Xtreme Gaming Network
Simulation Racing Leagues => Assetto Corsa => Topic started by: Wally on December 09, 2014, 12:49:44 PM
-
Guybrush posted a really good idea as a ballast alternative on the AC forum. Instead of adding weight to faster drivers, give them a lower track grip. The cars would handle better, and not be so disadvantaged with loss of straight line speed. I can't see that ever being implemented though.
But.
The server already sends the current track grip level to all clients. I wonder if it would be possible to intercept those packets, on the server, and alter the grip level for individual drivers before sending the packet on to the client. In theory, that would be possible, I guess. It might be time to crack out WireShark and sniff the packets coming out of the server and see if they can be interpreted.
-
heh, bloody hackers...:D
The idea is good, tho.
I'd so like to see what happens if you run on an 80% track with an 80% tyre....you need to also output some ice texture in this case I guess ;D
p.s. I'd be happy to be involved. Maybe you need help sorting the package format out...
It can be a biatch, all you hunting for is a number. We don't even know if it arrives with every package!
Unless they've been nice enough to use some non-predefined format, mayhap xml or json (however unlikely)
Um...isn't there a way to mess with these values - or the tyre grip - from python apps? That would be a slight bit simpler, and would not add latency (which the package manipulation can introduce)
-
I think its worth trying until Ballast is added officially, but as I wrote in the thread the end result would be near identical to ballast. Guy thinks that less grip will mean it just slows you in corners but straight line speed is all about exit speed. If you have less grip you will be slower out of the corner and so slower down the straight just the same as with ballast I think.
Ballast does make you lose some power on acceleration but for the reduced grip level needed to level things out I think the top speed differential would be quite similar. You can already try to see the difference of either option just put in fuel with no fuel burn fill the tank and run some laps, compare the difference between a full and empty tank. Next try and set grip level to a point where with an empty tank you are at the same lap times as with a full tank, then compare telemetry of both and see if there is any noticeable difference in how the lap is achieved being handicapped purely by weight or grip. ;)
-
Good idea Marty. Do you want to try that at a pretty "average" track (straights vs. corners) like Imola, and capture your telemetry.nkt files?
-
Yes can do, any car you think it would be best to do it in? I think running a gt car at Imola with medium tyres no wear and fuel off should be easiest to get good data. Just set a hotlap on low fuel then do the max fuel run, followed by reducing grip level on low fuel until it matches the same time. If I can put in a good lap in each at pretty much the same time within a tenth or two the data should show this fairly well.
Wally have you got a link to a proper working Motec or Aim setup for AC, I did dl Aim a while back but couldn't quite get it to work for some reason.
-
Good idea Marty. Do you want to try that at a pretty "average" track (straights vs. corners) like Imola, and capture your telemetry.nkt files?
When I get a chance I might try it at Mugello. I find Imola is a little unreliable as a lot of the lap time is derived from riding the curbs in a particular way.
-
I *think* I've just downloaded the latest version of AIM Race Studio. Have you used it before? What problem are you having?
-
I did run a few laps in the m3 GT2 no fuel use tyre wear on soft tyres. I turned on tyre wear because it just felt strange with wear off for some reason and running softs they are close to 100% from lap 2 to lap 4 here so it gives me a few laps.
Fuel weight surprisingly wasn't as big an impact as I expected with soft tyre and 1l of fuel with 100% grip level I got 1:42.3, 110l I got 1:43.4 (I think) then reduced grip levels to a fixed 97% which was probably a fraction slippery and got 1:43.5.
The difference with the added weight felt fairly minor and I likely could have gone a couple tenths quicker. The 97% grip it took quite a while for tyres to get to temp and it was quite a bit slower into a few corners. Braking was the main issue trying to fight locking up as I used no ABS or TC. But mid corner speeds with lower grip were clearly slower as I needed to drop down an extra gear in some spots. Also much harder out of corners due to lack of traction and even though I think id need a little more grip to match the 110l of fuel handicap id have more hope racing with the weight then the reduced grip I feel.
Overall there were no perfect laps in there but all fairly even amount of push, with maybe only a few tenths to gain in each. The main thing I noticed is this car would need a lot of weight for handicapping if you wanted to knock it down 4 seconds per lap. At that point it would likely feel pretty bad but grip wise it wont need as much though at the lower levels it still is much harder to drive then a heavy car most likely.
I havent checked the data but will attach it here for Wally to take a look and bring up some stats.
http://www.mediafire.com/download/sa7snffswwd3nkl/M3+GT2+weight+v+grip+handicap+test.rar
I will get the latest AIM studio but last time I tried it I couldnt find any way of importing the telem data, I had never used the software so also had no idea what I was doing. ;D
-
How would it effect dynamic track grip levels, though?
I like the way we have the track rubber in (even though I'd prefer if we started in the 8-% range to make it more extreme).
-
I don't think we need to start grip at 80% but maybe reduce the speed it grips up at, once there are 24 cars going around it doesn't take long to get to 100%. But when grip is very low its pretty pointless to even run practice laps at a level we wont be racing at. If the grip stay low even during qualy and maybe only gets closer to 100% at the end of the last lap of the long race it would be close to ideal. 97% costs me more then one second so 95% is probably nearly 2 seconds and quite slippery, I think the devs said somewhere 80% would be like light rain on slicks.
Combining individual grip levels and dynamic session grips I think just adds too many variables, the odds of making it fair for most is unlikely I think.
I tried installing Aim studio but the install crashes when it tries to install some USB driver, I tried a second time and pressed dont install on this driver but it just crashed the system again. I dont think it works in win 8.1x64 so I will leave the data analysis up to you.
-
I think the devs said somewhere 80% would be like light rain on slicks.
I guess you don't like wet weather as much as I do Marty. ;)
It would be the closest we can get to simulating a drying track. 8)
-
I dont mind it at times but if it starts wet in practice and I know its not going to be during qualy or race I wouldn't bother running laps in that prior to race nite. ;)
Managed to install race studio forcing win 7 compatibility mode on the installer, will see if I can get the data to come up.
-
I just don't like 100% grip.
In the real world a track is never at 100%.
-
The ballast we used to end up running in netKar was quite significant for the front runners, to even up the field. It was much, much heavier than a full tank of fuel. I know what you're saying about exit speed Marty, but with significant ballast, you lose straight line speed even say from a standing start, no corner exit speed to be considered. I am fairly convinced, after having driven heavily ballasted cars, that a less grippy track would feel much better.
-
Looking at the data it kind of shows there isnt much in either way but the effect of a full tank isnt much and so for a 4 second difference youd need nearly 4 times the weight or much lower track grip. Either way it wont feel great as your taking quite a chunk out of the cars performance.
A track that is 4 seconds slower is very slippery and your braking distance is going to be way ahead of someone with much more grip, corner exits you will also struggle to get traction. So either way its not going to be any better with less grip rather then less weight.
Guys main concern is not being able to pass others but losing so much mid corner and under braking and acceleration with less grip wont make that any easier.
I also think its just making it a bit silly if every driver is basically on another track with individual grip levels per driver. Weight is commonly used in real racing but giving everyone less grippy tyres is getting a bit too arcade I am starting to think.
-
You make some good points. The real test, as always, would be to try it, if it's even possible.
-
Best way to test would be get someones telemetry data that is a little off the pace, then get a quicker guy to find what grip levels they would need in order to match their lap times. This would be the handicap limit needed and then compare telemetry data between these 2 separate drivers. I can get data for myself running good grip and low grip but its better to compare against others.
If its done with an open mod car then we can also alter the weight of the car manually to provide ballast as a comparison between ballast and grip penalty.
-
I just don't like 100% grip.
In the real world a track is never at 100%.
In real world there's no such thing as 100% ;)
There's only more grip and less grip. But for the sake of simulation percentages are perfect. You can simply imagine 100% means "grip is ok". :)
-
But it's not "grip is OK", it's actually "grip is constant", which is the unrealistic bit. ;)
-
But it's not "grip is OK", it's actually "grip is constant", which is the unrealistic bit. ;)
That's entirely correct. I'd be happy with a racing line grip, but full grip level simulation is a feature of a future sim I guess.
-
But it's not "grip is OK", it's actually "grip is constant", which is the unrealistic bit. ;)
Well the way AC does it its still a constant grip and consistant through the track, 100% would be maximum grip but in the real world there is no maximum grip it can always get grippier. Still it likely wont ever get as grippy as AC 100%, there are also plenty of variables that effect real world grip that are missing from AC. Most basic is changing track temps through the session and also some parts of track getting more direct sun then others so heating up more.
Then there would be a live track type system like in RF2 but even more advanced, not just the line other cars drive on but also cars going off track bringing dirt on some parts. I cant imagine how slippery public servers would get with the poor standards of driving, but atm servers cant even set ambient temps and there is no temp change from the start of the first session to the end of the last.
I agree with Bacchulum in that it really should never get to 100% grip and even if its close 97 to 99% with a very slow rate of improvement it would add a bit more then starting very low then climbing to 100% quickly and staying there for the end of all sessions and races. Ideally I think it should slowly build from practice to race keeping most of the grip gained during each session and at the end of the last race is the only time it should even get close to 100% but still not reach it.
-
One final note before I let it go (if I can :P).
The track can lose grip, even in perfect conditions.
Once the tarmac has taken all the rubber it can, the tyres start 'peeling' it off causing marbles (they don't come directly from the tyre surface).
So in theory, it should oscillate between 98% & 99%, if it ever gets that high.
But your point, Marty, about the amount of variables is all the more reason we should be aiming for a lower level. 8)
-
We can certainly try it, you grip junkies. I just have to set the rate of increase much lower. It might be interesting for the long race next week. I can easily make it go from say 95 to reach 100 only a few laps from the end. But how realistic is this? I've never heard them complain in V8 Supercars, for instance, that the grip is really low and only getting good by the end of the race. They talk a lot more about tyre temp and condition rather than track grip.
-
From an F1 article:
“Early in a Grand Prix weekend a track surface usually has less grip than it will have after Formula One cars have been running on it. We call the track surface ‘green’ before it has rubbered-in,” explains Hamashima. “Some tracks get a lot of motorsport use and the change over a race weekend with these circuits is less extreme than at a track which is seldom used. When the track does not get a lot of use we see a more dramatic change of grip level provided by the track over the race weekend.”
No real clues there about how noticeable the grip change really is... but ”dramatic” makes it sound quite high. That's over the course of a race weekend though, not just one race.
Also:
Marcus Ericsson, car #9, chassis CT05-#01: "At the start of FP1 the grip level was just so low - the car was sliding around everywhere and I couldn't push anywhere around the track. I know the Hungaroring is always pretty green at the start of the weekend, but it was worse than I'd expected. You can go chasing the track when it's like this, trying to find setups that work, but we have to let the track come to us as the grip level improves overall, so for the next run we continued with the plan and were improving, little by little as the session went on.
From http://intelligentf1.wordpress.com/2012/06/12/canadian-grand-prix-the-role-of-track-evolution-in-hamiltons-victory
”Usually, I’m not big on the effects of track evolution during a Grand Prix. That’s because I can see the effects of the track rubbering in from the data in the opening laps (between 3 and 9 laps at most races) and then the behaviour is pretty much constant for the rest of the race. However, in Canada, the data shows big track evolution up to about half-distance.”
Implies grip changes stabilize rapidly.
-
From what I've been reading, a rubbering in up to 100% in the first third if the race seems to be realistic. That's what I'll set the next long race to. Up to now, the tracks reached 100% in about 2 or 3 laps.
-
I think F1 has more rubbering in then any other series, the drivers even complain if there are cars running very different compounds in the support races taking away some grip. Tracks like Melbourne do get quickest at the end of the race but even in the last qualy run they get lots of gain and you hear commentators say it will likely be last across the line that gets the best track.
In V8's they dont shed anywhere near as much rubber their tyres last more then 6 to 10 laps which is about all they get out of super softs in qualy and they are already fairly spent after 3 or 4. GT cars run very hard tyres which can last a full fuel run or even 2 with limited sets for the entire weekend. So these tyres wont be rubbering the track in as much as they are cleaning it and also why you wont see as many marbles as in a f1 race.
I think GT cars on permanent circuits used most weekends there wouldn't be a huge gain from qualy to end of race but on a street track things would be quite different. I dont think they start as green as they do in our practice but they may only gain 2% all weekend but grip changes quite a bit with time of day and weather. We cant set temps in a server but I think at least now everyone gets the same 26c in MP where in earlier builds whatever you set in SP mode you also got in MP, ::)