chromatix

The long freight trains don't work correctly with the automatic-dispatch version of the scenario. This is probably because the tail of the train is still occupying a section at the moment the script tries to set one of the routes to it.

The radio splitter is a new device in the locomotive, which wasn't modelled before. Since shunting is now possible, all the mechanisms associated with changing cabs had to be put in.

I think, though I could be wrong, that this is the permissible current to regenerate into the overhead line when braking. This does differ between countries, even those that nominally use very nearly the same electrification system (eg. Austria vs Germany).

OK, I think I can interpret that in terms of Absolute Block operations. If the interlock is presently set for trains towards you, you need to ask for line clear (aka "offer the train"). If accepted, the receiving station presses Poz to set the interlock direction towards them ("line clear"), analogously to semi-automatic block. Thereafter, the sending station only needs to press Po to signal "train entering section". The Ju-SDn section does not work like that, however. I tried it just now, and it looks basically broken to me. Or have I got it wrong, and it's the sending station that needs to press Poz for some reason?

There's one small point I'm still confused about, after reading the above and trying a number of different panels. Suppose you have a double-track line with manual block. Here, the Poz fields will normally remain white on the right track and red on the left, corresponding to the usual directions of travel. To send a train on the left track, obviously this must be changed, but there is no Wbl button to do it with, and the replacement Po button of course cannot be used until Poz is in the correct state. What is the correct procedure? Then, would this procedure be different in any way on a single-track line with manual block? I've noticed that the interlock is typically released after each train on single-track lines with semi-auto and automatic block. I haven't yet encountered any examples of this with manual block.

It's more that there's a fundamental difference between a "notch step" and advancing through a continuous range. In the former case, one keypress advances exactly one notch. In the latter case, the time the key is held down determines how far the control moves. A solution here would be to make the input system more independent of the graphical framerate, ie. in a separate thread so that it can monitor the precise time keys are held down for. This would also help normal keyboard players by making their control responses more consistent, so it doesn't immediately become redundant when joystick-type inputs for custom controllers are implemented.

That sounds very much as through you're running out of air pressure. Did you enable the converters and compressors? On this train you have to do that by hand.

The Steam Deck internally runs Linux on AMD hardware. So there's your access!

Lots of people run Linux with AMD GPUs, specifically avoiding Nvidia GPUs because they require the extra step of installing the proprietary driver to work at all. AMD GPUs just work out of the box because they have open-source drivers, but it seems to be with AMD GPUs that the grass is missing. It would be interesting to know how the grass is implemented. It could be that there's an Nvidia-specific feature being used (which would also be a problem on Windows, presumably), or maybe it's just an obscure but standard feature that the open-source devs have somehow missed and could use reminding about.

I can confirm that Linux "works for me".

In the UK, we also have two electrification systems (25kV AC overhead, and 750V third rail) - and historically a third (1500V DC overhead) which was retired in 1984 with the closure of the Woodhead Line. This doesn't include various metros and tramways which have their own special systems. On both the 750V and 1500V DC systems, electrical sections can be powered by multiple substations simultaneously, with most of the load of any given train effectively falling on the nearest substation, or on the two either side of its position if it's about halfway between them. In some places on the third-rail system, there are "section gaps" with the third rail removed over a distance longer than the connected shoes of any locomotive or multiple-unit in service, to ensure that the rail can be depowered without accidentally being shorted to its neighbour. 1500V locomotives routinely ran with both pantographs raised to reduce wear due to arcing, and there was one location where it was required that the three nearest substations were switched in and no other trains were operating in that area before particular trains could start away - due to extreme gradients, they were double-headed and also double-banked from behind, so a total of four locomotives were starting at once. It was apparently common for one substation to be switched out, but the section of line it normally powered would be supplied by feed-through from the neighbouring substations. Isolation switches were provided, but normally kept closed to allow this feed-through, which also made it easier for descending trains to regenerate power for ascending trains. On the 25kV system, there are neutral sections about every 15 miles (25km) where the supply is taken from a different phase of the National Grid, balancing the load. There is an insulated section between them which pantographs can run over without being lowered. Either side of this neutral section, there are APC "magnets" (they might actually be tuned inductors) beside the track, which automatically prompt the main circuit breaker to toggle off and back on as the pantograph passes over them, ensuring that no significant arc is drawn over the gap. It is considered good practice, but not strictly necessary, to reduce or cut power on approach to a neutral section, mainly to reduce the jerk to the train. There are signs which prompt this to be done early enough for the older type of tap-changer locomotives, of which some are still in service for container freight traffic. This system is designed to operate reliably at full express speeds, up to 140mph (225kph) in principle.