Why did the Mini win?

[PlanetJanet]

Not the beginning of a bad joke I'm afraid, just curious if anyone has any idea why the Mini was given the win in this race at Brands Hatch? For context, it was in a lobby with a few friends (neither car is me), the Pallas had been gaining down the straight. Initially I thought it must be to do with when the wheels cross the line (rather than the front bumper) but it's clear that the Pallas got their first. So what determines the result?

 

[OldandSlow]

Are there penalties in Lobbies? That’s all I can think of.

 

[Famine]

in a lobby

So... the nature of this sort of thing means that your version from your point of view is subject to all the latency that you're subject to and may not reflect the "real" events.

It would be interesting to see the same moment from the points of view of the Mini driver, the Citroen driver, and - particularly - the lobby host.

 

[toastet]

Netcode or bad connection. Not uncommon.

 

[Panoz]

Not the beginning of a bad joke I'm afraid, just curious if anyone has any idea why the Mini was given the win in this race at Brands Hatch? For context, it was in a lobby with a few friends (neither car is me), the Pallas had been gaining down the straight. Initially I thought it must be to do with when the wheels cross the line (rather than the front bumper) but it's clear that the Pallas got their first. So what determines the result?

View attachment 1346936

Did you save the replay or got a screenshot of the finishing results for this race? As others said already, its likely latency skewing the result of this race.

A league series I run in had a similar event in the photo below with the Toyota seemingly finishing ahead the Mercedes but they actually tied going off their total race time. The in-game results gave the Mercedes the higher finishing position.

 

[PlanetJanet]

So... the nature of this sort of thing means that your version from your point of view is subject to all the latency that you're subject to and may not reflect the "real" events.

It would be interesting to see the same moment from the points of view of the Mini driver, the Citroen driver, and - particularly - the lobby host.

Ah interesting. I was the host, but the replay is not mine (replay was only saved by the Mini driver so we can't check different perspectives). Both parties were surprised by the result, thinking that the Pallas had nudged in front.

@Panoz, there is a replay but only the one perspective, no screenshot of the result (which gave it to the Mini by 0.003 I believe).

Never happened to us before, it made for a fun evening where everyone was being accused of corruption.

 

[Meythia]

replay

Replay as Replay file should contain all information, shouldnt it?
Replay as "Videofile" will only be identical to the way the player has recorded it.

 

[PlanetJanet]

Replay as Replay file should contain all information, shouldnt it?
Replay as "Videofile" will only be identical to the way the player has recorded it.

I don't know, to be honest. That screenshot is from the replay file (saved after the race and shared via the showcase), not recorded using the Share button or anything like that.

 

[p78]

Ask the Mini driver to share the replay.

 

[Lomic]

Replay as Replay file should contain all information, shouldnt it?

As far as I can tell, replays (files) only replay from the point of view of the person who saved the replay. Any lag from other players will be present on the replay.

The other day I was watching a replay I've saved from a lobby race. During the actual race, some players' car were moving more or less jerkily. When I watched the replay those jerks were present even when I switched the car I was viewing. To the point where while watching the lap from bumper cam of the winner, the car would jump left and right in the turns, and it was really not usefull to analyze his driving to get clues of what he was doing. Accerator and brakes were all over the place too.

So to be 100% sure you need to compare replays from both drivers.

 

[Famine]

Replay as Replay file should contain all information, shouldnt it?

As far as I can tell, replays (files) only replay from the point of view of the person who saved the replay.

Exactly this. It's a telemetry file, and that includes the telemetry received from other players - and when - just as it was in the race you took part in.

I suspect that the "real" version of events in a lobby is the one the host's machine decides, hence my note about seeing the host's replay - but it might not be.

 

[PlanetJanet]

I wish I'd saved the replay to test this theory! Next time.

 

[Lomic]

I suspect that the "real" version of events in a lobby is the one the host's machine decides, hence my note about seeing the host's replay - but it might not be.

I'd be more inclined to believe that the replay of the host is the same as anyone partaking : just his telemetry + the telemetry received from other players during the race (with all the issues it can show where some players have a not so good connection).

But it would be interesting to check.

 

[Famine]

I'd be more inclined to believe that the replay of the host is the same as anyone partaking : just his telemetry + the telemetry received from other players during the race (with all the issues it can show where some players have a not so good connection).

But it would be interesting to check.

Definitely, but for timing purposes there must be a definitive version of events - and that's either some weird calculation based off several (or all of the) machines or one of the machines is the "master" timekeeper. In the latter case it makes the most sense it's the lobby host, but that's only why I suspect it rather than anything more concrete.

 

[Lomic]

I think that it is something easy to test. Next time we race with the "Titans" the host (@Pesselles ) will save his replay, and we will see if some cars are jerky or not.

 

[Dragonwhisky]

Because the Pallas was stapled to the Lag Monkey.

Ain't ascairt of bad jokes.

 

[Famine]

I think that it is something easy to test. Next time we race with the "Titans" the host (@Pesselles ) will save his replay, and we will see if some cars are jerky or not.

The jerkiness isn't really part of it though, because that'll happen regardless. We're not looking for cars jerking about or not, we're looking to see whether the host console is the timing master or if something else is.

Pretty much the only way to test it is to engineer a close-to-dead-heat (per the first post here) and compare what the results say to a frame-by-frame replay at the finish line for as many participants as possible.

If the host is the master timekeeper we should see the "correct" final result on their replay - but that doesn't mean that it is the master timekeeper, as some will show one car leading and some will show the other car leading and simply being correct isn't the arbiter (as several will be correct and they can't all be the master).

However if it shows the incorrect final result, it definitely isn't the master timekeeper and something else is at play: it could be that the console with the lowest ping is selected, or the one at the geographical centre, or it's aggregated from several (perhaps all) consoles, or something else...

 

[Dragonwhisky]

Famine = Master Troubleshooter

 

[P-Courty]

Pretty much the only way to test it is to engineer a close-to-dead-heat (per the first post here) and compare what the results say to a frame-by-frame replay at the finish line for as many participants as possible.

I guess the best way to test this would probably be a head to head drag race with each person driving the same completely stock cars on auto transmission (preferably a car that doesn't wheelspin so much) to get a very close finish.

Run this a couple of times and compare each session replay.

 

[D_Dragline]

You’re all making this way more complicated than it is. GT7’s timing uses the driver, not the front of the car. The Mini won because the mini driver hit the line first. End of Story.

 

[Meythia]

You’re all making this way more complicated than it is. GT7’s timing uses the driver, not the front of the car. The Mini won because the mini driver hit the line first. End of Story.

By perspective of the screenshot this isnt 100% clear, but I dare say the driver of the mini is further away from the line by looking at the hands of both of them.

 

[D_Dragline]

By perspective of the screenshot this isnt 100% clear, but I dare say the driver of the mini is further away from the line by looking at the hands of both of them.

The drivers aren’t at the line though. I bet if you take a screenshot at the point the drivers cross the line, the Mini driver is in front.

 

[PlanetJanet]

The drivers aren’t at the line though. I bet if you take a screenshot at the point the drivers cross the line, the Mini driver is in front.

Hm. Interesting that it's based on the driver, I didn't know that. I'll try to take another shot, but the Pallas was the one overhauling the Mini, so considering the drivers look pretty level in this shot, I'm not sure how the Mini driver would cross first.

 

[Clinton Ducks]

So what determines the result?

View attachment 1346936

I was gonna say, maybe it’s which driver’s arse crosses the line first - but then I read the rest of the thread.
Yea it’s probably latency.