Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
---
As a quick test of what happens in game, I've just run a quick race as Steinmann at Phoenix on the full oval - Harry Chapman underfuelled by enough he'll have to do the whole race on low, and Nina Holtz on full fuel tank and overtake until they pit; she's used an extra 5 and a half laps of fuel to be 9.5 seconds down on Chapman. (who incidentally, is now only 12 laps down on fuel; I guessed that it would be out of fuel if he started with only 59 laps (out of 74) but 20 laps in he's recovered just under a third of the fuel he's lost, so he could have underfuelled further.
Now redoing it again with the strategies reversed, and this time Chapman's lead was small enough they got stacked at the pitstop (mainly because I wasn't concentrating and left them on attack too long and ruined the tyres after 12 laps) which extended the gap massively between them - it was 3 seconds after the pitstop, it reduced down to 1.8 seconds while Holtz's car was quicker, and now at the point where Chapman has less fuel than Holtz and is now rebuilding the gap because Chapman simply thrashes her, the gap is 2.3 seconds.
Now redoing it again with them both on the same underfuelling strategy as a control test, and this time, Chapman leads Holtz by 10.4 seconds after 20 laps.
As this was a quick test, I was not controlling for all the variables you ought to control for - therefore, it must be stated that Holtz's randomised setup was much better on the full fuel and overtake strategy than it was on the equal underfuel strategy, and therefore, the fact she is down an extra 0.9 seconds cannot fully be placed on fuel/engine mode.
Essentially, I was expecting this track to prove the higher engine mode strategy better, and at that, it has comprehensively failed.
Generalising for other tracks, I'd expect the disparity to increase, given that the game's own data says each lap of fuel at Phoenix oval is only worth 0.078 seconds per lap, while at other tracks (ones that aren't pretty much full throttle everywhere) it can be worth 3 or 4 times that.
So if I understood right you are saying that:
- Chap underfuel trashed Holtz overfuel.
- Chap overfueld got just a small advantage over Holtz underfuel.
- Both underfuel and it was similar to Chap under and Holtz over.
Conclusion:
- Chap was definitely better underfueled while Holtz didn't had much of a difference when under or over (related to Chap).
- All in all, underfueled is better than overfueled or may not give a particular pilot/car a boost but won't hurt anyway, while overfueled may hurt.
I think I will give this a longer test. For, like, 3 or 4 seasons I been doing the overfuel strategy and while doing a single race wont prove much, I will try the underfuel for this whole season and see how I feel about it.
Also, as far I can tell: the races where I do a lot of pits (like dry > inter > wet > inter > dry) I does seens to pays off by A LOT (compared to the AI) to overfuel since you will be carrying a low amount of fuel anyway so not much extra weight. It allows me to win those races even with a inferior car, but I also may be the result of me consuming much more rubber than the AI (wasting a whole tyre in just 10 laps in Attack mode since I would have to change tyres anyway) but even before I start to eat the rubber I was doing better. It could be worse when you are doing just one or two pits in a race so carrying around all that extra fuel can be bad.
"- Chap underfuel trashed Holtz overfuel." Yes. (of course, Chapman trashes Holtz anyway; the Steimann season is meant to be the Hammy and Rosberg show, and as it took everything Rosberg had to beat Hamilton, so it takes Holtz everything she has to beat Chapman)
"- Chap overfueld got just a small advantage over Holtz underfuel." I wouldn't say this at all; this massively worked in Holtz's favour, given we know on equal strategies Holtz is thrashed at a rate of half a second a lap. It was a terrible strategy really, Holtz was actually 3rd after the pitstops.
"- Both underfuel and it was similar to Chap under and Holtz over." Yes - but, at the 20 lap mark, third place was 24 seconds behind Chapman, rather than 15 seconds behind Chapman as per the 1st test. They both gained hugely from the strategy, it was just I omitted the fuller picture for the sake of brevity, and the fact I have no idea what strategy Lipponen/Wexler (1st race/3rd race respectively) were using as I didn't bother tracking that, and therefore, a comparison is even less able to be drawn.
Basically, I did a very poor test, as I assumed the overfuel/overtake strategy would work brilliantly for an oval, so well in fact that it would be clear; even without controlling all the variables. As it stands though...
Trying to correlate for a refuelling championship would mean that effectively you'd only be adding fuel while you change the tyres, assuming a fuel efficient enough engine.
The time gain per lap of fuel is listed in the car setup fuel options screen.
Twelve seasons, man. Twelve seasons and I hadn't noticed it. I'm blind or stupid, hahaha! Anyway, it doesn't awnser my question but it does give me a lot to consider.
I did a very good Qualifying due to the weather. First driver 4th, second one 8th. Last season most of the time they are between 8~12th. Again, just because of the weather and most of the AI ruining in a wet track while me and some other run in a briefily dry track.
The whole race was dry with 1 pit. Starting with softer tyres and pitting for medium tyre (with a extra rubber to burn).
I will just speak about my best car for now on:
What I expect based in the last season overfueling+overtake: in a pretty normal race (no crashes, no crazy weather, AI qualifying decently). I usually start 10~12th and lose or gain few positions in the first lap, and in the following 2 laps I gain back my positions and most of the time some more. In the following 2~3 laps I'm not able to keep any extra higher positions if I grab them before (4~6). I generally blame the tyre since I'm colling them down after the aggressive start. So starting in 4th I expected maybe keep this position or win some in the early laps, but fall to 6~9 before lap 10. That's not how it went.
Early on was the usual knife fight and I drop to 5th by lap 2. On the following laps I was able to keep myself pretty close to the front cars using Push and in a couple of laps and before lap 10 I overtook a car putting me back at 4th. For the end of the stint my pilot held his position in front of a good team car and close enough to the good team car in front of it. I overtook car in front of mine during pits (he pit before me, I pit after and still came out in front). For the rest of the race I managed the exact amount of fuel I need (Pushing for some laps, Medium for others, if I have more fuel I burn, saving just half-lap for an eventual overtake) my driver hold his position always around 8 secs after the 2nd and 3 secs ahead of the 4th. In the last laps, with a lot of rubber to burn and some fuel to spare, I finished around 4 secs behind the 2nd and 6 secs ahead of the 4th. I expected high rank drivers and teams that got messed in the qualifying to catch up but they really didn't (I mean, I wasn't slow enough for them to do it, they did won positions behind me).
That was a pretty solid show that I didn't had similar last season. Ok, first race and I didn't have any mechanichal problems (for that matter relevant AI drivers also didn't saw to have anything serious), but still a much better start than last season. My second driver, by the way, hold his 9th position for most of the race and in the final laps when burning extra rubber he did overtook the car in front of him from a team that I really tought my second car have 0 chances to keep up against.
First races always are a anything-can-happen, but I'm impress so far. It does seen that carrying less fuel and smartly managing it is better than overfueling for a regular race. Ah, worth noting: I only used the blue engine mode in the whole race for two laps with my second car, near the end of the race, when I making sure he would have enough fuel to push in the last laps when burning extra rubber (and paid off).
At Doha, according to the numbers the game presents, on a 42 lap race 1 lap's worth of fuel is worth 5.8 seconds at the end of the race, which is an overtake and a bit - especially in a refuelling championship, as maintaining that reserve means more time spent stationary in the pits during refuelling.
At the moment I'm running a Rossini save to test it, and my usual strategy is overtake until 3 laps below fuel delta, then fuel save on medium, which in effect means you can underfuel by 3 laps; I know this works, it is how on my other Rossini save I had Wexler and Saarinen 1-2 for the championship in 2016. Now, on low for the whole race, I can underfuel by 8. (Wexler was marginal on fuel at the end but still won after manipulating Chapman into pushing his tyres too hard and therefore needing to pit, Saarinen had 0.30 of a lap left as he had a more fuel efficient engine, which means in effect he could have had a quick burst of overtake at the start, which might have helped with his terrible qualifying...)
This means that 29.2 seconds can be saved from not hauling around as much fuel (low vs medium) which means the question for me is how 3 laps of overtake and then 39 on medium compares to 42 laps on low.
----------
Sigh. What would be incredibly useful is having the option to use the test track you can build for testing; run a full race simulation in a far more consistent environment than a real race track is, and then you can get real, usable data, rather than having to do analysis of less reliable data, as there are hundreds of ways race-derived data can be tainted for usability if used for the purpose of abstract fuel usage calculation. (even by save/reload, you can't effectively control for testing unless you are willing/have the time to do a whole season's worth of driving for one race)
The key things that need verification are 1. how reliable the time saved per lap of fuel number is, and 2. time gaps between all the engine modes over race distance.
Yup, that seens to be a big problem. It's pretty hard to have a perfectly undisturbed race to test out. Traffics, weather, track grip, pilot form, AI performance, qualifying, mechanichal failure, parts improvement, car and tyre wear are all factors in the race so it's pretty hard to put out a decent test, that's why I'm going for a whole season of tests and even so it's not accurate by any mean. Even pratice season isn't really reliable because we can directly control the fuel amount, there still is some traffic and track grip is a huge factor there.
Also, about the extra lap for overtake, I don't save it to the end. I slowly spend it during the stint, so at the pit stop or end of the race I will have as few fuel as possible. Example: if it's the last stint of the race, I may start burning it right away or not even load it. If the car ahead of me is far away, I also don't save it. I only keep it if I'm really close of another car, ahead or behind me, so I can push/overtake when needed and if the time don't come I just burn it before in the last few laps. I think you understand it as much (or more) than me: golden rule is to finish your stint with no extra fuel or rubber.
There are also no testing restrictions until you actually enter the championship; for example, Porsche, with it's 200 former-LMP1 program guys, could be benchmarking a potential 2020 F1 team fairly easily. (they'd need to expand on the employee count a bit for F1, but if I was Porsche, it is what I'd be doing if I intend to keep those guys in their jobs, the 919 EVO project, while it is great, I strongly suspect is setup for something else)
Frankly, I'd rather the sport went back to the old ways when you were allowed to test freely, but I digress...
----
The main thing I was getting at is that an overtake takes place when a faster car catches a slower car, and makes the pass - the crux of the matter is can the several seconds saved by not running that fuel make the pass happen in a different way to high engine mode + slipstream + last on the brakes.
The more I go, more convinced I'm that overfueling and overtaking all the time does not worth and actually hursts your race time. I just keep doing it in very short stints (<10) like in changing weather and in this situation it does seens to pay off. It also paid of in a race where I miscalculated my tyres and had to pit with only 8 laps left and by overtaking and attacking the whole time I even finnished better than before the pit (worth noting that the AI had pretty ♥♥♥♥ tyres at this point). This last thing made me consider doing 2 pits, not all the times, but in some races that I usually went for just one.
Hi!
For question #2, the answers are in the DesignData.txt files within the resources.assets file in the game's data folder. This database file can be extracted via the Unity Asset Extractor tool.
Within the DesignData file, there are sections about the different time gaps--coded as "time costs"--for each engine mode. These numbers vary based on the length of races you are playing (Short, Medium, Long). The section begins with the maximum time delta between the highest engine mode (Super Overtake) and the lowest engine mode (Low).
Below is an excerpt from the SingleSeaterDesignData.txt file, for Long Races:
Running the engine mode performance values through the calculation, you'd get the amount of time added to a car's lap time--per lap--due to the engine mode.
So, theoretically, that answers the effect of the engine mode's performance on a car's laptime. The next part is to calculate the effect of the fuel load on the laptime, and account for the variable fuel burn rates per engine mode. This gets a little tricky because of the way the simulation calculates fuel burn; the engine mode is all relative based on the number of laps. Rather than having an absolute fuel volume (or fuel mass) number that is being burned off, it's all based on percentages of laps, which can make the calculation a bit tricky.
I've made an assumption that the time penalty due to the fuel load is calculated by this equation:
So, to use an example, if Car 1 plans for a full 25-lap stint using the Overtake engine mode (~1.35 laps / lap fuel burn rate), it will need to be fueled for 25 * 1.34 = 33.750 laps to just make it across the line at the end of the stint under its own power.
Let's say the fuel time delta per lap is 0.073 seconds per lap (just a random number I remember based on my playthrough). I'm assuming, therefore, that the time cost for having a 33.750-lap fuel load will be 33.750 laps * 0.073 seconds = +2.464 seconds slower during the first lap because of the fuel.
So, theoretically, assuming that the only factors affecting the lap time is engine mode performance and fuel weight, Car 1 will lap 0.400 + 2.464 = 2.864 seconds slower on the first lap compared to its final lap of the stint.
On Lap 2, the calculation is repeated, but since Car 1 is using the Overtake mode, it's burned off more than 1.00 laps of fuel (1.35 laps of fuel in this example). So the time cost for the fuel load is (33.750 - 1.35) * 0.073 = +2.365 seconds .
So, with that in mind, the fuel stint for Car 1 (Overtake mode) and Car 2, under perfect conditions, should look something like this:
Elapsed Time: 2122.429 seconds
Elapsed Time: 2131.149 seconds (+8.720 to Car 1)
So, unless I screwed up my math somewhere, having a car on Overtake mode for an entire fuel stint is able to gain a substantial time advantage (almost 9 seconds in this example) over a car in High engine mode and carrying less fuel. The effect of the Overtake engine mode (the time cost value in the DesignData) is larger than the correspondingly heavier fuel penalty.
But if the car is overfilled too much, that advantage can be erased due to a higher fuel time cost penalty. There is a "tipping point", if you will...
For example, if Car 1 in Overtake mode had filled up with an extra 5.000 laps of fuel, that extra fuel time penalty would cause it to have a total elapsed time of 2131.919 seconds, which would put it 0.421 seconds behind Car 2, which was only using the High engine mode. All of that advantage Car 1 would have had, would have been squandered.
Of course, this doesn't account for the driver mode time cost system, which has its own set of values, nor does it account for the car differences, the driver variability, tire wear effect, and so on. That's where the simulation gets far more complicated... and fun!
EDIT: Formatting
I mean, while "theory" says one thing, the pratice is telling me another. Like you said, I think the simulation is way more complex than just what engine mode is faster in the spreadsheet.
To have greater accuracy, the data for the WMC calendar (2016) is this:
Track - fuel delta - race length
Doha A - 0.139 - 42
Yokahoma A - 0.124 - 47
Beijing B - 0.092 - 63
Tondella A - 0.122 - 48
Black Sea A - 0.117 - 50
Munich A - 0.122 - 48
Cape Town A - 0.091 - 64
Guildford A - 0.137 - 40
Milan A - 0.147 - 40
Ardennes A - 0.143 - 42
Phoenix C - 0.107 - 55
Vancouver A - 0.134 - 44
Singapore A - 0.104 - 56
Sydney A - 0.143 - 41
Rio de Jenero A - 0.138 - 43
Dubai A - 0.161 - 36
While Beijing changes the situation least, in all cases the overtake car will be worse off when the maths is redone; the disparity between the two of you is probably because the number remembered is for a different category of car, or a different track layout. (I've been presuming the 2016 calendar as that is the only fixed point, there can be a lot of changes over 12 seasons...)
Either way, this is data pulled ingame, long race setting, all from Chapman/Steinmann. (as, by virtue of the car's dominance, they are the easiest to test stuff with)
Taking out that much fuel makes a huge difference. I can't bring myself to do the Big Overfuel. My teams run light, and pit as few times as possible. I usually design cars with 4 star fuel efficiency in the WMC and hope for a dilemma to push that to 5, and I go for race engineers like Toby Hart that have lightfooted perks (soft is the most consistently useful I think). Everything is geared to running lighter and longer.
Of course a wet race can change this. If you know rain is coming after 5 laps, overtake is better, but maybe massively overfilling at the start is better yet because the pit stop will be shorter. All this would be dependent on fuel tank size, ERS rules, pit lane speed limits, pit crew size.