Misc Physics Pipeline + tools

[Kyuubeey]

Kyuubeey submitted a new resource:

Physics Pipeline + tools - Documentation, advice and tools for physics creators

08.04.2021 Version 0.1

- Initial 0.1 release
- Added Pipeline
- Added Car Analyzer
- Added Damper Sheet
- Added Stabilizer Index
- Added Inertia Solver


Links

Pipeline: https://docs.google.com/document/d/1gZ7NetA2dHOI3mfc06jnQcJqRCPcOhc3FObf6CaxReU/edit?usp=sharing
Car Analyzer: https://docs.google.com/spreadsheets/d/1LAlUbTERg1SaVz9D-6dRqKsdZ7IqjsY2TF506e2F1YU/edit?usp=sharing
Damper Sheet...

Read more about this resource...

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

Load curve tool, more documentation, more advice, typos

09.04.2021 Version 0.2

- Added Tire Load Curve Helper
- Added first documentation of drivetrain.ini to Pipeline
- Expanded tire chapters in Pipeline
- Better documentation of FLEX_GAIN in Pipeline
- Corrected writing in some parts of Pipeline
- Fixed typo in front and rear bump travel default parameter in Car Analyzer

Read the rest of this update entry...

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

Inertia solver removal

09.04.2021 Version 0.3

- Deleted Inertia Solver by author request

Read the rest of this update entry...

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

Engine, suspgeo

10.04.2021 Version 0.3B

- Added first documentation of engine.ini to Pipeline
- Expanded Suspension geometry section of Pipeline
- Added definition of LINEAR_STEER_ROD_RATIO to Pipeline

Read the rest of this update entry...

 

[Kyuubeey]

Just now I noticed that I forgot engine.ini from the table of contents in Pipeline.

These kind of things will happen very much until 1.0. Just bear with it. I'm not even going to include page numbers or linking to chapters before everything is relatively finished, so as to not waste time redoing things.

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

Engines, suspension, tires

11.04.2021 Version 0.4

- Corrected some tire documentation inside of Pipeline
- Added Advice for engines chapter
- Expanded suspension geometry chapter
- Added missing engine.ini entry to table of contents

Read the rest of this update entry...

 

[Kyuubeey]

Beware, there is an error in the Car Analyzer CG calculator. The height is correct from what I can tell, but longitudinal is not. The sprung mass is being split at the wrong moment so the calc ends up off.

I will need to fix that for the next version, for now don't use it. In-game "telemetry" tool outputs wheel loads.

 

[Tennisbiki]

Not sure if this is the right place to discuss this, the steer_car or tyre position.

I think I am just starting to understand what you stated about not ideal to put the mount on the axis of lower A-arm, because when adjusting for toe the mount gets moved laterally, along x-axis. Consequently creates a toe change over the travel, bump-steer. This example here is a Stratos I am experimenting. (I love this car, fyi. But apparently the mod, looses camber and slides out easily.) I am wondering how those rear strut or DWarm with those "N" arm are modeled by others.

 

[Kyuubeey]

Not sure if this is the right place to discuss this, the steer_car or tyre position.

View attachment 465417 I think I am just starting to understand what you stated about not ideal to put the mount on the axis of lower A-arm, because when adjusting for toe the mount gets moved laterally, along x-axis. Consequently creates a toe change over the travel, bump-steer. This example here is a Stratos I am experimenting. (I love this car, fyi. But apparently the mod, looses camber and slides out easily.) I am wondering how those rear strut or DWarm with those "N" arm are modeled by others.

Adjust the toe arm up or down on the CAR until you find a good position. Usually you have to go 0.1mm at a time. It won't be perfect but I've found some which are basically 0 bumpsteer at most toe angles. Maybe move it 0.1mm downwards in your geo and check every time.

About the Analyzer, I did some testing to determine the behavior of some things having to do with STRUT, so I will also include a fix for those later. The calc is just a bit annoying so it needs more verifying.

 

[Tennisbiki]

Adjust the toe arm up or down on the CAR until you find a good position.

Thank you for the reply.

I am assuming "up or down" refer to to my attach view which is increasing or decreasing z value in mm. Or did you mean Y value, as in elevation?

Seems like a iterative process. Can you share the tool you prefer to observe this bump steer or camber curve?

For me I test drove the car and watch the suspensions app. I could see in real time, that the camber went from negative to zero or positive as the rear went into skid. I wish I can see them in replay. Toe change is a feel thing. I found that car sometimes does this swaying back and forth at speed, like fish-tailing sort of. Adjusting the toe arm position sometimes fix that.

 

[Kyuubeey]

X side to side, Y up and down, Z forward and backwards.

Download Stereo's "test pad", then put in rod length adjustment to setup. Move the assembly up and down and check the angles, that is your curve. I do it ingame so I see what it, for sure, is giving me.

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

STRUT corrections

17.04.2021 Version 0.5

- Corrected (?) some misinformation about HUB_MASS for STRUT in Pipeline
- Corrected (?) STRUT toggle in Car Analyzer

Read the rest of this update entry...

 

[SDX]

Very exciting to see such tools shared.
I collected a bunch of other calculators so far, but will have to test yours when I regain free time

Thank you for your pedantic physics tuning! Game developers do not have a chance to get this detailed and put as much time into research as enthusiasts like you.

 

[Kyuubeey]

Kyuubeey updated Physics Pipeline + tools with a new update entry:

Updated Pipeline, more sheet functionality

18.07.2021 Version 0.6

- Updated Analyzer, made default sheet generic
- Updated Damper, made default sheet generic
- Updated Stabilizer, added more data
- Updated Load Curve Helper, added more data
- Corrected misleading statement in Pipeline engine.ini section
- Corrected or removed misleading statements in suspension design sections and replaced with new information
- Removed some bad examples of suspensions and added better ones to suspension design sections

Read the rest of this update entry...

 

[Kyuubeey]

@Tennisbiki

Hey, a followup to the last semi-trailing arm discussion:

You are correct from the start, actually. All of that was a misunderstanding due to some error in how AC displays the wheel alignment in the setup and telemetry. It is not accurate.
It seems like caster gain will produce error currently.

Maybe it will be fixed later, if the current testing fix makes it to public. For now use external software to check the curves.

Make trailing arms the right way, with the toe arm on the instant axis! Ignore that crap I said earlier. The toe arm deflection from toe changes doesn't throw it off too much.

 

[JackCY]

Is it possible to get the inertia solver? Or at least more information about it?
I have my own approximation when using the default inertia parameters in AC.

From the documentation:

“INERTIA” determines the SPRUNG inertia of the car in the form of a dimension box that has mass specified by TOTALMASS. Unsprung mass generates its own inertia separately. You must NOT input in the total car box dimensions. A solver to solve for sprung inertia from total car inertia has been developed and may be shared in the future. Input is in meters. Typical values are smaller than the physical constraints of the vehicle.
Example: BMW E30 1.450,1.260,4.200 total, 1.179,1.339,4.149 sprung.

That looks to me same as the default AC inertia in car.ini. It's 7am so I don't remember what the other inertia definition method was called.

The vanilla AC inertia is this KS box thingy.

Ah yeah explicit inertia:

- MODDING REQUEST: Added explicit inertia instead of box based. Use
[EXPLICIT_INERTIA]
INERTIA=1300,1400,500
in car.ini

Was it a calculator for that explicit inertia? What units are those numbers in? For KS it's meters (dimension) but what units is it for the explicit inertia? I actually found a calculator for inertia so I need to convert the units. Or make my own calculator for it if this one isn't complex enough.

 

[JackCY]

Not sure if this is the right place to discuss this, the steer_car or tyre position.

View attachment 465417 I think I am just starting to understand what you stated about not ideal to put the mount on the axis of lower A-arm, because when adjusting for toe the mount gets moved laterally, along x-axis. Consequently creates a toe change over the travel, bump-steer. This example here is a Stratos I am experimenting. (I love this car, fyi. But apparently the mod, looses camber and slides out easily.) I am wondering how those rear strut or DWarm with those "N" arm are modeled by others.

You can calculate the ideal position for steering arm. So you get minimal toe change with suspension travel. At least for DWB, don't think I have it for STRUT since I can't move STRUT.

Which Stratos mod? I did adjust one a bit long ago, an old tarmac rally Stratos. May still have it somewhere.

 

[Kyuubeey]

No, the solver is to solve for sprung-mass inertia, because AC uses sprung inputs and not total like everyone (including Aris...) thought. It has nothing to do with CSP. You can ask JPG for the solver. Don't ask him if you don't have real inertia values you're making your box from.

If you don't have real data for it then it doesn't matter. Just know that your car should be a bit narrower generally, because the unsprung is not included in it.

Exclipit inertia is in kg-m^2 like all of the IRL data is, but it's buggy, don't use it. I'm not sure if it even works currently. It's not a CSP feature, it's a KS feature.

EDIT: Oh and forget inertia calculators. You can't do it. You need at least a CAD model with CG and mass for every single object, otherwise forget it.

 

[JackCY]

No, the solver is to solve for sprung-mass inertia, because AC uses sprung inputs and not total like everyone (including Aris...) thought. It has nothing to do with CSP. You can ask JPG for the solver. Don't ask him if you don't have real inertia values you're making your box from.

If you don't have real data for it then it doesn't matter. Just know that your car should be a bit narrower generally, because the unsprung is not included in it.

Exclipit inertia is in kg-m^2 like all of the IRL data is, but it's buggy, don't use it. I'm not sure if it even works currently. It's not a CSP feature, it's a KS feature.

EDIT: Oh and forget inertia calculators. You can't do it. You need at least a CAD model with CG and mass for every single object, otherwise forget it.

Thanks, I noticed it's not CSP and edited it a little later. I just browsed my template and it seemed from my file name it could be but it's the "AC extras from changelog" that rarely if ever even get used by KS. Ah good that inertia calculator also seemed to be in kg m^2, yes it does need to have all these various many parts weights and their locations. But if I can figure more of these parts or approximate then I may get at least something out of it for testing. See how much it differs.

I vary (for each car) the "default box inertia" based on car dimensions, roof presence, width of main body (road vs formula style), CG location, that sort of thing. Maybe I could also remove the ground gap, not sure I did that, that is if the unsprung is to be removed from it.

 

[Kyuubeey]

Find actual inertia values from similar cars to use as a base then correlate it from footage or telemetry. Otherwise you're mostly just wasting your time trying to write a sheet to automate something that you can't automate with so little parameters.