[EricF]

I'm starting this as another topic; it's similar to the problem that Markus is reporting, but on a totally different route, and in a slightly different way, so I didn't want to throw the other post off in the wrong direction.

I'm running the Northeast Corridor 4 route in Explore mode with the default Acela Express trainset. I'm in full-screen mode. Starting at Philadelphia 30th Street Station, I can run for a minute or two until the mainline opens up near Darby, and Open Rails freezes totally. I'm not operating any controls at the moment. No errors pop up; the sounds continue normally. If I Alt-Tab out and back, I can get the mouse cursor and click, which causes the Windows message to pop up that the Open Rails Activity Runner has stopped running and I can exit. (Fast Alt-Tab switching option must be turned on, or Alt-Tab won't even work; the whole interface locks up -- no mouse, no keyboard at all.) Nothing significant appears in the log that I can see. I tried different trains and get the same result at the same location.

Then I tried something else -- I switched to windowed mode. Same route, still in Explore Mode, back in the default Acela, starting from the same location. This time, I can go far past where the freeze happened. I got all the way to Wilmington and a little beyond, and then the freeze kicked in same as before. This time, there was a bit more in the log, but I can't make much out from it if it's relevant to the freeze or not.

I'm attaching two sample logs, one from when the freeze hits early in fullscreen modeand the other from when it hits much further down the line when I'm in windowed mode. Not sure if this is more of an issue with the route or if it's something in Open Rails?

Attached File(s)

•  OpenRailsLog - Fullscreen mode.txt (8.68K)
  Number of downloads: 195
•  OpenRailsLog - Windowed mode.txt (14.3K)
  Number of downloads: 179

[Sid P.]

I just tried X2304 with the same consist on the same route (NEC4). and also got a freeze. Same symptoms.

ORTS in Windowed mode froze when approaching Highland Avenue (before Wilmington). Process Explorer showed that Runactivity.exe was still using approximately 26% of CPU (nothing else running of any significance), GPU activity was 0%. Dispatch Viewer also frozen. Sound was still running.

(Sorry, I had forgotten to turn on the log.)

Sid.

[markus_GE]

When I brought up my Freeze issue in the other thread mentioned, it was suggested to turn of the Dispatcher Window (completely, by un-checking it in the options, I think under General) and try it again... Didn´t help in my case, but maybe here?

Cheers, Markus

[EricF]

markus_GE, on 14 June 2014 - 08:24 AM, said:

When I brought up my Freeze issue in the other thread mentioned, it was suggested to turn of the Dispatcher Window (completely, by un-checking it in the options, I think under General) and try it again... Didn´t help in my case, but maybe here?

I was thinking that, too, and I already had it un-checked in the options.


Sid, like you I noticed that Open Rails was only using about 1/4 of my CPU resources if I brought up the Windows Task Manager while the frozen runactivity.exe was still up. Memory usage wasn't anything to speak of, either. I was hoping to catch some sort of runaway thread eating up CPU availability, or something that drove memory usage way up, but no such luck -- no obvious "smoking gun". Puzzling...

I'm not sure exacty where this behavior crept in; I noticed it earlier in the week before I updated to X2313 the other day, but it may have been there before. I pulled up the NEC4 to test out some of the cab display font improvements on my Acela and AEM7's and discovered it.

[Sid P.]

EricF, on 14 June 2014 - 08:39 AM, said:

Sid, like you I noticed that Open Rails was only using about 1/4 of my CPU resources ...

Eric, before the freeze ORTS was using 32-35% of CPU and 100% of GPU at about 140 FPS on my system. Version was X2304.

Sid.

[EricF]

Well, I just tried something and got interesting results. I ran the same route and consist in V 0.9, in fullscreen mode. (That should get me to the first suspected freeze point fairly quick.) I got past the point where the freeze happens, but at exactly where X2313 halts, v0.9's signalling shows the next block go red. If I proceed, I get stopped by the red signal. In X2313, the signals keep staying permissive ahead of my train, but the sim freezes at the point where the next block gets triggered. So it may have to do with how the signals are being handled.

Unfortunately, I'm no master of signal scripts and setup... :birthday: So I can't determine exactly what's going amiss...