So we are talking about two different things then. I thought the discussion was the rate of tire wear and like others my tires are shot in 30mins at race pace. You swing in and say no, if you drive fast and clean you can go much longer. I'm obviously misunderstanding something because that is what I'm calling b.s. on. But you are talking about an online race with humans at Poznan. Ok. Good to know that if you go slow enough, against humans, at Poznan, clearly, tires last much longer. Got it, sorry for my inability to read. Working on it.
They probably shouldn't have called it GT World Endurance mod if this is so. The general concensus in this thread and among those I chat with is that the tires in these mods is too extreme to be useful in endurance racing- even though this is supposed to be an endurance mod. IMHO even 1 hour on a set of tires is too little for an endurance mod. These tires are suitable for touring car racing and not much else.
Sorry I sounded a bit harsh but I was totally lost when I read your post that had nothing to do with our discussion or my posts. Managing the tyres is about the driving. How you drive not how fast you drive. You can drive 2 s faster than anyone else and yet still have them in better condition than guys that were driving slower and "trying" to preserve them. It's all about small details. Really small. Changing even the smallest thing in your way of driving, relearning to do even minor things can lead to totally different expierieces. It has some valid info even for today's racing. "The smoothest way is the fastest way". He talks about Monte Carlo but it still holds on other tracks. So being smooth doesn't mean slow. It means efficient. That's what it is all about. Efficiency.
I'm lost as well because I misread and thought your post was about tire wear. Anyway man no worries on any harshness, but I think Alonso might take issue with someone saying he had to pit one extra time over Kimi because his driving and not his speed in the F1 race in Spain. He pushed harder and went faster, yes, faster, (an example that doesn't include Poznan so don't get confused) thus requiring one more set of tires over Kimi who drove slower for a 3 stop but came in second behind Alonso's Ferrari. Will poor driving wear tires? Yes of course. Does lapping faster wear tires? Yes of course, exactly what happened in F1 this week. As Guy said, the concensus is the rate of degradation is too high for these types of tires considering these are GT cars. I hope ISI will address this soon.
Not trying to get into anything contentious here but I was curious if Panigale's assertion that you can't win a 1-hour race at LRP with AI set to 97% is true, so I tested it. I dearly wish the sim hadn't crashed post-race so I could upload the results file but unfortunately that seems to be happening all the time to me with this build. I started 3rd and moved into the lead after 3 or 4 laps. I led the remainder of the race until 5mins to go when someone distracted me and I spun out of the downhill. (This was on the uphill chicane version in the C6.R, btw). That damaged the tires a bit obviously but I didn't hit anything and dropped to 4th place. I wasn't able to regain any but I was still closing on the leaders by about .5 seconds per lap until the end. Now I don't want to give the impression I was leading the race easily. The AI is strange. They seem to go through some sort of cyclic crisis of confidence because they'll be catching me up quite nicely for several laps and then suddenly their pace drops and they go backwards. When they caught me I had to fight but I didn't do anything cheap to retain the lead. It was obvious I was going to win the race if I hadn't spun at the end. The left front was down to 2 bars remaining and it was the worst. And really this was done with my Sebring setup which is very low downforce and hardly ideal for LRP. But I agree with Ronnie that making comparisons between the player and AI cars is mostly irrelevant since we know they don't use the same physics. If it's off then all it means is that the AI's tire wear and other physics stuff needs to be better balanced with the player's. The real issue is that you guys are saying the tires can't last for a 1-hour race while pushing at full race speed but I didn't see that. I came within a tenth of my fastest race lap with 4 laps remaining (must have been just before I spun). Anyway, I know...pics or it didn't happen. I might do it again with a better setup and hopefully the results will save. Last thing I'll say is I don't disagree that the GT cars need more durable compounds, but I do hope the softer ones are kept for sprint racing.
Pics not needed I take your word for it and that was the sort of feedback I was interested in. Although I assume you ran AI 97% means lap times in the 50/51s for the C6Rs? And it was a field of how many AI? The AI example is just there to keep us honest, i.e want to prove an hour of quick laps not slow, slow we know is very possible. Curious, what was your fast lap with 4 laps remaining?
But they don't use same physics which begs another question which is how does ANYTHING really work with AI if don't use same physics? Can anyone explain how AI behavior and especially fuel, tire wear, etc., etc. is programmed if not using same physics? How can we expect ANYTHING to make sense if using different rules?
51.320 was my lap with 4 remaining, I believe. 20 AI cars. My aero settings though, 1 on the front and 13 on the rear wing, so I don't think that's at all good for LRP. I started on full fuel as well, btw. I'll just reiterate that I do think the GT cars should have harder tires that are capable of 2- or even 3-stinting for the purposes of endurance racing, because that's what they do irl. And while I can definitely make these last for a full stint without laying back, there's not much room for error and of course it will vary by track. But LRP is probably a pretty hard case since there's only one long straight and the Uphill and Downhill put a lot of stress on the LF. You don't spend much time not cornering there.
Very few racing games or sims have the AI using the same physics as the player. The Papyrus sims, which I worked on, never did. As Tim has mentioned a bunch of times, the computational intensity of the player's car - specifically the tire model - is such that on most machines you'd be lucky to run even a couple of AI cars if they used the same physics. So instead they probably run a simplified but still similar version of the player's physics. This does mean that they exist "in their own world," so to speak, and thus someone has to spend a bunch of time testing and tuning to get things like their tire heat and wear to somewhat resemble what the player sees. The problem is that tends to end up being skewed toward matching the performance of whoever is doing the testing, which was what I always struggled with at Papyrus. It might be that the AI's wear rate matches mine very well, but then you stick someone else in the sim and it seems way off to them. It isn't an ideal situation but if you're going to push the limits on tire and physics modeling then you inevitably end up in this place. It's also one of the reasons I don't care about racing AI at all.
Another small data point for you, Panigale. I was regularly seeing 90C+ on the LF coming out of the Downhill, at least when I was paying attention toward the end. I do find that with the Corvette I need to make sure the fronts are wearing a good bit more than the rears if I'm going to do a long stint, otherwise when the rears get some wear on them I find myself getting too much power oversteer and that tends to be a compounding thing. Just keeps getting worse until I'm either super slow or halfway to wrecking most of the time. But I think that's the hardest corner on the tires at LRP and so that's the temp I was seeing. What you can't do, as you probably know, is overcook the entries on any consistent basis and expect the tires to last. If you slide into the corners then not only is it scrubbing rubber off at a high rate but it also instantly shoots the temperatures up and the entire rest of the corner is then run with those elevated surface temps. And I believe the tires are wearing faster at higher temps.
Of course it's a customized setup based on a customized setup I don't think stiffness is the problem but maybe the chamber value (I prefer less negativ values) With the default setup I must reduce the ai strenght to avoid a desaster but I will give this a try and post the result. It would help to have more results to compare the expirience. Feel free to post your expirience. Thx for this tip! This will be the second case I test. If this works well and if the reason for the different tyre wear is not my setup or driving style it's of course a solution better then everything else I know current. I have not enough expirience to exclude this thoughts so maybe rFactor2 simulates more and better I thought? And the next challenge after learning to be fast is preserve the tyres from wear? We will see.
Well Matt my hat is off to you man. 51.3 is moving good for race pace, I could not do that on one set of tires for 60 minutes. I'm running a lot more downforce (default aero setup basically) but my lap times aren't much better so this will drive me to drinking tonight. I'll have to try those aero settings but honestly, I don't want to see how slow I will be with that little downforce...
Add real road, weather changes, etc, and matching AI to player will become impossible task. I truly believe so and can only hope those who race against AI would understand that they may be asking more than is possible to do. What could work is bunch of computers connected via network so there would be enough computing power for AI to use same physics. That would be cool. Indeed and same here.
Just FYI - Not a big concern for you since you don't race the AI but Santagoss in another thread identified some interesting settings to change in the plr for the AI. One was something about inside outside line, set that to 3. The other was something about AI limit or some oddness, basically set that 0 so the leading AI doesn't make mistakes to keep the pack together. This provides a more realistic spread of the faster AI.
Interesting, thanks. There's definitely something weird with them speeding up and slowing down in some sort of cycle so maybe that improves it. So yeah, bean's getting similar wear. His wear rates front to rear are more even than mine so that's letting him run longer, plus Spa has a lot of long straights.
I have had these settings for a while now and the AI is deffinently better now. They are much faster in the race. I've had some good races with them for example on Toban and Pukekohe 2013 and their level was only 95/55.
