No_OBsT33R's LSD Calc UPDATED Version 1.1

[No_OBsT33R]

We have gone off discussing the fundamentals of LSD's using real world examples. The clutch LSD is most likely while not certainly the type used in GT5.

There's to many variables aside from the differential to claim what power is "usable".

I never said the LSD claimed what power is useable, that's a variable you input in the calculation.

Determining how much power the tire can handle, is the next step.

I think this calculator would be awesome if you found a way to calculate how much wheel torque difference causes the LSD to do its thing. And how exactly the LSD shifts the power around. 👍

Agreed, it could be much better if we could calculate the other variables at play, this is the reason I posted this to begin with.

Let's work on figuring that out and generating a calc for the other variables. More heads working on it the better.


Note, still stoked RJ has begun to see the logic used. Good show RJ your getting it.

I think Da_Slump and I have been on the same page just conflicted over my explanation (I did the best I could, this is a complex discussion) once clarified it seems we are on the same page.

 

[Adrenaline]

Adrenalin. Why can't you seem to post in any thread without denigrating someone? Everything you have to say is just so "ur wrong, and I'm right. If you don't understand why, you must be stupid". Why? If you have such knowledge of these topics, maybe you should TEACH us about them instead of telling everyone they're stupid for being wrong?

I think you're mistaking Me for noobster.
Here are the related quote by noobster made in this thread alone.
I challenge you to find a single one of my quotes that's even close to the same level. Not to mention, none of my posts are edited, whereas noobster edits 90% of his, so I miss half of what he says, because I replied to his post before he went back and changed it.

Come on dude read up. Don't be so closed minded, the math is sound, even if you don't get it.

If your confused, entertain the thought it might be you that doesn't (fully) understand what the LSD is doing.

You are wrong, it is locked until it slips

Wrong, it's always locked until a difference in wheel speed causes it to slip.

Wrong, there is no effect at all until a difference in wheel speed, until then it remains locked

Wrong, it doesn't slip until a wheel speed difference is detected.

WRONG again

If you say so

I was hoping for some ideas from the peeps here but it seams when you shine true light on something it tends to piss off a few comfortable living in the dark.

nobody has been able to disprove it beyond the usual babble that goes on on this site.

if I'm so wrong and RJ is so right, PLEASE show me where exactly my calculator is wrong, if not, with all due respect STFU......

Whatever, truth stings.

I'm practical attacked by people who don't fully understand (Like you, you obviously don't because you don't realize there is ALWAYS equal power given to both wheels, yet your making statements about who is right and wrong? STFU and listen if you don't know yourself, your in no position to weigh judgment),

What have you done to help besides copy past a bunch of people repeating the in game text in their own words into a thread?

I gain NOTHING helping people out and putting my time into it, so F-it, Ill keep my stuff to myself 👍👍👍 AWESOME FORUM!!!!!!

Don't tell me I'm flat out wrong if you don't agree with my calculations

Its one or the other, not both at the same time, comprehension issues?

is there something wrong with you?

It's hard to fathom the abuse you must go through in reality for you to have this overpowering urge to be an Asshole online

Anyways, to answer your stupid question

incredibly intelligent question asshole.

I don't have to do a good damn thing to prove anything to anybody

GTFO!!! Grow a set.

What a joke!!

running your mouth through the safety net of the Internet.

Get Bent Adrenalin

stop being a tool

No moron

WRONG!

It also disproved your stupid conclusion

you Dumbass.

If you disagree with this simple FACT you need to STFU

 

[No_OBsT33R]

Evil give it up this guy has no life... He enjoys it when he gets a response, I've been sucked in to responding, I've recently learned a great technique to deal with Trolls like him..

 

[Rotary Junkie]

Note, still stoked RJ has begun to see the logic used. Good show RJ your getting it.

Not at all. You still don't see what I'm saying... And that thinking on preload came up in discussion long ago from me.

Would mean that 84% lock is possible but then... That would mean certain vehicles have an astounding 90% maximum lock (most RWD Nissans, 2000 Cobra R, etc) under acceleration... And I've yet to hear of an LSD capable of providing that. So like I said, I'm still leaning towards the initial = minimum lock thing.

And your idea of how it works scares the hell out of me... Because it's sure as hell nowhere near how a real LSD works.

 

[No_OBsT33R]

Not at all. You still don't see what I'm saying... And that thinking on preload came up in discussion long ago from me.

Would mean that 84% lock is possible but then... That would mean certain vehicles have an astounding 90% maximum lock (most RWD Nissans, 2000 Cobra R, etc) under acceleration... And I've yet to hear of an LSD capable of providing that. So like I said, I'm still leaning towards the initial = minimum lock thing.

And your idea of how it works scares the hell out of me... Because it's sure as hell nowhere near how a real LSD works.

Naw it's Preload (you were going in such a good direction with that thought, it's the right one) it's at 10% with a setting of 10, to get 90% is impossible so is 84%. In the game the highest is 60% plus at 90% or 84% it's going to be extremely hard to turn, if it hypothetically were possible to get those values.

 

[Rotary Junkie]

Naw it's Preload (you were going in such a good direction with that thought, it's the right one) it's at 10% with a setting of 10, to get 90% is impossible so is 84%. In the game the highest is 60% plus at 90% or 84% it's going to be extremely hard to turn, if it hypothetically were possible to get those values.

I'm not saying 90 or 84% preload... I'm saying total lock.

Since you can never have less accel lock than preload, the accel/decel values have to be on top of preload... Hence my 84, 90% examples. 84% is the maximum possible lock with the FC LSD (60 initial, 60 accel) if this is true, but then the odd one-way (50/80/0) diff used in Nissans, Cobra R, etc would have a max of 90% lock... Which is mad high given the hardest locking diffs on the market in the real world are around 80%.

My present theory on it that I'm thinking is more accurate says maximum lock with FC LSD is indeed 60% and that initial is a "minimum lock" of sorts... Which is to say that it's the absolute least the differential will lock, done as a percentage value of accel/decel.

 

[EivlEvo]

I think you're mistaking Me for noobster.
Here are the related quote by noobster made in this thread alone.
I challenge you to find a single one of my quotes that's even close to the same level. Not to mention, none of my posts are edited, whereas noobster edits 90% of his, so I miss half of what he says, because I replied to his post before he went back and changed it.

You sir are correct. Noobsteer could also stand to relax the denigration as well, though, when you do something designed to help people out, and an entire forum (seemingly) goes on the attack rather than being constructive, it tends to wear on someone.

Your first post in this thread which I believe started with "for the lulz" or something to that effect. It isn't exactly warm and fuzzy? I don't believe that he mentioned you prior, so if you had some information to suggest that his calculator may be at fault or have an error, why not explain things nicely or without quips? And whether you've stooped to his level or not, an insult is an insult. There are no levels.

Anyway... noobsteer is correct. I probably shouldn't be sticking my nose where it doesn't belong. But frankly, most people could stand to be a tad more polite when criticizing in this forum (myself included, in case you missed the SC/TC thread). It is a video game, or "simulation" if you will.

 

[Motor City Hami]

Noobsteer - I'm not going to argue numbers with you. Your intent is pure - to build a tool to help people understand and better set up their LSDs - right?

Assuming that your calculations are indeed right on the money for how LSDs really work and how GT5 programmed them, can you write more about how people should be able to use this tool. I am having a hard time figuring out how to apply it to setting up a car in the game. How did this tool help guide you into picking the LSD settings that you chose for the four cars posted in your garage? Maybe that could help clear up what the tool does.

LF-A Initial 12, Accel 36 and Decel 18
M3 Initial 8, Accel 32 and Decel 19
Trans Cammer Initial 16, Accel 49 and Decel 18
Yellow Bird Initial 14, Accel 54 and Decel 27

 

[No_OBsT33R]

I'm not saying 90 or 84% preload... I'm saying total lock.

Since you can never have less accel lock than preload, the accel/decel values have to be on top of preload... Hence my 84, 90% examples. 84% is the maximum possible lock with the FC LSD (60 initial, 60 accel) if this is true, but then the odd one-way (50/80/0) diff used in Nissans, Cobra R, etc would have a max of 90% lock... Which is mad high given the hardest locking diffs on the market in the real world are around 80%.

My present theory on it that I'm thinking is more accurate says maximum lock with FC LSD is indeed 60% and that initial is a "minimum lock" of sorts... Which is to say that it's the absolute least the differential will lock, done as a percentage value of accel/decel.

Uhhh whut? It's all good...

350ft-lb going straight gets 175ft-lb per wheel.

Can we agree that the higher the INT setting the harder the car is to turn, while at the same time the car is more stable????

Okay 350ft-lb one wheel slipping, in the air, on ice, take your pick.

A locked Diff = 175 / spinning

Open Diff = 0 / spinning

60/60 = 42 / spinning
Hard to turn, not much power retained, very stable.

50/80 = 70 / spinning
This is interesting, it should indicate a tough car to turn with decent power retention, & very stable.

40/90 = 94 / spinning
Hard to turn however much power retained, power retained looks promising.

80/50 = 17.5 / spinning
Fight to get her to turn to turn, little power retained.

90/40 = 7 / spinning
Hardest to turn, massive understeer. Barely any power at all retained, waisted LSD. The car is extremely stable though o_O

50/50 = 43 / spinning
Tough to turn, not much power retained, very stable.

40/40 = 42 / spinning
Not as hard to turn, not much power retention, stable car

35/40 = 45.5 / spinning
Still tough to turn, loosing stability, not much power retained

25/40 = 52.5 / spinning
On the right track, getting easier to turn, retaining more power, but stability is going down.

15/40 = 59.5 / spinning
Easier to turn, good power retention, loosing stability.

10/40 = 63 / spinning
Very easy to turn, not very stable, good power retention.

5/40 = 66.5 / spinning
This is easiest to turn and has great power retention, but is highly unstable.


What's the issue? Too loose? Too much? Power retained?

Then there is adjusting the transition point (where the accel overcomes the loss from the int (throttle snap)) thats done closer to 70-80 percent useable power.

You adjust the useable power to match the int and progressively reduce the useable power to see the progressive effect.

 

[dr_slump]

My present theory on it that I'm thinking is more accurate says maximum lock with FC LSD is indeed 60% and that initial is a "minimum lock" of sorts...

Yes could be

Which is to say that it's the absolute least the differential will lock, done as a percentage value of accel/decel.

Disagree with your percentage theory

 

[Rotary Junkie]

Uhhh whut? It's all good...

350ft-lb going straight gets 175ft-lb per wheel.

Can we agree that the higher the INT setting the harder the car is to turn, while at the same time the car is more stable????

Okay 350ft-lb one wheel slipping, in the air, on ice, take your pick.

A locked Diff = 175 / spinning

Open Diff = 0 / spinning

@Bold: No. A locked diff will have both tires spinning at exactly the same speed... So either both are spinning or neither is.

I only partially agree with the Int = harder to turn/understeer thing... The results from changing it seemed more consistent with what you're saying in GT4 but in GT5 I've not had the same effects... I've maxed out initial with absolutely zero change in decel behavior (but change in the transition to on-throttle) in GT5, in GT4 it caused understeer.

60/60 = 42 / spinning
Hard to turn, not much power retained, very stable.

I still don't get what makes you think a high initial creates power loss... If initial torque is indeed preload (as you claim) then that just means there is always a certain amount of resistance to wheel speed differences and less change in resistance between neutral throttle and accel/decel. Again, the clutches only handle power transfer from one axle to the other, not from the ring gear to the axles.

50/80 = 70 / spinning
This is interesting, it should indicate a tough car to turn with decent power retention, & very stable.

Not entirely the case. It's very understeery at moderate to high throttle with good traction but a high tendency to go from understeer to snap oversteer in the blink of an eye due to the high accel lock. I'd love to be able to try it on a FWD though... Ever driven the XF in LFS with a spool? It'd be like that but with less push on entry. Would destroy tires but be absolutely blazingly quick.

40/90 = 94 / spinning
Hard to turn however much power retained, power retained looks promising.

Would suffer much the same as above on the rear diff... Push, push, loose.

80/50 = 17.5 / spinning
Fight to get her to turn to turn, little power retained.

Would have an extremely sudden and fairly stiff lock, may even be possible for it to spin the inside rear until wheel speed passes that of the outside, followed by a violent lock and snap into further oversteer.

90/40 = 7 / spinning
Hardest to turn, massive understeer. Barely any power at all retained, waisted LSD. The car is extremely stable though o_O

Wasted LSD indeed... No lock, no lock, no lock, 40% lock nearly instantly. Can't see how you can say the "car is extremely stable" seeing as you can't test it though.

50/50 = 43 / spinning
Tough to turn, not much power retained, very stable.

Fairly late and rather strong lock, the initial "hit" would be like a xx/25 getting hit instantly followed by very quick ramp-up. Snap understeer on-throttle.

5/40 = 66.5 / spinning
This is easiest to turn and has great power retention, but is highly unstable.

Given I run diffs like that all the damn time (7/35/5 is my usual go-to in fact, a carryover from GT4 days) I must disagree with the "highly unstable" bit. In fact, it's probably close to the most stable you can get while maintaining decent corner exit pull in a lowish torque car. More accel will help exit speeds at the cost of increased likelyhood of snap oversteer.

The "most stable" diff is generally 5 or 0 accel... Because inside wheelspin is safe as hell compared to lighting both up.

Edit:

I totally agree

But what happens with a setting like this: 30/20/20?
Overall 30% lock?

Max lock of 20% on accel or decel, minimum lock of 6% on either. Open until there's enough input torque to cause 6% lock in other words.

30/20/5 would have...

Accel min: 6%
Decel min: 1.5%

 

[No_OBsT33R]

But what happens with a setting like this: 30/20/20?
Overall 30% lock?

Same conditions but;

30/20 = 24 / spinning

Tougher to turn, relatively stable, but poor power retention.

Closer to 15% retention from looking at it.

I don't know what your getting at RJ are you agreeing with the result of the calc but disagree with my visual assessment? Have you tested any of the results right now other than your usual go to values?


Ohh stable is tougher to turn IMO and description, unstable cars are faster because of a greater ability to change directions rapidly. This is harder to do with high int settings as apposed to high accel settings.

Ohh and when the locked diff is

175 / spinning the wheel is either slipping or not on the ground therefore it has ZERO useable power (did you forget that fast?)

Can't even throw you a bone.

 

[dr_slump]

30/20/5 would have...

Accel min: 6%
Decel min: 1.5%

How can you have different lock without accel or decel (rolling trough the corner)?

 

[Rotary Junkie]

How can you have different lock without accel or decel (rolling trough the corner)?

This is a bit problematic.

You can't have preload be less than accel or decel yet, if IT is preload, you can with GT5.

This means that either accel/decel go on top of initial (donut test seems to preclude this) or initial is not preload... And that preload is possibly a constant or arbitrarily assigned (perhaps an average between accel and decel?).

My thinking is, as I said, that initial is the minimum lock that the diff will engage at (torque required to activate LSD as mentioned in game menu). My interpretation of this is that it's a percentage of accel/decel value as I have already detailed. I could be wrong but it seems to be a sound explanation for what I've seen/felt.

 

[No_OBsT33R]

This is a bit problematic.

You can't have preload be less than accel or decel yet, if IT is preload, you can with GT5.

This means that either accel/decel go on top of initial (donut test seems to preclude this) or initial is not preload... And that preload is possibly a constant or arbitrarily assigned (perhaps an average between accel and decel?).

My thinking is, as I said, that initial is the minimum lock that the diff will engage at (torque required to activate LSD as mentioned in game menu). My interpretation of this is that it's a percentage of accel/decel value as I have already detailed. I could be wrong but it seems to be a sound explanation for what I've seen/felt.

Dr_Slump could you please, I can't take it anymore. I'm going to play some GT5

 

[dr_slump]

I think accel and decel are surreal numbers and depend on a constant (Initial).
Cause everytime you change the Initial you've to change your accel and decel settings too.

 

[VAjeeper]

As far as the game goes... in your calculator, is the first number the int torque and the second number the accel ? As in 50/50 = int/accel?

I'm well versed in real world applications of diffs, since I rely on them for traction offroad. I'm currently running air lockers front and rear, but I'm not stepping into that conversation, haha

 

[No_OBsT33R]

I think accel and decel are surreal numbers and depend on a constant (Initial).
Cause everytime you change the Initial you've to change your accel and decel settings too.

If I said it I would be wrong in some way.

RJ the accel is dependent on the int. The Decel is dependent on the INT, the accel and decell are independent of each other. En tous cas, C'est la Vie.

 

[Nicktune]

No_obst33r since you're going to play with GT5...can you test stock diff setup vs your calculated diff setup...would be nice to see the results(time)..tsukuba maybe?
Thanks.

 

[No_OBsT33R]

No_obst33r since you're going to play with GT5...can you test stock diff setup vs your calculated diff setup...would be nice to see the results(time)..tsukuba maybe?
Thanks.

Stock Diff would be (generally)

7/30/15

With one wheel spinning 500ft-lb

7/30 = 48.83 / spinning

Locked = 175 / spinning

Open = 0 / spinning

What exactly would you like me to test? Lap time after adjustment? It's not really an ideal test to show result, I could be getting variations due to too many other factors for tangible results.

Maximum benefits will be realized when we can calculate the limitations of the tires.

It's always informative to have a visual representation of the forces at play even if only to help our train of thought while adjusting to feel.

I was hoping for this to lead to other things, doesn't seam like that's the direction it's going.

 

[chuyler1]

If you run 10-20 laps on each setup at Tsukuba, in practice mode, it should reveal something. The track is short enough that it doesn't take long to get clean laps in. If there is an issue with traction entering or exiting turns you'll know as your ghost flies by you.

Honestly, if you don't test, how do you know your theory works?

 

[Leonidae@MFT]

Show us a driver that is consistent enough to try out your way..

 

[No_OBsT33R]

If you run 10-20 laps on each setup at Tsukuba, in practice mode, it should reveal something.

What set up? The tool is to provide info to the tuner for when they are adjusting the LSD, it doesn't adjust it for you.

We will be able to use the info it provides to greater extent when we can calculate the limitations of the tire compounds.

 

[chuyler1]

Sorry, I wasn't able to view your spreadsheet.

Here are some basic notes...
1) If you have an updated version, update your OP with the new version. I see version 1.1 is on page 2.
2) Learn how to save as an older excel file (choose "Save As...").
3) Use Zip instead of Rar
4) Explain what it is supposed to do on the calculator.

Honestly, this calculator does nothing for me. A better approach would to provide typical stats on a car and have it pump out a suggested LSD setting. For example, if I were to enter a car's tires, torque, weight, downforce, and drivetrain class, it could give me a starting point for the settings.

 

[No_OBsT33R]

Sorry, I wasn't able to view your spreadsheet.

Here are some basic notes...
1) If you have an updated version, update your OP with the new version. I see version 1.1 is on page 2.
2) Learn how to save as an older excel file (choose "Save As...").
3) Use Zip instead of Rar
4) Explain what it is supposed to do on the calculator.

Honestly, this calculator does nothing for me. A better approach would to provide typical stats on a car and have it pump out a suggested LSD setting. For example, if I were to enter a car's tires, torque, weight, downforce, and drivetrain class, it could give me a starting point for the settings.

I use excel 07 and winrar, why should I "learn" how to save in an outdated format (like there is anything to learn about choosing a save option, lol) to accommodate people who need to "learn" to update their software (97 come on) winrar is what I use, don't like winrar? Not my problem, or concern. People can use whatever zip tool they want. If your tool can't handle rar files, get a better one.

It doesn't have to do anything for you, it's free, and only for those who wish to use it, or help make it better.

If you don't want to help, it's all good, feel free to not use it, don't help, do your own thing. Why say anything at all?

If you get no use out of it, what difference does that make? It's too bad, but no big deal. It's not like you paid for a app to solve your LSD woes and all you got was a excel calc showing what the wheels are doing. If your unable to use the info provided to help yourself, have your own methods and they work for you, use them

If you have a better approach by all means, make a thread and discuss it.

 

[chuyler1]

I am trying to help. I have Office 2K on one machine at home and OfficeXP on the other. It works fine and I've never had a reason to update it. Actually, aside from the background image, you could probably create the same spreadsheet using Google Docs. If you want lots of feedback, you have to provide your tool in a format that lots of people can use.

Anyway, my point is, you obviously want to help the community out by contributing a calculator. My suggestion is to calculate something useful for newbs to tune with. The experts already have their theories and don't need this to fine tune the LSD. If you want to sell something to them (sell being figuratively, not literally), you've got to have something concrete that produces faster lap times. Otherwise people are going to pick holes at your theory until you provide evidence.

 

[Motor City Hami]

... those who wish to use it, or help make it better.

Ok. You have convinced me that you are onto something, so I am going to try to help. I will leave the numbers side of things to you and others who want to argue with you. Your intent is pure here - to build a tool to help people understand and better set up their LSDs - right?

Assuming that your calculations are indeed right on the money for how LSDs really work and how GT5 programmed them, can you write more about how people should be able to use this tool or how you see it helping the newby. I am having a hard time figuring out how to apply the findings to setting up a car in the game. How did this tool help guide you into picking the LSD settings that you chose for the four cars posted in your garage? Maybe that could help clear up what the tool does or is intended to do.

LF-A Initial 12, Accel 36 and Decel 18
M3 Initial 8, Accel 32 and Decel 19
Trans Cammer Initial 16, Accel 49 and Decel 18
Yellow Bird Initial 14, Accel 54 and Decel 27

 

[No_OBsT33R]

I am trying to help. I have Office 2K on one machine at home and OfficeXP on the other. It works fine and I've never had a reason to update it. Actually, aside from the background image, you could probably create the same spreadsheet using Google Docs. If you want lots of feedback, you have to provide your tool in a format that lots of people can use.

Anyway, my point is, you obviously want to help the community out by contributing a calculator. My suggestion is to calculate something useful for newbs to tune with. The experts already have their theories and don't need this to fine tune the LSD. If you want to sell something to them (sell being figuratively, not literally), you've got to have something concrete that produces faster lap times. Otherwise people are going to pick holes at your theory until you provide evidence.

I'm not doing anything to empress the "experts" and I use that term in the loosest possible way. I don't see anybody else providing tools with tangible results, so far I'm the only one. Just theories, my calc is not a representation of my theories on how to tune an LSD, it calculates what the LSD is doing when activated.

So it's all good, if you don't want to help, don't. Why continue on as such on your opinion of the validity of the work I've already done? Are you interested in helping, or are you just trying to illustrate your opinion as an elevated one?

 

[chuyler1]

I am interested in a tool that actually provides tangible results. This is not it. For it to be considered a calculator, it would have to in some way aid the average user in configuring the LSD. To use the various spring rate calculators available I don't have to understand suspension theory, I just have to know the current weight, and weight distribution of my vehicle.

You spend hours arguing your theories on this site and I think you just need to sit down with the game and run some of your own tests to create equations for optimal LSD settings. The input variables are drive type, torque, tire compound, and possibly weight and weight distribution. The output should be the 3 LSD settings. If you want to contribute to the community, that's the best way to do it. No one else has done it yet because we haven't bothered to spend the time testing equations. Turn your theories into fact by creating a tool that turns those inputs into optimal LSD settings. Until you do that, your theories are...well...theories.

 

[No_OBsT33R]

You keep missing the point, it's not a LSD settings calculator, but a LSD action calculator.

It doesn't tune LSD it provides info. The level of usability of the info is minimal at the moment, but can be further beneficial if we figure out how to calculate other info to use with the info provided by this tool. The goal at this point is to figure out ways to calculate the other info required (tire compound limitations, as an example) to get the most out of what we have here.

You can either help or waist time calling it useless...

When we can figure out how to calculate the amount of useable power (in a 1G conering enviroment for example) instead of manually inputing the variable, the tool will come to life.