Happy Holidays - A December Update

Dude, sounds like you should get outside more. I have done 25 000 steps through snow today, got from 1 to 11 strict pullups this year, cycling, walking, calisthenics, lifting weights. Thats all really awesome, I recommend it to everybody.

Would be cool to have a new video game that would be worth spending time while resting from physical activity and work. But there isn't. Lucky those who feel differently about games and technology of these days. To me it is indeed more like grinding and getting enslaved than freeing accessories of life.

 

Wow. Didn't take long for this thread to turn into something it's not. Merry Christmas and thank you for this excellent game. Peace.

 

I don’t know what you mean. I just said that because not everyone knows it and thinks rf2 FFB is most realistic, while in fact it has some canned effects. I believe Raceroom, ACC and I think iRacing FFB is purely physics based. I still subjectively prefer rf2 FFB feel (this and the tire model are basically the only advantages it has over the other sims for me) But it doesn’t mean rf2 FFB is the most realistic.

 

It's the most responsive and informative FFB, that's why it's considered "the best"...

FFB effects through software are made to inform you about what's going on with the car, since you don't have that "pants feeling", so i don't know what you mean "realistic", but if it's a real car wheel effect that would be mostly dumb and hard (depending on the car)...

 

ACC is the least physics based out of all of those. It might even be 100% canned effects.

 

What is rF2 adding to the physics?

The FFB output is pretty much a 1:1 match for the steering column torque, so any additions must be incorporated within the physics engine itself (a potentially troublesome approach, but possible). What have you seen/found that confirms this?

 

There are some effects configurable in controller’s JSON file I.e.
"Jolt magnitude":0.5, "Jolt magnitude#":"How strong jolts from other cars (or walls) are. Suggested Range: -2.0 to 2.0.", "Off-road multiplier":0.2, "Off-road multiplier#":"Temporary test variable to reduce force feedback strength off-road (0.0 = zero FFB, 1.0 = full FFB)", "Other spring coefficient":0.2, "Other spring coefficient#":"Static spring effect rate (-1.0 to 1.0) for any other FFB-capable controllers", "Other spring saturation":0.7, "Other spring saturation#":"Static spring effect peak force (0.0 to 1.0) for any other FFB-capable controllers", "Rumble strip magnitude":0.3, "Rumble strip magnitude#":"How strong the canned rumble strip rumble is. Range 0.0 to 1.0, 0.0 disables effect.", "Rumble strip pull factor":0,

 

Source?

 

Have you tested what these do?

 

Not that I want to make this thread about me, but I'm currently renovating a house, restoring a classic car, building a second race car to go rallying in, and also making a living as a pro photographer. So trust me I get out a lot... I do a lot of different things every week and meet loads of people from all walks of life (most of which I don't like lol)! I have a pretty full and satisfying life, even if I do occasionally get grumpy about how many modern day things have totally gone to sh*t.

I'm 35 years old dude, I've done many many things in my time and tried a lot of different stuff! I have 80 games on Steam, but one thing I like about rF2 is that I can boot it up and be enjoying it in minutes, and if I don't have much time I can play it for as little as 5 or 10 minutes and I get a pretty good fix. It just does what I need it to do. I could also put something like Counter Strike in a similar catergory to this. But boot up a mainstream AAA game, and I'm watching bloody cinematics for 10 minutes and it can be something like 30 minutes before you're actually into the core gameplay and you can actually decide if you like the game!

As far as trying racing games go, my system is as follows: choose a RWD car with 200+hp, turn off all assists, take it around a slow to medium speed corner, gently turn the wheel 90° and mash the throttle. If I don't think that what happens next is reasonably realistic, I uninstall it straight away and boot up rF2! I don't have time for things that don't suit my preferences. It dosn't mean it's a bad game, it's just not for me.

 

You were lucky to choose a good car when you did that with rF2 when you tried it.

 

I just meant that we surely have evolved as humans. Subtopic - we sure have evolved as simracers at evaluating quality, realism etc....

And when I use the word "evolved" I mean different word. Because as a general rule we mustn't mean the same thing as we say, because the second rule - we don't need to know what we are talking about.

 

Source?

 

Please guys, remember this thread is about....

 

For Raceroom it's shown in the game itself e.g. when loading a track. Plus it's pretty easy to find:
https://www.bsimracing.com/raceroom-brand-new-force-feedback-system-coming/
and also on official Sector3 forums.

For ACC it's a bit harder. I mostly seen several comments on forums that Stefano or Aris said that ACC FFB is purely based on physics. Of course you have the ability to enhance some Road effects with a slider. I think even original AC FFB was physics based but you also had 3 sliders with fake additional effects (slip, kerb, road effects).

https://community.granitedevices.com/t/assetto-corsa-competizione-and-simucube-2/2784/184
"As mentioned before by Aris and Stefano, we have no product-specific code going on in the FFB, hell we barely have an FFB code, the whole point is to transmit the physics without any artificial tinkering. You highly overestimate our capacities if you think we have time to write code for specific hardware (apart from supporting LED features but that’s not exactly FFB)."

https://www.racedepartment.com/threads/assetto-corsa-1-competizione-ffb-comparision.159514/
"Having followed the various FFB discussions in the ACC forums, I'm under the impression that the Dev's haven't ruled out implementing more FFB settings but, they want to keep feedback as close to being on a level playing-field as possible and that requires keeping things simple in terms of user-adjustments. Basic physics-based FFB is what we get in ACC (for now) and I think that makes sense in terms of player-feedback at this point in development"

https://www.bsimracing.com/assetto-corsa-competizione-sro-e-sport-series-trailer/
"Stefano claims, the FFB can't be improved, because it's pure physics based and he can't tweak the outcome."


"Kunos have explained their approach to FFB and keeping it "pure" and why it is like it is. I support this approach totally and find as a result the FFB feel more real to me than any other sim (even if it lacks some of the canned effects of other sims). I'm using a Fanatec clubsport 2.5 and McLaren rim and honestly find I have all the FFB information I need to drive on my personal limit and enjoy the game."

As for iRacing I also saw it somewhere in forums, but don't have time to search now.

 

Sorry my last off topic reply:
Couple of them, yes . There are descriptions for each effect e.g.:
"Rumble strip magnitude#":"How strong the canned rumble strip rumble is. Range 0.0 to 1.0, 0.0 disables effect.",
"Rumble strip pull factor":0, How strongly wheel pulls right/left when running over a rumble strip. Suggested range: -1.5 to 1.5.

For sure I tested "Rumble strip pull factor" and it gives quite strong pull effect into a rumble strip.
Interestingly: "Jolt magnitude":0.5, "Jolt magnitude#":"How strong jolts from other cars (or walls) are.

changing this one didn't alter FFB from hitting walls for me, so perhaps some of them are unused

 

Paul Jeffrey enters a random building and goes into the elevator:
-Good morning.
-But ACC, what about the AI? We need a hybrid implementation!

Next day Paul Jeffrey took the stairs.

 

LMAO!

 

Hahahahaha love it and soooo true!

 

Small disclaimer: this post is NOT about what's realistic what's not. I don't know, I don't care.

Ok, I know where the misunderstanding is coming from.
We know there are two "famous" approaches to making a physics engine: "physical" and "empirical". At least for rubber/tires physical tries to simulate the physical characteristics (to some point) of rubber as close as possible and get the outcome. Empirical knows what the outcome should be and creates artificial equations that grant this outcome. In simple terms.

iRacing, rF2, madness engine games (PC2, AMS2) use "physical".
AC (vanilla), ACC use "empirical".

Now, when we have FFB derived from a "physical" physics engine I'd call this physics based FFB.
When we have FFB that's derived from an empirical physics engine (at least in the way AC and ACC implements it) I wouldn't call it a physics based FFB. It may be derived from the physics engine they have, but the engine itself is kinda "canned" and so is the FFB.

The FFB effects themselves can be implemented manually or can use Direct Input builtin collection of simple and plain effects. All physical FFB implementation use manually steered FFB. All current racing games use manual implementation of those forces and effects (except two, more on that in a second). The Direct Input ones are used for instance in ETS2/ATS and some old simple games.

The smoking gun for all of this for me has been how the tires behave when car is stationary. Try rF2 and AMS2. You feel the rubber when rotating the wheel. You feel the moment when you cross the point of static friction. The rubber is springy when you release the wheel and the wheel moves back a little. I presume iRacing would do the same (although I never driven it so I wouldn't know for sure).

Now try this in AC/ACC. And voila! Stationary wheel feel is implemented using the simplest and sh&ttiest Direct Input damping effect (see screenshot, you can test it yourself if your wheel allows you to change the strength of this effect). This is as far from physical based FFB as it can be. This is ETS2/ATS level of FFB (the rest of AC/ACC FFB does not use those DI effects, only stationary car, I know this, but the fact they resorted to this is telling).


If your physics engine only simulates specific phases of the simulation using vast approximations everywhere else I wouldn't call FFB derived from it as physical based. Another smoking gun are those famous ACC slides where you loose control for a moment waiting for wheels to regain grip. FFB doesn't do much in those situations as well. Contrary to rF2 and AMS2.

Again, disclaimer: I don't care which of those games or physics approaches is better, more realistic, etc. Every model can be made very realistic or not realistic at all. This post is NOT about it.