Yeah the old LeMans looks stunning (and very atmospheric) with much better framerates than the SimBin tracks. Going over the humb before Mulsanne, properly airborne at 180mph I just love the whole presentation and feel of the LeMans track. The level of attention to detail and obvious time and effort spent optimizing it are something I haven't seen since GPL, take a look at the site: http://www.virtua-lm.com/lemans/index.php
As much as I like the Virtua LM Le Mans track, I wish they hadn't added artificial vibration effects to model undulations in the track surface along the straights. I understand that modeling the entire track surface in sufficient detail to capture subtle undulations would have been an enormous undertaking but the artificial effects mess up the RealFeel force feedback, in my opinion. Thankfully, they can be adjusted in the TDF file.
That C06 is better good actually, much better than I thought it would be, and much better than anything else I have tried in rF, especially with the RFFB plugin.
Feels fantastic.. But I don't know if it's just rFactor in general but all the cars, this car less so but still, have a horrible snapping tendancy when the weight shifts from once side to the other...
But great fun all the same, and with 900 degrees + G25 + realfeel plugin = a lot of fun.
Yes but because ReelFeel takes the data from the forces acting upon the steering rack the suspension needs to be realistically modeled using Carfactory, which is why most mods before the current releae of Caterhams are sucky with realfeel. Don't ask me why as I don't know the first thing about modding tweaking squirreling. I'm just going by what the guy who created the Realfeel plugin says.
CarFactory makes suspensions with certain characteristics that you enter:
- roll centre (and how much it moves about)
- roll camber rate
- travel..
- ackerman
- anti dive
and it just makes a geometry that has these properties. Since its just a sim and there are no physical bits getting in eachothers way, you might end up with long A arms that would in a real car, be mounted in the middle of the engine..... Luckily space is no problem for a sim..
One of those things I used to find about rFactor is how little tye tyres communicate with sound. Especially with GTR you'd hear that dreaded bad loop of a 'skid sound', almost regardless of the magnitude of your 'slide'.
I looked for a way to really make the tyres sound different from slight scrub to peak grip, but then still increasing in noise and pitch a LOT beyond this peak grip. A maniac slide or locked tyre tends to make a lot more noise than driving at the peak, yet I never really heard that in ISI!
Until more random parameters got tweaked and sounds edited!
Preview attached, its not about the absolute sound quality of the skid, but more to show the bigger dynamic range of tyre sounds.. notice the mistake at ~30secs (big slide) and the locked tyre at ~1:30 (not too pretty).. and the intentional sliding after 1:40..
Very early days, but chuffed I am for sure. If you look at rFactor as an unpolished early Beta version its not too bad!
Well, not to be harsh, but there are many addons/mods (what you call it) to LFS to, wasn't it your sounds that Scawen put into the V patch?
And that guy that made hi res textures for the tracks.
Something can be done better, but it doesen't mean that the devs are useless
Edit: Unless you did mean the rFactor devs, I'm not sure, anyway if you do, well they have provided a good base for modders to work from, but they are still making money off other peoples effort (the rFactor devs). I always remember one of the rFactor devs say something like "you can only drift a car with an appropriate setup" which is just not true..
I remember that and I remember also when I did show them videos of street cars drifting without special setups and how they agreed that it is very possible to drift such car
There is many people on their team and certainly not all of them shares same views so that causes some issues for sure in development and PR too.
LFS devs have easier situation, of course if they don't have already very divided personality, which surely can be with us making silly posts like this here
I think that I can paste here what rFactor dev did comment to me about drifting issues in rFactor:
Just WOW. Who the hell are these guys? Do they care about simulating stuff or are they in for the free beer or what? I'm tempted to think that statement was a bad joke from a frustrated developer or something...Oh well.
... wow...erm rather than join in on the rfactor dev bashing which is all-too-easy... i think i'm gonna do a few laps around suzuka with the corvette and. 5:21am... insomnia is awsome. i have uberlow expectations from rfactor devs... i regretted buying the game... but now... notsomuch, thanks to realfeel and carfactory. nvm i think i'll go for a run... its a beautiful cloudy rainsoaked day.
Also thanks to Niels and Bristow from hard work, even there was quite many nay sayers.
rFactor is not made with same princibles as LFS, imo their intention is not to create most realistic sim, but sim that is rather ok, but still such where reallife data would provide something that we would except.
I would say that there is not much of bashing going on, only few minor comments, more it is critics how their own mods(vehicles) are made quite quickly without such dedication what we see in LFS cars. Of course rfactor is made with different idea, it is just meant to be platform and cars devs have made are just samples, or so it at least seems to be.
Also about comment of rfactor dev, it is surely only single opinion from 10-12 people team, maybe bit bad PR management, but something you can't really take directly as it is because you don't know whole discussion.
Also we should remember how little each version has progressed so when looking on that light it is not so big of surprise if they have not tested streetcar drifting in their sim as they have made it to be racing sim with race cars in first place.
race drivers do mess it up from time to time though and the reason why they dont end up in a wall i not because they can countersteer before the tyre curve drops
the basic problem with rfactor is quite simply that its not physics based in many core parts
where lfs appears to have formula based behaviours and brute force physics models where forumlae dont exist rfactor mostly has a table in behaviour out method of working
with devs and modders having as little knowledge about physics as they apparently do leads to a garbage in garbage out phenomenon
looking at your and niels' work the underlying table analysis of rf doesnt seem all that bad if you feed it properly
Yes, pretty much all is true, also sometimes I get feeling that table does contain too little values, causing rough behaviour, but that is just feeling, can't really say if that is the case as I don't know sizes of tables at all.
Tables do work just fine for most of situations as seen on RBR, if tables are enough good resolution. Sometimes however there are possible to get into situation that is not thought and then everything can go berserk, that is biggest issue with tables, imo.
Anyway, what ever way we look it, rFactor is best platform to build your car into at the moment (Until we can make content to LFS it is about to remain so).
same thing can happen with physics or forumla aproaches
well kown examples are the collision physics of lfs and 90° dorfito in netkar ... push the forumla past the range for which it was fit or the physics model into a point beyond or between what its built for and things might turn out to be rather ... interesting
after seeing your and niels' work im starting to believe that this is in fact true and rf isnt all that bad after all
the problem with rf really is modders who knowingly or not turn rf into tdu or nfs (not necessarily a bad thing) and try to sell their cliff tyred front engined 911s as an accurate representation (this is where things turn bad)
i dont mind modders using rf as a platform to build arcadish cars as long as they are aware of what theyre doing and dont try to make users believe what theyre releasing is a simulation
LFS Collision physics are bit different matter actually, formulas do work perfectly or at least near perfectly, but what is missing is one crucial part of deforming. All energy is put now to motion, where IRL huge part of energy is spent to deforming of objects and most surfaces IRL do have friction too where in LFS there is many objects where is no friction at all.
I certainly don't know how netkar physics do work, all I know is that sometimes there can be really odd things going on, but I'm still hoping to see fixed version of it someday
That 911 part was funny as there is really lot of people that has voted such to be one of best in physics wise too, yet again one mod that is pretty and has nice sounds etc. However hardest part is made as well as those cars ISI has made to be as examples. But it is possible to fix, too bad often mod authors are not too receiving for new ideas how to make mods more realistic.
My work is really small part really, just spring and damper area and I guess I started something with playing tires a bit too, but it is nice to see more clever people have managed to point out issues and some ok(ish) mods are possible.
i being someone who was brought up to live in discrete time think the problem with collision physics happens at a much more fundamental level
to really make it work youd need some pseudo continuous time methods rather than just fudging things back to a workable state after you find yourself in a huge mess
Well, surely it might be that it is needed to be done completely from scratch, with proper damage system there is so many things affecting collisions too that current quick&dirty system is not sufficient I think. ut then again, I'm not a programmer, my brains like to take idea, twist, turn and mix it until it is totally weird, after that it is turned insideout. So I'm not really compatible with any programming where you need to be so logical
I'm not noticed that with my G25 - what MaxForce, SteeringDamper and Overall Effects Strength are you using? Are you using the [ForceFeedback] section from the controller.ini included with RealFeel?
With the G25 I find I can run with no damper (i.e. set SteeringDamper to the same as SteerForceInputMax) which makes it more direct and connected than with damping.
SteerForceInputMax is an interesting one. The default (in controller.ini) is 11500 (i.e. force sent to the wheel must be in the range -11500 to 11500) but I have discovered you can increase this as far as 65535. You must then set the same in the RealFeel.ini (so you get forces in the correct range) and set SteeringDamper also to 65535 (no damping, changes in force as high as 65535 are allowed through to the wheel).
This effectively gives the FFB system around 6 times more fidelity. On a DFP you would probably be hard pushed to tell the difference but with the G25 I'm almost sure it feels smoother. There is a caveat with this if you want to use any of the default FFB effects as they will still only output up to 11500 force and hence hardly be noticeable on the wheel (I'd say that's a good thing:P).
I just love the whole presentation and feel of the LeMans track. The level of attention to detail and obvious time and effort spent optimizing it are something I haven't seen since GPL, take a look at the site: http://www.virtua-lm.com/lemans/index.php
(Until we can make content to LFS it is about to remain so).