Jump to content

Is FFB ever going to get further refinement/improvements

Recommended Posts

Again, how many sims you have experience with?

I guess none, otherwise you would know no sim has perfect ffb while there are a few existing being able to smash DR2's FFB easily.  It's not because of CM devs laziness but due several reasons like available development resources, computing resources (remember DR2 runs on consoles too), target audience, ROI but at first complexity of tire model. At this point you should know the tire part is most complex one in car simulation and in turn most simplified in titles like this., because most studios cannot afford studying thermodynamics of tires to create complete approximation. So they implement simplified thermo-mechanics (forget about thermo part in DR) which must be then augmented. In the past by canned effects, now often by some physical inputs.

At the end most important is, simulated car MUST provide enough information to a simdriver considering lack of other ways to satisfy his senses. And back to the question about your experience in this area. Because it would say a lot about your expectations.

BTW do the experiment: get a car and try to run its one of front wheels over small stone or curb. Then come back to say if it causes steering wheel to turn or not.

 

Edited by MaXyMsrpl
  • Disagree 1

Share this post


Link to post
Share on other sites
6 minutes ago, MaXyMsrpl said:

Again, how many sims you have experience with?

I guess none, otherwise you would know no sim has perfect ffb. while thera are a few smashing DR2 easily.  It's not because of devs laziness but due several reasons like available development resources, computing resources, target audience but at first complexity of tire model. At this point you should know the tire part is most complex one in car simulation and therefore most simplified in titles like this., because most studios cannot afford studying it. So they implement simplified mechanics which must be then augmented by caned effects.

At the end most important is, if simulated car provides enough information to a simdriver considering lack of other ways to satisfy his senses

 

Like I said, the devs at codemasters are exceedingly blessed with people like you to set them straight and tell them exactly what canned effects are needed to satisfy your senses... and only your senses. 

Share this post


Link to post
Share on other sites

For sure they will not change their mind only because of you 😉

Personally I think current FFB is far from what I would expect. I didn't do direct comparison to DR1, but I'm sure DR1 was providing more information about what's happening between car and a surface. But for sure FFB right now is a bit better to what was available before patch4. To mi liking of course. You are free to feel differently. But don't mix personal preferences with mandatory requirement about how FFB should be made and the latter has been already established during recent 20 years of computer simulations

 

Edited by MaXyMsrpl
  • Disagree 1

Share this post


Link to post
Share on other sites
4 minutes ago, MaXyMsrpl said:

For sure they will not change their mind only because of you 😉

Personally I think current FFB is far from what I would expect. I didn't do direct comparison to DR1, but I'm sure DR1 was providing more information about what's happening between car and a surface. But for sure FFB right now is a bit better to what was available before patch4. To mi liking of course. You are free to feel differently. But don't mix personal preferences with mandatory requirement about how FFB should be made

 

The ffb is for sure a bit worse than what was available before patch 4. You are free to feel differently but don't mix personal preferences with mandatory requirements about how FFB should be made. 

  • Agree 1

Share this post


Link to post
Share on other sites
10 minutes ago, MaXyMsrpl said:

But don't mix personal preferences with mandatory requirement about how FFB should be made

 

The same applies to you, doesn't it? DR1 is said to have some FFB effects that don't come from the physics, so it's some kind of fantasy vibration. How's that supposed to give any information about what your car/tyres/whatever are doing? Also, people forget that SAT wasn't present in DR1, so why is DR1 providing more information than the current game? If you try to answer these questions you'll see the contradiction. 

Edited by richie
  • Agree 3

Share this post


Link to post
Share on other sites

It's not my preferences only. Content of FFB forces has been established during recent 20 years of FFB powered simulations existence.

BTW I've asked you ( @salvador ) for your experience in sims area. No answer means you have none. Not wrong or bad, but definitively influences your expectations. I asked you to do experiment with a real car. Hope you have a driving licence. Do it slowly to not damage/scrap a wheel. It will answer you a question about forces from bumps which might appear on steering wheel.

Edited by MaXyMsrpl
  • Disagree 1

Share this post


Link to post
Share on other sites
10 minutes ago, richie said:

The same applies to you, doesn't it? DR1 is said to have some FFB effects that don't come from the physics, so it's some kind of fantasy vibration. How's that supposed to give any information about what your car/tyres/whatever are doing? Also, people forget that SAT wasn't present in DR1, so why is DR1 providing more information than the current game? If you try to answer these questions you'll see the contradiction. 

Not at all. FFB might/should emulate torques coming from various sources. If one out of ten is missing, it doesn't mean FFB sucks. And likely will be better comparing to other FFB which simulates only 5 out of 10 (of course depending on what is missing).

Speaking about caned effects. Let's imagine asphalt which obviously has micro-bumps which we can feel it IRL. At first likely tire physics will not be able to process those bumps. At second generated forces are too low to drive weak FFB in our steering wheels (considering all torque is scaled down). Game developer has to decide if add such noise artificially and amplify it making present in FFB. Without this noise you cannot feel the road, cannot distinguish you are in touch with surface or airborne. So... some kind of augmentation is needed even if you don't like it. Even amplifying or scalling down in non-linear manner is a must with all FFB except of DD (which is obvious for you guys I hope). But it's still augmentation. Something fake, right?

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites

I never said it's a reference. In fact it lacks a lot too. I'm saying that in practice (regardless what was lacking or what was augmented/caned), DR1 was giving me more information about what is happening to a car comparing to its successor. Especially on tarmac but also on loose surfaces. For example in DR2, going over ice in Monte Carlo feels like getting airborne. It really sucks, regardless how much more advanced is FFB in DR2 comparing to Dr1. It's simply feels unrealistic, because in real going through ice makes no feeling of loosing contact with a surface.
TBH I don't know if SAT is most important thing. It's important to feel centring forces: yes. If they are calculated from physics or pre-calculated... if tuned properly I could live with this as long as such torque exists. Of course the closer to reality the better. But IMO lack of feeling a surface is something which is more impacting competitiveness than proper SAT.

BTW do you have a reference about missing SAT in DR1? I never seen that, so happy to read details.

BTW there are a lot more compromises in racing/rally simulations: exaggerated impact of mass transfer, breaking effects, oversteering effects or even simulating torque generated by driver arms. No need to argue those kills immersion since they are helpers. The best is to have options to tune those effects to own liking. But too much options is not always good (ie PC1) tought

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites
8 hours ago, richie said:

The same applies to you, doesn't it? DR1 is said to have some FFB effects that don't come from the physics, so it's some kind of fantasy vibration. How's that supposed to give any information about what your car/tyres/whatever are doing? Also, people forget that SAT wasn't present in DR1, so why is DR1 providing more information than the current game? If you try to answer these questions you'll see the contradiction. 

Dirt rally 1 had a very simple physics model and consequently the ffb was very simple as well. Consequently the devs added a lot of fake effects and vibrations to make it feel richer and to cover up the physics and ffb deficiencies. How DIrt Rally 1 ffb became the pinnacle of rally game ffb boggles the mind. Absolutely boggles it. 

  • Agree 1

Share this post


Link to post
Share on other sites
8 hours ago, MaXyMsrpl said:

It's not my preferences only. Content of FFB forces has been established during recent 20 years of FFB powered simulations existence.

BTW I've asked you ( @salvador ) for your experience in sims area. No answer means you have none. Not wrong or bad, but definitively influences your expectations. I asked you to do experiment with a real car. Hope you have a driving licence. Do it slowly to not damage/scrap a wheel. It will answer you a question about forces from bumps which might appear on steering wheel.

oh I have a driver's license, and several cars and motorbikes. And I've been sim racing with ffb wheels for more than a decade. Name a sim and chances are I've got it either on my ps4 or my PC right now. 

But even if I didn't drive and DR2.0 was my only racing game, how would that prove that the devs who made the ffb for DR 2.0 are incompetent idiots who need people like you to set them straight? What makes you better qualified than the devs over at codemasters to design a great ffb. Evidently with expertise like yours I'm surprised they haven't offered you a job yet. I truly am...not. 

Share this post


Link to post
Share on other sites
9 hours ago, MaXyMsrpl said:

Not at all. FFB might/should emulate torques coming from various sources. If one out of ten is missing, it doesn't mean FFB sucks. And likely will be better comparing to other FFB which simulates only 5 out of 10 (of course depending on what is missing).

Speaking about caned effects. Let's imagine asphalt which obviously has micro-bumps which we can feel it IRL. At first likely tire physics will not be able to process those bumps. At second generated forces are too low to drive weak FFB in our steering wheels (considering all torque is scaled down). Game developer has to decide if add such noise artificially and amplify it making present in FFB. Without this noise you cannot feel the road, cannot distinguish you are in touch with surface or airborne. So... some kind of augmentation is needed even if you don't like it. Even amplifying or scalling down in non-linear manner is a must with all FFB except of DD (which is obvious for you guys I hope). But it's still augmentation. Something fake, right?

You're just not getting it. Adding a canned micro bump/vibration effect covers up the more subtle and more informative aspects of the ffb. It effectively creates a mesh of forces on the surface of the stages that interferes with the feel for the tyres digging into the surface, thus giving a more pronounced top-of-the-surface feel. 

This is a rally sim and is meant to simulate driving over soft deformable surfaces. This is not a Tarmac simulator like assetto corsa which can get away with heavy bump effects. And even in that sim the bumps are physics-derived and there are no canned vibration effects.

Edited by Salvador17

Share this post


Link to post
Share on other sites

You are getting close to territory of saying Newton's formulas differ depending on a discipline you are competing in. Be careful.

IMO "rally" simulator applies to rules of the competition not to physics being applied. At the end rallying cars moves over various surfaces. Not loose only.
Similar mistake you are doing in case of racing simulations. Yes.. they focus on tire physics on tarmac, since this subject is so complex, no company is able to realistically simulate tire on loose surfaces to extent comparable to hard surface simulation. What you know from rally game is rough extrapolation (a.k.a. guess) comparing to physics formulas applied in "tarmac" based racing. Recently there are attempts to create universal advanced tire model (PC2). But it's possible thanks to computing power of recent PC technology. Different physics in racing and rally sims has nothing to do with differences in disciplines. It's just consequence of insufficient of resources.

So to repeat: loose surface simulations uses simplified tires interaction formulas to simulate (forget about thermo-dynamics, tire flex, flatspots etc). This fact is very important since it leads to a need of putting augmentation on next stages of simulation. Doesn't matter those are caned effects, forces artificially translated between objects, but still based on physics or values changed by some functions (like amplification of some forces). For example: if there is no tire flex, you have to cheat on suspension. BTW do DR2 simulates so much important body flex?

Keep in mind, that while those cars have to go over various surfaces, the physics must be maintained as consistent as possible between them. Also there are dozens of cars to simulate. A lot of those cars are road ones! Why they should behave differently just because mounted into rally game?
To make it all works you need consistent physical model to drive all of them on all surfaces using all sub-optimal equipment of gaming rigs. Compromises must be done.

And we reaching up to the end point: why do you think that rally cars don't need to output forces coming from surface structure ? If a car goes over a mud, there is a hard surface under loose surface, hit by a wheel hard every time. Those are not sand dunes. So why are you constantly suggesting to ignore those forces when creating FFB torques? It's felt in reality though no reason to skip it in simulation. Especially since it helps. And if you have experience with racing sims, you should know to what extent 'feeling the surface' helps to be competitive. Yes, those all forces/toques interferes each other. It's nothing new - it is how physics works. In real world too. I can get you have your own preferences. But you cannot say "it doesn't exists in reality"

BTW you are expressing the problem like those surface-based rattles were majority of forces generated by FFB. On my end those are still very weak, I can barely feel that (hopefully there are some). So definitely it will be an issue with configuration. For some reason I can feel it well balanced while you feel them too much destroying your immersion. So maybe this is a reason why we don't understand each other.

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites
24 minutes ago, MaXyMsrpl said:

You are getting close to territory of saying Newton's formulas differ depending on a discipline you are competing in. Be careful.

IMO "rally" simulator applies to rules of the competition not to physics being applied. At the end rallying cars moves over various surfaces. Not loose only.
Similar mistake you are doing in case of racing simulations. Yes.. they focus on tire physics on tarmac, since this subject is so complex, no company is able to realistically simulate tire on loose surfaces to extent comparable to hard surface simulation. What you know from rally game is rough extrapolation (a.k.a. guess) comparing to physics formulas applied in "tarmac" based racing. Recently there are attempts to create universal advanced tire model (PC2). But it's possible thanks to computing power of recent PC technology. Different physics in racing and rally sims has nothing to do with differences in disciplines. It's just consequence of insufficient of resources.

So to repeat: loose surface simulations uses simplified tires interaction formulas to simulate (forget about thermo-dynamics, tire flex, flatspots etc). This fact is very important since it leads to a need of putting augmentation on next stages of simulation. Doesn't matter those are caned effects, forces artificially translated between objects, but still based on physics or values changed by some functions (like amplification of some forces). For example: if there is no tire flex, you have to cheat on suspension. BTW do DR2 simulates so much important body flex?

Keep in mind, that while those cars have to go over various surfaces, the physics must be maintained as consistent as possible between them. Also there are dozens of cars to simulate. A lot of those cars are road ones! Why they should behave differently just because mounted into rally game?
To make it all works you need consistent physical model to drive all of them on all surfaces using all sub-optimal equipment of gaming rigs. Compromises must be done.

And we reaching to the end point: why do you think that rally cars don't need to output forces coming from surface structure ? If a car goes over a mud, there is a hard surface under loose surface, hit by a wheel hard every time. So why are you constantly suggesting to ignore those forces when creating FFB torques? It's felt in reality though no reason to skip it in simulation. Especially since it helps. And if you have experience with racing sims, you should know to what extent 'feeling the surface' helps to be competitive. Yes, those all forces/toques interferes each other. It's nothing new - it is how physics works. In real world too. I can get you have your own preferences. But you cannot say "it doesn't exists in reality"

BTW you are expressing the problem like those surface-based rattles were majority of forces generated by FFB. On my end those are still very weak, I can barely feel that (hopefully there are some). So definitely it will be an issue with configuration. For some reason I can feel it well balanced while you feel them too much destroying your immersion.

You're still not getting it. There are limitations to what a lever mounted to a torque inducing motor can communicate, and there are limits to the amount of information that the human brain can make sense of before it becomes noise. 

For that reason you can't just keep cramming effect upon effect and expect the whole thing to work. Instead, you have to discriminate and prioritise. So you can either have surface vibration or tyre flex. The more pronounced the surface vibration the less pronounced the tyre flex. Likewise the more pronounced the surface vibration the less pronounced will be the tyre slip. So if you want more tyreslip feel you have to live with less surface texture and vice versa. That's just the way it is. Similarly you can either have more surface vibration or more tyre digging into surface feel. If you want more of the latter you have to live with less of the former and vice versa. 

In the case of DR 2.0 the devs made absolutely the right decision in prioritising surface feel and tyre digging into surface feel over a canned vibration because the former is useful info while the latter is useless noise. It's just a shame that the whiners wanted useless noise and forced Codemsters to give it to them. 

Share this post


Link to post
Share on other sites

I think you constantly ignoring the fact repeated by me a few times: on my end surface-based vibrations are very weak comparing to other forces. What if this is only you who suffer from this? Or there are very few players over the world whos HW reproduce FFB in less balanced way?
IMO the problem is not because DR2 projects surface bumps to FFB but low dynamic range of our FFBs. Amplification of some forces dumps others down. This is the fact. But not a reason to completely disable some forces just because some player don't like it.
The FFB changed a bit (even with using configuration sliders)? It's possible. It's called a COMPROMISE which originates in weakness of steering wheels, since the game must support whole range of input devices.

or

Quote

and there are limits to the amount of information that the human brain can make sense of before it becomes noise. 

My brain has no problems with processing those information 😉

And I'm still not sure about 'digging into surface feeling being most important. It has importance only on low speed, while most track you are flying over the surface with high speed. While tire is still working toward those loossy surfaces, there are mostly no forces felt on steering wheel on high speed.  So most time this effect would be useless. But whatever.

Ahh And please be consistent. Don't call ffb effects caned just because you don't like them. IMO surface-based effects are not caned since they comes from surface mesh.

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites
16 minutes ago, MaXyMsrpl said:

I think you constantly ignoring the fact repeated by me a few times: on my end surface-based vibrations are very weak comparing to other forces. What if this is only you who suffer from this? Or there are very few players over the world whos HW reproduce FFB in less balanced way?
IMO the problem is not because DR2 projects surface bumps to FFB but low dynamic range of our FFBs. Amplification of some forces dumps others down. This is the fact. But not a reason to completely disable some forces just because some player don't like it.

or

My brain has no problems with processing those information 😉

And I'm still not sure about 'digging into surface feeling being most important. It has importance only on low speed, while most track you are flying over the surface with high speed. While tire is still working toward those loossy surfaces, there are mostly no forces felt on steering wheel on high speed.  So most time this effect would be useless. But whatever.

Ahh And please be consistent. Don't call ffb effects caned just because you don't like them. IMO surface-based effects are not caned since they comes from surface mesh.

It ia not a case of dynamic range or even subjective perceptions. When, for example, you turn a surface into a fine mesh of forces to create a vibration effect, those forces will interfere with the tyres sinking into the surface. It's not merely a matter of perception, the actual sinking in will not be properly simulated via ffb. That's why you have to prioritise and that's why DR2.0 launched without any surface vibration whatsoever, and that's why the subsequent addition of surface vibration produced a more pronounced top-of-surface feel. 

Edited by Salvador17

Share this post


Link to post
Share on other sites

It is the matter of dynamic range.

Mentioned forces interfere to significant extent only if ratio between them is unartistically low. And this is what happens in our gaming steering wheels.
Consider, in reality 2 forces generating torques differs by let's say 5Nm and additionally by frequency. Remember that such steering wheel has still headroom for greater torques (20 or even 30Nm). In such hardware configuration you can clearly distinguish those forces and their origin.

Now imagine what will happen on steering wheel which is able to generates maximum torque of 3Nm and is not able to reproduce signals with required frequency? At first those are compressed reducing their dynamics into available range. At second they will be even lowered down to leave a headroom for stronger ones (ie self-aligning ones). But cannot be lowered under some level because our FFBs are highly not linear in very weak torques range or even are not able to reproduce such low forces. In result mentioned to signals ends up in the same range let's say about 2Nm both. On top of that some forces will not be "replayed" or will be affected (interference, aliasing etc) by FFB response times.

And here we go: mess in forces, inability to distinguish their origin...  something you are talking about.

But, it's not because caned effects or wrong simulation. It's because our steering wheels are way weaker than real ones. Problem is, that to be competitive, we need all of those driving information on steering wheels, since we have no other way to get it. And some additional too, translated from other cars dynamics. And yes: you have to learn how to read them.
I can say "we" since after 20 years being in simulations world needs of community are pretty obvious to me. As you know already it's impossible to make all forces 'loud and clear' at the same time. Therefore devs have to make compromises, eventually providing all possible forces with options influencing a ratio between them given to a user. They're doing the best trying to pull maximum out of a game engine. At this point remember what I've said about imperfection of all simulations and reasons why must be augmented (inlc. your effect of digging into surface).

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites

BTW.. I think I understood you are expecting sinking into loose surface effect to be major because of your expectancy of what rally roads consists of. But TBH all rally tracks I saw in reality (and in media) are not butter-slick  ironing board covered by half a meter of soft sand. Most of them are very bumpy with a bit of loose surface over the top. Doesn't matter it's sand, gravel, dust. And if it's a mud, for sure the road is damaged, bumpy with stones. Simply, going over such surfaces is brutal.

So how did you get to idea of smooth FFB?

Edited by MaXyMsrpl

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×