The Sim Index

The "floaty" sense and problems with grip some mention - any similarity between rF2 and pCARS then? And about the fixed center point thing?

Center point still exists with pcars, but the problem with rf2 is that ISI have decided to ignore G27's, in fact, first rf2 build was undrivable on the road itself, let alone going over curbs etc, but a modified ini by and large fixed it.

The problem now is that the latest F2 feels horribly disconnected, so I now have to find another ini solution which will more than likely cock up every other car, baring in mind that Watkins Glen is still undrivable for me.

Also, overall progress on this game is bogus as well.

Netkar and other sims are deemed inferior by ISI forum genius's who either have a different wheel and experience, or who think that controller quality is 2nd to unusable FFB.
Anyway, I'm a patient individual and I've been patient with this game, but I now look forward to their forums being swamped by 1000's of dissatisfied G27 owners if they dare release this as is.
 
Center point still exists with pcars,

Going back to GPL and GP4, I can't remember a self respecting racing sim having this problem you guys describe. You name it: ISI based sims (any!), GPL, GP4, NR2003, Racer, XMR, LFS and iRacing, even open source ones, find one that exhibits that behaviour. So why would a dev like SMS, using Eero's work, produce "centre point pseudo-physics"?

Imho, this is an odd, shady area.


but the problem with rf2 is that ISI have decided to ignore G27's, in fact, first rf2 build was undrivable on the road itself, let alone going over curbs etc, but a modified ini by and large fixed it.

The problem now is that the latest F2 feels horribly disconnected, so I now have to find another ini solution which will more than likely cock up every other car, baring in mind that Watkins Glen is still undrivable for me.

[...]

Netkar and other sims are deemed inferior by ISI forum genius's who either have a different wheel and experience, or who think that controller quality is 2nd to unusable FFB.

Are you referring to ISI possibly neglecting the most basic aspects of the interface with G27s? Or are you referring to the FFB you have with G27s?

If it is FFB...

It seems the more I and a handful others explain to people FFB does not define the physics engine the more people mistakenly believe otherwise.

Give you an example, some people (not you, I think) seem to believe, rF's physics "miraculously" improved since the appearance of the realfeel plugin. I suggest to them "turn FFB off and try to judge rF's physics as you would have without FFB". The result is invariably the same: good physics are there, with or without FFB.

Handling? Indeed, handling changes because you have extra feedback from the sim (not just visual and sound cues) - but that is all.

Stefano once stated that NKP's FFB was 20 lines of code. Should we compare that with the implementation of most sims "combined grip" code (only a small part of the whole)? ;)


Netkar and other sims are deemed inferior by ISI forum genius's who either have a different wheel and experience, or who think that controller quality is 2nd to unusable FFB.

I don't know who you're referring to.

The most knowledgeable people about sim racing physics (especially tire physics and isiMotor) are those that have come all the way since F1C - Niels, Bristow and a couple of others.These guys don't deem other sims as inferior, I am certain they, as I, see the value in other physics implementations and respect them. Even pCARS. As a matter of fact, the "old timers" have already raised important questions about both the physics and the modding abilities of this current crop of sims (especially rF2).

The others, the loud types (we have them in every sim) are the ones that really don't know a tire from a rubber band. Let them have it their way.

As you said very well, one must be patient and give devs time to finish their job.:thumbsup:
 
So why would a dev like SMS, using Eero's work, produce "centre point pseudo-physics"?

I can only guess it's cheaper than a proper physics model, but I thought the whole point of the community funding was to pay for that aspect in particular.

As far as tin tops go, GSC Camaro is probably the best at feeling like it's got 4 tyres and almost no pendulum effect.....as good as RACEON US muscle cars are, they do have a bit of the pendulum effect at high speed tracks, granted they're 1800kg cars, but so is the GSC Camaro.

If it is FFB...

I mean that the F3.5 for example feels quite connected at low speeds, but the new F2's FFB all but disappears at times, and just feels downright bad compared to FISI/F3.5 etc, but this is more than likely due to my modified G27 ini....trouble is, it must remain modified as default rf2 FFB is undrivable.

I asked Tim to transfer my rf2 payment to rf1.....let's see what he does, my guess, nothing but continue to admonish me.
 
I can only guess it's cheaper than a proper physics model, but I thought the whole point of the community funding was to pay for that aspect in particular.

Hmmm...

A few years ago a co-worker (self appointed simracing aficionado of the project I was working on) bugged me on end until I tried Need for Speed Carbon. I tried it (I had sworn I would never touch a NFS title after my mental hiccup with NFS Most Wanted, go figure) and obviously failed miserably. The game had the most basic physics engine (a simple 3d kinematics representation) and basically everything rotated around a single centre point. Complex interactions of car and road were wasted, though, in a title like that.

Such a physics engine, in my humble opinion, is not only cheaper, it is a rotund fail for racing sims.

In my opinion, of course.

As far as tin tops go, GSC Camaro is probably the best at feeling like it's got 4 tyres and almost no pendulum effect.....as good as RACEON US muscle cars are, they do have a bit of the pendulum effect at high speed tracks, granted they're 1800kg cars, but so is the GSC Camaro.

Race On muscle cars feel good. Never noticed the pendulum effect and much less the centre point physics (properly used, isiMotor2 doesn't produce these effects).

Anyway, I reworked them. Inertias, drag, downforce figures, downforce dropoff with sideways movement, yaw multipliers, downforce COPs, suspension rates, springs, brakes - all this needed to be changed as per the data collected. Then came work on tires (dropoffs, slip curves, dry friction coeffs, speed effects, load sensitivities, camber influence on lat and long forces, temperature and pressure effects) - significant changes that bring very good results.

From experience, all I can say is that SIMBIN did a very good job with Race 07. :thumbsup:

I mean that the F3.5 for example feels quite connected at low speeds, but the new F2's FFB all but disappears at times, and just feels downright bad compared to FISI/F3.5 etc, but this is more than likely due to my modified G27 ini....trouble is, it must remain modified as default rf2 FFB is undrivable.

I see.

Well, fortunately for me (or is it unfortunately, who knows) I can live without FFB and because of it my wheels last considerably longer than those of friends. side effect is that I can concentrate solely on the pure physical side of things - FFB calibration comes later if at all.

I asked Tim to transfer my rf2 payment to rf1.....let's see what he does, my guess, nothing but continue to admonish me.

Tim W. is a good guy, his patience doesn't wear thin easily. Maybe he will attend to your request, who knows? ;)
 
Race On muscle cars feel good. Never noticed the pendulum effect and much less the centre point physics (properly used, isiMotor2 doesn't produce these effects).

Centre point physics only apply to pcars, however, some simbin cars and even the f1800 from netkar give me the impression of a degree of pendulum.
But netkar overall is extremely good, as is simbin, but simbin have so many cars, there's bound to be some that miss the mark a bit.
 
Even Codies gave up on centre point physics some time ago. Shift's physics engine was basically ISIs physics engine with extra features like turbo modelling and a dynamic tyre model to replace the Pacejka-based tyre model. I'm not a member of pCARs but don't believe for an instant that SMS would develop a centre point physics engine.

People are entitled to their opinion but it would be nice if they stuck to subjective comments like - "I prefer sim As physics model", maybe with reasons such as "IMO sim Bs tyre model is too forgiving." But unless you really know the physics engine "under the hood" its best to avoid getting any more technical.

I can remember Niels Heusinkveld visited the RSC forums from his usual haunts in the LFS forum loudly declaring rFactor was arcade because of the flaws in its physics model. When he actually took time to understand the physics engine he changed his mind and is now hailed as an ISI physics "guru".

The term "arcade" or its little brother "sim-cade" and "semi-sim" should be banned in all sim forums. Lets be honest, they are intended as insults rather than descriptions. The term arcade should be resticted to MarioKart forums where it belongs.
 
As I can't edit the post above...

Even Codies gave up on centre point physics some time ago. Shift's physics engine was basically ISIs physics engine with extra features like turbo modelling and a dynamic tyre model to replace the Pacejka-based tyre model. I'm not a member of pCARs but don't believe for an instant that SMS would develop a centre point physics engine.

Exactly my thinking.

People are entitled to their opinion but it would be nice if they stuck to subjective comments like - "I prefer sim As physics model", maybe with reasons such as "IMO sim Bs tyre model is too forgiving." But unless you really know the physics engine "under the hood" its best to avoid getting any more technical.

I'd bet we think similarly on this, but I don't know if by "really know the physics engine "under the hood"" is meant as understanding how it operates, what it requires, what are its effects, its virtues and flaws, or if you are referring to the source code of the physics engine.

I don't know the source code of isiMotor2's physics engine, or XMR's or Racer's. But we certainly don't need to be a programmer to actually understand the underlying physics of any racing sim.

You have a point, one that you, me and others have tried to make people see: if people don't really understand the physics engine, or even at least the physics involved, people ought to refrain from making evaluation-type comments pertaining to sim physics. Many people don't get this, though, and mix "handling" and "physics" and believe they can produce worthy comments on the subject.


I can remember Niels Heusinkveld visited the RSC forums from his usual haunts in the LFS forum loudly declaring rFactor was arcade because of the flaws in its physics model. When he actually took time to understand the physics engine he changed his mind and is now hailed as an ISI physics "guru".

It took some time, but after a ton of emails and PMs, a certain modder stopped criticizing ISI/SIMBIN physics and is now making delightful progress in several areas - to the point of letting go of iRacing and NKP. :D

This shift (no pun intended) happened in the last 12 months, so I can imagine how people resisted isiMotor2 based physics when it was relatively new.

The term "arcade" or its little brother "sim-cade" and "semi-sim" should be banned in all sim forums. Lets be honest, they are intended as insults rather than descriptions. The term arcade should be resticted to MarioKart forums where it belongs.

You know, back in 2001, we (flight simmers) used that term to refer to 2 different kinds of flight "simulations":
- those of the 80s arcades (oh yeah, REALLY arcadey back then or rather, the true arcade games)
- console type combat sims (Ace Combat and the like pre-2005)

Now, some people don't insult others in the vein of "idiot, imbecile, dumb", they just say "arcade gamer" - in reference to other simracers.

I saw dozens of posts at iRacing forum with this kind of nonsense, same thing at LFS forums - one such post referred to the Pacejka-era of NKP as being a "sham". Go figure...
 
Re-Sim racing.

IMO, a sim must have a mix of good physics and FFB, the FFB allows me to feel what the car's doing and allows me to learn to drive closer to the limit by the feel of it.
If the FFB is weak/limited or botched, how can I control the car at race speeds without guessing/memorizing tricky moments???

Do real life drivers guess that they can take sweepers at 240kph, or have they built up to that speed and "felt" the car near the limit of traction...I'm thinking it's obvious they inch their way to the limit, in fact, you can ytube many of the great drivers speak about this, so it becomes a battle near the "limit".....but without proper consistent feel, how could you go further without terrific risk...

Put simply, no matter the GFX, sounds, series, quality of MP etc, if a driving game has poor or limited physics and FFB, it can't be judged a sim, or at least a very good one because you're not really doing the driving, the computer is, and when we refer to driving sims, the sim aspect must be based on the driving.

People aren't elitist for wanting good physics and FFB, they're just plain jane sim racers, as it's the physics and FFB that define the game......GFX and sounds etc all help, but they're desirable extra's, not prerequisites to the games defining element.

Those who prefer GFX/sounds/fancy helmet views etc over physx and FFB aren't arcade snobs either, they're just arcade gamers, and nothing wrong with that, just as there's nothing wrong with wanting good physx/FFB.

Good news is that Assetto Corso could set a very high standard whereby physx/FFB, GFX, sounds and other elements of realism are met, such that we no longer have to distinguish racing games by GFX.
 
The RF2 tyre model is almost industrial grade in its intricacies and complexities, like a true proper peice of simulation software, rather than a video game sim, maybe this is whats giving him a hard time.

http://www.virtualr.net/simhq-motorsports-interviews-niels-heusinkveld

An interesting comment from some dude at VR.....trouble is, if someone with an engineering degree and responsible for GSC and some of rf1's best mods is having a hard time, imagine the trauma the typical modder would suffer.

To rate as a good piece of sim software, the first test is that the sim works precisely with my G27, and that the cars feel and drive in a believable manner, ie, based on my own driving{not racing} and what I see and read about race cars.

To me the solution is more factory built cars, ie, ideally in version 1.00 to encourage people to buy it in the first place, then if the sim rates well amongst the fans, paid DLC at the same quality as the original game.

As I've said before, I'd pay $20 for each of Neils cars for GSC 2012, just as long as they're my type of car, for example, C6R, race trim Camaro etc.
 
Interesting interview also that he prefers rF1, unfortunately I almost haven't spend any time at all with iRacing and rF2 yet so I can't say much about that for myself. Pretty much sums up the current situation for some people I know or read in forums though. iR and rF2 are not yet where a perfectly configured rF1 or other ISI sims are.

The problem is if you model a 225/45 michelin sports tire there's only one solution within the model that perfectly fits, a few which model a similar enough tire performance, and as he said there's no way to verify you actually have found or not found that compound. So tire forces telemetry and real time rig testing should be included, after all the "x at y" graphs can be extracted from any model at a specific time and condition. Entering "x at y" points will potentially create issues unlike the rF1 MF model which can't be ever inconsistent in physical model and parameters, because no (more or less) full physical model exists in the first place, instead the parameters are always right, wether they are actually right (realistic) or not.
 
Interesting interview also that he prefers rF1, unfortunately I almost haven't spend any time at all with iRacing and rF2 yet so I can't say much about that for myself. Pretty much sums up the current situation for some people I know or read in forums though. iR and rF2 are not yet where a perfectly configured rF1 or other ISI sims are.

The problem is if you model a 225/45 michelin sports tire there's only one solution within the model that perfectly fits, a few which model a similar enough tire performance, and as he said there's no way to verify you actually have found or not found that compound. So tire forces telemetry and real time rig testing should be included, after all the "x at y" graphs can be extracted from any model at a specific time and condition. Entering "x at y" points will potentially create issues unlike the rF1 MF model which can't be ever inconsistent in physical model and parameters, because no (more or less) full physical model exists in the first place, instead the parameters are always right, wether they are actually right (realistic) or not.

It's quite easy to do this with the TTool actually. And things like micro/macro friction is easy to relate to a slip curve (drop-off specifically) with a WLF equation and a simple brush model.
 
Pretty much sums up the current situation for some people I know or read in forums though. iR and rF2 are not yet where a perfectly configured rF1 or other ISI sims are.

.

The problem I'm having at ISI forums is people accepting rf2 as it is.....it'd be one thing if virtually everyone felt the same way as I do{at least with a G27}, but many claim it's fine/the best, so what incentive does ISI have to fix the FFB/phyx issues{on some cars}.
 
http://simtorque.wordpress.com/

A strange article on Sim vs Arcade, which oddly enough sounded like an apology for pcars state.

The article is strange as it considers ultra arcadey Dirt3 as a sim, even though you can slam on the brakes on dirt, under change of direction for no penalty.....try that in most sims and you'll be in a world of hurt as the game calculates the result of hard braking at the wrong time/manner and load point.

I want sims to be as realistic as possible, but I don't object to fun/arcade modes built into the game just as long as the sim side is genuine.

The best sims have the best FFB/physx, whereby both the FFB and phyx stand up to real world scrutiny, not that a sim has to be perfect, but then again, it's limited by our consumer gear and smaller market=smaller budget, so no sim will be perfect in 2013, but blend of the best of rf2/EVO and GSC together and you're not far off.

I like the fact that sims have a higher degree of difficulty that responds to good driving technique, IOW, the more precise you are with your driving{brake/gas/steering inputs}, the faster you can go.

I can understand people objecting to absurd difficulty levels{as I do}, but I can barely see the fun in allowing the PC to do most of the work....in a good sim, YOU have to learn to drive fast and control the vehicle, it's unforgiving and increases in difficulty the faster you go, but there's such awesome precision built in that if you're quick and precise enough, you can max out your ride, ie, reach your personal peak and maintain that pace.
 
Hi David,
Thanks for acknowledging my "strange" article :)
It is in no way an apology for the state of Project CARS. Although I touched on PCARS quite a bit, mainly because a lot of chatter gets devoted to that game and it's current state, I have no affiliations with that or any other game with which to be angling for an apology from.
My articles are based purely on my own opinion with no sway or bias towards anything. I hope that can continue to be true. I do invite your responses and feedback.
I have seen some great discussion here in this thread on the subject too.

To clarify the DIRT 3 issue, yes, it is arcade in as much as the sim-racers mentality labels it so. The point was that a lot of sim-racers label racing games in two boxes, arcade or sim. If they don't like the physics, its arcade. Simple. One or the other, no middle ground and the labeling of whatever game is simply dependent on how good the physics are deemed to be compared with games that are called sims.
However, DIRT 3 was termed as a sim in my article because it would be classed as one if you went by the dictionary definition instead of the sim-racers opinion.
You seem to have missed the point that bad physics doesn't necessarily make a game "arcade". It also points out the absurdity of such a term these days. Arcades are long gone.
DIRT3, does 'simulate' rally driving. Maybe not very well, but it still simulates it.
That is how it can be called a simulator. But I do realise that this term means something entirely different to the community than what it does to the dictionary.
 
You seem to have missed the point that bad physics doesn't necessarily make a game "arcade". It also points out the absurdity of such a term these days.


To me, the key elements of a sim are the combination of physics/FFB and obviously controller precision.......and even though a game like LFS sux on GFX, sounds and feels somewhat wooden on FFB at times, there's no question the aim of that game was a good mix of physics/FFB and controller precision, so I think most people accept it as a sim regardless of how good or bad it is these days.

Same deal with any racing game, especially something claiming to be a simulator, ie, the better the mix of physx/FFB/controller, the better the sim, cause real world cars feel alive under brakes, have the feeling of an independent front and rear, and aren't high on pendulumy/CPPP, and if you have GSC or GTR EVO well set up, you should notice these games are extremely good by that criteria.

Dirt3 operates well with a wheel, but comes with arcadey physics and average FFB, as such, I disqualify it as a sim as it's driving model is too forgiving and quite frankly, this should be common knowledge to anyone who owns cars, however, many people don't own cars, yet have firm views one way or the other, or they default to whichever paid pro driver's providing longwinded "forced" feedback.

Btw, glad to hear you're a independent entity.
 

Latest News

Online or Offline racing?

  • 100% online racing

    Votes: 76 7.3%
  • 75% online 25% offline

    Votes: 111 10.6%
  • 50% online 50% offline

    Votes: 151 14.4%
  • 25% online 75% offline

    Votes: 288 27.5%
  • 100% offline racing

    Votes: 418 39.9%
  • Something else, explain in comment

    Votes: 4 0.4%
Back
Top