One example is Aston Cadet reversed where they brake downhill into the first corner. It seems they can't handle downhill braking as this happens elsewhere.
I have uploaded a WIP version of my new Lotus 49 mod in case you have time to use it (getting a skin ID was overdue anyway
) - I think this car is an ideal stress case for the AI as it is difficult to drive (in a fun way!).These cars were hard to drive in-period and the 49 has a very rear biased weight distribution plus it is on Road Supers
For me with my mods, the AI are fine (not perfect) through the lap but the problems are in the braking area (my vintage cars are high power and low grip). They tend to brake too late and too hard. So they are racing well but then they all lock-up and spin / slide into each other. Maybe this is to do with estimating grip as discussed so maybe the new physics will fix it. As an interim could the ABS be enabled for the AI?
Some suggestions from my car development thread explained here:-
- Definable upright geometry (eg. lower length & upper length)
- Roll centre heights displayed on the suspension diagram
- Full 3D positioning of inner wishbone pivot points
- Oval / round selectable per tube
- Material selection for suspension tubes and eventually texturing
- Definable points that move with the upright in special draw (not just the default three)
And here:-
- A bigger range of movement for the disc offset
- Definable upright geometry (eg. lower length & upper length)
- Roll centre heights displayed on the suspension diagram
- Full 3D positioning of inner wishbone pivot points
- Oval / round selectable per tube
- Material selection for suspension tubes and eventually texturing
- Definable points that move with the upright in special draw (not just the default three)
And here:-
- A bigger range of movement for the disc offset
New sub-object type
Suspension
Uprights
Because I can only define "half upright length", it is not possible to replicate the 49's geometry where the front upright is very "top heavy" and the upper ball joint falls outside the brake disc diameter and the lower within the "cup".
Initially I thought this would only affect the physics in camber gains which I could recover in other ways. As I learned more I realise the bigger effect is in relative roll-centre heights which are critical to handling balance.
The obvious difference is graphical where the cowling on the body needs to be lower so I can't quite get the nose shape the same. And because I wanted the same KPI of 12 degrees (to keep a reasonable scrub radius) the lower ball joint sits within the actual disc. I have tried to hide it but it is a bodge!
I thought rear end would be less of a problem but the packaging is very tight around those exhaust headers!! (TBD)
Pivots
On this and other cars the rose-joints at the inner end of the wishbones are at different widths which is not an option in LFSE.
Correct 3D positioning of all suspension pivots would obviously have some physics implications. Graphically, there is a big impact in packaging in tight areas - in this case there is a lot going on with suspension arms, drive shafts, shocks, exhausts, all in close proximity. Luckily my shocks are invisible
Tubes
Because some of the arms on the 49 are pressed and some are tubular it would be better if oval / round was selectable per tube.
It would be nice if material colours were available for the suspension arms eg. black / white / chrome or even better full textures.
Imagine if a thick tube could be added between thrust bearing and lower upright you could texture it with a spring and the whole thing would expand and compress as a simplified version of a coilover.
Additional
If you could define points in special draw that move with the upright then things like steering arms and anti-roll bars would be possible.
Improvement Suggestions
- (MAIN) Definable upright geometry (eg. lower length & upper length)
- Roll centre heights displayed on the suspension diagram
- Full 3D positioning of inner wishbone pivot points
- Oval / round selectable per tube
- Material selection for suspension tubes and eventually texturing
- Definable points that move with the upright in special draw (not just the default three)
Because I can only define "half upright length", it is not possible to replicate the 49's geometry where the front upright is very "top heavy" and the upper ball joint falls outside the brake disc diameter and the lower within the "cup".
Initially I thought this would only affect the physics in camber gains which I could recover in other ways. As I learned more I realise the bigger effect is in relative roll-centre heights which are critical to handling balance.
The obvious difference is graphical where the cowling on the body needs to be lower so I can't quite get the nose shape the same. And because I wanted the same KPI of 12 degrees (to keep a reasonable scrub radius) the lower ball joint sits within the actual disc. I have tried to hide it but it is a bodge!
I thought rear end would be less of a problem but the packaging is very tight around those exhaust headers!! (TBD)
Pivots
On this and other cars the rose-joints at the inner end of the wishbones are at different widths which is not an option in LFSE.
Correct 3D positioning of all suspension pivots would obviously have some physics implications. Graphically, there is a big impact in packaging in tight areas - in this case there is a lot going on with suspension arms, drive shafts, shocks, exhausts, all in close proximity. Luckily my shocks are invisible
Tubes
Because some of the arms on the 49 are pressed and some are tubular it would be better if oval / round was selectable per tube.
It would be nice if material colours were available for the suspension arms eg. black / white / chrome or even better full textures.
Imagine if a thick tube could be added between thrust bearing and lower upright you could texture it with a spring and the whole thing would expand and compress as a simplified version of a coilover.
Additional
If you could define points in special draw that move with the upright then things like steering arms and anti-roll bars would be possible.
Improvement Suggestions
- (MAIN) Definable upright geometry (eg. lower length & upper length)
- Roll centre heights displayed on the suspension diagram
- Full 3D positioning of inner wishbone pivot points
- Oval / round selectable per tube
- Material selection for suspension tubes and eventually texturing
- Definable points that move with the upright in special draw (not just the default three)
Drive Shaft U/Js
The introduction of "Hub" sub-objects allowed me to define a shape to represent a rotating universal joint going into the rear uprights It is not perfect but looks good from behind when racing. Ended up as 78 triangles but I think it is worth it. Now if I had an object that rotated with the wheel but didn't move with the suspension then I could do the gearbox end. 
Improvement Suggestion
- A new type of sub-object that rotates with the wheel but doesn't move
Improvement Suggestion
- A new type of sub-object that rotates with the wheel but doesn't move
Last edited by teeembo, .
Brake Discs
One of the distinctive features of this car, and prominent for the in-car view, is the inset brake discs. I wanted to replicate the vented design (even though the drivers apparently didn't like them!).
I couldn't use the LFS editor discs because the 49's disc are inset too much (circa. 130mm front and 190mm rear) so I needed a solution.
This was before Scawen made the "Hub" sub-object available but I am glad I didn't have that option otherwise I wouldn't have found this ...
I defined the disc faces as "Brake" sub-objects which do not rotate - this may seem counter-intuitive but it means that I could use an image of a real disc with all it's reflections and shadows which would have looked wrong rotating. The illusion of rotating is provided by the ventilation holes which I defined in the wheel "Spoke" definition. The result looks really good in action!!
As I further develop the car I intend to redraw the ventilation holes as "Hub" sub-objects and also create a disc _ALP to allow a metal overlay so that the discs reflect the environment.
Improvement Suggestion
- A bigger range of movement for the disc offset (even though I didn't use them)
I couldn't use the LFS editor discs because the 49's disc are inset too much (circa. 130mm front and 190mm rear) so I needed a solution.
This was before Scawen made the "Hub" sub-object available but I am glad I didn't have that option otherwise I wouldn't have found this ...
I defined the disc faces as "Brake" sub-objects which do not rotate - this may seem counter-intuitive but it means that I could use an image of a real disc with all it's reflections and shadows which would have looked wrong rotating. The illusion of rotating is provided by the ventilation holes which I defined in the wheel "Spoke" definition. The result looks really good in action!!
As I further develop the car I intend to redraw the ventilation holes as "Hub" sub-objects and also create a disc _ALP to allow a metal overlay so that the discs reflect the environment.
Improvement Suggestion
- A bigger range of movement for the disc offset (even though I didn't use them)
F67L - A certain classic F1 car
After watching this video and then trying most of those games, physics were as unconvincing as I remembered (in different ways) so I thought LFS must be able to represent the car better.
It started with wrestling a coffin-shaped experiment around Blackwood, trying to get the right feeling. And then it snowballed ...
This is what it looks like so far.
I hope to use this thread as both a progress report and to contribute some improvement suggestions and their justifications. Hopefully at the end of it there will be a released car!
It started with wrestling a coffin-shaped experiment around Blackwood, trying to get the right feeling. And then it snowballed ...
This is what it looks like so far.
I hope to use this thread as both a progress report and to contribute some improvement suggestions and their justifications. Hopefully at the end of it there will be a released car!
Public Release
Despite there still being many things to do I have the car in a good state and am happy for public release!
The car drives well and (IMHO) looks cool
Please vote it up and I'll see you on the track!!
Cheers,
T
The car drives well and (IMHO) looks cool
Please vote it up and I'll see you on the track!!
Cheers,
T
Thanks for the comments ...
Most of these cars had a front-mounted oil cooler and the additional cooling helped reduce the size of the rear-mounted radiators. Some examples are this Matra that won the World Championship in 1974 or this Alfa Romeo that won it in '75 - both have mid-mounted water cooled 12-cylinder engines. Also this arrangement was still used on some F1 cars in period, for example this Shadow that was on the podium in Monaco in '74.
That was my thinking
Some progress
Nearly a month and I've managed to do quite a bit. It has been rewarding completely designing the car from scratch but it means that many of the design elements rely on each other so it kind-of evolves.
I wanted to do longer side intakes and tried some large NACA ducts but didn't like it. Ended-up with this version which re-uses the side cowls from the R69 - they were my least favorite part of that car but I like them here. I also started doing some shut-lines but they need some more work:-
Created some 70's-ish rear lights which look good on track:-
Overall I'm happy! I've tried to put myself in the place of a mid-1970s designer but maybe the car is more like late 70s - early 80's
Anyway I will keep going.
I wanted to do longer side intakes and tried some large NACA ducts but didn't like it. Ended-up with this version which re-uses the side cowls from the R69 - they were my least favorite part of that car but I like them here. I also started doing some shut-lines but they need some more work:-
Created some 70's-ish rear lights which look good on track:-
Overall I'm happy! I've tried to put myself in the place of a mid-1970s designer but maybe the car is more like late 70s - early 80's
Anyway I will keep going.Wessex R74SP - a mid 70's sports prototype
Winter is approaching and I have decided to do a sequel car to my first effort ...
By the mid-70s slick tyres were available for sports prototypes. The chassis had evolved for a wider track and there is more downforce.
The R74's engine bore increases taking it to 3.0 litres and it revs higher and so is now producing 446bhp @ 10238rpm.
I've chosen to model the car form scratch to my own design in LFS Editor (I'm still too lazy to learn Blender!). Here are some early shots ...
I started by deleting most of the old car and creating "channel sections" over the wheels (all I have to do is fill the gap between them right?!).
With the splitter added and a basic dashboard shape.
Finished wheel design (probably).
Front nearly done. I'm happy with the way it is evolving!
Chassis development is mostly done - it is great to drive and lap times are only slightly slower than the Formula V8. I will keep at it and post again as I continue to get more of this thing out of my head and onto the screen.
Cheers,
T
By the mid-70s slick tyres were available for sports prototypes. The chassis had evolved for a wider track and there is more downforce.
The R74's engine bore increases taking it to 3.0 litres and it revs higher and so is now producing 446bhp @ 10238rpm.
I've chosen to model the car form scratch to my own design in LFS Editor (I'm still too lazy to learn Blender!). Here are some early shots ...
I started by deleting most of the old car and creating "channel sections" over the wheels (all I have to do is fill the gap between them right?!).
With the splitter added and a basic dashboard shape.
Finished wheel design (probably).
Front nearly done. I'm happy with the way it is evolving!
Chassis development is mostly done - it is great to drive and lap times are only slightly slower than the Formula V8. I will keep at it and post again as I continue to get more of this thing out of my head and onto the screen.
Cheers,
T
Epic!! I was wondering if my car was being actually RACED? That is what it is for but I accept it is tricky to drive on the limit. I really enjoyed watching this and all of the drivers' skill.
In future I hope to improve the driving experience when the new physics arrive. And also maybe bring out the R74SP with slicks and downforce. 😃
In future I hope to improve the driving experience when the new physics arrive. And also maybe bring out the R74SP with slicks and downforce. 😃
I got this wrong...
Doubtless the physics situation contributes as well as the lack off anti-dive (and anti-squat) geometry.
But the AI are locking the FRONTS. Then when they release the brake with some steering angle already applied, the sudden grip causes the rear to come around (I think).
This is happening with other high powered mods on road tyres and so not specific to the R69 and so I cannot do anything - it makes AI racing a bit of an anticlimax when you are battling and then they just lose it.
Excellent - more power to the Devs!
Sorry for the late response and glad you enjoy the car
The CofG is 0.287m above ground with the editor stating the minimum as 0.212m. I think there are some complex things going-on with jacking and the rear roll center but it is too complicated for my tiny mind right now!
The AI seem to lock the rears under braking - I think this is a combination of the known problem with the tyres / physics (transferring too much weight) and the low engine inertia / lack of springing in the clutch simulation - they just seem to dump the clutch on the downshift while fully on the brakes and lock the rears. I could move brake bias forward but then you couldn't trail brake effectively.
Hopefully the car will come even more into its own with the new physics and also the headlights will allow good endurance racing through twilight and into the night!!
You need a paint program that supports png with alpha - my old one didn't so I downloaded GIMP for that purpose.
Thanks M3THANOL - you said it perfectly as to what I wanted for this car but hopefully also really raceable because LFS should always be about the battle!
Thanks to all for the positive comments and constructive criticism. I will feed the ideas into version 2.
Version 1 uploaded for review.
To do:-
- Interior detail improvements
- Exterior smoothing and tweaks
- LOD2 triangle reduction
- Skin support
- Setup improvements
... plus all those other things I will notice that will start to annoy me
To do:-
- Interior detail improvements
- Exterior smoothing and tweaks
- LOD2 triangle reduction
- Skin support
- Setup improvements
... plus all those other things I will notice that will start to annoy me
Most of these cars were right-hand drive and many had two seats so I have made it switchable and also have allowed a passenger (which is a crazy idea but at least you have the option
).Wessex R69SP - A vintage sports prototype
For some stupid reason I decided to do an LFS car and now I only dream in triangles …
The car I went for is a fictional late 60s style sports prototype, most closely related to the Porsches (particularly the 910/8) but also with influences from Ferraris, Alfas, etc.
I decided to do the mechanical development first so that I could wrap the body around the resulting suspension travel, etc – I think this was the best approach. I already had a 917K schematic I found online and so I measured the geometry from that and scaled it down to ninety percent for this slightly smaller car. Rear suspension has particularly short upper wishbones to fit in with the wide boxer engine and intakes, making for quite high camber gains.
The engine is a water-cooled 2.8 litre flat-8 making 374bhp @ 9300rpm. Weight is currently 644kg which makes for a fun power-to-weight ratio!
Slicks were not available in the 60s so this car has road normal or road super – I think authentic grip would be somewhere between the two.
Aero is very primitive and has a heavy rear bias which leads to aero understeer in the high speed corners if you put both ends on max. Downforce is limited to the extent that top speed is barely affected – think of it more like “aero trim”.
For the body I wanted a starting point as I had zero experience. I found a low-ish poly model "Porsche 910 Bergspyder" on Sketchfab with a CreativeCommons Attribution (BY) license which was close to the idea in my head. The model is by dagtholander so thanks to him, ref https://sketchfab.com/3d-model ... 2ba42477b9e73a4a6895104bc
There were a lot of adjustments needed to shape and proportions and I have made subtle changes all over and created many parts from scratch to hopefully make it distinct enough to be its own car.
I wanted to make more realistic lights because I think this lifts the realism and particularly with the rears you are looking at them a lot (start line, slip stream, braking manoeuvre, etc). I realised that modelling the internal reflector would not add that many triangles and would actually save on textures. Rear reflectors ended-up at 34 triangles and the bulbs are 8 each – I think it works pretty well.
There are still lots of details to improve but I hope to get version 1 uploaded soon!
…The car I went for is a fictional late 60s style sports prototype, most closely related to the Porsches (particularly the 910/8) but also with influences from Ferraris, Alfas, etc.
I decided to do the mechanical development first so that I could wrap the body around the resulting suspension travel, etc – I think this was the best approach. I already had a 917K schematic I found online and so I measured the geometry from that and scaled it down to ninety percent for this slightly smaller car. Rear suspension has particularly short upper wishbones to fit in with the wide boxer engine and intakes, making for quite high camber gains.
The engine is a water-cooled 2.8 litre flat-8 making 374bhp @ 9300rpm. Weight is currently 644kg which makes for a fun power-to-weight ratio!
Slicks were not available in the 60s so this car has road normal or road super – I think authentic grip would be somewhere between the two.
Aero is very primitive and has a heavy rear bias which leads to aero understeer in the high speed corners if you put both ends on max. Downforce is limited to the extent that top speed is barely affected – think of it more like “aero trim”.
For the body I wanted a starting point as I had zero experience. I found a low-ish poly model "Porsche 910 Bergspyder" on Sketchfab with a CreativeCommons Attribution (BY) license which was close to the idea in my head. The model is by dagtholander so thanks to him, ref https://sketchfab.com/3d-model ... 2ba42477b9e73a4a6895104bc
There were a lot of adjustments needed to shape and proportions and I have made subtle changes all over and created many parts from scratch to hopefully make it distinct enough to be its own car.
I wanted to make more realistic lights because I think this lifts the realism and particularly with the rears you are looking at them a lot (start line, slip stream, braking manoeuvre, etc). I realised that modelling the internal reflector would not add that many triangles and would actually save on textures. Rear reflectors ended-up at 34 triangles and the bulbs are 8 each – I think it works pretty well.
There are still lots of details to improve but I hope to get version 1 uploaded soon!
Last edited by teeembo, .
I have tried to record and use wavs before but it is difficult to get one bang without any other noise. Also if there is any echo it causes issues. I like the concept though and hope it can be refined.
My concern is that it seems to make the loudness on the power closer to loudness on the over run (once I've compensated with the volume knob), which is less realistic ...
Its a bug bear of mine in race sims that the loudness doesn't vary enough. I know this is difficult to achieve but LFS has always been always quite good. Obviously IRL engines are quieter when not burning fuel and race engines massively so
.I've long thought that loudness should be more closely linked to power and maybe torque (via some clever maths) but I suppose that's another discussion.
FGED GREDG RDFGDR GSFDG