I want to make a track which would have patches of racing line which would give greater grip in dry than the rest of the surface, and then same patches to give less grip in the wet than the rest of the track. At this point I wish I could use pure "wet=x.x" values, but they are obsolete now, because "roughness=x.x,y.y." has replaced it as I have read in "rain effects" thread. At first I thought that it is going to give great effect being more technical parameter and cotaining better logic than just simply "wet=x.x.". I have read a post by a member that roughness parammeter x.x is for adhesion, and y.y is for macro texture. To my knowledge, adhesion is major grip element in the dry conditions, and macro texture becomes greater grip influencer in the wet as adhesion falls off massively as surface gets lubricated by water. As much as I have tested right now making x.x value zero initiates slides/slip very easily, and y.y value at zero makes sliding/slipping more severe and at this point to my understanding it fits adhesion and macro texture effects. However, it has same effect in dry, surface due to more adhesion becomes much more grippy in the wet as well as in the dry, wet only has slightly enhanced effect. In my best attempt I have achieved that alternative racing line is very similar pace to normal racing line in the wet, doing it more makes it unnatural in the wet, and even more unnatural in the dry. I'm yet to try this with official S397 cars. But I don't think that results will be different. Could I possibly just not find magic ratio of y.y/x.x yet ? I suppose sum of x.x+y.y is also important and should be around 0.75 ? Maybe I don't understand something about Roughness value ? It is possible, but I think I have tried all variations by now. Originally I expected to make one surface with lower x.x and higher y.y to have an advantage in wet and disadvantage in dry, and other surface the opposite way. But it does not work. I thought aswell to post it to issues, as it is issue IMO not only related to my modding project. Here is a slide from some slideshow about braking traction: Source: https://www.slideshare.net/hebronashraf/braking-performance-4
Have you tried lowering both values, and then raising both values, and seeing how they compare? Also, be wary of fully wet tracks (wetness above 0.50) as then the TemporaryGripLossForWetness parameter in the tyres will start to override the roughness effects which apply to a damp track (wetness between 0.0 and 0.5). *(and since the tyre parameters can have a big effect on what happens, consider testing using official recent content)
Don't remember exactly, since I didn't do anything with that since this thread, but as far as I can remember I tried all combinations. There is no noticeable grip "inversion" as tarmac gets wet. Will refresh the subject today. I still haven't learned how to detect what is the wetness percentage, I was said that some different hud shows it. I am also curious what is the height of water level at 100 percent. By the way, the first thought that comes to my mind is why would you ever override roughness, to my logic roughness will always be important nevermind water level, as long as tire is not completely aquaplaning, but perhaps I don't understand something. Though if it is being very progressive then maybe maybe... Does it over ride roughness completely at 100 percent, or just partially ? Thanks for information about "TemporaryGripLossForWetness parameter" I'll look at it, also will try the track with latest official content.
There isn't any water depth. There isn't real aquaplaning, just that parameter to facilitate more grip loss with a wetter track. For testing I would use devmode, so you can set the dampness directly. Perhaps use devmode to generate a suitably damp realroad and load that into the main game with static realroad effects. And the full wetness (0.50 - 1.0) doesn't override the dampness effects, but it will mask it because every surface will become more slippery. Up to half saturation you should only be getting the effect of the tyre's dampness parameters and have the most ability to judge what it's doing. I've never investigated the actual grip effects of various roughness settings, or what each parameter means (maybe I read it somewhere in the past, but I can't remember), the tyre pdf mentions that the sum of the parameters represents the surface roughness, and it's that roughness that determines wet grip level (it took over from the old wet= parameter in the tdf) so I'd imagine it's possible to achieve somewhat correct results that way. To put it another way - if the roughness is the only thing that differentiates the behaviour of different surfaces as they become damp, it must be possible for it to produce a required result (or at least a trend in that direction; it probably can't overcome starting points that are too distant).
The percentage definitely represents water depth, even though it is unknown what the actual depth is and we don't have dynamic hydroplaning. This thread is actually about the thing called viscous hydroplaning, that happens doe to surface being smooth, this tarmac has a smoother polished texture if it is worn and rubbered. I just did the test again. This time with Skipbarber and static 0.6 wetness. Four laps - normal line - alternative line - normal line - alternative line. I have bumped normal surface to extreme values of roughness (1.0,1,0), and left the normal values for patches that represends smooth texture tarmac. Tested with static 0.6 wetness, and static 0.0 wetness too. Results: //0.0 wetness Dry 1.32 Wet 1.32.4 //0.6 wetness Dry 1.33.1 Wet 1.33.5 Conclusion is that even with roughness pumped up to such insane level outside is not faster. Moreover, there is no inversion between wet and dry track. Only difference was a second slower pace(at least 4 too little). So I think roughness parameter for track surface is not a substitute to simple "Wet" parameter. To me it is quite obvious that feature does not work, and I hope it could be adjusted properly without damaging anythign else in the game. One more thing is that by using alternative line I loose the most time (about 0.4s) in lavant corner, ironically in reality it is probably the most rewarding to use alternative line when in wet. Lavant curve: Also the thing in the dry is that it made the grip so high, that it rolled my car in few occasions, while in my mind it should be more slippery than smooth surface when in dry, and more grippy in the wet, but it is always more grippy, and not more grippy enough in the wet to make alternative lines a logical choice. Rolling over at madgwick after gettign too much sideways because I clipped grass.
I don't know if actually anybody is interested anymore, but I just drove on heavily used track with marbles, and thought that marbles shouldn't be slippy in the wet, I think they shouldn't be so... I remember developers of ACC sharing knowledge that they got from the drivers, that they even try to pick marbles on purpose, because if they manage to get it stuck to tires for a while, then it improves grip in the wet. It is strange, but I suppose it makes sense, if marbles lifts the tire slightly, or creates a gap for water to escape. Though I suppose thats more about slicks, as I'd imagine marbles could also close the tread.
Parameters regarding wet effects of rubber line and marbles are long overdue. Unless there is a movement toward tyre surface modeling, but that seems very unlikely. I didn't consider your laptime tests wholly scientific, so I tried doing some friction tests on surfaces of varying roughness and dampness but had no success in getting logical results that the roughness parameters would suggest are possible (apart from overall grip change, wet or dry). I'm not totally satisfied that I tried a wide enough range of values (or more accurately, enough intermediate values) to draw final conclusions, but I think it would be good to have s397 verify these parameters do work as intended for wet vs dry grip and give some examples of surfaces and their roughness values. There are too many variables open to interpretation for a modder to achieve their desired results, let alone have consistent effects across the game's entire content. And that's important for an aspect of driving that relies so much on instinct and feel.
Yes it was not scientific, the only science part was laptimes, which also depends on my driving. Also even though my confidence was completely destroyed after placeboing over .tbc file and learning about it, I still think that the feel, which I expect from a track that would require choosing alternative racing line in the wet, has to be not too far from being correct. I agree that it would nice to have if S397 would give it some light in future, maybe after current big tasks gets less consuming ? However, there are examples - like roughness of paint surface. Less interpretations would be welcome for sure. I did personal discovery by looking through skippy TGMs in devmode, I found that 3rd file had lots of TGM values explained, or briefly explained. I am not with my rF2 computer now, but IIRIC roughness has three values actually, even though only two shows in .tdf. IIRIC default is 0.5 for adhesion, 0.25 for macro deformations, and third is for micro deformations, it seems to sum up to 1 and whats left from adhesion + macro deformation is for micro deformations. The only thing which I'd expect to happen would be adhesion disappearing quickly as surface gets damp and then wet. While micro deformations getting less effect, and then macrodeformations getting the least impact from wetness. So I'd expect a surface that generates more grip through macro deformations to have noticeably better grip in wet. It would be nice to get free from interpretations. I'd expect roghness .tdf value of (0.0,1.0) to have similar wet grip as in the dry. Then value of adhesion (1.0,0.0) to have good grip in dry, but poor sliding resistance and terrible grip in wet. I think I have already tested these values, will test again. Also, @Lazza , if TemporaryGripLossForWetness parameter is not presented in TGM, then roughness values goes for whole wetness levels just the same, or gets still switches to some default TemporaryGripLossForWetness value ?
I wouldn't even be sure the roughness has more effect after 0.50 wetness - everything may be static at that point except the grip loss for wetness (damp is 0 to 0.5, wet is 0.5 to 1.0). Can't really say without some tests, but it hardly seems worth the effort. I would be wary of extremes when testing roughness values (1.0, 0.0 vs 0.0, 1.0) as that can't really exist and is probably well outside any values they considered. The TGM you found is 'the' reference - if you weren't aware of that then apologies, it's the main source and help in trying to understand most of the parameters.
I think macro rougness should always be in effect, except for fully hydroplaning contact patch Yes, extremes are nonsense, they are only to see if function is performing as expected. But perhaps I am not aware of some side effect. I just checked TGM, and thats not "StaticRoughnessEffect" that give information about roughness values for .tdf. Next to " SlidingMacroDeformationCurve" this explanation is written: "to get the totals the adhesion, micro and macro curves are then multiplied by the surface types as defined in the TDF files, the defaults of which are 0.25 for adhesion, 0.5 microroughness, 0.25 macroroughness" So when I wrote the previous post today I remembered many things incorrectly. Adhesion seems to be only 0.25 of the totals, however microroughness is probably very simmilar and should have to take strong effect even only from dampness just like adhesion. Also there is nothing about suming up to 1.0, memory let me down on that, mixed up with somethign else, perhaps because those three adds up to 1.0 by default. There seems not to be any control for adhesion from .tdf, or could it be controled by "Dry" ? Then "Wet" could have been adhesion level for wet, but it is obsolete now, so it probably is just global multiplier. Usual values in .tdf are Roughness=(0.50,0.25) Road Cement Patchs: Roughness=(0.56,0.15) Paint: Roughness=(0.45,0.02) P.S. paint roughness values were one of the first that I tried when begun this topic. The fact that I can't alter the adhesion might be the whole problem. The whole idea is to take away from adhesion and give to roughness, so in the wet surface would have more grip, than default surface, and less grip in the dry than default surface. However, I would expect microroughness to have a lot of similarities with adhesion regarding wetness effect on it.
Remember in terms of the game, dampness/wetness are different phases of the same thing. Wet values 0.50 and below are 'damp', whereas 0.50 to 1.0 are 'wet'. The dampness effects are at full effect at 0.50 wetness, from there they don't change anything (I think). Conversely the wetness parameter (temporarygriplossforwetness) doesn't do anything until 0.50 wet, and has full effect at 1.0. Again, rF2 doesn't actually do water depth. This damp/wet crossover is a placeholder for aquaplaning. So when I say roughness has no more effect after 0.50 wetness, I'm not saying it no longer has any effect - I'm saying its effect doesn't change. As the track gets more wet (above 0.50) the roughness is still doing the same things to grip it was at 0.50, while the wet effects are reducing all grip as wetness approaches 1.0. That's really confusing to read, sorry. We know that overall surface roughness, as the sum of the two roughness parameters in the tdf, influences static grip. (from TGM quickstart guide: StaticRoughnessEffect = Terrain roughness influence on static friction. Rough terrain reduces static friction, by reducing surface contact. Roughness is the sum of the Roughness= variable in the track surface .TDF file, the defaults being (0.50,0.25), or 75%.) From this blog entry, you may have luck playing with ttool to test the effect of these parameters in a very scientific way: https://www.studio-397.com/2018/05/realtime-tyre-analysis-improvements-build-1110/ Note the comments on the test parameters, which place a limit of 1.0 on the total roughness. It also strongly indicates the values are for macro and micro roughness. Good luck with it...
