[steamer_ctn]

A new patch has been added to the curve friction function.

This patch models curve friction based upon the radius of the curve, speed, and rigid wheelbase of the vehicle.

The patch can use a new ORTS value, or assume some relevant approximations for different types of rolling stock.

The attached document describes the patch, and the effects can be seen in the Force Information in the HUD.

A small bug is still present. This being that some wagons "don't have curve resistance" when going around a a curve. This appears to be due to the curve reporting function not being "robust" for every wagon. I will continue to look into it.

Cheers

Attached File(s)

•  Curve_friction.pdf (122.91K)
  Number of downloads: 553

[captain_bazza]

A couple of explanations re steam locos, please:

"ORTSRigidWheelBase ( 3in )"

What does the ( 3in ) represent? 3 inches.....that figure doesn't equate to a real world figure for wheelbase?


"WheelBase = 1.25 * (axles - 1) * DrvWheelDiameter"

What does the 1.25 represent.....and the - 1 figure? Say my loco has a 48in wheel diameter, what would the required data be
for that, for example?

Thanks.

Cheers Bazza

[steamer_ctn]

Hi Bazza,

captain_bazza, on 24 March 2014 - 01:01 PM, said:

"ORTSRigidWheelBase ( 3in )"

What does the ( 3in ) represent? 3 inches.....that figure doesn't equate to a real world figure for wheelbase?

This wasn't meant to be a realistic value, but rather to show the typical format of the statement.

Values, can be in any standard distance format, eg 1.5m, 45in, 5.5ft, and also 5ft 4in.

captain_bazza, on 24 March 2014 - 01:01 PM, said:

"WheelBase = 1.25 * (axles - 1) * DrvWheelDiameter"

What does the 1.25 represent.....and the - 1 figure? Say my loco has a 48in wheel diameter, what would the required data be
for that, for example?

This is an "internal formula that OR uses to calculate a default approximation for a steam locomotive. This is only used if the ORTSRigidWheelBase statement is not found in the wagon section of the ENG file.

The ideal would be for the player to enter the actual wheel base value from relevant plans.

Cheer

Peter

[cjakeman]

When I run the default Marias Pass activity Cutbank Grain Car Sorting, I can take of the train brake and, within a minute and without opening the throttle, I receive a warning:

"You are travelling too fast for this curve."

Is this a bug or do we need to adapt the default MSTS model?

[steamer_ctn]

Hi Chris,

cjakeman, on 07 April 2014 - 10:13 AM, said:

When I run the default Marias Pass activity Cutbank Grain Car Sorting, I can take of the train brake and, within a minute and without opening the throttle, I receive a warning:

"You are travelling too fast for this curve."

Is this a bug or do we need to adapt the default MSTS model?

Thanks for the pickup.

Yes it appears to be a bug. I will need to look into it, as the speed value seems to be high, even though the locomotive is not moving.

Thanks

Peter

[R H Steele]

http://www.elvastowe...ishes-in-x2127/

The above thread appears to have relevance to this thread. AI traffic was vanishing when rounding a curve. Problem was traced to having the "curve resistance speed dependent" box checked. When the AI vanished the activity effectively stopped because signals and switches did not clear for the player train.

Are there any "control" engines and wagons (freight) with correct physics (mass, friction, resistance, etc.) to build a control consist? I ask because it would be beneficial to run this control consist in this activity and see if it made the curve.

(another edit): I just found FCalc, so I suppose by using this I can plug in the numbers from a specific eng or wag file and determine what are the correct values. How is one with little railroading knowledge (like myself) to know if the values used in the eng or wag file are real or bogus.

[cjakeman]

R H Steele, on 08 April 2014 - 02:51 PM, said:

The above thread appears to have relevance to this thread. AI traffic was vanishing when rounding a curve. Problem was traced to having the "curve resistance speed dependent" box checked. When the AI vanished the activity effectively stopped because signals and switches did not clear for the player train.

You're right. Seems to be associated with AI.

I'm helping Steamer_stn sort it out.

[James Ross]

I am a little concerned that curve resistance is enabled by default when it's causing a few big issues (the AI disappearing and NaN in Track Monitor crash in particular). Would people mind if I changed the default to be 'off' for the moment, whilst people work to resolve these issues? (I think the feature might also be better categorised as experimental.)

[Matej Pacha]

James, I believe the first version was off by default. There has been some changes and now, because of some problems, it's reasonable to disable it by default. Some fixes are needed in the calculation...
The feature is placed within the Physics section, what is ok for me - experimental category is not necessary since all the problems are caused by unconditional division (the code doesn't check inputs for zeros = it's a bug).

Matej

[steamer_ctn]

James Ross, on 11 April 2014 - 12:40 AM, said:

I am a little concerned that curve resistance is enabled by default when it's causing a few big issues (the AI disappearing and NaN in Track Monitor crash in particular).

As far as I can tell, it is still set for default being off.