"World Record" times unrealistic?

[R8pilot]

I practice on the Nordschleife most of the time. I'm relatively new to sim racing, but I've put in enough laps to have matured in the sense that my better laps are all within a few seconds of each other. I'm still improving, but by now it's a fight to lower my times by more than a second or so.

However, I note that the per-car "world record" times seem unrealistically fast. On my current favorite car (a generic 80's-ish F1 car), the record time is almost a minute faster than my best time, and that story is about the same on every other car I've tried. I do run the cars stock, but I'm skeptical that tuning would deliver such radically-faster times.

So, are these record times actually realistic? Are they achieved in hacked cars? Or am I just lame? (Don't answer that last one! )

 

[wolftree]

I practice on the Nordschleife most of the time. I'm relatively new to sim racing, but I've put in enough laps to have matured in the sense that my better laps are all within a few seconds of each other. I'm still improving, but by now it's a fight to lower my times by more than a second or so.

However, I note that the per-car "world record" times seem unrealistically fast. On my current favorite car (a generic 80's-ish F1 car), the record time is almost a minute faster than my best time, and that story is about the same on every other car I've tried. I do run the cars stock, but I'm skeptical that tuning would deliver such radically-faster times.

So, are these record times actually realistic? Are they achieved in hacked cars? Or am I just lame? (Don't answer that last one! )

I doubt there are any fake times. The FClassic gen 1/2/3 allow to push much more than you initially think. On corner exit when you think the rear tyres are slipping you should just give it more throttle and it will actually stick. The Nordschleife has 170 corners so if you lose just 0.2 sec per corner, your final lap time will be 34 sec behind.

 

[R8pilot]

My question is an attempt to calibrate my expectations (though I don't ever expect to actually match or beat the records) and the time divided by the number of corners came to me as well.

But "170 corners" is likely high when so many of those are just little kinks that can be taken at very high speeds, and often flat-out, meaning that significant gains on those would be very hard to come by.

I've seen many people claim something like 73 corners, meaning I'd have to improve by something like 0.8 second per corner. That feels unrealistically out of reach, though of course corners aren't everything either.

Another way to look at it is comparing the speeds of historical cars. So the Lotus 49 ran the historical ring (which had fewer chicanes than AMS2's ring, I think), driven by one of the worlds greatest drivers of all time, in just over 8 minutes. Unless I've misplaced it, there is no original 49 in this game, but the record in the closest Lotus (49C, which admittedly has some early aero), in 6:50. That's about 1 minute 14 seconds faster! It doesn't seem right.

Then, a "Formula Vintage Gen 1", arguably close to an original Lotus 49 in configuration, power and weight, etc., also has record times more than a minute faster than the real 49. To my thinking, it's just too much faster to make sense.

Or is it!? That's my dilemma.

 

[wolftree]

Another way to look at it is comparing the speeds of historical cars. So the Lotus 49 ran the historical ring (which had fewer chicanes than AMS2's ring, I think), driven by one of the worlds greatest drivers of all time, in just over 8 minutes. Unless I've misplaced it, there is no original 49 in this game, but the record in the closest Lotus (49C, which admittedly has some early aero), in 6:50. That's about 1 minute 14 seconds faster! It doesn't seem right.

Then, a "Formula Vintage Gen 1", arguably close to an original Lotus 49 in configuration, power and weight, etc., also has record times more than a minute faster than the real 49. To my thinking, it's just too much faster to make sense.

Or is it!? That's my dilemma.

The Nordschleife was different in the 60's, more than a mile longer than the modern configuration (without the GP circuit). Also those guys didn't have infinite respawns and weren't racing in a sim rig in their house.

 

[cpcdem]

@R8pilot why don't you just load the ghost of one of those top times, to see where they are gaining on you? Or, because those ghosts will disappear into the distance very quickly, maybe load a ghost that's 10 seconds faster than you and see if you can catch it. Usually that's how it works, at first you think it's impossible to improve by 10 seconds, but you start figuring out how to do it, then put another target for another 10 seconds faster etc.

 

[Gevatter]

Don't compare in-game world record to real life records, the in-game ones will always be faster. There will always be differences to the real world, since no game currently is able to completely 100% realistically simulate every aspect of whats's going on in a real life car.

Adding to that, simracers control every aspect of their run and can set them to maximise their times

• Set Time of Day for best sunlight, time never changes.
• Optimally rubbered in track, consistent across the whole lap
• Consistent low fuel load and perfect tire quality, never changing
• Constant perfect temperature for maximum engine power and/or tire grip
• Virtually endless seat time to practice the car and the track - the 60s F1 drivers weren't able to run around the Nordschleife for threee weeks straight for several hours a day trying to eek out the last hundreth
• Virtually endless time to refine setups since it takes no time and team to actually do the changes
• No risk to limb and life
• No monetary cost
• Damage can be disabled and even when enabled, the worst that happens to you when you crash is a lost lap and four clicks to be back running
• No other drivers if you don't want them there

Real life records aren't usually set in all those conditions, at least you cannot control all of them like you can in a sim.

 

[R8pilot]

Ok, so the consensus is that:

(a) The track has changed far too much for direct comparisons.
(b) Tires, fuel, suspension, environmental conditions, etc., can be set up ideally in sims, in ways that actual cars in the real world can't expect to experience.
(c) The people piloting them have run potentially thousands of utterly berserk laps during their attempts, and they likely "died" hundreds of times before finally getting that record that no actual driver in fear of life and limb could rationally achieve.

It does make sense. Thanks!

(Back to improving by a second or two at a time, apparently with a long way to go. )

 

[Amanda]

Most sims nowadays are unrealistic when it comes to the actual racing overall.
"Back in my day" there was no "Time Trial" that put you on the last corner, with infinite 5L fuel and infinite tires. You had to lap the track like in real life, use fuel and tires.

Still, even sims like ACC and rFactor 2 seem to be quite unrealistic when it comes to lap times. Perhaps not because of their physics interpretations, but a combination of factors:

- Allowing the driver to go beyond the white lines even though FIA regulations prohibit this (most sims allow this in qualifying, just look at lap records in Spa, for example, look how much the "aliens" cut on Radillon, at the exit of no-name, and at Stavelot, it's ridiculous);
- Allowing for all new tires at all times;
- Allowing for maximum grip everywhere;
- Allowing for unlimited 1 to 5 liters of fuel;
- Allowing weather control, always nice weather.

If you put all of these together you don't get a proper "simulation", you get a toy that allows drivers to feel like they're accomplishing something special. They're not.

Here's what I would do if I could develop a simulator:

- If the driver crashes their car, their career could be compromised, making them return to lower classes to earn money and respect before allowing them back up;
- Have a limited set of tires for practice, qualifying, and race;
- Always use current weather information;
- NEVER allow for full grip on the track;
- Put some leaves and oil in certain places, reducing the grip on these areas a bit;
- Force the drivers to use and calculate fuel;
- Force the drivers to drive the lap;
- Limit the number of laps or time to get the laps done.

There. Now we actually have something that actually simulates real life, not an unrealistic all-around setting that we get on all "simulators" these days.

 

[Amanda]

There is a "however", however.

I went back to Sim Racing after 2 years away with no wheel and I was able to do a 2:15:1 in Spa with a GT3 car in AMS2. I thought "how ridiculous" that was, since the 24h even (Blancpain) usually have qualifying laps of 2:18:X.
However, the GT3 lap record for Spa was in the GTOpen event and it was a 2:13:7 IIRC. So depending on the event you pull the lap info from, they might use different tires and have entirely different regulations for BoP (ballast, power limits, etc). So who knows. If the 2:13:7 driver didn't cross any whitelines during his lap, 2:15:1 doesn't look so unrealistic.
But to add a "however" to this, what's up with the leaderboard guys in AMS2 doing 2:11's?

The same for Nurburgring. The Blancpain event usually registers qualy laps of 1:53-1:54, but sims are doing 1:50 to 1:51. If we use the same case as the Spa record, 1:50 might actually be realistic.

 

[b)b]

I would not take leaderboards times as a reference for anything. It's about abusing physics (setups) and track limits as much as possible, which might be fun but not super representative of regular lapping.
Also, top times may have been achieved with different physics and Reiza plans to wipe them at some point.

 

[wolftree]

There is a "however", however.

I went back to Sim Racing after 2 years away with no wheel and I was able to do a 2:15:1 in Spa with a GT3 car in AMS2. I thought "how ridiculous" that was, since the 24h even (Blancpain) usually have qualifying laps of 2:18:X.
However, the GT3 lap record for Spa was in the GTOpen event and it was a 2:13:7 IIRC. So depending on the event you pull the lap info from, they might use different tires and have entirely different regulations for BoP (ballast, power limits, etc). So who knows. If the 2:13:7 driver didn't cross any whitelines during his lap, 2:15:1 doesn't look so unrealistic.

As you point out, GT3 pace depends a lot on the BoP, since some cars in the class can go much faster without it while others would struggle to keep up. AMS2 GT3 cars are based on the Endurance Brasil series, which doesn't use as strict BoP as the GT World series, hence the cars can be quite a bit faster. Also the 1.2 update changed the tyre model and this made the cars even faster.

 

[R8pilot]

I don't quite know what changed in recent updates, but the driving has improved dramatically. Not with faster times, necessarily, but with much better handling or tire feedback or something. Whereas before I'd frequently skid out while pushing hard on a hot lap, now I rarely do, and driving confidence has increased enormously. I'm tempted to take credit for the perceived improvement myself (I'm still improving somewhat rapidly), but it's probably mostly the sim.

Still, I'm under 6 minutes on the Nordshleife (in a retro F1 car), while the record is around 40 seconds faster still. That doesn't seem impossible, but it still feels improbable without some advantage that I'm lacking (aside from a difference in skill). How are the (ghost) cars able to pull away as they do on the straights, for example? Drastically reduce aero, minimum fuel, close all vents, etc?

A related question: Most, but not all, track times have a "wrench" symbol next to them, which I'm guessing indicates that the driver changed some parameters. What triggers this? Only "inauthentic" aids, such as traction control in a car that didn't have it? More? Any changes at all from stock?

 

[R8pilot]

I've knocked a lot of time off the Nordschleife since my OP above, and now am generally within a handful of seconds from world records in the 90's F1 cars I favor (who is that guy I keep seeing on top!?). What I've learned since my naive question about records:

Who needs more than a liter of fuel!?
Or open brake or engine vents?
Or tires that wear out?
Or an engine that might blow?

These were mentioned by others above, and are indeed taken advantage of in most of the fastest lap times. I congratulate those souls who put in their awesome laps without these cheats.

But then the more realistic yet difficult one: Drive with as little aero as you can manage. As much as the Nordschleife benefits from ample downforce in the corners, lap times benefit even more from the enhanced top speeds you get without it. The fastest times are generally achieved by ice-skating through the track with almost no downforce, and learning to do this is almost a necessity to be competitive on Page 1.

 

[Kurupt CDN]

Who needs more than a liter of fuel!?
Or open brake or engine vents?
Or tires that wear out?
Or an engine that might blow?

Sadly this carries over to 99% of the MP lobbies too, where they turn everything off ?

It would be nice to see realistic settings in time trials.

 

[R8pilot]

That brings up a question I have: What does the wrench symbol mean?

I've thought it would make sense for it to indicate modified/unmodified controls, e.g. automatic transmission in a car that was manual, or traction control on in one that never had it. But maybe it's more mundane than that?

Also, the car symbol?

 

[GeekyDeaks]

IIRC The car symbol means there is a ghost you can use. The wrench seems to show once you adjust any of car setup from default. I never tested to see if assists cause it to show though.

I'm actually pretty new to TT and found a quirk due to what I was using it for. Normally when I selected a ghost and showed the replay of a lap, the ghost seemed to be stationary, however I found that if you do a lap, then come out of TT and re-enter, you can select both your ghost and the one you want to compare to and then after launching the TT, but before entering the track, click on the ghost details box and it lets you view the lap, albeit without full replay controls. I usually just record the session and view it on a standard video player. Would that help you with your final few seconds at all?

 

[Leynad777]

That brings up a question I have: What does the wrench symbol mean?

I've thought it would make sense for it to indicate modified/unmodified controls, e.g. automatic transmission in a car that was manual, or traction control on in one that never had it. But maybe it's more mundane than that?

Also, the car symbol?

The wrench-symbol means the sim-racer was not using default-setup. You can just pick the ghost-setup if there is a ghost- and a wrench-symbol. If there is a ghost but no wrench symbol, he or she was using default setup.

About the initial post: If you shift h-pattern F1-cars with flappy pedals, I guess you loose 10 sec. alone through the shift-delay. The fastest way is using h-shifter with auto-clutch in AMS2 in those cars to left-foot brake, but without getting the shift-delay that's built in the game if using sequential.

After almost 8 years of sim-racing and the first three years almost just hot-lapping Nordschleife in AC I can tell, mastering the Nordschleife is more than managing it. You need to drive on the limit all the time using the optimal line in every corner and technique to get close to the best times or beat them. I got better around Nordschleife by practicing other tracks instead. The Nordschleife is IMO NOT a great teacher and especially modern GP-tracks like Hockenheimring or Silverstone do a better job for the driving-technique because some corner-combinations are awkward and unnatural and force you to think outside the box.

I've set some dozens 'WRs' in AMS2, but I'm not the talent to beat aliens probably half as old as me, but in AMS2 and rF2 I get close enough to tell that the lap-times are legit and there is no setup-BS that makes you faster. In rF2 there is more setup-BS than AMS2.

F1-cars are very tricky at exploiting their potential because the faster you drive them, the more downforce they produce, the more grip. This is convenient for beginners since there is less understeer and shorter braking-distances, but the learning-curve is much better with low to no-downforce cars. The typical F1-driver started with karting (spool), than Formula Ford/BMW (F Trainer) or Vee in the past (open diff) and it's IMO mandatory to be quick in those cars.

What I love about AMS2 and rF2 as well: Being quick at hot-lapping doesn't make you win races and could be even a handicap. I just had a LFM-race in rF2 around Nürburgring Endurance and managed to qualify 6th out of 25 while the pole setter was incredible 11 sec faster. After 8 race-laps I won and wasn't even overtaking anybody, just doing roughly my qualifying-time over and over again.

The problem with hot-lapping is not just the lack of risk-management. Fresh tyres allow a different driving-style (late apexing) that's quicker with fresh tyres, but prone to fail with tyre-wear kicking in. My apex (slowest point of the corner) is maybe a car-length earlier which makes me a bit slower, but I don't have to adapt my line during the race. There are people that can do both and win more races than I do, but most people can't. I was doing far more TimeTrials in the past and noticed that my race consistancy went down. And the pure pace is usually meaningless if you fail just once during a race.

The best way of learning AMS2 is watching real onboards and observing where and how they brake and what's the slowest point of a corner. This might not put you on P1 of the hotlap-leaderboards, but shouldn't be far off either.

 

[Durge Driven]

Sadly this carries over to 99% of the MP lobbies too, where they turn everything off ?

It would be nice to see realistic settings in time trials.

Exactly why I proposed 2 versions of every sim since like GTR2.
So the Pro version should have absolutely no aids what so ever.
If you want to race 60's I think you should have no info other then pit board.

I race AI with 100% damage in rF2 full stop, real mirrors because " I did not see him" is part and parcel of real racing incidents, having virtual mirrors to avoid them imho are aids as well.

If you online and feel offended maybe look at yourself. I hosted a online room for years never used virtual mirrors ever, they look ugh, destroy immersion and are unnatural IMHO.

But hey if you like them I say good luck to you and I happy for all of you.
But what I say will just be viewed as arrogant, trolling, disrespectful or some other condition. Typical, have a radical view and you ostracized.


For topic my answer is race with AI so you can out brake them and go from there.

 

[Deleted member 1066209]

What I love about AMS2 and rF2 as well: Being quick at hot-lapping doesn't make you win races and could be even a handicap. I just had a LFM-race in rF2 around Nürburgring Endurance and managed to qualify 6th out of 25 while the pole setter was incredible 11 sec faster. After 8 race-laps I won and wasn't even overtaking anybody, just doing roughly my qualifying-time over and over again.

The problem with hot-lapping is not just the lack of risk-management. Fresh tyres allow a different driving-style (late apexing) that's quicker with fresh tyres, but prone to fail with tyre-wear kicking in. My apex (slowest point of the corner) is maybe a car-length earlier which makes me a bit slower, but I don't have to adapt my line during the race. There are people that can do both and win more races than I do, but most people can't. I was doing far more TimeTrials in the past and noticed that my race consistancy went down. And the pure pace is usually meaningless if you fail just once during a race.

The best way of learning AMS2 is watching real onboards and observing where and how they brake and what's the slowest point of a corner. This might not put you on P1 of the hotlap-leaderboards, but shouldn't be far off either.

I wholly agree with you.

My simracing experience largely lies in Assetto Corsa, but I've come to the same conclusion as you, too.

I recently started to hotlap exclusively, and discovered that I raced poorly despite qualifying very well. It was quite the wake-up call to realize that I shouldn't have put so much effort into hotlapping when I wanted to win races.

I also watched Lewis Hamilton's superb lap at Monza and was shocked when I achieved a new PB by a large margin, after applying some things that I've learned from watching his onboard video.

You give some really good advice - which isn't only applicable to AMS2 - and I hope everybody who reads it will take it to heart.

 

[R8pilot]

Interesting conversation!

I'd agree that the Nordschleife is not the best for learning, in my opinion simply because it's too long to be able to repeat your way through certain difficult corners often and conveniently. A short track that you can zen your way through over and over making tiny incremental gains is likely better.

A question, then: which tracks are more ideal for learning to corner?

About leaderboards, I would think that only default (period correct) setups should be permitted to maintain a level playing field. However, that could block some people from participating, such as those who don't have a manual shifter or a clutch for a car that nominally requires one.

As for me, I can see it's going to take a long, long time to knock off those last 5-10 seconds. It's already been about a month since my last new record, which was a near-miraculous 4 seconds faster than my previous one. (I don't have flappy-paddles and have been thinking that I need a pair one day, too.)