The problem is that the FF rate is tied to the frame rate. If you run VSync you're limiting the FF to 60Hz as well and it feels a lot different. Having said that it may be better to have consistency at 60Hz than for it to deteriorate when you have a lot of cars around you.
In what way does it feel a lot different Paul? Gary made the same point to me recently when I said I always use vsynch. But I've tried with vsynch off and tbh I can't feel any difference with my wheel or any of the other controls. I find the advantage of a smooth, non-tearing image much greater than any faster physics processing.
Gary also got me to try something called D3DOverrider. This enables triple buffering, which I think allows the game to run the physics faster even with vsynch enabled. Is that correct Gary? I have tried it (after having some trouble getting it to actually run on my PC) but I can't really feel any difference. In fact I'm not sure it's actually working. I do get the 'beep', which indicated the programme has detected a D3D process running, but I have found that unless I select vsynch in my gfx card settings then the Overrider does not force it as it is meant to. So if it isn't forcing vsynch is it actually forcing triple buffering??
It sounds like you've got VSync set up optimally Ken. Triple buffering is essential to avoid lag when using it. The problem is that you're still restricting the frame rate and I have good reason to believe that the FFB rate is directly linked to the frame rate. GPL's certainly was, I've seen the code and I have no reason to believe iRacing is any different. Try the GPL 60Hz patch and see the difference.
Since upgrading my PC I've been getting absurdly high frame rates with a single screen, running on my own at Laguna I was seeing peaks over 600fps! Now obviously the screen isn't updating at that rate, it's limited to 60Hz but I believe the FF is. With the frame rates so high the wheel feels tighter, more responsive and far less susceptible to FF lag effects causing tank slappers. After experimenting I found that limiting to 168 gives 95% of that high rate FF feel without pushing the machine too hard. I certainly don't see any point in going above the core physics rate of 288Hz (IIRC) as you'll just be doing the same work twice.
Now I've gone to 3 screens and frame rates are back to normal levels that's gone, the steering feels less in touch with what's happening at the wheels. More importantly it's not consistent. On my own I can still get 168fps but last night at a few places round the circuit and with cars around the rate was dropping to around 120 and the steering felt different.
GPL used an averaging formula over recent force samples to smooth out FF spikes. I think it's a fair assumption iRacing does something similar though it's obviously been refined, remember the asymmetric camber FF spikes on the straights at Monza? At lower frame rates it will have fewer samples to work with and is more likely to get it wrong. It could be that they're increasing the smoothing at lower frame rates?
Most of this is supposition and experiment, apart from the GPL bits I've not seen the code. IMO though there's a very noticeable difference in FF feel between 60fps and 600fps. It's all about finding a compromise. It's almost certainly better to get used to the slightly inferior FF at a consistent frame rate than having it vary when you most need it in a tight racing situation. I'm certainly going to experiment further with Vsync, visually it's super smooth and I've not had time to get D3DOverrider working on the race machine yet.
Standard disclaimer - this is all just my opinions based on many year's experiment and a few known facts. I could be completely misinterpreting what I'm seeing or imagining the whole thing.
Paul
