1.7 TMV10 Issue

burrito · Jul 16, 2016

@kunos

TMV10 is a nice step forward in terms of the new SAT code (Much better FFB in open wheel cars) and the tyre radius growth. However I have to report a small bug (I am not sure if it was intentional), the slightly abstract input method of a lookup table for dx and dy mu (grip) that looks like this;

DY_CURVE(0=2.25|1000=2.0597|2000=1.8775|3000=1.705|4000=1.5438... etc
DX_CURVE(0=2.23|1000=2.0283|2000=1.8415|3000=1.6702|4000=1.5147.... etc

that was originally added in 1.5 (I think), has stopped working. Changing the header version to 10 and hopping in-game creates a scenario where the tyres have zero mu for dx as seen here (the same happens for dy);

I am hoping this was not done on purpose so new effects could be added and that the lines of code can be re-instated reasonably easily as it provided a much higher degree of control of mu against tire load especially for cars with high downforce.

P.s sorry for curved VR screen capture

Chris Kennedy · Jul 18, 2016

If you didnt catch the stream, stefano said they didnt implement it yet in v10 because they dont personally use it, but it sounded like he might

burrito · Jul 18, 2016

No. Didn't have time, thanks for that. Hopefully they do add it in one of the hotfixes. The same kind of thing happened before with some other variables and that was fixed reasonably quickly so I was hoping it was the same in this case.

The few of us that do make use of it would be greatly appreciative

mclarenf1papa · Jul 19, 2016

burrito said:
@kunos

TMV10 is a nice step forward in terms of the new SAT code (Much better FFB in open wheel cars) and the tyre radius growth. However I have to report a small bug (I am not sure if it was intentional), the slightly abstract input method of a lookup table for dx and dy mu (grip) that looks like this;

DY_CURVE(0=2.25|1000=2.0597|2000=1.8775|3000=1.705|4000=1.5438... etc
DX_CURVE(0=2.23|1000=2.0283|2000=1.8415|3000=1.6702|4000=1.5147.... etc

that was originally added in 1.5 (I think), has stopped working. Changing the header version to 10 and hopping in-game creates a scenario where the tyres have zero mu for dx as seen here (the same happens for dy);

I am hoping this was not done on purpose so new effects could be added and that the lines of code can be re-instated reasonably easily as it provided a much higher degree of control of mu against tire load especially for cars with high downforce.

P.s sorry for curved VR screen capture

Does the .lut style of input work (the only cars I've tested V10 on are ones with Kunos formulas, but I have others with .lut style lookups), or are both the in-file and out-of-file inputs gone atm for V10?

burrito · Jul 19, 2016

The lut style physically doesn't work. However you are correct. All cars on TMV10, regardless of method of mu calculation, output 0 to the telemetry app (I wasn't aware of this when I posted the picture and thread). But is has been confirmed by the devs that the line method is not available in TMV10 currently.

PhilS13 · Jul 19, 2016

Has the tyre app returned in 1.7 ? iirc it was supposed to be back. You might get Dy,Dx there and much more

Qazdar Karim · Jul 19, 2016

No in-game reads DX/DY values on TMV10, i hope someone reports it to Kunos ...

And also (a bit OT) :GRAPHICS_PITCH_ROTATION doesn't seem to work anymore, at least on my side

Chris Kennedy · Jul 19, 2016

Qazdar Karim said:
No in-game reads DX/DY values on TMV10, i hope someone reports it to Kunos ...

And also (a bit OT) :GRAPHICS_PITCH_ROTATION doesn't seem to work anymore, at least on my side

graphics pitch thing is reported and should be fixed next patch.

can you clarify that both the inline LUT method AND referencing a LUT file are not working? if so i can make a thread on the ac forum

Qazdar Karim · Jul 19, 2016

On TMV10, DX/DY can't be read by any app apparently, it's just reported as 0 ... the LUT method is not working on TMV10, but it's the same issue with the other methods.

Another OT (sorry): there is definetely something wrong with the dampers, on the MP4/4 mod i'm working on, current values cause the car to bounce weirdly, lowering values solves it, but they used to work fine before 1.7.

mclarenf1papa · Jul 19, 2016

Qazdar Karim said:
On TMV10, DX/DY can't be read by any app apparently, it's just reported as 0 ... the LUT method is not working on TMV10, but it's the same issue with the other methods.

Another OT (sorry): there is definetely something wrong with the dampers, on the MP4/4 mod i'm working on, current values cause the car to bounce weirdly, lowering values solves it, but they used to work fine before 1.7.

Might make sense...Stefano said TM10 is a different section of code, so it might just use different outputs (i.e. the apps to use the <TM7 values, which since they aren't used, the apps return "0").

Avo77 · Jul 19, 2016

From my tests no method ever worked as intended from 1.3 regarding mu vs load variation.
I sent some telemetries to Aris but never received a precise answer.

He is probably still trying to find some free time to check.