge jumps or coming to a stop after hitting a tree.

The video is not disabled, it is just watchable on youtube, here is the link : 
www.youtube.com/watch?v=0RUIcESAMI8

I'm pretty sure hitting a tree will generate much more than 4G.

Share this post


Link to post
Share on other sites
Gordouxk said:
I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.
Then if hitting 3g to 4g on gravel for a rally car on gravel is normal, I'm sure there is plenty of onboard videos with telemetry on gravel that shows that. Could you show us some ?
I think i need to go back to rallyschool, i forgot to take my third lesson...   :#

bogani said:
I'm pretty sure hitting a tree will generate much more than 4G.
Indeed, close to 30G for a fraction of second sometimes.

Share this post


Link to post
Share on other sites
Accro2008 said:
Gordouxk said:
I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.
Then if hitting 3g to 4g on gravel for a rally car on gravel is normal, I'm sure there is plenty of onboard videos with telemetry on gravel that shows that. Could you show us some ?
I think i need to go back to rallyschool, i forgot to take my third lesson...   :#

bogani said:
I'm pretty sure hitting a tree will generate much more than 4G.
Indeed, close to 30G for a fraction of second sometimes.
I don't think it's normal. I think normal is 1.2 to 1.8g but sometimes the g force can go even above 3 g according to the book. But in Dirt 4 I'm not getting above 1.5 very much unless the corner is banked towards the car and I throw the car in very hard and the surface is soft, so I think it's realistic in the game.  

Share this post


Link to post
Share on other sites
Some more proof that the G-Meter in the previous video is broken.
Same driver, same car, same rally, same roads and yet the g-meter never go over 1.5g :
https://www.youtube.com/watch?v=6u1FSK43zlk
And an interesting onboard with telemetry for comparison with Dirt4 with a Lancer evo 9 in Wales  :
https://www.youtube.com/watch?v=eIgz7TF9UCQ


Share this post


Link to post
Share on other sites
Gordouxk said:
Accro2008 said:
Gordouxk said:
I know that the grip on gravel can be very, very high because in the book it said that rally cars can get close to F1 cars with their g forces on gravel, and I rememeber that part well because I couldn't believe it at the time.
Then if hitting 3g to 4g on gravel for a rally car on gravel is normal, I'm sure there is plenty of onboard videos with telemetry on gravel that shows that. Could you show us some ?
I think i need to go back to rallyschool, i forgot to take my third lesson...   :#

bogani said:
I'm pretty sure hitting a tree will generate much more than 4G.
Indeed, close to 30G for a fraction of second sometimes.
I don't think it's normal. I think normal is 1.2 to 1.8g but sometimes the g force can go even above 3 g according to the book. But in Dirt 4 I'm not getting above 1.5 very much unless the corner is banked towards the car and I throw the car in very hard and the surface is soft, so I think it's realistic in the game.  
How old is that book? Not trying to be offensive or anything, just wondering, what kind of F1 is it comparing the rally cars with, because they pretty much increase the G-forces year over year. There's a video with 2017 F1s reaching 6.5G's, whichs is ridiculous, that's getting close to jet plane teritory
https://www.youtube.com/watch?v=OR5Tjfpi-Kg

Share this post


Link to post
Share on other sites
Accro2008 said:
Some more proof that the G-Meter in the previous video is broken.
Same driver, same car, same rally, same roads and yet the g-meter never go over 1.5g :
https://www.youtube.com/watch?v=6u1FSK43zlk
And an interesting onboard with telemetry for comparison with Dirt4 with a Lancer evo 9 in Wales  :
https://www.youtube.com/watch?v=eIgz7TF9UCQ


It doesn't look like it is broken. But it doesn't matter because in dirt 4 it's very hard to go above 1.5g on gravel unless the corner is banked. Also in the video you posted the drivers are driving very smoothly. They are trying very hard not to take off any speed in the corners. In dirt4 if I drive very smoothly I can't go above 1g. The only time I go above 1g is if I throw the car very hard into a corner and the tires sink into the gravel, but that loses a lot of speed, I don't think a good rally driver would drive like that. 

Also so in the bottom video you posted I see about 1.7 g at 3:11 and abou 1.3 to 1.4 g when braking a few times. 

Also I will post some more videos with telemetry. I'm on my mobile right now and I don't know how to copy and paste. 

Share this post


Link to post
Share on other sites
Guys just because something is written on paper doesn't make it true. Gravel grip on pair with F1? Cmn... We can do better than that. Even if that was true it doesn't apply to D4 locations. Or WRC rounds either. It would have to be a very specific local rally none of us heard about. I posted two videos showing deep ruts in gravel (recent rally Poland was one of them) and you could observe behavior somewhat similar to D4, except braking distances were much longer and cars could powerslide in those ruts with no problem. So even if we assume there are invisible ruts and banks in D4 grip is still too high all around. 

Share this post


Link to post
Share on other sites
SamRWD said:
Guys just because something is written on paper doesn't make it true. Gravel grip on pair with F1? Cmn... We can do better than that. Even if that was true it doesn't apply to D4 locations. Or WRC rounds either. It would have to be a very specific local rally none of us heard about. I posted two videos showing deep ruts in gravel (recent rally Poland was one of them) and you could observe behavior somewhat similar to D4, except braking distances were much longer and cars could powerslide in those ruts with no problem. So even if we assume there are invisible ruts and banks in D4 grip is still too high all around. 
But in the video I posted of the Audi Quattro in Finland the car doesn't power drift very much. Maybe this is because the grip is high and it would cost too much time to go sideways . It also explains why the g forces are so high in that video. So maybe it just depends on the type of gravel. A lot. 

Share this post


Link to post
Share on other sites
Is that Latvalas onboard? If it is he was saying he didn't use turbo. So not enough power to powerslide all the time. I am on mobile, so I cannot check myself. Or just check other on boards. Rally cars in general have no problem powerslidng at all. 

Share this post


Link to post
Share on other sites

×