RailDriver problems with some new assets

[Machinist]

I am posting this brief report to draw attention to this problem that has emerged lately and could worsen.

RailDriver levers are not working well with some new assets. I did some tests with old and new locos for comparison purposes.

* CLASS 450 (eletric) from Portsmouth Direct Line - RW2 Ready:
No problem

* CLASS 455/9 (eletric) from Portsmouth Direct Line - RW2 Ready:
No problem

* CLASS 04 Shunter (diesel) From Hatchet Hill Quarry - TS2012 Compatible:
No problem

* ICE (eletric) - TS2012 Ready:
No problem, in fact a minor problem: moving RD lever doesn't move in-cab lever, but anyway brake is working fine.

* CLASS 421 - 4CIG (eletric) - TS2012 Ready:
Only Reverse lever is working. Throttle and Auto Brake (train) aren't.

* P42DC Genesis (diesel) - TS2012 Compatible:
Reverser, Throttle and Independent Brake levers are working. Only Auto Brake (train) isn't.

* EMD AEM-7 (eletric) from NEC - TS2012 Compatible:
Reverser and Independent Brake levers are working. Throttle and Auto Brake (train) aren't.

It doesn't seem to be a problem with F4 Virtual Console. Most of recent assets have problem (except ICE and CLASS 04). I notice that AEM-7 has changed the keyboards for Engine* Brake: "[" apply and "]" release, which is the opposite of normal commands: "]" apply and "[" release. May other creators content made changes like that, or changed keyboard related usual commands? Maybe a clue...

I am reporting this to RailDriver Support, but would be interesting creators have a look on these problems, and work it out, especially if new problems are being caused by improper command assignment in conflict, or not in accordance, with the standards established.

I know RSC crew is very busy with new release, but....

Thanks in advance.

* Edited later, due to a mistake.

[Kali]

It is because Raildriver doesn't work with scripted controls properly, it tries to bypass them and use the "raw" direct parts of RW that live "beneath" the script layer. I've already posted about it here. The ones you've listed include some quite complicated scripted physics, especially the CIG. You will find lots of other stock, like the 86, the 455/8s, the EPB-based units also probably won't work, and some of those have been around ages.

As someone who spends quite some time scripting engine behavior I'd really like it fixed; RD just has to duplicate the F4 controls ( and give us a reverser script authors can virtualise ) and everything would be fine again. Until then it's just going to get worse, I'm afraid. I will compromise to an extent to keep support, but if I ever had to program an engine that did some odd things, like say a camshaft-controlled resistance controller ( quite common in older electric stock ) or a tapchanger, then I'm afraid given a choice between simulating it well, and keeping raildriver support and not being able to simulate it, then the simulation is going to win.

The train brake is by default the ; and ' keys, btw, [] are the default loco brake keys.

Actually now I look at the other post, you've already read that, heh. Still, comments still stand, it is a raildriver interface problem more than anything. It would be helpful to know who put the RD support into RW.

[Machinist]

Hi Kali,

Thank you again, for your explanatiom. I open this new thread in order to not miss that post among so many subjects.

I tried what you suggested (to program manually RD console by pressing keyboards) but I had forgot the levers are not programable, only other switchers and all the blue buttons. So it didn't work. I can run around programming the "A" "D" "," ";" "[]" keys to the blue buttons, but it's like use keyboard, and therefore the imersion in simulation goes into the drain (IMHO)...

Now that you explained very well what the problem is (especial script for especial engine), I can only appeal to RD support and hope they can solve this. Of course more realistic physics simulation had to win the battle. But I am very sad with this, and really afraid that the things become worse.

About "[]" was my fault, I meant Loco (independent) Brake (is already edited in original post). But why, for the AEM-7, RSC has inverted the command in the opposite way of default?

Well, what I'm seeing is like entering in a very long tunnel, I hope there will be a light at its end.

Let's see.

Regards.

[Kali]

I think it also needs some support from the RSC end, I'll have to explain a couple of things about RW first. I don't know if you're at all familiar with basic computer programming terminology, so I'll try and explain from the ground up. For anyone who does know RW internals, I might bend things slightly for illustration, so don't get too irate :P

RW has to store the value of what the driving controls are set to somewhere; it does this by storing the number in what it calls - somewhat confusingly for us here - a control. I will refer to it in bold when I'm talking about the RW storage space.

I'll take the throttle as an example: in a very basic RW setup the throttle control is called "Regulator" - this is what Raildriver accesses to control the sim, and by default what the big red lever on the F4 panel also accesses. However, if I make a new control myself called "VirtualThrottle", by magic the F4 red lever starts changing this instead. Because I've made it myself though, it doesn't actually do anything to the game unless I tell it to, and this is where the script comes in. Basically what you do in a script is read the value of VirtualThrottle, process that somehow ( maybe varying it depending on speed or wheelslip or something ) and then set Regulator with the final result. In turn the game internals read Regulator and move the engine.

There is a similar relationship between "TrainBrakeControl" and "VirtualBrake" which is why you found some brakes not working either. There is no such pre-defined thing as "VirtualReverser" though, but scripters are free to make up any control they want and bind keystrokes and bits of the 3d cab to it, so you will find us doing that too. Unless someone provides us with a predefined magic control for the reverser, it's unlikely to work with either RD or the F4 hud.

Now, it's not quite as simple as saying "oh, instead of RailDriver affecting 'Regulator', let's make it work with 'VirtualThrottle'" and so on, because not every engine has 'VirtualThrottle'. RSC recognised that when they made the little bit of magic with the red lever; they would need to give whoever sorts the RD interface access to that code in some form, and of course that means they need to schedule a little dev time - even if it's just to check something isn't broken - and I suspect that's in rather high demand right now. You might get lucky and Paul will read this and offer bonus biscuits, never know :)

[Machinist]

Around 2006/2007 I started to create some scenarios, using a especial tool (not in game) that directly manipulated these basic commands, so I could understand what you said.

As you can see in my short report, Reverser was never affected in RD yet, it still works indeed. Independent (engine) Brake wasn't affected also. As you very well explained again, problem is for sure due to new "Regulator" and "TrainBrake" controls, related to virtual console (F4), or something alike.

In RD site, they have a free tool called RailDriver® Software Development Kit - SDK, where they say: "rddevkit.zip (2.3 MB) for Microsoft Visual Basic 6.0, Visual Basic Microsoft Visual Studio .NET 2003, C++ Microsoft Visual Studio .NET 2003, Microsoft Visual C++ v6.0, Borland C++ Builder v6.0. Our P.I. Engineering Software Development Kit (SDK) contains code examples to help developers integrate applications with the RailDriver Desktop Train Cab Controller. Please visit the Developer's Page on piengineering.com for a free download of our SDK".
http://www.piengineering.com/developer/splatsdk.php

May be missing more information and a better interaction, for integration purposes, between artists (creators), RSC and PI Engineering.

Thanks again for your detailed and patient explanations.