| Pengarang |
Topik: 0% probability phase |
27 balasan
|
|
|
|
#1 dikirim Feb 24 2018, 03:21:22 (terakhir disunting Feb 24 2018, 03:26:28 oleh Akshay Dhole)
|
Sebut
|
I observed Elite group race very closely. Rain start stop is dependent on their lap times, as we all know this.
So, after end of 1st hour, they finished 51 laps. Which is when rain probability becomes 0 means no rain after 60th minute. It lasted for their 69 minutes which is 9 minutes over. This is flaw in the game. If they say rain probability is totally 0 during those 30 minutes (which is 3rd half), they should code in a way rain is stopped right near that time (+/- 1 or 2 laps) it lasted for 5 more laps(rain stopped at 56th lap). This costed me (5*1.8)= 9 more seconds over my opponent who won the race ( unfortunately I used less risk during rain and he gained ~2 secs per lap), which is lesson learned for me.
But, GPRO admins should closely consider this reviewing. Especially 0% probability phase. I understand more than 0% phase like 15-25% or 35-45% etc.
As they say 0% probability is 0000000, nothing else.
|
|
|
|
|
|
#2 dikirim Feb 24 2018, 03:29:15
|
Sebut
|
|
If I'm not mistaken, that 0% is 0% probability that rain will start during that period. If it is already raining starting that period, it is guaranteed to stop some time during that period.
|
|
|
|
|
|
#3 dikirim Feb 24 2018, 03:40:28
|
Sebut
|
|
|
|
|
|
#4 dikirim Feb 24 2018, 03:40:38
|
Sebut
|
Quote ( Adrian Summers @ February 24th 2018,03:29:15 ) f it is already raining starting that period, it is guaranteed to stop some time during that period.
I am still not convinced with this. I know there must be so many threads regarding this.
But need to understand to help myself backup with right strategy in future.
|
|
|
|
|
|
#5 dikirim Feb 24 2018, 03:44:35
|
Sebut
|
I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
I see cars overlapping each other while overlapping around the corner or coming out of pit. We make so much effort to make this game realistic and it still looks weird at times.
Honestly, I am big fan of the game and want to contribute to improve as much as I can
|
|
|
|
|
|
#6 dikirim Feb 24 2018, 03:51:43
|
Sebut
|
Quote ( Akshay Dhole @ February 24th 2018,03:44:35 ) I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
No one can know a weather forecast so accurately before a race starts. To the exact laps where rain may start/stop. I think the "close to within a half hour bracket" is realistic enough for GPRO. Although everyone will tell you here that GPRO is not F1 and it will get annoying very quickly, in this regard I'd treat it as gospel. In GPRO we do not track the weather changes live. We predict them in advance, before the race starts.
Quote ( Akshay Dhole @ February 24th 2018,03:44:35 ) I see cars overlapping each other while overlapping around the corner or coming out of pit. We make so much effort to make this game realistic and it still looks weird at times.
The live race viewer was only quite recently introduced. It's come a long way since then and gets better and better each season. And since there is only one guy working on it with limited funding, I think it has done pretty well in this short time. :)
|
|
|
|
|
|
#7 dikirim Feb 24 2018, 03:53:40 (terakhir disunting Feb 24 2018, 03:53:57 oleh Stuart Foster)
|
Sebut
|
Quote ( Akshay Dhole @ February 24th 2018,03:44:35 ) I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
and do what instead? Where it says 0% it should stop immediately when the period starts? Cos the problem with doing that is those who've played the game for many seasons can then anticipate a weather change...which actually makes it more unfair for new players than it stopping "at some point" within a period.
|
|
|
|
Quote ( Akshay Dhole @ February 24th 2018,03:44:35 ) Quote ( Lyee Chong @ February 24th 2018,03:40:28 )
Adrian is right as of the explaination in the wiki, where it even mentioned that the rain can stop after one minute or after 29 minutes.
https://wiki.gpro.net/index.php?title=Weather
I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
That is what makes things interesting and your "improvement" would actually make things worse.
If everyone knew for sure the exact moment the weather is going to change in advance, the strategies would conform.
Quote ( Akshay Dhole @ February 24th 2018,03:40:38 ) But need to understand to help myself backup with right strategy in future.
Quote ( Akshay Dhole @ February 24th 2018,03:21:22 ) is lesson learned for me.
That is a good idea, learn how to read the forecast and make yourself a strategy which can cover multiple possibilities.
Next time you might get it done better.
After all, predicting weather is not exact science.
|
|
|
|
|
|
#9 dikirim Feb 24 2018, 04:04:10
|
Sebut
|
Quote ( Stuart Foster @ February 24th 2018,03:53:40 ) Quote ( Akshay Dhole @ February 24th 2018,03:44:35 )
I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
and do what instead? Where it says 0% it should stop immediately when the period starts? Cos the problem with doing that is those who've played the game for many seasons can then anticipate a weather change...which actually makes it more unfair for new players than it stopping "at some point" within a period.
Best way is to map it with real weather and leave it like that. i.e. if next race is in istanbul use the real life prediction of weather, so no one will complain about how its done. Atleast we have some base . My thought
|
|
|
|
Ok....but....
Quote ( Mikko Heikkinen @ February 24th 2018,04:01:46 ) If everyone knew for sure the exact moment the weather is going to change in advance, the strategies would conform.
The result i'm afraid Akshay is everyone who has many seasons of play would accurately predict the weather changes by using future forecasts...
This really is not improving things buddy, least of all for players with fewer races :/
It really is better that the weather is random in these mixed weather races, as for many its their best chance of achieving a result that maybe in other races they could not. It's better that people have something to moan about aswell....its healthy that everything is not linear...we might all not like this some times but it does actually make you think more about these kinds of races for the next time....even after 300+ races you'll be learning stuff from these kinds of races.
If people moan about trivial matters in a game like weather....actually, the game has to be doing a lot of things right if that is one of the main things moaned about - which the weather quite often is. But, that's actually a healthy thing for the game. The weather model really shouldn't be perfect for us to prepare for.
|
|
|
|
Quote ( Mikko Heikkinen @ February 24th 2018,04:01:46 ) Quote ( Akshay Dhole @ February 24th 2018,03:44:35 )
Quote ( Lyee Chong @ February 24th 2018,03:40:28 )
Adrian is right as of the explaination in the wiki, where it even mentioned that the rain can stop after one minute or after 29 minutes.
https://wiki.gpro.net/index.php?title=Weather
I am still not convinced with this approach. I know it is adding another uncertainty to the race. But there are other things that can be improved instead of keeping this rule.
That is what makes things interesting and your "improvement" would actually make things worse.
If everyone knew for sure the exact moment the weather is going to change in advance, the strategies would conform.
Quote ( Akshay Dhole @ February 24th 2018,03:40:38 )
But need to understand to help myself backup with right strategy in future.
That is a good idea, learn how to read the forecast and make yourself a strategy which can cover multiple possibilities.
After all, predicting weather is not exact science.
Totally agreed with you Mikko.. I am not against maling it completely predictable.
Simple example is.. if a guy pits 1 pit prior to rain...you can see rain has come but you have set up for rain tyres only when rain starts.. it makes it so unfair.
In real life they will straightaway change to rain tyres.. as they know for sure rain will come in 1 lap( this happened in our group in this race)
|
|
|
|
Quote ( Akshay Dhole @ February 24th 2018,04:10:20 ) if a guy pits 1 pit prior to rain...you can see rain has come but you have set up for rain tyres only when rain starts.. it makes it so unfair.
In real life they will straightaway change to rain tyres.. as they know for sure rain will come in 1 lap( this happened in our group in this race)
True but you can already strategise for this scenario....pit later in the race to avoid the double stop (more fuel) or even choose to pit for wets at your stop.
all of us have been in that situation at some point btw Akshay....quite a few times I pitted one or two laps before rain. Hell, this race went terribly for me with a 1 stop strategy really...but there's no point moaning cos I could've picked XS or soft and had a better outcome...was my fault in the end that I failed to second guess when it might rain and being overfueled for pretty much half the race cos when it rained and then stopped I was driving the equivalent of a fuel tanker round the track :) Of course, had it started raining on lap 35 I'd have probably won had I avoided a mid-race pit :)
|
|
|
|
Quote ( Akshay Dhole @ February 24th 2018,04:10:20 ) Simple example is.. if a guy pits 1 pit prior to rain...you can see rain has come but you have set up for rain tyres only when rain starts..
Again... If you're expecting that rookies (specially on first season) get it right, you might wanna lower your expectations
Quote ( Akshay Dhole @ February 24th 2018,04:10:20 ) it makes it so unfair.
Nothing unfair when the person in question has chosen to make such strategy.
Quote ( Akshay Dhole @ February 24th 2018,04:10:20 ) In real life they ... as they know for sure rain will come in 1 la
Now that would depend on what kind of real life you're talking about.
I don't know if it's gonna rain in 5 minutes. When I'm karting I don't know if it's gonna rain.
But sure, in F1 they have a better idea of the weather than I may have.
Also... you can't compare the whole of gpro to F1, At best you can compare Elite to almost F1, but rookie would be closer to Kids Karting where dad changes tyres in pit and mom offers juice while it's done :)
|
|
|
|
Quote ( Akshay Dhole @ February 24th 2018,04:10:20 ) In real life they will straightaway change to rain tyres.. as they know for sure rain will come in 1 lap( this happened in our group in this race)
Well, not really, they generally change to wets when it's raining already, if the rain's heavy enough to get onto inters or wets. There has to be a massive downpour coming up for them to change to rain tyres before the rain starts, and they have to know for sure it will come on the next lap. And even then they tend to stay out til it's actually raining; how many times have you seen the F1 cars on slicks struggling to make it around back to the pits because the rain is so heavy that they just aquaplane everywhere? There is a big uncertainty level in RL racing too.
Weather is unpredictable, even sometimes you know the rain's coming, you see it on the radar, but it holds on for another lap, then another, then it may be wet in one part of the track, and so on. That's when the teams make the decision when to pit. We can't make those decisions here, and we only have a choice between dry and wet track, like an on and off switch. So we need some unpredictability from time to time, to mix it up a little and give people a chance to put their thinking caps on.
What the gpro weather forecast says is pretty clear, if you read carefully, so there are no ambiguities there. Rain probability refers to the rain starting during that time, not whether it will be wet or dry at all. With 0% probability, there is no chance for the rain to start if it's dry coming into the particular half-hour period, but if it's raining already, then all it says is that it will stop raining during that time (between 1 minute and 29 minutes into that half-hour block), and if it does stop raining, it won't start again. If we had certainty about when the rain will stop, like you propose, then everyone would just make sure they have the right amount of fuel to make it to the end with nothing left in the tank and are on the optimal tyres. There would be no need to think how to cover different scenarios. The weather forecast could just then say that the rain will start on lap x and finish on lap y. What's so interesting about that?
|
|
|
|
|
I get frustrated when when I get the weather prediction wrong, but love it on those rare occasions when I get it right. If we would know exactly when the weather would change and could pick the perfect strategy along with everyone else, then we may as well have all dry races.
|
|
|
|
|
I think we need to put this in the game rules... not in the wikia...
|
|
|
|
You guys must realize that 'the weather' is probably not input for each race by Admin, right?
It must be a 'weather program' it spits out what the algorithms select within certain parameters.
As for cars 'on top of each other' imagine two layerrs on the screen lap 5 on the bottom / lap 6 the leader on top.
As long as they don't touch.. its cool. :)
|
|
|
|
Quote ( Roy Mitchell @ February 24th 2018,18:09:22 ) You guys must realize that 'the weather' is probably not input for each race by Admin, right?
It must be a 'weather program' it spits out what the algorithms select within certain parameters.
hi roy,evrn if it is randomly generated, as with any random generator the results are diferant each time you run the program, so response to this i doubt admins can manipulate any script but whats to stop the script being run many times with varying results and then cherry picked out to the public? im not accusing them of it but the possibility is there.
No weather is guarenteed even in real life,you just have to try to cover all bases in this and take youre chances.
last race i went with slower tyre to stretch out my tyre stint and missed it by 2 laps wich meant the extra stop but if it had rained 2 laps earlier id have gained a full pitstop and its time in the race,hey ho,thats how it goes. you just have to try to cover bases like i said.
theres many random things in this game and annoying as it is can or cant work in youre favour, the plus side of it is that 100% managers all face the same randomness so its equal all round.
keep up the good work gpro and no need for slating anyone for making theyre points,they obviosly feel strong enough to discuss it.
happy racing guys.
|
|
|
|
Quote ( Roy Mitchell @ February 24th 2018,18:09:22 ) You guys must realize that 'the weather' is probably not input for each race by Admin, right?
It must be a 'weather program' it spits out what the algorithms select within certain parameters.
Weather forecast is randomly chosen from a set number of forecasts (hence why it isn't totally uncommon to get the exact same forecasts in different races). The race engine then generates weather from that forecast in however it does it's race-forecasting magic (which is why the same forecasts don't mean the exact same weather).
|
|
|
|
Quote ( Kevin Parkinson @ February 24th 2018,19:54:44 ) Weather forecast is randomly chosen from a set number of forecasts (hence why it isn't totally uncommon to get the exact same forecasts in different races). The race engine then generates weather from that forecast in however it does it's race-forecasting magic (which is why the same forecasts don't mean the exact same weather
random. Thanks kevin :D
|
|
|
|
|
|
#21 dikirim Feb 25 2018, 22:58:36 (terakhir disunting Feb 25 2018, 23:08:49 oleh Roy Mitchell)
|
Sebut
|
You said it and explained better than I, Kevin.
The same weather forecast from an earlier race on the same track... will not guarantee the same racing weather.
edit: Mikie it being 'random' still does not support the admins. 'cherry picking' the weather script.
Just try to imagine how many scripts they would need to review after 63 seasons. lol
Random yes, 'within parameters' as Kevin points out. :)
|
|
|
|
|
|
Weather forecast (no reliable):
1. Weather forecast (°C);
2. Relative humidity (%);
3. Rain probability (%).
Weather fact:
1. Interval: +1/0/-1; <--- Is this true?
2. Interval: >=+-1 and 0.
3. Interval: unmeaning.
Example (S63R7 A1-Ring); forecast/fact:
I. 24-28 °C/24-26°C
II. 24-28°C/24-28°C
III. 22-25°C/22-27°C (Meteorology was a bit wrong. No worries.).
Example (S63R7 A1-Ring); humidity/fact:
I. 46-55%/46-48%
II. 43-51%/44-46%
III. 44-52%/42-45%. (Every time within a range - good.)
********************
I know. GPRO is not the reality.
1 °C temperature change - almost - never causes a relative humidity change of 2-3%. This must be refined. Rough contrast.
********************
I will not go on further because you are stoning.
|
|
|
|
Quote ( Tibor Szuromi @ February 26th 2018,00:10:54 ) I will not go on further because you are stoning. Not yet, give it a few minutes.
|
|
|
|
Quote ( Tibor Szuromi @ February 26th 2018,00:10:54 )
1 °C temperature change - almost - never causes a relative humidity change of 2-3%. This must be refined. Rough contrast.
1°C with all other things constant (closed room for example) will not generally create that much change in RH,but in an open space with, Sun, clouds and most importantly wind.
|
|
|
|
Tibor,
I think you are looking in the wrong places re humidity.
I have developed my own humidity v's temperature formula many years ago.
Look at the bigger picture and think about the physics of weather patterns.
Cheers
|
|
|
|
Quote ( Tibor Szuromi @ February 26th 2018,00:10:54 ) Weather forecast (no reliable):
1. Weather forecast (°C);
2. Relative humidity (%);
3. Rain probability (%).
Weather fact:
1. Interval: +1/0/-1; <--- Is this true?
2. Interval: >=+-1 and 0.
3. Interval: unmeaning.
Example (S63R7 A1-Ring); forecast/fact:
I. 24-28 °C/24-26°C
II. 24-28°C/24-28°C
III. 22-25°C/22-27°C (Meteorology was a bit wrong. No worries.).
Example (S63R7 A1-Ring); humidity/fact:
I. 46-55%/46-48%
II. 43-51%/44-46%
III. 44-52%/42-45%. (Every time within a range - good.)
********************
I know. GPRO is not the reality.
1 °C temperature change - almost - never causes a relative humidity change of 2-3%. This must be refined. Rough contrast.
********************
I will not go on further because you are stoning.
My thought was on the similar lines by trying to relate it to real weather at that point. You took it to different level Tibor, which is good for a discussion.
|
|
|
|
If we based it on real weather since pretty much all of Europe and America have around 5C weather currently (in England it's currently minus temps) then everyone would pick the same tyres as they know it'll be a cold season. Same with summer everyone would pick the same as they know every race will be hot. Having real weather makes it say too predictable temperature wise. Having it currently means noone can just guess what every race temperature is
Plus humidity has nothing to do with temperature in real life. Humidity can drastically xhange regardless of changes in temperature
|
|
|
|
Kyle!
True and good (temperature).
True (humidity) - Open space.
---------------------
1 circle ÷ 1-1.5 minutes.
Budapest, January 2018, Relative humidity:
https://imgur.com/a/jRivn
|
|
English
Deutsch
Español
Română
Italiano
Português
Português
Français
Polski
Русский
Български
Македонски
Српски
Nederlands
Vlaams
Suomi
Lietuvių
Čeština
Magyar
Eesti keel
Slovenčina
Türkçe
Shqip
Ελληνικά
Hrvatski
Dansk
Pirate
