LaFerrari vs P1 vs 918

So, first, a story to head off the judgement and comments before I get to the point of this post.

Spoiler: Long story

I was working on hotlapping the 918 at Spa last night. All these people here and on Reddit say, "Turn off the suggested line! It slows you down and teaches you bad habits! You don't really learn the track!" And it makes sense—the argument seems sound. Especially in Assetto Corsa with its static (generic?) racing line.

So that's what I was doing. Beating my head against Spa in the 918 without the suggested line. Kept messing up and missing braking zones over and over again and I kept spinning ever other lap at the approach to Blanchimont. Finally managed a 2:33.895 (which put me in 45th place on the RSR leaderboard). Then, just for the heck of it, I turned the suggest line on for one last lap. Got a 2:31.398 (-0:02.497), which moved me up to 41st place. (Here is that lap.) So... so much for that idea. I guess I'm faster with the suggested line on.

Now that we've gotten that out of the way...

The Nürburgring GP is my go-to test track from back in Gran Turismo 6. I'm not especially good at driving it but I know it well. When 1.8 came out, I tested the P1 and LaFerrari there, so I already had times to compare the 918 to and they're rather surprising:

2:09.442: LaFerrari (2016.08.28)
2:09.212: Porsche 918 Spyder (2016.10.26)
2:03.618: McLaren P1 (2016.08.28)

Why I'm so much faster there in the P1 than the LaFerrari or 918, I have no idea. (I recorded the LaFerrari and P1's laps, but I was annoyed enough at the best time I set there with the 918 that I didn't bother recording it.)

Now, specifically for this thread, I went out tonight in an attempt to record hotlaps on the Red Bull Ring. Are they good laps? Nope—I'm under no illusions that I'm a good driver. (My average laptime on RSR is 108% of the top time.) I used the default setups for all three cars—yes, including Trofeo tires and TC (based on this thread, I tried the 918 without it and I hated it since I apparently have a lead foot). I put the 918 into Hotlap mode and I did use the DRS on the P1. I did the 918 first, then the LaF, then the P1. Here's the times I got:

1:34.382: Ferrari LaFerrari
1:33.244: Porsche 918 Spyder
1:32.341: McLaren P1

Wow. So this is more what I expected from these cars (rather than the insane disparity I got on the Nürburgring GP).

For videos, the recording of the 918 is 100% from the replay (which means my mods were only half-working)... in part because I was honestly embarrassed to record myself using the suggested line. But I got over that so the LaF and the P1's videos were recorded "live" from ShadowPlay, mods on. With the 918, I have the external video first, then the onboard with the HUD on; the LaF and the P1 videos are in reverse order.





Finally, subjective experiences with the cars?

I'll start by saying that I expected to enjoy the 918 the most, since I'm a bad driver and the 918's supposed to be the easiest to drive. Next, I expected to like the LaFerrari, as automotive journalists almost universally rate it the most fun to drive; and finally, I expected to find the P1 as unpleasant to drive as the McLaren F1 is in GT6—fast but too hard for me to handle.

In fact, that wasn't how they are at all. The 918 is frankly unpleasant to drive. It's very slow to stop and jittery under braking. I missed braking zones with it constantly and was understeering off the track with it more times than I can count. It felt the slowest even though it wasn't. As a huge Porsche fan, I found this quite disappointing.

After the 1.8 patch, the LaFerrari is now nicely balanced and pretty fun to drive. Its brakes could be better but, overall, I like it. It's like a super 458.

The P1 is a bit more lively—apt to yaw in some corners badly enough to spin, actually. It feels closer to a street-legal GT3 racecar than a regular roadcar. Honestly, I enjoy it about as much as the LaFerrari but for very different reasons. My biggest complaint is that it's incredibly ugly.

So, for whatever it's worth, there's my 10¢ on the hypercar trio. As it stands, if the 918 gets patched and ends up driving better, I wouldn't be terribly surprised. But, OTOH, I would imagine that Porsche has been keeping a close eye on the development of their cars (they will be using AC for their own simulators), so this may well be what the 918 drives like without Porsche's driver aids.

Speaking of which, I do wonder what it would feel like to drive these three cars with accurate representations of their electronic nannies. I would imagine Porsche's PSM would make the 918 much better to drive for a noobcake like me. I think the LaF's ESM is supposed to be pretty good, too, so I'd probably do better with it than not. Harder to say for the P1. Obviously, I'll never know.

 

the randomness is definitely there in SP. most defaults have pretty low gain as well. id use optimal for this sort of testing or make a preset with 0 gain and randomness.

 

I did a few laps with the slicks at RBR just for some rsr laps.



I also ran some laps on my server with the Road tyres, server has factory TC enabled but I dont bother running it. I enabled tc hoping it may keep more in the server but nope, nobody runs more then a few laps before giving up lol. So many here claim to like the hyper road tyres yet put them in a server and most stay away or leave fairly quickly once they find out they have to use these for some reason. Same track conditions hyper road 1:31.591 and the Trofeos 1:28.700 the roads do lose a bit after 2 or 3 laps, battery wise I dont think you have any issues at RBR as with hotlap mode and using kers on most exits Im still pretty much equal with what it gets back after 7 laps.

The more laps you do in the 918 the better it starts to feel but I havent driven the other two cars at RBR recently so will give them a run too at some point. For people having trouble with the 918 just brake a little earlier and its also no need to floor the brake even if it has ABS, that seems to unsettle it a bit more once ABS is kicking in. It has such great acceleration on exits that you just need to get slowed to hit the apex then nail it out. General rule with any car if you cant hit an apex and are understeering then brake sooner, hypercars are rather quick with not much aero and relatively heavy compared to a gt car on slicks. Even the Trofeo tyres arent going to help it brake that much sooner its only a few meters difference really.

 

Didn't know the 918 had KERS and driving modes so I tried them out at Nords and it didn't make that much difference to my lap time.
In "hotlap" mode I was only about half a second quicker which isn't much in a 6:50 minute lap.
Managed to gain 1.5s in Race mode which is supposed to be slower than "hotlap" mode but it still isn't much.

 

I watched you pull off some of those turns and I still don't know how you did it. LOL. (If I were braking where you seem to be, I swear I'd just end up putting it into the gravel.)

Well, one thing is counter-steering. I can't manage to do that at all. If I lose the back end, it's just game over for me.

I wish I shared your experience with it. I find the more laps I do in the 918, the more I question why I'm still doing them and not driving something else.

 

Today I will try uploading my laps on trofeos and hyper road tyres. I am not an alien, but did 1,28:900 on trofeos, so u will see my line.

 

I've done some laps at Nords with these cars and these are my thoughts:-

Corner entry, the Ferrari is stable, the Porsche is tricky in the high speed corners and the McLaren can get loose in the slow corners. This makes the Porsche the hardest to drive out of the three and the Ferrari the easiest.

Corner exit, the Porsche is planted, you can really step on the throttle in that one. The McLaren is quite stable as well but the Ferrari can be a bit twitchy. I didn't use TC with the Porsche and the McLaren ; TC (3/4) on the Ferrari.

Speed, Ferrari wins on outright speed reaching 354 kph, McLaren was 341 kph and the Porsche could only reach 329 kph. Acceleration in the Porsche is good but the slower cornering speeds and having to brake earlier cancels out that benefit.

 

Tried the P1 yesterday for the first time in a while....its come a long way, feels amazingly stable now, dont have to fight with it no more ......

 

Great video, Stefan! And I think you've summed up the differences in the driving character of the three cars excellently.

Situations like this make me think about how nice it would be if we could load multiple hotlaps into a replay and have the game spit out a "virtual race" (albeit one where the cars were ghosts that could clip through each other) to really see them duking it out. But I've never seen such a thing in a racing game before so I certainly never expect to see it in AC.

Interestingly, I believe the 918 has the best real-life performance at the Nordschleife. I don't know if that's due to the driver Porsche hired (I thought it was Walter Rohrl, but here's the video, and it's not him) or the car, but I suspect a good deal of it is the latter. IIRC, the reason why McLaren could only get a "sub 7 minute" time (they wouldn't say exactly what it was) was because their hybrid system ran out of juice while the 918's didn't.

 

Was that with no manual kers use? I'm still working on my time when I get a chance. I still see mistakes and time i can make up without using the kers manually but its tough for me to tie it all up in a neat bow in a single lap. Right now I'm at 1:31.933 on the road tires, no manual kers and completely stock settings. I think, if I can lock up a really good perfect lap there is another half second in there that I can extract. Perhaps an "Alien" as people have been saying could take a second or more out, who knows, but I doubt I can. I know my lap time will make a fairly large improvement of I start using manual kers on every exit but knowing I still have more on the table without using it bugs me.

 

In game TC uses the exact same inputs are RL TC....

 

Clearly it doesn't work how its supposed to then. Something doesn't work or add up. Example, the new Porsche 911 Carrera S. IRl, thing can pull 11.5-11.6 sec 1/4 miles. Thing is insane. A little bird told me that the in game version needs about 470-490hp to accomplish that same feat. The car in game launches slower with the TC on then it does off, which shouldn't make sense. In fact with the in game version I really don't see how its possible for the 911 to launch so hard as to pull 11.5-11.6. Even with slicks it might not manage it. Was also tested at 3.1 seconds to 60mph. You couldn't pull that time in Assetto Corsa with a 20 degree downhill.

 

3.9 seconds is quoted online....

 

Now where'd that come from? I haven't seen/heard of the devs commenting anytime about any cars in AC having the TC the real car has. Really, really, wish that was widely communicated (on Steam page, on forums) which cars have their true TC, SC, ABS, anything like that.

 

And yet, here it is beating a Viper ACR on barely legal slick tires. Keep in mind as well, that nobody has come close to matching or beating that Nurburgring record. Everyone knows Porsche underrates their engines, question is by how much.

 

If it did no one IRL would use TC even to save tyres considering how slow and limiting it is in-game.

 

LMPC car traction control was literally a direct carryover of inputs from real life to AC.

 

Not only that but the way it changes the way the car handles all together. To me with it on you get these weird responses during some turns and this only happens with tc on.

 

And the inputs are?
From my understanding AC uses just the longitudinal slip ratio as a threshold for 'on-off' torque modulation at a certain frequency.

Even the most basic TC we used on two strokes 250cc moto gp bikes was much more complex than this...

 

I'm not saying the implementation is the same, just that the inputs are.