Re: Proton Persil: v1.2 available 2008-12-05

[C14ru5]

Great car!

[Tomte]

Ha, I messed up. The white texture was flipped. Legacy will be updated soon.

[Tomte]

With the X-Bow around, the Atom needed an update to keep the balance. I hope it works. Let me know.

Changelog in first post, but basically I made the Atom a bit slower and a bit more loose at the rear.

Ariel Atom v1.1

[slow]

a bit more loose at the rear.

nice. amped will love it!

[See Flat]

With the X-Bow around, the Atom needed an update to keep the balance. I hope it works. Let me know.

I really did not have a harder time with the car. I was sure I had the wrong one, and it stills pulls out in straights like a rocket. But this might all be right... as my disclaimer always states, I know nothing about cars. As val mentioned.. lighter car. I just know that most people will pick the atom as both cars are now.

[Tomte]

I really did not have a harder time with the car.

Actually, I'm having quite some difficulty setting up the Atom how I want it.
I know it's supposed to have quite a soft suspension, but I feel it's too soft.
I'm having troubles with the rear end. I'd like it to be more gradual in the way it slips with improved feel at the front end. I also don't like the way the rear handles bumps.

I tried stiffening the suspension (via damperStrength), but that makes matters worse. Softening it makes you seasick. Raising suspension friction makes the suspension not react fast enough on bumpy surfaces.
The swaybar balance is already so far to the front (10000 to 4000), that if I raise the rear, I get snap oversteer at the slightest chance.

For now I reverted to use rear loadsensitivity and downforce (about 25kg in total at 200 kph in comparison to 200 kg of the X-Bow).

I'm kinda out of my depth here. Thomm's Caterham uses quite a lot of loadsensitivity and wheels stickyness to keep things in check with a similar light car, but the Cat is also better balanced. What I have is basically an old 911 without weight (and the lift...).

Any suggestions are welcome.

[Tomte]

Looks like I believed the marketing hype of the Atom. It was advertised as having a weight of about 470 kg, which appears to be fron an early version with a lighter engine, no fluids, driver, lights and so on.
A road ready Atom with the 245 hp engine appears to weigh in at about 620 kg, includig fuel. Adding 75 kg for the driver makes it about 695kg, which is much closer to the X-Bow, but further away from the Radical and the Caterham.

But then again, in this car class the weights which are given by the manufacturers often don't match up with reality. Furthermore, Thomm's Caterham tips the scale at 615 kg with driver, while the official stated curb weight seems to be 575 kg. 40 kg for driver and fuel?
This opens up the question how we handle weight in Redline. Sure enough, with a 1500 kg car it makes hardly a difference. At 500 kg, that's a slightly different matter.

[DonaemouS]

i is not real, but i use weight from car makers' sites. Perhaps, these values don't include driver and full tank fuel. I know the driver weight is a main part of the car. But, how we should give him a 75 kg? I mean, my weights is 65. Why I need to get fat of about 10 kgs once i seat inside a car?

Also, Why I should race with a full tank? Also in F1 they choose the right amount of loaded fuel, before start...

[Tomte]

it's mainly a question of if we want to play on a common ground.
In F1, the weight is limit of, dunno 550kg is with the driver, all fluids and a tiny bit of fuel.
Let's just say the 75kg is including the racing combo and the helmet. Plus the wallet, because you need to carry your credit card for all the tires you're smoking!

[DANDY]

weights etc

I typically add 75kg driver weight (which is considerably less than my weight [redline is not helping here]) and no fuel, to partially make for the unsprung weight factor. And I never use a TommyGun as reference for anything.
Unless it's an endorsed race class, I also try to build cars to be what they are, no more no less, so I won't be disappointed if the Atom (or anything else) is faster or slower at any particular track, especially if they're assembled with realistic data.
In this case, we probably won't get a good idea of comparative performance until we get some big races on a variety of tracks. See you there!

[See Flat]

Unless it's an endorsed race class, I also try to build cars to be what they are, no more no less, so I won't be disappointed if the Atom (or anything else) is faster or slower at any particular track, especially if they're assembled with realistic data.

That's just Dandy! I read the article in the link you posted at ASW board and then followed to the article on Atom. It sure seems like you guys are nailing these cars. From a noob's perspective
Gotta say D, I just admire your stance on achieving realism in your creations. The question always is....
can we have races where everyone does not pick the car that has an edge on a particular track.
The whole reason I set up RLTCC tournament the way it was.
SF

[DonaemouS]

ok... i will align with other plugin developers, I will add 75Kg on Caddy CTS too.

[Tomte]

ok... i will align with other plugin developers, I will add 75Kg on Caddy CTS too.

Well, that depends where you got your numbers from. Sometimes the driver is already included (German regulations). I think in the EU every manufacturer should give the weight will all fluids, a 90% filled petrol tank, 68kg driver and 7kg luggage. Other nations handle it differently.

[Tomte]

I'm having bit of a hard time setting up the Atom in a way that is realistic, drivable and fun.

I tried out the settings Dandy proposed in another tread, unfortunalty they did not work as I'm now fighting with a more realistic rearward CoG. The weight distribution is actually 36/64 (like other similar cars), not 45/55 I used in v1.1.

I know I should be happy with what what I'm doing and that I cannot please everybody.
Still,

• looks like that swaybars are out because the original car does not have any.
  Downforce is out because the original car is very likely not to produce any.
  
• loadsensitivity is generally regarded as cheat and deadening the feeling.
  
• High values differentialLock is out because it induces understeer, artificial straightening up the car and masking other handling problems.
  
• Wheels stickyness is already set to an agreed limit (and I'm actually happy with it).
  
• Z CoG is at -0.57, because I cannot justify anything less rearward
  

This leaves me with the question what I can actually use to balance the Atom.
What I have atm still uses front swaybars and load sensitivity and some differential lock.
Here's the .car file
Link to 1.2 beta

I would really appreciate all the comments and help I can get on this.

[DANDY]

[*] loadsensitivity is generally regarded as cheat and deadening the feeling.
[*] High values differentialLock is out because it induces understeer, artificial straightening up the car and masking other handling problems.

Everything in moderation.... loadsen~ty is probably the answer as compensation, it might represent the magic of camber, design, lowered air pressure and the mysteries of ingenuity rolled into one setting, which can lead to dependancy. On a well-balanced lightweight car, differences of .01 can be huge.

The right difflock number is critical for r/r cars I think, it completely solved the 911's early ills. My test is to fly through the last sweepers (Blanchimont) at Spa... if I can't get through without losing the rear, I crank it till I can. That being said, the locking-to-weight ratio is on same as the 911, though striking the right balance here might take some finer tuning.

[Tomte]

Ok, update wars. Well, looks like I fight alone, but anyway.

v1.2 is now up at Legacy (first post or here[/url]).
Link removed, plug pulled
After raising the weight and shifting the CoG quite a bit rearwards, I removed the swaybars on Dandy's suggestion, which gave me quite a bit of trouble.

After some liberal use of loadsensitivity and with tweaks to the LSD and body inertia, I think I got it working. I then reduced loadsensitivity to a level that the Atom is still drivable. It still doesn't like bumps, feels rather soft and don't even think about using the kerbs.

I think this will even things out with the KTM a bit, but I'm not sure anymore.

[DANDY]

Ok, update wars. Well, looks like I fight alone, but anyway.

v1.2 is now up at Legacy (first post or here).
After raising the weight and shifting the CoG quite a bit rearwards, I removed the swaybars on Dandy's suggestion, which gave me quite a bit of trouble.

After some liberal use of loadsensitivity and with tweaks to the LSD and body inertia, I think I got it working. I then reduced loadsensitivity to a level that the Atom is still drivable. It still doesn't like bumps, feels rather soft and don't even think about using the kerbs.

I think this will even things out with the KTM a bit, but I'm not sure anymore.

Those long-race clubbers are gonna *love* you!

[Tomte]

Those long-race clubbers are gonna *love* you!

Oh bugger. A quick run on Rocky Pass revealed that the Atom is undrivable there. The rear comes round much too easily. And that with silly amounts of loadsens and wheel travel.
I'm out of ideas.

[DANDY]

day off tomorrow, I'll see if I can't figure something out.

[Cowgirl]

Oh bugger.

oh boy, there he goes again

[Tomte]

day off tomorrow, I'll see if I can't figure something out.

Thanks a lot, I think I need more than one pair of eyes now.
I asked around on an Atom web board to get some more information. What I got until now was that the suspension travel is about 3" and that the car should pull 1.37g on a 300ft skidpad with Yokohame A047.
This is what I have now.
v1.2b 081127

My test tracks are Mountainside, Rocky Pass and Limerock. The rest hopefully falls in place.

[DANDY]

Z CoG is at -0.57, because I cannot justify anything less rearward

Are you quite sure about this? The KTM is close, with a similar wheelbase, quite sure I calculated based on the axles, not the entire length of the car. I tweaked a few things, and set CoG to -.175 (to the XBOW's -.153), and its handling much better...still off the pace, however.

[See Flat]

My test tracks are Mountainside, Rocky Pass and Limerock. The rest hopefully falls in place.

I hate to see you torturing yourself like this. Right now, I cant drive the atom anymore. It's really hard. I mean.. really hard, as in, harder than the radical.
I just tested on
Lime= had a hard time getting up the hill, took me four laps not to lose control on the climb, and then the downhill is dangerous. The rest of the track is gave the false sense of stability until one of those two came up.
HLS= spun out everytime on the S

Obviously did not give myself a chance to get used to the new handling... but immediate results are a comparison to the much easier to drive X-Bow and Radical.

[Tomte]

Are you quite sure about this? The KTM is close, with a similar wheelbase, quite sure I calculated based on the axles, not the entire length of the car. I tweaked a few things, and set CoG to -.175 (to the XBOW's -.153), and its handling much better...still off the pace, however.

Ok, checking...
well, the model is not placed so that the axles are at the same distance from Z=0. I placed the whole car centered. The front axle is about 0.93m from Z=0, the rear axle about 1.41m from center. So with a weight distribution of 50/50, CoG Z would be at -0.24.
I never had a car with such a large offset, maybe Redline does not like my model placement.
I will quickly move the model in MI so that the middle between the axles is a Z=0. Then adjusting all Z coordinates in the .car and so on.
Apart from that, I know that the car has a weight distribution of about 37/63 with fuel but without driver. What I learned from the atom forum is that the driver shifts the balance a bit to the front, so I now assumed 38/62 (Z=-0.52), I could probably argue a tiny bit more.

[Tomte]

Ok, I moved the whole model and .car 0.24m to the front, recalculated the inertias, but left the rest unchanged compared to the version from this morning (081127).

I think I'm seeing ghosts, but can you feel a difference? Might be that I have that much loadsens in there that is it difficult to feel anything...

081127b
BTW, that's the whole plug, which can reside beside 081127 and v1.1.

[Sponsored content]