PDA

View Full Version : Weather progression



DeliCreep
30-09-2017, 12:34
When you set weather progression to a multiplier it behaves a bit strange. For example when you set it to 60x, then each weather slot will last a minute. For example when you have a storm and a light cloud weather slot and progression is set to 60x, both storm and light cloud lasts a minute, which is as it should be, but the strange part is that during that one minute of storm, 1 hour's worth of rain falls and it dries up almost instantly, which means that both the amount of rain and drying rate of rain are boosted to 60x of the original. I think it should just rain for a minute but with no boost to rain amount/drying rate, so it should rain as much and dry as fast as it normally would in that time, it would be more realistic. Right now if rain is set to anything other than "real time" it will dry up unrealistically fast. So for example you can't have rain for 10 minutes without the side effect of super fast drying. What do you think?

Oomph
30-09-2017, 12:47
there is two components at play. there is time progression and weather progression, check on those, and try again. i have previously tried 4 slots at bathurst with time progression on "real time" and weather progression on X20, so weather change every 3 minutes, and the puddles and track was still wet during the dry period.

DeliCreep
30-09-2017, 13:18
It's the same regardless of time progression.

PeteUplink
30-09-2017, 13:38
I've told them about this before. I raised the issue when PCars 1 released and I'm quite disappointed that it's still being done the same way.

I think the weather progression should really only affect the weather and how long it lasts, and the track drying should remain in real time. But what we currently have is a system where the rain and track drying = 1 weather slot (which is 1 hour in real time), but if you speed up time it will speed up the weather slot by the amount you selected, including the track drying. SO if you choose 60x, the rain passes 60x faster and the track dries 60x faster too. This is why, if you use sync to race for weather, you get odd stuff happening like it being totally wet at turn 1 on lap 4, but a lap later the rain has moved to turn 4 and it's wet there and completely bone dry at turn 1. Which is very silly as water just doesn't drain away that quickly.

I think it should be changed and be more like the F1 2017 system where weather slots are evenly spaced through the race, but the track still dries in real time. But it's probably way too late for that now.

Oomph
30-09-2017, 13:54
well i went and tested on brands hatch indy, 2 weather slots, 1, light clouds and the other thunderstorm. time progression on real time and weather progression on 60x. whilst some part of the track do dry a quicker than expected, most of the track remained wet, and at those areas that did dry, puddles remained..

241420

241421

Roger Prynne
30-09-2017, 14:09
Also don't forget that it can be raining on different parts of the track and dry on others.

PeteUplink
30-09-2017, 14:10
I've just done the same thing at Oulton Park. I ran some test laps with a 10x weather progression and with a 60x weather progression. This is what I found.

10x: The weather started in thunderstorm conditions and was set to go to fully clear. I managed around 4 laps in torrential rain, but on the 5th lap the rain started to ease. The track started to dry and was bone dry, aside from deep puddles, by lap 7.

60x: The weather started in thunderstorm conditions and was set to go to fully clear. This time I completed one lap before the track was completely dry aside form deep puddles.

So, what's happening is the standing water/puddles is drying realistically, but the surface water is drying as soon as the rain stops. The surface water really should evaporate slower.

Trippul G
30-09-2017, 14:11
I never noticed this behavior in PCARS1, as I never really did much custom weather racing. I always found it to be more of an unfair annoyance rather than a challenge. AI would always seem to magically know exactly the right time to pit, or they wouldn't pit at all and would stay out, running perfectly fine on the wrong tires...there was never an appropriate time to use inters, as the entire track would dry instantly as soon as it stopped raining, and you could go straight back to slicks...the whole thing was a bit of a mess IMO.

I haven't properly jumped into exploring the full range of weather options in PCARS2, but Iet me see if I'm understanding this correctly...you're saying that track drying speed is tied to the speed of weather progression (as opposed to either remaining constant, or being tied to time progression, which would seem the most logical to me)? If so, that does dampen my enthusiasm a bit (no pun intended).

I'd guess there must have been a good reason from a technical standpoint why track dying speed couldn't be decoupled from weather progression speed, because it certainly does seem odd to design it that way if they had the choice.

PeteUplink
30-09-2017, 14:12
Also don't forget that it can be raining on different parts of the track and dry on others.

yes, this is true. But with weather progression sped up, and the surface water drying so unrealistically, we get situations where the rain will be at turn 1 and the track will be wet on lap 3, but by lap 4 that rain has moved so quickly that it's wet at turn 3 and bone dry at turn 1 only one lap later. This is very unrealistic. If rain actually dried up that quickly we'd never need to delay F1 races in real life, because as soon as it's stopped raining the track will be bone dry in one lap. That simply doesn't happen.

Look at the Itailian Grand Prix qualifying form this year. It was delayed for a long time due to the weather conditions. The rain stopped, but the track took a long time to dry. By the time it was looking like it might be getting towards driveable, the rain started again and we got another delay. Even when the rain did stop and we got cars on track, it still took a long time for the track to start to dry. Session 1 in F1 qualifying had 13 minutes left to run, and by the end of it the track was still wet. Then the rain started again. But if it was PCars, the rain would stop and the track would be dry in a lap.


https://www.youtube.com/watch?v=-eTFNSalcmI

Oomph
30-09-2017, 14:31
I haven't raced with weather progression, and never used it in pc1,, but with a couple of practice runs I have done with it, it seems reasonable. I will definitely play around with it some more. But also on a realistic note, I have never experienced 60 weather change in an hour even thought I have lived in Melbourne at some time :p

Could it make a difference by choosing winter and/or maybe heavy clouds. Possible that a cooler track would dry slower.

Roger Prynne
30-09-2017, 14:50
Yeah but speeding up Time and Weather is not realistic anyway.
If you want realistic then use real settings.

Trippul G
30-09-2017, 14:59
Yeah but speeding up Time and Weather is not realistic anyway.
If you want realistic then use real settings.

Ideally, yes. But the problem with that statement is that we don't have any control over when weather transitions happen, other than by accelerating time. Otherwise, we need to race for a minimum of 15 minutes to see any kind of change if everything is set to real time and we're using four slots. That's just not practical in a lot of cases, and shouldn't be a requirement for realistic track drying.

Roger Prynne
30-09-2017, 15:05
Sorry I was talking more tongue in cheek really, just ignore me :rolleyes:

Trippul G
30-09-2017, 15:08
Sorry I was talking more tongue in cheek really, just ignore me :rolleyes:

Sorry, can't hear you, you're mumbling. Take your tongue out of your cheek. :p

PeteUplink
30-09-2017, 15:22
Yeah but speeding up Time and Weather is not realistic anyway.
If you want realistic then use real settings.

The problem with that is every weather slot is set to exactly 1 hour (which in itself is unrealistic). So if you've set the weather to go from dry to wet and you're doing, say, a 9 lap Formula Rookie race around Oulton Park, which is over in 15-20 minutes, you'll never see the weather transition.

Basically the surface water shouldn't evaporate as soon as the rain stops, which is what currently happens with weather time acceleration on (video below), and the surface water should dry at a comparable rate to the puddles and standing water.

Also, on the subject of standing water and aquaplaning. I was entering the hairpin at Oulton Park when I was doing my testing and there was a big puddle there. I was driving at 10mph and not accelerating, and the car still aquaplaned. This is actually unrealistic as the weight of the car at such a slow speed would press the car into the ground under the force of gravity and ensure that there's enough traction to drive through the puddle safely. I could understand it if I was travelling faster, or if I was accelerating through the puddle, but I was not.

Edit:

in this video I'm running a few test laps around Oulton Park in a Formula Rookie at 60x weather progression. As you can see, every time the rain stops, the surface water evaporates almost instantly but the standing water stays. As someone who's been a racing fan since I could walk, and have seen a lot of wet races, this is very unrealistic. Pay special attention to Druids, the corner that leads onto the start/finish straight. On lap 1 it's wet. On lap 2 it's drier and on lap 3 it's completely dry except for large puddles. Same with the pit lane. When I started it was soaking, when I finished it was bone dry. Tarmac doesn't dry in 3 laps in real life. The surface water evaporates too quickly in the game because it's tied into the weather progression speed. The standing and surface water should always dry in real time independently from the weather progression speed.


https://www.youtube.com/watch?v=MT0bGzL8xzI&feature=youtu.be

Sadly, PCars weather simulation is rather flawed.

DeliCreep
30-09-2017, 15:51
This video shows the problem. We should be able to set how long it's gonna rain but the track should always dry at the same realistic speed (modified by temperature and other thing of course).

N0body Of The Goat
30-09-2017, 17:00
A LiveTrack progression multiplier slider would be a great addition, which can be set at a different rate than the weather progression.

Trippul G
30-09-2017, 17:06
This video shows the problem. We should be able to set how long it's gonna rain but the track should always dry at the same realistic speed (modified by temperature and other thing of course).

I mainly agree, but I wonder if we really would want to have the track always dry in real time, or if it wouldn't be better to have the drying speed be tied instead to time progression speed? For example, say you're doing 60x time progression at Le Mans to simulate a full day/night cycle in 24 minutes. Say the race starts at 3pm, and it's raining. Say it rains for 4 minutes in real time, so 4 hours at 60x, which would be 7pm in the race. So now the game, and the track, are dealing with 4 hours worth of rain, which IRL, would take maybe an hour or more depending on conditions, for the track to start drying out. So, if the game was forced to simulate water drying in real time only, this water that finished falling at 7pm in the game, would still be around all through the night, the following morning, and into the afternoon when the race ends at 3pm...because our game time is only 24 real life minutes.

I'm not sure that would be any better than what we have now, just different. I think tying it to time progression is the way to go.

PeteUplink
30-09-2017, 17:16
This video demonstrates the issue better


https://www.youtube.com/watch?v=_JrM7_-fhiM&feature=youtu.be

In this one I do a 5 lap race against 60% AI with the weather progressions set to "sync to race". As you'll see, the track goes from absolutely soaking wet to virtually bone dry (aside from puddles) in 5 laps. I think it would be better if the weather progression controlled how quickly the rain passes and once the rain stops the track should dry in real time. Currently the track drying time is determined by the weather progression speed.

Also a couple of other things from this video:

1) The AI were way too slow in places when it's wet, especially in the hairpin. I know it was only 60%, but still.
2) At one point I accidentally hit another car into turn 1. I have full damage switched on, but we both got away with it unscathed. The impact was wheel to wheel, so we should both have had suspension damage.
3) The game looks fantastic in the wet. I just wish the water would stick around a bit longer.