Sherman Hill Car Weights

[TDHenderson]

I'll be happy to test.

Trevor

[GaryG]

Hi

Another big item that affects train handling is the Buckeye couplers definition. Kali posted a fix a year or so ago that leaves some slack but removes all the excessive stretch (coupler drawbars pulling a foot or two out of the pocket when under heavy pulling forces). Yet another set of files to replace after updates.

I would be pleased to test your car physics settings.

GaryG

[dfcfu342]

Latest update for everyone: I've run three scenarios with my updated physics and 2 out of the 3 ran to completion.

1) Introduction to Sherman Hill: Ran to completion with no issues.
2)Granite via Wycon: Ran to completion with no issues.
3)Arrival of the Thunderheads: Crashes as soon as the scenario starts due to "multiple problems" with the player train. I have no idea how to troubleshoot scenarios so I am of no use correcting that.

I'm releasing the files into public beta but before you proceed to download and use them I have 3 warnings:

1) I am in no way responsible for what the files do to your computer. Please run a virus scan on the folder before you even open it (You never know what its picked up along the way).
2) There is a possibility that this may break ALL scenarios using these cars with the stock .bin files that you already have installed.
3) Before you do anything, backup your Assets/RSC/ShermanHill/Railvehicles folder in a safe place! You may want to revert back to the original .bin files at a later date.

With that out of the way now to some better physics! I don't have any fancy installers or drag-and-drop thingys so you will have to manually copy paste all of the .bin files into their respective folders. Once you open the RSC/ShermanHill/RailVehicles/Freight folder open each car and drop the updated .bin file into the Default--->Wagon folder.

As an example the updated Autorack.bin file would be dropped into:

C:\Program Files (x86)\Steam\steamapps\common\railworks\Assets\RSC\ShermanHill\RailVehicles\Freight\AutoRackCar\Default\Wagon

When you are prompted to overwrite the file say yes (You backed up the entire RailVehicles folder first right?).

If you do not feel comfortable altering important files please do not proceed any farther. If you are then you can download the updated .bin files here: https://docs.google.com/open?id=0Bzrx0SnzP2QRdERqcWNPWkF4Z3M

Start Train (slightly better) Simulator 2013, clear the blueprint cache, and enjoy the Sherman Hill with some more realistic train workings.

Please provide feedback and any errors/strange things you notice when using the new files. Once any bugs are worked out I will see about adding them to the library with a proper install procedure.

Have fun and be prepared to slide right through your first few red signals

[GaryG]

Hi

The storm scenario crashing at start is definitely being caused by your replacement file. Don't know what the cause is but replacing with the original file allows that scenario to start.

GaryG

For clarity, by "replacement file" I mean the autorack file only. As that's the only file used in that consist, I only replaced that one file.

[dfcfu342]

Hi

The storm scenario crashing at start is definitely being caused by your replacement file. Don't know what the cause is but replacing with the original file allows that scenario to start.

GaryG

For clarity, by "replacement file" I mean the autorack file only. As that's the only file used in that consist, I only replaced that one file.

I know it is because of my file but I don't know enough about scenarios to determine what is causing the crash and to potentially come up with a fix.

If anyone is good with debugging scenarios it would be a great help!

[GaryG]

Granite via Wycon - I like the feel of the train handling. The train actually felt heavy and I also noticed that the AI trains appeared to accelerate from a stop at a more realistic rate.

Tomorrow (Sunday) I'll take a peek at the differences in the auto cars and see if I can get an idea of the cause. It could be a problem due to the number of that type of cars in the train and a sum of the new mass/weight values causing an overflow when a consist is set up for a scenario.

GaryG

[BNSFdude]

Once I get internet again, I can send you my changes to the locomotives on Sherman for the brake controls.

[dfcfu342]

Once I get internet again, I can send you my changes to the locomotives on Sherman for the brake controls.

What are your changes? Right now I'm using the ones included with RW Tools and I find them to work well. If yours are even better I would gladly include them as well with the final package with your permission of course.

We could create a RWA Community physics upgrade I would be more than happy to put it all together and see if PapaXpress will accept it for the file library. We could even add a few demo scenarios to show off the upgrades if anyone is good with writing scenarios.

[GaryG]

Hi

Well, a bit of modding of your changes shows it is the "MaxForcePercentOfVehicleWeight" value causing the problem.

With the other three changes at your new setting, and "MaxForcePercentOfVehicleWeight" set to 50, the scenario starts but if at 40 I see the failure, still a long way from the wanted 13.

The strange bit is that with that value set to 12 for the boxcar, there's no problem.

I'll continue digging...

GaryG

[GaryG]

I am starting to think you may have found a programming logic error in the program...

This train is made up entirely of autoracks. If I decreased the car Mass to 50, I could get the scenario to run with the "MaxForcePercentOfVehicleWeight" set at 42 but if I had the Mass at 80, the scenario startup fails with 42 for the percent.

I am guessing that the program's consist force calculator might be summing some values and in this scenario we might be causing an overflow and the scenario startup fails. If this is the case, we have hit another brick wall at least with this method of trying to achieve more accurate train handling.

I know other scenarios will run with these new settings for the autorack but there aren't as many of that type of car in those consists. Perhaps someone could start modding the failing consist by removing cars to see how many autoracks could be used in a consist with these new settings.

For now, I will use the 43% which is somewhat better than the original 70%

This is all the testing I can do today...

GaryG

[dfcfu342]

Hi

Well, a bit of modding of your changes shows it is the "MaxForcePercentOfVehicleWeight" value causing the problem.

With the other three changes at your new setting, and "MaxForcePercentOfVehicleWeight" set to 50, the scenario starts but if at 40 I see the failure, still a long way from the wanted 13.

The strange bit is that with that value set to 12 for the boxcar, there's no problem.

I'll continue digging...

GaryG

That is interesting that it only effects the auto rack car. All of the cars are set to between 12-13% of the loaded weight of the car but is based on the unloaded weight of the car (If that makes sense) which is why the loadable cars have significantly higher values for the maxforce.

I appreciate your looking into what exactly is causing the problem but I'm still curious as to exactly what the problem is. I tried setting the auto rack weight back to its "RSC" weight and it made no difference so it does appear to be tied entirely to the maxforcepercentofweight value.

[dfcfu342]

I am starting to think you may have found a programming logic error in the program...

This train is made up entirely of autoracks. If I decreased the car Mass to 50, I could get the scenario to run with the "MaxForcePercentOfVehicleWeight" set at 42 but if I had the Mass at 80, the scenario startup fails with 42 for the percent.

I am guessing that the program's consist force calculator might be summing some values and in this scenario we might be causing an overflow and the scenario startup fails. If this is the case, we have hit another brick wall at least with this method of trying to achieve more accurate train handling.

I know other scenarios will run with these new settings for the autorack but there aren't as many of that type of car in those consists. Perhaps someone could start modding the failing consist by removing cars to see how many autoracks could be used in a consist with these new settings.

For now, I will use the 43% which is somewhat better than the original 70%

This is all the testing I can do today...

GaryG

Interesting.

"Arrival of the Thunderheads" uses a train with 37 auto rack cars in the consist. I created a free-roam scenario with a train of 50 auto rack cars to try and force the overflow again and it starts just fine and operates correctly. Do the standard and free-roam scenarios load the consists differently?

[Kali]

Published figures from various US rail manufacturers give about 14% of vehicle weight as the top end for brake force. The issue with RW arrives when you have a loadable vehicle, you have to decide if you want it braking correctly when it's loaded or empty because the load mass isn't taken into account in the brakeforce figures. I've never managed to break the game by putting strange physics values in - it worked fine accelerating & braking from 2000mph, even - but it's quite easy to screw up blueprints even when they look ok.

I would be careful when you say "These real life resistance figures say this, and RW has something else!" because you might well not be looking at the same resistance calculation. There are tons more factors given RW does not have a very sophisticated physics model anyway. Brakeforce seems ok because it would seem the physics engine is doing a reasonable attempt at modelling real values, but even then ( clasp ) brakes work differently at different speeds, mostly due to the difficulty of shedding heat fast enough at higher speed.

[dfcfu342]

Published figures from various US rail manufacturers give about 14% of vehicle weight as the top end for brake force. The issue with RW arrives when you have a loadable vehicle, you have to decide if you want it braking correctly when it's loaded or empty because the load mass isn't taken into account in the brakeforce figures. I've never managed to break the game by putting strange physics values in - it worked fine accelerating & braking from 2000mph, even - but it's quite easy to screw up blueprints even when they look ok.

I would be careful when you say "These real life resistance figures say this, and RW has something else!" because you might well not be looking at the same resistance calculation. There are tons more factors given RW does not have a very sophisticated physics model anyway. Brakeforce seems ok because it would seem the physics engine is doing a reasonable attempt at modelling real values, but even then ( clasp ) brakes work differently at different speeds, mostly due to the difficulty of shedding heat fast enough at higher speed.

I set all the vehicles to brake properly when loaded because I assumed that's the way most people run their trains. I'm most likely going to end up making both loaded and empty cars in the end for the utmost realism for all situations.

As far as using the correct values I can assure you I checked and double checked everything with the developer's wiki and the rolling friction coefficient is used correctly and inputted correctly in the new bin's.

What interests me is that the auto racks only have a problem on the player train in the one scenario. They work perfectly everywhere else in the game on both player and AI trains which is what leads to the conclusion that something in the game engine is having an issue in that scenario, not that the blueprint has been altered improperly. On top of that I altered all of the blueprints in the same manner and in the same order and all of the others came out perfectly and work wonderfully in game. So something specific in that scenario is not adding up and GaryG seems to have narrowed it down to the "MaxForcePercentOfWeight" value which seems like a strange thing to cause a scenario to fail on start-up.