Discussion in 'General Discussion' started by dadaboomda, Nov 4, 2017.
There is no consensus.
Interesting, source? You are not referring to QSA model “simplification” in
OK maybe not among pro drivers(especially when not too many of them compare sims).
But at least when it comes to tire model, consensus is the most advanced one is in rF2. I know that more complex doesn’t necessarily mean more accurate, but still you would think that with detailed, real input data you should get “deeper”simulation than with simplistic model.
Anyway I’ve been interested in (tire)physics models in major sims, and from the articles and forums I’ve read rF2’s is usually considered as one of best( if not best) in the business. Many people say AMS’ tire model is also one of the best and I can agree with that. It feels quite good, despite being simple.
Besides, I guess what’s important is what a sim user feels, and for me rF2 tire model (coupled with a good car) is most realistic and surely more fun to drive over the limit than in AC. Unfortunately mods in rF2 are inconsistent(especially third party ones) and manky of them don’t use CPM or have older CPM or something else feels wrong (I.e. aero). But I guess similar things could be said about mods in other games.
The AC has a few flaws and some very simple canned physics like flatspots. They are only sampled into the ffb and not related to the driven speed. Driving 200 or 300km/h does not create different vibration frequencies. Sliding sideways does not create flatspots, whats really stuff. You can´t notice vibration in the suspension geometry while driving with flatspots, so thats clear that there is no physical background. Driving with massive flatspots without loosing your theets is boring for me. In rFactor 2 you have to change tyres with massive flatspots, in AC you can drive with it like going into holiday.
By the way, AMS is much closer to rFactor2 than AC. What i can say ist that its really important using a real FOV to compare phyics. Using a single screen and not and VR HMD oder triple screen hides to many details.
The reason for many simracers choosing AC are, better graphics, more content, easier to drive for many people.
The street cars are not bad in the AC but i prefer race cars and compete to others and like shown in the video, Ad stands for drifting over all and tyre with massive slip curves so that even the GT cars slip sideways much to easy with very slow reactions caused by high inertia values.
I would say there is no guarantee that greater model complexity and increased real data accuracy will produce a more accurate representation of reality.
I would say the bigger factor would be the people generating the models, their understanding of real world physics and the techniques and theory with which they apply it.
OK so complex, accurate physics model plus good mods (like you would expect most of ISI/S397 are), should result in most realistic experience.
A model of physics + realroad etc. complex does not guarantee its realism, yes but.
But the most realistic simulation will always have a great complexity in its model physics and in all the other elements that affect realism in the global sense.
We can have tracks and fictional cars very interesting to drive.
But to get the highest realism (what I look for 99.999% in a real simulation), the real data (tracks + car) are totally essential.
The good thing about games/simulations is that there is something for everyone.
I would have just enjoyed having the choice between two simulations, but there is only RF2 which largely corresponds to what I am looking for.
Greater model complexity and increased real data accuracy WILL produce a more accurate representation of reality... Only if it's done right. However if you take a fantastic guy, passionate, knowledgeable, hardworking genius and give him little data and lots of data for another car. Guess which will end up being better. Sure, having more data is no guarantee of quality of a car/track. But if one is smart enough having more data ALWAYS ends up with faster made and better quality product.
Well yes. If one doesn't have really in depth education in physics, sometimes a common sense (which is not so common sadly) is good enough. However being a physics expert always helps and it is really needed to keep an open mind. I've seen professors in many fields that were as closed to fresh ideas as McDonald's for Burger King's coupons.
With lots of real accurate input data, does it really need to be genius physics modder if there's physics calculation template:
Sure. I've seen it more than once when someone screws it up even with full data. Or better yet, some don't want to use already done and correct tyres, because it's not made by them and they have to create their own (which usually end up being, well... crap). Pride leads to nothing sometimes.
But yes I still think one needs to be smart enough even with lots of data. Having a genius helps when you have less data and I actually would take genius one with half the data than a halfwit with entire data any day. Genius in my previous post is more of a hyperbole than anything to be honest.
Yes. It's not a matter of plugging in real data and it produces realistic results, unfortunately. One must understand how the data is used and the limitations of the models.
Try it for yourself. Place in as much accurate real data as you can and see how the car drives.
I thought that rFactor2 was the most accurate in this regard. I've read it somewhere( but maybe it was regarding rFactor Pro).
Otherwise what would be the need of the physic calculation spreadsheet if you need to input fudged data to produce realistic handling and physics.
Thats interesting. S397 already told they are not dumbing down cars. Just updating, probably, to get closer to reality (i hope)
Since reality is since reality is something "subjective" , I think it would be nice if motec users could give data-based answers whether the cars are getting any easier to drive or not. My general feeling is yes, cars are getting easier to drive.
Here is a source of ISI cars before s397 takeover for anyone interested:
I think you must not install both at the same time because folder structures and names. Maybe if you had both installed, they share some files
I m going to unsubscribe the car from steam workshop and install previous version from launcher.
I think that in general you don't even need telemetry for that. If you know car/track combo and after a change you achieve similar lap times and cars feels easier to drive then it's indication . I know it may not be that simple - there could be exceptions i.e. car is faster on straights and that's why it feels "easier" to achieve similar lap times.
Anyway it's hard to tell if car is more realistic unless you have a pro driver who drives the same car , with same setup, tires , track conditions etc. and has good experience in the sim(plus good sim rig). I don't think there was any comparison done like that maybe pro sims like rFactor Pro had something simmilar. For commercial sims there was some telemetry comparison in rFactor 1's and it was pretty close to reality.
Yes, i know. Thats why i said my general feeling is that cars are easier to drive. I´m not saying that about lap times, but almost flat out turns i can make in Interlagos with new fisi that i cant with old fisi. And gaining speed without spin exinting corners
Its not about most hard / slippery sim A is most real than less harder/slippery sim B. Its about any sim will be harder to control than real life because we only had visual and some ffb wheels. Real life we got the visuals, ffb, 3d, gforces, better sounds... so any sim that can reproduce real life data will be harder to master than the real thing (unless when we had extreme forces like f1, flight simulators, high speed cars.. which requires you be phisically prepared to deal with). Imo.
edit: unless you are Ari Antero with his fancy wheel and sim motion rig . But the vast majority of players use at most 3 screens and a steering wheel with ffb only
Separate names with a comma.