Initial Torque: What do you do?

[Poppins]

i just plain found 5/60/5 to be the best bet because like i said accel sens modifies/affects torque required to lock under accel and decel sens modifies/affects torque required to lock under decel

and from what ive calculated it changes the initial torque under accel to 43 aka 57% of torque to one wheel before lock

you have to look at initial torque by itself

and then you have to look at accel and decel purely modifying the initial torque while braking / accelerating, while they dont actually CHANGE it, it just affects it when accel and decel.

but the thing about 60/60/60 is that it can only go up to a maximum of 60% lock total no matter what, even if the calculation goes over that. which causes 40/60/60 to behave the exact same.

 

[budious]

but the thing about 60/60/60 is that it is pretty much 100% lock at all times.

... only by your theory... I'm still working on the plate density in the clutch assembly of the LSD theory, in which case the car turns great at 60/60/60 for me... it typically understeers more off throttle than on it, so if you can keep 20% or light throttle on it in corners it will naturally power oversteer and drives quite well and in a controlled manner.

 

[Poppins]

... only by your theory... I'm still working on the plate density in the clutch assembly of the LSD theory, in which case the car turns great at 60/60/60 for me... it typically understeers more off throttle than on it, so if you can keep 20% or light throttle on it in corners it will naturally power oversteer and drives quite well and in a controlled manner.

its not quite 100% lock i dont think its 60% max no matter what im pretty sure so 40/60/60 will act the exact same as what you have. changed other post before u posted that haha i was just not thinking right.

 

[Disavowed]

@Adrenaline:
Sounds about right...

All those percentages and calculations mean nothing if you have no way of measuring the torque @ the wheels. It's really all up to trial and error. You'll have to grind each car to find out what is the best LSD setting for it since each car reacts differently. What I can say is that I've tried driving some of the cars that I have driven in the past and I can say that the guys who made GT5 really did their homework. When you start modding the cars, well, that's where some of the inconsistencies start to appear.

 

[Poppins]

if you do the calculation opposite of what i said, you can get the exact same answer.

if the one tire wants 95% torque to it which means (initial at 5) before locking, but accel sens is set to 60 (40% slip) which means the slow wheel MUST rotate at 60% speed of fast wheel, it causes the wheel getting all 95% of the torque to not be able to get that much because the other one must spin 40% of its speed, which is causing that 95% to only get 60% of what it should while not affecting the speed at which it spins.

95 * 0.6 = 57, yet again. which makes the other tire get 43%.

 

[Poppins]

@Adrenaline:
Sounds about right...

All those percentages and calculations mean nothing if you have no way of measuring the torque @ the wheels. It's really all up to trial and error. You'll have to grind each car to find out what is the best LSD setting for it since each car reacts differently. What I can say is that I've tried driving some of the cars that I have driven in the past and I can say that the guys who made GT5 really did their homework. When you start modding the cars, well, that's where some of the inconsistencies start to appear.

all cars react the same with the same lsd. it all depends on power suspension settings, gears, everything as well though but the lsd reacts the same.

 

[Disavowed]

all cars react the same with the same lsd. it all depends on power suspension settings, gears, everything as well though but the lsd reacts the same.

Yes, the LSD always reacts the same no matter what. Thanks for bringing that up.

 

[budious]

Yes, the LSD always reacts the same no matter what. Thanks for bringing that up.

The problem with 5/60/5 that 50/60/50 or 60/60/60 do not show is that when you pair the larger numbers with each other you don't get the nasty snap oversteers and side to side play that comes with mixing extremes of low initial and high acceleration.

As for the 40/60/40 theory, wheel spin is far more present than 60/60/60... actually I found 49/60/49 to be about optimal to kill it. So first car was 46/60/46, this one is 49/60/49...

 

[Disavowed]

The problem with 5/60/5 that 50/60/50 or 60/60/60 do not show is that when you pair the larger numbers with each other you don't get the nasty snap oversteers and side to side play that comes with mixing extremes of low initial and high acceleration.

As for the 40/60/40 theory, wheel spin is far more present than 60/60/60... actually I found 49/60/49 to be about optimal to kill it. So first car was 46/60/46, this one is 49/60/49...

I'll give that a try later tonight.

BTW, just a heads up for those who don't know. Some 4WD cars are better off with their standard diffs like the GTR, Gallardo, etc...

 

[Disavowed]

Ok, did a quick test using a stock Ford GT40 '69.

Set downforce to max, brakes 7/4, and everything else default.

Using 60/60/60, 46/60/46, 49/60/49, the car was very difficult to control. If you don't catch the first tail slide and/or overcorrect it, the car will spin out. It definitely had a lot more traction coming out of corners though when compared to the default setting.

Now try my setting of 10/9/30 and let me know how it feels to you. It can still be tweaked to be better but I don't have the time right now.

Note that the car still has a tendency to autosteer under braking...

 

[budious]

Ok, did a quick test using a stock Ford GT40 '69.

Set downforce to max, brakes 7/4, and everything else default.

Using 60/60/60, 46/60/46, 49/60/49, the car was very difficult to control. If you don't catch the first tail slide and/or overcorrect it, the car will spin out. It definitely had a lot more traction coming out of corners though when compared to the default setting.

Yeah, that method doesn't exactly work unless you tune the suspension first to make full use of it... you might want to check the anti-roll bar settings, if they are set to 7/7 that can be a little jerky sometimes. If it's not set at max, then use anti-roll bars 4/5, 5/6, with +1 on the rear ideally but below 7 and this will counter some of the understeer. Full overhaul of dampers and spring rates is recommended though.

 

[Disavowed]

Yeah, that method doesn't exactly work unless you tune the suspension first to make full use of it... you might want to check the anti-roll bar settings, if they are set to 7/7 that can be a little jerky sometimes. If it's not set at max, then use anti-roll bars 4/5, 5/6, with +1 on the rear ideally but below 7 and this will counter some of the understeer. Full overhaul of dampers and spring rates is recommended though.

I used to setup the Zonda R like that but it just wouldn't drive consistent enough for my taste which is why I went to the other end of the spectrum. I lost a little bit of speed in high speed tracks but the consistency of the lap times was a lot better with the added benefit of being easy to drive in tight tracks. No more white knuckle driving for me! LoL!

 

[Sail IC]

What an interesting thread on a very confusing subject.

I'm a decent driver/tuner. With a stock power GT500, I lap on the sliperier online at Suzuka in the 1'53.s on soft and in the 1'58.s on hard tires.

I prioritize control over ultimate traction. Let's say that I start tuning with the LSD set to 10/10/10. If I understand what has been written correct, if want more acceleration traction, increasing the initial or the acceleration will both give me more traction.

But in which way will it differ ? Still confused.

 

[nomis3613]

If I understand what has been written correct, if want more acceleration traction, increasing the initial or the acceleration will both give me more traction.

But in which way will it differ ? Still confused.

Yep. If the only problem is wheelspin on the inside wheel, increase accel.

But if you also have a stability problem (eg car wants to spin out, especially for MR drivetrains), then increase initial to kill 2 birds with one stone.

 

[Disavowed]

10/10/10 is a good starting point imho. Start by tuning the acceleration trying to equalize wheelspin on both left and right wheels. After that is done, I usually just leave it alone and adjust Initial according to the track. I'll usually use a lower Initial for higher speed tracks and higher for low speed/tighter tracks. This value will differ greatly from car to car.

 

[Poppins]

i have it completely figured so its all good for you guys to have more theories and such but nothing is going to be able to show me that what i said is wrong or anything else is right

 

[Slurm]

Disavowed, Scottster, I'm on completely the same page as you guys. Nice to know I'm not imagining things, haha.

i have it completely figured so its all good for you guys to have more theories and such but nothing is going to be able to show me that what i said is wrong or anything else is right

👍 /care!

 

[Sail IC]

Thanks for the answers. I've been leaving the initial at around 10 and played with the acceleration for traction but I will start to use the Intial more and see what happens.

Another question. Decelaration is easier to tune but ..... Does the initial have any impact on the car when coasting and/or breaking (ie decelaration setting). I been thinking it hasn't but I'm not sure

Thanks

 

[Disavowed]

Thanks for the answers. I've been leaving the initial at around 10 and played with the acceleration for traction but I will start to use the Intial more and see what happens.

Another question. Decelaration is easier to tune but ..... Does the initial have any impact on the car when coasting and/or breaking (ie decelaration setting). I been thinking it hasn't but I'm not sure

Thanks

I haven't noticed any changes in coasting or braking behavior when changing the Initial so it's probably safe to say that it doesn't. Deceleration definitely has a huge impact though.

 

[budious]

if you do the calculation opposite of what i said, you can get the exact same answer.

if the one tire wants 95% torque to it which means (initial at 5) before locking, but accel sens is set to 60 (40% slip) which means the slow wheel MUST rotate at 60% speed of fast wheel, it causes the wheel getting all 95% of the torque to not be able to get that much because the other one must spin 40% of its speed, which is causing that 95% to only get 60% of what it should while not affecting the speed at which it spins.

95 * 0.6 = 57, yet again. which makes the other tire get 43%.

You are aware that there are some stock differentials on cars in this game with much higher values. I believe some RX-7's have 50/80/0 so that would be... I assume 0 deceleration means it's a 1-way slip.

50 * 0.8 = 40?

So equip a customized slip with 5/60/5, 33/60/5, and then compare it to the stock 50/80/0 differential (if you find the right car to test) and tell me which turns better, or why it would be necessary to have values of 50/80/0 if 33/60/5 gave the same result by product of your formula?

 

[Adrenaline]

i have it completely figured so its all good for you guys to have more theories and such but nothing is going to be able to show me that what i said is wrong or anything else is right

Funny, I'm pretty sure you said something along these lines 3 days ago... and then you changed your mind on what 'the best lsd is' 13 times, and now you've got it completely figured out again.

Wanna race online?

 

[budious]

A Limited Slip Differential is very similar to an Open Differential, but it adds a spring pack and a set of clutches. Some of these have a cone clutch that is similar to the synchronizers in a manual transmission. The spring pack pushes the side gears against the clutches, which are attached to the cage. Both side gears spin with the cage when both wheels are moving at the same speed, and the clutches aren't really needed - the only time the clutches step in is when something happens to make one wheel spin faster than the other, as in a turn. The clutches fight this behavior, wanting both wheels to go the same speed. If one wheel wants to spin faster than the other, it must first overpower the clutch. The stiffness of the springs combined with the friction of the clutch determine how much torque it takes to overpower it.

http://www.stangnet.com/mustang-forums/438617-trac-lok-vs-limited-slip.html#post3949300

This is pretty much in line with my theory of the sensitivity values as representations of plate density in the LSD clutch mechanism.

The quoted source also includes definitions on how other models of popular differentials operate and is a recommended read.

The stock differentials in this game can also be quite deceiving, those with clutch type limited slip differentials usually have numbers accompanying them as grayed out values. However, others display only zeros; this is not because there is no differential, it's because these differentials are modeled on other designs than a clutch operated limited slip differential.

Here's a simple test car to try:

[Premium] Ford Mustang Mach 1 '71
----------------------------------------
* purchase new at ford dealer
------------
+ Front Aero
+ Rear Aero
+ Wing Aero
* Aero @ 0 / 5
+ Engine Tune 1
+ Single Plate Clutch
* Stock Clutch (may be best clutch for this car overall)
+ Sports Flywheel
+ Customizable Tran @ 180MPH
+ Customizable LSD
+ Sport Soft
* Sport Hard/Medium (to track preference, testing conditions)
-----------------------------------------

Run the car on a course first with all parts except the custom LSD, make note of how little the rear wheels slip and how well the car turns, record a few lap times as a benchmark. Now equip the custom LSD and do your best to try and find any combination that is better than the stock differential. I bet you won't. I was testing on a custom Mt Aso based course I use for this type of testing, but a technical circuit such as Trial Mountain, Deep Forest, or Nurburgring should show some similar results. Ford's signature traction-lok differential at work?

If it's not pronounced enough on the test track in use, switch back to Sport Hard/Medium. Once I moved over to Deep Forest, I found the stock suspension a bit too spongy there, so adding fully customized suspension at default settings will be sufficient for this test. Another interesting note is that stock clutches and flywheels can be as deceiving as their upgrade counterparts, typically I find single plate clutch fastest, but on this car the default clutch puts up consistently faster lap times than the single plate. Also forgot to spec the aero kit upgrades, original specs updated slightly to reflect this.

Phase II Testing
-----------------
+ Racing Air Filter
+ Supercharger
+ Fully Customized Suspension (@ defaults)
-----------------
* Sport Soft

Lapped Nurburgring Nordschleife in 7:45" (red lining 180 MPH tran so need to adjust...) with Phase II test spec using stock clutch, sports flywheel, stock differential. This is definitely a case of less is more with this car. I love it...

 

[Disavowed]

Tested the car as you specified above with and without the LSD. In all tracks, imho, the stock suspension was way too soft but I stuck with it. Ran the car on Grand Valley Raceway, Trial Mountain, and Deep Forest (didn't have time to run the Nurburgring). The car was consistently more than a second faster per lap with the LSD. My LSD settings were 12/12/10. It wasn't very noticeable on the Grand Valley track but on Trial Mountain and Deep Forest, traction was definitely an issue without the LSD. The car was spinning it's rear tires hopelessly around the tighter corners. The LSD cured that problem with the added benefit of additional stability when trail braking.

Please note that I setup the LSD like that to more mimic the way a real LSD works. You could probably come up with a LSD setting that gives better lap times but I found this setting to be closer to how the real thing works. It's also very fun to drive.

Budious, I haven't had time to test out that clutch thing you mentioned but I'll be sure to try it out later tonight. If anything, this setup of yours really shows how less can be more. The car was a pleasure to drive.

 

[budious]

I tried a numerous bit of settings, but I'll give 12/12/10 a try later... I was trying increments of 5/5/5; 10/10/10; 15/15/15... so forth looking for a sweet spot and did not find one. The stock differential to me was far more predictable when it did slip and more recoverable under full throttle. The car is good though, I did have to switch out to custom suspension for other tracks though; the stock suspension worked well enough for the phase I testing on my custom Mt Aso track, I'll see about grabbing the specific parameters needed to input to generate it so others can try it. I find it to be much more of an LSD test than the included tracks in the game.

 

[oopssorryy]

So is everbody in agreement here? If so thank you all for all your hard work, if not.... wel I guess it would be back to the drawing boards.

 

[Disavowed]

It really is up to each driver and car really. I prefer to setup my cars neutral with a touch of understeer while using a LSD setting that allows me to steer with the throttle easily but that's just me.

 

[oopssorryy]

What I mean is, is everybody in agreement on what Initial Torque does, and how it does it? If so I might just go and try to compile all the info everyone agrees on and post it on the first post so that more people will see it.

 

[Slurm]

^ hahaha. The funny thing is even after this massive dialogue, I really don't think anyone can pin it down exactly, in the game at least. Personally, I would just say it's the overall strength of the differential. I would also say that IMHO Poppins is completely wrong.

 

[Disavowed]

Initial Torque is supposed to be like preloading a real LSD but it seems to be reversed. Lower values give more preload. In simpler terms, use lower values if you want the LSD to activate earlier and higher values to allow more wheel slip before it engages. Since each car's torque band is different, the IT value will vary from car to car.

 

[oopssorryy]

^ hahaha. The funny thing is even after this massive dialogue, I really don't think anyone can pin it down exactly, in the game at least. Personally, I would just say it's the overall strength of the differential. I would also say that IMHO Poppins is completely wrong.

Maybe one day this will all be cleared up, I find Poppins stuff overly complicated and hard to read.