Weight Bias - Redux

[Blooze]

Here on the eve of the much anticipated upgrade to our world, I thought I'd take a minute to organize my thoughts on the weight bias deal.

First, it is important to remember that much of this discussion is centered around game decomposition more than actual race tuning the spring components of a race car.  What I am trying to devine is how Turn 10 is dealing with the values in the game.  Here are a few of the behaviours that originally set me to looking into this.

• Adding parts that altered weight always altered the spring rate, and only the spring weight.
• What we came to call the SWR (Spring Weight Ratio) was never altered by the game.  Buying lighter wheels had the same affect as buying a lighter hood.
• The Front/Rear ratio was never altered by the game.  Adding a splitter had the same affect as adding a wing.
• The published Wb (Weight Bias) was never altered, even when adding something like AWD to an FWD car.
• The Wb and the the ratio of the front/rear spring rates were normally quite close.  However, when there was a difference, in the case of cars that were "Tuners" or a high performance version, they always leaned toward a more balanced car.  For instance, the VIS Integra had a published Wb of 58%, but the spring rates were actually 53%.  In other words, cars that were supposed to represent a high performance tuning and drove better than their stock counter parts, had spring ratios closer to 50%.
• Weight changes were always distributed according to the spring ratio and the SWR, rather than the published Wb.
• There is no way in the game to actually modify the Wb of the car, like Balast or something, and that seems to be a serious short coming in a game that allows you to take a street car all the way to a ALMS GT2 level, almost.

All of these things led me to determine that the physics engine looked at the front/rear spring ratio to determine weight bias rather than the published Wb value, and accordingly - if you wanted a car with a 50% balance, all you had to do is set the front and rear springs to the same value.  And for me, the Wb of a car came to refer to the front/rear ratio of the spring rates, rather than the published value.

Now, discussions that tried to relate the Wb and spring rates to the Damper and ARB values came later, and we'll try to avoid them in this thead, other than to relate that the intial settings for these parts always reflected the published Wb.

So, first thing for me will be to see how many of the anomalies in that list found their way into Forza III.  Perhaps by Wednesday or Thursday, I will know.  

Or...

...one of you Brits could check for us now...

Oh, I am so ready for this game...

$

[Tonka Crash]

I thing that will put a serious kink in my plans to investigate the new settings is cash.  At least initially, I'm going to have to spend a lot of time just generating income to afford the cars to look at the settings.  Last go around this wasn't an issue by the time I got around to participating in the weight bias tuning conversations. 

One thing that would really help is if we could generate a default settings data base with as many cars as possible.

[Tonka Crash]

For discussions sake, do you want to continue with our current definition of SWR, (FSpring+RSpring)*2/Weight or change it?

From the 2 cars I've seen this leads to SWR being closer to 100% and we should see some of the cars approaching 200%. 

If we drop the *2 multiplier this drops SWR values to the range we see in FM2 to compensate for the doubled rate springs.

[TheTechnobear]

Funny, i was thinking about this today... well pete publishing those spring values (and the non WB ratio) , made me grab the spreadsheet
I should have the game on Monday - postie failed today
first thing im going to do is try to mass some credits, and gets as many cars with spring values as possible, the add some weight and see how it distributes.

petes values seem to indicate things are a little different this time, but i think adding the weight will show us if they have switched the calcs around on us.
(i.e. using springs to compensate WB and not imply it) ... but we need more data.

what i also want to check is if the telemetry is a little more useful this time around ... well we can hope.

as tonka says, perhaps we should post settings here? initially,im going to be looking at spring rates to keep costs down.

[Spiny Anteater]

I'm afraid to say that I have failed to note down any of the inital spring rates. From memory though, the initial setup on the Clio V6 (big V6 engine in the middle, rear wheel drive) started with stiffer front springs than rears, but my memory may be faulty on that so take it with a pinch of salt.

Telemetry is more of the same (see demo) but is more useful just because they have brought in the slow advance and rewind.

Also, had a go with the Shelby GT500 (new one) last night - those initial spring settings were both over 1000lbs/in so I suspect the settings may be for two wheels. Certainly, FD's old rule of springs/100 for the bump damping and bumpx2 for the rebound don't work any more - they don't fit on the scale

[Blooze]

My initial (re: gut) feeling is that you are correct and the x 2 multiplier in the SWR calc needs to be removed.  That would appear to put us in the ball park if it is the case that they are reporting both wheels now.

As for what to do about the Dampers and ARB...      more data I guess.

The thing to keep in mind is that I don't think that they will have strayed too far from their orginal algorythm, and there will be a basic car that the physics engine is dealing with - we just need to find it.

I am going to sort through the test data we all worked up a couple years ago and see if I can meld that and my car sheets with the New Car List so that we will have the foundation of a comparitive data base that can be updated over time, i.e., when we all have more money.

Just a gut feeling here - looking at the Spring Data Sheet and BBQ's initial numbers...  the only justification I can find for the very high R Spring value is to produce a 51% Wb.  I think...

$

[Blooze]

Well...

My other intial gut feeling was that I shouldn't make any jumps or come up with conclusions until the game gets here.

It turns out that this time, you can make descisions that change the Wb, and the spring rates and resultant SWR is not too close to what it was in Forza II.

One of the things we established the Tonka Toy on was the relatively fixed condition of the R# cars.  Judging by the number of X Classed Peugeot 908s I saw when checking out the leaderboards. that may no longer hold up either.

so...

We'll get back to this when we know more.

  $

[Fit4aking]

I'm trying to wrap my head around tuning at the moment and I've finally come up with a car that gets worse the more I tune it.  This evaluation only applies to the spring/damper/anti-sway values not the alignment specs.  I've slowed down on almost every track and the others I'm struggling to make the same, pre-adjustment, times.

The car is my F-class VW Corrado. 
Weight - 2749lb
Initial Front Spring Rate - 705.9
Initial Rear Spring Rate - 658.5
This makes the initial SWR very near 100% and the initial bias is 51.7%

My only change to the car at this point is reducing the SWR to the FM2 78% and moving the bias to 54.5%.  I never liked those settings in FM2 based on my prefered car characteristics but in all fairness they were the 'starting' point for most so I indulged on this car.  Not a good choice as the softer suspension definitely slowed it down by inducing off throttle oversteer and on throttle understeer.  Quite the handful. 

More testing is in order on this one.  I'm sticking to my guns this time.  No need to try and apply faulty tunes to my higher class cars until I can sort something out on my initial batch of F-classers.  For comparisons sake I broke down and purchased yet another car last night, making my grand total of purchased cars 3!  An F-class RWD to use as comparison to my two FF monsters.  Still looking for a MR and AWD to complete my stable.  I'll pay close attention to the initial spring rates and post them so that they may be helpful to others.

[Silvo 1981]

Firstly, this wasn’t a technique I used in Forza 2.  Although I did read about it a little while before the release of 3 I never got round to using it so please bear with me if I’m not quite up to speed yet.  This is also my first post on here so please go easy on me if I have missed something very obvious.

It dawned on me at the weekend that you don’t need to own any of the R-class cars to get the data you need to determine the SWR.  I used the hotlap section as all the cars are unlocked from the start to gather the info needed to get this post moving.  I’ve collected data for 5 cars in each R-class to see if this can tell us anything useful for tuning the cars ourselves.

CLASS: R1

Make: Acura
Model: #66 de Ferran Motorsports ARX-01b
Drive: MR
Weight: 1,984lb
WB: 50%
Springs: 658.5/658.5
SWR: 50.00%

Make: Acura
Model: #15 Lowe's Fernandez ARX-01b
Drive: MR
Weight: 1,764lb
WB: 45%
Springs: 526.8/643.8
SWR: 45.00%

Make: Audi #2
Model: Audi Sport North America R10 TDI
Drive: MR
Weight: 2,039lb
WB: 41%
Springs: 555.0/792.3
SWR: 41.19%

Make: Audi
Model: #2 Audi Sport North America R8
Drive: MR
Weight: 1,984lb
WB: 45%
Springs: 592.6/724.3
SWR: 45.00%

Make: Bentley
Model: #7 Team Bentley Speed 8
Drive: MR
Weight: 2,024lb
WB: 45%
Springs: 604.5/738.8
SWR: 45.00%


CLASS: R2

Make: Aston Martin
Model: #007 Aston Martin Racing DBR9
Drive: FR
Weight: 2,425lb
WB 50%
Springs: 696.7/696.7
SWR: 50%

Make: Audi
Model: #8 Audi ABT TT-R
Drive: FR
Weight: 2,381lb
WB: 50%
Springs: 790.2/790.2
SWR: 50.00%

Make: Chevrolet
Model: #3 Corvette Racing C5.R
Drive: FR
Weight: 2,510lb
WB: 49%
Springs: 706.6/735.5
SWR: 49.00%

Make: Ferrari
Model: #11 Larbre Competition 550 Maranello GTS
Drive: FR
Weight: 2,425lb
WB: 49%
Springs: 682.7/710.6
SWR: 49.00%

Make: Honda
Model: #8 ARTA NSX
Drive: MR
Weight: 2,601lb
WB:
Springs:
Calc:


CLASS: R3

Make: Chevrolet
Model: #31 Whelen Engineering Corvette Z06
Drive: FR
Weight: 2,950lb
WB: 50%
Springs: 729.7/729.7
SWR: 50.00%

Make: Dodge
Model: #126 Team Zakspeed Viper GTS-R
Drive: FR
Weight: 2,976lb
WB: 47%
Springs: 803.7/906.3
SWR: 47.00%

Make: Ferrari
Model: #71 Tafel Racing F430 GT
Drive: MR
Weight: 2,635lb
WB: 45%
Springs: 681.1/832.5
SWR: 45.00%

Make: Ford
Model: #25 Britek Motorsport FG Falcon
Drive: FR
Weight: 2,987lb
WB: 55%
Springs: 812.8/665.0
SWR: 55.00%

Make: Ford
Model: #10 tiger Racing Mustang
Drive: FR
Weight: 2,900lb
WB: 53%
Springs: 760.4/674.3
SWR: 53.00%

Some things I noticed were that pretty much all the R-class cars had exactly the same SWR as their published WB.  Many of the cars of the same drive type had the same WB; for instance a lot of the FR cars were 50% and MR 45%.  When I gathered the info for the R3-class, I started to look for cars with more varied WB which is why my results appear more diverse than those for my R1-cars (at the start I was simply taking the info of the first car available).  I had to scroll through quite a lot of R3 cars before I got ones that weren’t FR-50% or MR-45%.

After typing this up this up this morning I noticed that I missed the data for my last car in R2-class and that there is only 1 car that the SWR does not match the WB reading precisely.  I need to check this reading is correct and get the info for the other car later but I’m at work at the minute so I am unable to check it.

I also had a look the other day at the effects of changing the WB by installing parts to the SWR and it does appear to alter when they are fitted.  I need to check this again really as I have no figures noted on it.

I also checked the SWR of the non-race car I was working on last night after collecting this data which was my Porsche 911 GT3 RS (997).  It’s published WB was 39%, the springs were 790.2/903.4 giving a SWR of 46.67%.  All the R-cars seemed to have their SWR at exactly that of their published WB so it was interesting that this non-race car was out by so far from its published figure.  I then used the published WB 41% of one of all the R2-class Porsche’s with a RR drive type and re-tuned it to 694.5/999.0.  It felt really good to be honest and much better than previously. I haven’t had time to do further investigating with any other cars on this yet.

I think there could be something in the drive type being used to determine what SWR should be used.  Has anyone else been working on anything similar to this yet?

Phil

[Fit4aking]

Good write up Phil. Welcome to the board!  I've been struggling with bias and SWR a bit this time around and I can't wait to see what happens when the pieces start falling into place for me.  The more information on this the better.  Thanks for your input.

[Silvo 1981]

Found time to pop home for "lunch" and checked if altering the WB affects the SWR.

Make: Honda
Model: NSX R-GT
Drive: MR
Weight: 2,601lb (I had added a few parts but re-applied stock tune)
WB: 43%
Springs: 797.5/851.5
SWR: 48.36%

I then added Race Exhaust, Roll Cage and Weight Reduction to increase the WB to 45%. The springs changed to 676.7/722.6 but retained the SWR of 48.36%. I then went the other way and removed those parts and added the Race Oil and Cooling and Twin Screw Supercharger to reduce the WB to 42%. The springs changed again to 816.1/871.4 which still kept the SWR at 48.36%.

Therefore, changing the published WB by adding parts to the car that alter the published WB does not affect the SWR of the car as I had wrongly believed in my last post.

I ran out of time to check the #8 ARTA NSX (missed details off my notes) and the Audi Sport North America R10 TDI (WB does not match SWR). Hopefully do that tonight and maybe look at a few more cars.

[fndrbndr]

Hm...seems like the race cars are tuned in a manner similar to Choccy's approach. 

[Blooze]

Thanks for the info - good job!  And let me second Fit, "Welcome to the Forum."

First, let me clear something up.  The value you are showing as SWR is acutally something more along the line of "Spring Bias", or Sb.  The calculated SWR for this particular car is 132.76%.  This number is far and away beyond anything we ever saw in Forza II, where the typical SWR for an R1 car was around 84%

This is a link to the original white board post concerning the term SWR and why we were "looking" for it.
http://forza-tuning.net/index.php/topic,12.msg6147.html#msg6147

I do find it interesting that the Wb and Sb match on the R# cars...  I have yet to find a "match" on any of the production cars that I have built so far.

The SWR was our way of coming up with a value that could tie the Spring/Weight in to the ARB and Rebound dampers - the key, so to speak, to a one button click setup.  Having the SWR would give you your ARB and Damper values.

In just a few cars now, it is obvious that the relationship between springs and ARB and Dampers for Forza III is no where near what it was in Forza II.  i.e., the Tonka Toy is busted (removing the "x 2" modifier doesn't help either).

However, the good news is that the Wb thing is still in play... I think - at least a good sized chunk of evidence points to it being the case.  I too ran into the same anomaly that Phil did on changing the Wb with parts.  The Wb would change, but the new weight distribution was according to the Sb, not the Wb as you would expect.  I was using an Acura Integra with a Wb of 62% and a Sb of 53%.  Changing the Wb does not change the Sb, nor does it change the bias of the ARB or Dampers.  In fact, as in Forza II, the ARB and Dampers are not affected by weight changes.  At least, those are my observations at this point.

So then, the working Theory on Weight Bias...

• The Weight Bias used by the physics engine is derived from the Sb, not the published Wb.
• The published Wb is nothing more than a ruse, along with the new "changing" Wb.

The word "ruse" may be a bit harsh, but I consider it nothing more.  For one thing, it is not applied across the board - buying a wing makes the whole car heavier, not just the rear end - the Wb is not altered.  Buying a splitter does not change the Wb as you would expect it to.  And, once again, the concept of Unsprung Weight is not, or does not, seem to be addressed.

All that, and Phil's experience with his Porsche leads me to believe the Wb Theory is still in play.

The Tonka Toy, on the other hand is going to need some work.

$

[Tonka Crash]

From what I remember of FM2, the spring bias, Sb, always agreed with the published Wb of an R-Class car.  As did the bias in damping and ARB settings. It was when you looked at the actual bias of the production class cars you really started seeing deviations from the published values.  I always took this to be the way T10 introduced individual character into each car.

I fully agree with published Wb being a ruse.  I've assumed that the published Wb has nothing to do with the actual settings of  the car beyond giving a crude ballpark number of the balance.  It would tell you if the car was close to balanced or had a fwd or aft bias, but not much more.

Has anyone stuck on a race suspension and then looked at default settings as weight reduction is added?  In FM2 the SWR and Wb remained constant as weight changed.  Since the published Wb moves with weight reduction I'm curious if this is still the case.

Also, springs are just one part of the story.  For the complete picture we really need ARB and damper settings as well.

[Blooze]

I have checked the Sb as weight is added and the Wb changes.  The Sb stays the same, no matter what the Wb does.  Also, the ARB and Damper values do not change with weight changes as before.

I checked the ARB and Dampers of the first car on Phil's list, the DeFerron Acura (Yaay, Acura!).  the Dampers are stronger in the new version, totals being 20.2 vs the old of 18.5.  The ARB are significantly softer now, 25.62 vs the old of 42.70.

The ARBs have a different ratio than the Sb, 45%.  Strange, compounded by the fact the ARB ratio been 45% on every car I have built so far.

The Dampers for the R car are wierd also in that the Bump is about 50% of the Rebound.  However for all my production cars it has been around 75%.

The ARB and Dampers have me bamfoozled a bit...  any changes I make to them are going to be real slight and without reguard to relationships until I understand them better.

$