Train Simulator Classic – Core Update & Route Refresh

How do consist IDs end up the same anyway? Is this only an issue with QDs?

 

A Source ( & Solution) Of Some Crash To Desktops
This is the link to the above article which is excellent.
https://forums.dovetailgames.com/threads/a-source-solution-of-some-crash-to-desktops.63153/

 

In my case they end up the same because I find it quicker and easier to clone QD consists and (mostly) haven't bothered renumbering the consistIds (my bad as they say). I tried using the inbuilt consist editor once and then gave up on it. Directly editing them works far better for me.

 

Progress report for QD OOMs. Having deduplicated the QD fragments I have now tested the first 19 QD scenarios - these range from small (50 consists) to quite large (280+ consists) and I have had zero OOMs. Before making this change I used to close down all other running apps, kill stray tasks in task mangler etc just to minimise other memory use, in the hope that this might help TS to run cleanly (but still got OOMs). Now, I'm running Youtube alongside plus other apps and the game is (so far) solid as a rock. Thanks Richard!

 

Update#2 - so an hour after I posted I got another OOM. However I haven't (yet) completed editing all of the QD consists. Overall, my quickdrives are much more stable. Once I have completed all the edits I will try another set of tests and then release my (route-specific) preload file for people who use my route.
Please be aware - this fixes a particular problem with QD consists for some people (like me) who have been a bit careless in cloning QD consists in the past. It doesn't fix any of the other causes for OOMs.

 

> who have been a bit careless in cloning QD consists in the past

Wouldn't a fresh and clean insgtall fix that ?

 

The structures maybe more solid, some cars on the street “sink” into the road, and some passengers on the platform go through the walls and pillars.

 

Not in this case. The proximate cause (may) have been my habit of using RWTools to clone QD consists to create new ones -without amending the ConsistIds. This resulted in multiple duplicate ids. It didn't 'seem' to be a problem - however (following Richard Maxted's post) by making them all unique i appear to have vastly improved the QD situation (for my route).

 

I find this really handy in these situations > https://www.guidgenerator.com/online-guid-generator.aspx

 

Nice tool Reef - in this case the ConsistIds are only an 8-digit number so I've gone old-school and created a set of rising sequences (8 apart).
At the moment I still get the occasional OOM - though I still have a few more to do.

 

Now I have got used to it I do find that using the .xml mapping in excel to extract GUIDs from .xml files and then getting excel to generate and random set of codes that it checks against the GUID's I already have listed to also work quite well. Its a faff to set up but it does make sure I never duplicate a GUID. The boring bit is serzing all my consists so they can be read into excel but once that bits done its really easy to spot duplicates and generate completely unused GUIDs

 

DTG Jamie

Fix for this coming in next update?

 

What is the problem ?

 

Page 2 of the QD AI spawnpoint is inaccesible. The bug has always been there, seems to happen when the memoryusage of the scenario tracks.bin exceeds 4gb, but after the update it is even worse.

DTG has acknowledged the bug on this forum, and since the update making quick drive content nearly impossible, i just want to know when i can get on doing what i do in TSC. Make seriously advanced Quick Drives, with random signalling etc….

 

Also the schedule and the time of arrive at station seems two thing, how the game engine count the time?