Message Areas - "AI Traffic "cheat" on physics?"

Train-Sim.Com Message Areas

"AI Traffic "cheat" on physics?"

Email this topic to a friend
Printer-friendly version of this topic
Bookmark this topic (Registered users only)
Previous Topic | Next Topic Conferences MSTS Activity Design (Protected) Original message
sniper297 (1003 posts)
Nov-28-01, 10:08 PM (EST) "AI Traffic "cheat" on physics?" I've been playing around installing signals and testing AI traffic meets for the next version of my route, and after I got out of the flatlands into the mountains I wanted to see if the AI trains would stall on a grade or not. Surprisingly, they didn't, I expected a human engineer finessing the throttle and using the sander would make a steeper grade with the same consist, but the AI train had no problems at all. That, of course, got me into "how far does this go?" mode, so I tested with the 2 dash-9, 49 mixed car consist that I knew from previous trials was impossible to get up that 4.2% grade on the South Pass.
Just to double check my ancient memory, I drive the consist myself, 2 dash 9, 49 mixed. Most long freight default consists won't make it up this hill.
I think I can, I think I can...nope.
So now I set up an activity with the same consist as an AI train, all the defaults, starting speed zero ending speed zero.

Waiting on the helper shed track for the AI train to pass, chasing with an SD40 with no cars.
Passing Fiddletown/Combs Landing branch line cutoff, Most of the train ahead is on the grade, doing almost 30.


Middle of the train is on the steepest part of the grade now, speed dropping just below 20. I didn't make it this far with the same consist even with a running start.


Last few cars on the 4.2%, engines around muleshoe curve by this time on a 2.1 % grade, still doing 12 mph.
At the summit, crossing the bridge into 77.I'm wondering if anyone has done any really psychotic testing on this, like a GE 44-tonner pulling 300 loaded coal cars or something. If not, I'll either give it a go or refer it to Boweavel, to find out if there actually is a limit to what an AI train is capable of. Alert Edit | Reply | Reply With Quote | Top
Table of Contents

Subject Author Message Date ID RE: AI Traffic "cheat" on physics? LawrenceKellie Nov-28-01 1 RE: AI Traffic "cheat" on physics? lnghairedwizard Nov-29-01 2 RE: AI Traffic "cheat" on physics? LawrenceKellie Nov-29-01 3 RE: AI Traffic "cheat" on physics? nw_fan Dec-05-01 4 RE: AI Traffic "cheat" on physics? sniper297 Dec-06-01 5

Lobby | Topics | Previous Topic | Next Topic
Messages in this topic
LawrenceKellie (841 posts)
Nov-28-01, 10:26 PM (EST) 1. "RE: AI Traffic "cheat" on physics?" I'd say go for it. You already have the process worked out, just a matter of substituting another consist.Lawrence Remove | Alert Edit | Reply | Reply With Quote | Top
lnghairedwizard (343 posts)
Nov-29-01, 05:55 AM (EST) 2. "RE: AI Traffic "cheat" on physics?" hey jim, I appreciate you sharing that with us. BTW, did you ever make it onto the end of the train?...doesnt sound like you needed too. Ive been wanting to do some activities based on helper service. Ive know of several different ways to achieve this but I'm wondering if anyone knows if, in real life, a train must stop before helpers are put on the rear, or if helpers can pull up behind an already moving train and carefully couple on the move?Longhairedwizard Remove | Alert Edit | Reply | Reply With Quote | Top
LawrenceKellie (841 posts)
Nov-29-01, 07:26 AM (EST) 3. "RE: AI Traffic "cheat" on physics?" http://krugtales.50megs.com/ has some great articles about helper engines.Lawrence
Remove | Alert Edit | Reply | Reply With Quote | Top
nw_fan (133 posts)
Dec-05-01, 00:38 AM (EST) 4. "RE: AI Traffic "cheat" on physics?" From reading stories on the N&W Historical Society's mailing list, the engineer in the helper/pusher could see what the front engin was doing by watching the air/brake gauge. So this means the trains stopped to couple the airline. Some roads put the caboose behind the pusher, so as not to telescope the caboose frame. N&W did not, but some did.Mike
Norfolk & Western
Precision Transportation Remove | Alert Edit | Reply | Reply With Quote | Top
sniper297 (1003 posts)
Dec-06-01, 07:22 PM (EST) 5. "RE: AI Traffic "cheat" on physics?" I think a lot of that had to do with the caboose, the early ones were built by the road's carpenters, and the later ones came with a steel frame and it was an actual car rather than a toothpick-construction trailer on steel wheels. Even without the crunching possiblility the worst trouble with switching with the caboose on is bumping hard enough to knock over the conductor's coffee pot, earning his wrath, and in the earlier versions there was a pot-belly stove that would get knocked over and set the wooden crummy on fire, which tended to be a distraction for any further switching movements. Hell of a way to run a railroad.
Well, I've also noticed that the speed the AI train runs depends on the consist too, for some reason on level track the single gp38 with 17 cars runs about 30, the 2xgp38 with 35 cars crawls along at a more prototypical speed. Odd, when you think about the weight to horsepower ratio being almost the same, and it makes it kind of hard to sequence different AI trains over a route. Remove | Alert Edit | Reply | Reply With Quote | Top

Lock | Archive | Remove

Lobby | Topics | Previous Topic | Next Topic