[WIP] Porsche 935/K3

Just wants to say that current public version is still v0.50 with "boaty" physics.
This has never meant to be the end result.

For a long time I have been struggling to build a better physics, but since this is my first real deep dive into this and the complex physics spreadsheet, it takes a while to get a reasonable result, and I will not update, until it's closer to what I would subjective think it should feel like.

When physics frustrates me too much, I make liveries (up to 19 now)
index.php
 

Attachments

This might not be the best advice, but I don't think that physics spreadsheet helps to make better physics that much, especially for beginners. It is probably best method for someone who is highly professional and highly accurate and exact, and already can build very good physics without a spreadsheet, and then use it for extra accuracy and professionalism. I do look at physics spreadsheet frequently to understand how some parameters works, as there is that much information in it. Other than source of clues and a tool for high end professional physics I think that the spreadsheet curbs people in making physics, especially the more new and amateurish they are. Even for professionals IMO it is inevitable to just simply have lots of quick tweaks straight into HDV and other files. Without that physics will end up dead and lacking soul, not everything is just data rules. One has to look deep into real car, and understand what needs to be achieved, and in what ways real car ends up having certain behaviors. For normal physics crafting to me spreadsheet is more like a distraction, I would probably use it for fine tuning and polishing, once physics already are configured decently and ready for next level.

Besides that IMO the old 911 are one of harder cars to make physics for. Well maybe this particular one is little bit easier, because it will naturally get more performance from good downforce, big tires it will also be stiffer and I think it is easier to make physics for stiffer chassis and suspension, there are less dynamics. It will still be challenging one to do.

I could offer my 2.8RSR physics as basis to from, if any use of that could be seen. If not then not. Basically adding turbo, giving different aerodynamics and tires would already be ̶h̶a̶l̶f̶ ̶o̶f̶ ̶t̶h̶e̶ ̶j̶o̶b̶ half of the result. I hope to update 2.8RSR later this week, I found and completely solved an issue causing excessive cockpit view sway. Various other things were solved. For turbo engine, I have a template that I use for my own Yellowbird physics version, just would need to add one more turbo, as I see rF2 can have two... it shouldn't be too hard to get reasonable physics for an initial build, and if needed then can try getting professional and methodical with physics spreadsheets as an evolution.

Anyway, IMO aiming for a professional methods right from beginning is inefficient, and doesn't offer any advantages other than perhaps having a better insight of how certain parameters relates. I don't know how big physics guys works, but I imagine it would be: crafting decent physics without spreadsheets (probably up to 60-80% done) > increasing accuracy and quality of physics with spreadsheet (I assume up to 95% done)> and finally fine tune physics without spreadsheets (get over 95% done). I wouldn't be surprised if some high tier physics guys (official developers) does everything in spreadsheets and refrain from any artistic observational decisions, and constrain themselves with data and general rules, which I think is bad idea, inevitably there has to be some decisions done based on common sense, logic and observations.

Either way, good luck with your project, it inspires me, I might try something similar. Car looks great :)

And the way car drives IRL is amazing, I am sure everyone has seen it alreadu, but worth having it still:
 
Last edited:
mantasisg said:
Anyway, IMO aiming for a professional methods right from beginning is inefficient, and doesn't offer any advantages other than perhaps having a better insight of how certain parameters relates.
Click to expand...

A large part of the physics spreadsheet is understanding what parameters you have good data for vs. what you must guesstimate. Looking at the older ISI cars is useful, even from rF1.
 
@mantasisg thx for your advise.
I have tried for a long time to "guess" a chassis setup and tires, but didnt get where I wanted to go, so now I'm trying to use the spreadsheet for chassis, spring & dampers etc and maybe more if possible. I think, I also have learned that tires comes after the chassis, damper etc. so all that time spend on tires in the beginning, I probably have to do it over again.
 
If you have got reasonable tire already, it shouldn't need major tweaks again. If they do, then the tire is not that good yet.

Certainly chassis and aerodynamics comes first, tires might be after that, but not before suspension stuff.. Usually I think the tire would be something that is absolute first thing in vehicle design, thus should be in simulation design as well, unless you have knowledge of a car that is so much better than tire, that you are basically reverse engineering into the tire. Even then you must have great sense of isolating things that tire does by itself, from things that are majorly influenced by chassis and aerodynamics.... And you can only do that when you know if some particular thing can't be at fault.

The things that are usually alterable in setups is something that makes chassis, aerodynamics and tires a happy singular community. And takes care, so that they would remain friendly at every scenario.

Even some major changes in chassis and aero, probably shouldn't wreck the rest of the physics. Even recalculating subbodies, massively changing some of geometry, stiffness. Altering aerodynamics. Nothing should entirely wreck the rest of the physics if they are fundamentally sound. It should be possible to get away with minor rebalancing, something that would usually be available or almost available to do in garage setup. Unless you redo things so much, that fundamentally it is entirely different car.

Anyway, do what feels best for you. And you should eventually discover how much good discovery emotions rFactor2 provides as this whole puzzle starts showing a sensible picture for you.
 
thx for your feedback, I will look into the dead spot center. I have tested this car for so many hours that I probably have gotten too used to how it acts that something like this have gotten under my radar. Will test it.
 
I gently disagree, it may be different to the Porsche style we´re used to but it has its own beauty. The rear lines around the narrow rear wing are beautiful, look at the side diffusers. I personally love it.
 
v.0.72.0 - tire improvement update

Changes
* Adjusted tires some more, mainly in tbc
* Added some turbo lag
* Added incar adjustable front antiswaybar
* Fixed small left headlight flicker
* Added rain tires, might need more adjustment for dry condition.
* Fixed IBL tire shader, visual more realistic (quick fix)
* Fixed wiper mesh, removed transparent from inside
* Redone dirt on inside windows on all skins
* Fixed racing team names
* Fixed typo in group name
* Added more groupnames GRP5_LMxx GRP5_IMSAxx GRP5_DRMxx
* Changed default FFB to level "18"
* Changed digital speedo to default, and "Classic no speedo" as an option
* Added #11 DRM 1980 Team Willeme.
* Added #50 IMSA 1983 Kathy Rude.
* Added more details to rpm dial and added curves dial glas with scratches.
* Added "max" needles to rpm and boost dial.
* Adjusted engine life to start increase wear at 7500 rpm. Recomendation, sprint race max 8500rpm, endurance max 7500rpm.

steamlink in first post
 
You must log in or register to reply here.
Back
Top