[beresford]

When playing MSTS scenarios the steam loco brakes are too weak. I've been told in the past that this is because OR scales the braking parameters differently and I should edit the loco ENG files. If this is a simple multiplier, can't we have a toggle to say you are using an MSTS ENG file, or isn't it that simple?

Of course for passenger trains most of the actual braking is done by cylinders on the coaches rather than the engine.

[copperpen]

Open Rails does not yet properly support Vacuum brake operation. The brakes themselves I fine adequate in slowing and/or stopping the train, it is just the vacuum propogation and release that is far too slow.

[dennisat]

copperpen, on 11 November 2017 - 05:18 AM, said:

Open Rails does not yet properly support Vacuum brake operation. The brakes themselves I fine adequate in slowing and/or stopping the train, it is just the vacuum propogation and release that is far too slow.

For some years I've used the attached patch which for me improves vacuum brake pressure propagation.
I noticed that full pressure does not fully propagate from front to back of the train until the pressure difference between cars reaches zero. The algorithm being used, however, is exponential and theoretically cannot reach zero. Propagation just gets slower and slower the longer the train. This of course is true in real life but the current algorithm in OR just seems too slow. The basis of my patch is to say that pressure has been fully propagated between cars when the difference is less than 0.01psi. I experimented with this value and found on my system that 0.001psi hardly made any difference and 0.1 was unrealistically fast. With long trains, this improves braking response for me (and reduces cpu cycles!). However purists may (will?) object to my logic. I've also examined whether assumptions about brake cylinder and pipe volumes are reasonable. I'm not an expert but all the resources I could Google seemed to confirm these assumptions are reasonable.

Dennis


 Vacuum_Brake_Response.zip (573bytes)
Number of downloads: 277

[Smecty]

dennisat, on 11 November 2017 - 06:45 AM, said:

For some years I've used the attached patch which for me improves vacuum brake pressure propagation.
I noticed that full pressure does not fully propagate from front to back of the train until the pressure difference between cars reaches zero. The algorithm being used, however, is exponential and theoretically cannot reach zero. Propagation just gets slower and slower the longer the train. This of course is true in real life but the current algorithm in OR just seems too slow. The basis of my patch is to say that pressure has been fully propagated between cars when the difference is less than 0.01psi. I experimented with this value and found on my system that 0.001psi hardly made any difference and 0.1 was unrealistically fast. With long trains, this improves braking response for me (and reduces cpu cycles!). However purists may (will?) object to my logic. I've also examined whether assumptions about brake cylinder and pipe volumes are reasonable. I'm not an expert but all the resources I could Google seemed to confirm these assumptions are reasonable.

Dennis


Vacuum_Brake_Response.zip

Dennis,

First, thank you for addressing this problem to the point of uploading a patch!

As someone who has a keen interest in vacuum brake physics for the pre-grouping routes and trains I enjoy, I've written in the past about braking problems in ORTS with this now-abandoned form of braking.

So it is with great interest that I downloaded your patch. However, I'm stumped as to how to apply it to my ORTS installation ... does this require some sort of compiling?

Regards,

Dave

[beresford]

copperpen, on 11 November 2017 - 05:18 AM, said:

Open Rails does not yet properly support Vacuum brake operation. The brakes themselves I fine adequate in slowing and/or stopping the train, it is just the vacuum propogation and release that is far too slow.

The time taken to change the control positions is unrealistically long as well. To get from release to apply you not only have a slow progress through release but there is this big band called 'Running' which I don't recognise from engines I've driven. It takes far too long to shut the regulator. Shifting a screw reverser can take a while, but the game is about half the speed of real life.

[dennisat]

Smecty, on 11 November 2017 - 02:08 PM, said:

So it is with great interest that I downloaded your patch. However, I'm stumped as to how to apply it to my ORTS installation ... does this require some sort of compiling?

Hi,
Yes, it's a source patch so you need to have downloaded the OR source and be familiar with patching and compiling it. If you're technically minded and familiar with programming environments, the details of how to do this are on the OR website. I'm not an OR developer and thus don't have a "trusted" background, I could be trying to pass malware on to the unwary. That's why I didn't supply an executable. At least if you have the OR source and the patch you can check there isn't anything underhand going on. If there's a favourable response to the patch, I can pass it to the OR team as a suggested improvement.

beresford, on 12 November 2017 - 05:38 AM, said:

The time taken to change the control positions is unrealistically long as well. To get from release to apply you not only have a slow progress through release but there is this big band called 'Running' which I don't recognise from engines I've driven. It takes far too long to shut the regulator. Shifting a screw reverser can take a while, but the game is about half the speed of real life.

All of this can be overcome by alterations to the eng files.


Dennis

[copperpen]

beresford, on 12 November 2017 - 05:38 AM, said:

The time taken to change the control positions is unrealistically long as well. To get from release to apply you not only have a slow progress through release but there is this big band called 'Running' which I don't recognise from engines I've driven. It takes far too long to shut the regulator. Shifting a screw reverser can take a while, but the game is about half the speed of real life.

Speed of controls is something that can be changed in the eng file for brakes by specifying where the different sections take effect between 0 and 1 ( first set of figures on each token line).

The throttle can be closed fast by using CTRL and A

[Csantucci]

Here is the .dll file generated from Dennis' patch to be inserted as replacement into x.3981.

If it will be considered an improvement I can commit it as bug correction.

21/11/2017: file deleted because new file for x.3987 available

[dennisat]

Csantucci, on 13 November 2017 - 11:16 PM, said:

Here is the .dll file generated from Dennis' patch to be inserted as replacement into x.3981.
Orts.Simulation.dll.zip
If it will be considered an improvement I can commit it as bug correction.

Thank you Carlo.

Dennis

[Csantucci]

Is there any feedback about the file I uploaded deriving it from Dennis' patch? As there was a good number of people reporting problems with OR vacuum braking, it would be nice if some of them would test the proposed solution and report the results.