I finished this epic map in 2009. It did take me a week! As I said, I added some extra restrictions on my play. One of the main ones being that I was going for LTD speed goals. As a consequence of this I was a freight only railway up until 1960. Except for the first decade or so, my LTD average freight speed was always 23mph up till that point. The other thing was that I looked in the editor to see which steamers I had to play with. In doing so I saw and fixed the bugs in the events that were displayed in Event Validation. This meant that the 2nd World War cost me an extra $20M and the Korean War events fired.
Because I was operating on low revenues and the combined $50M hit from the 2nd World War was too much, interest payments completely wiping out any potential profits and then skyrocketing towards the huge payments just before the aid event that resets cash thus wiping out all that debt. The Korean War only added a tiny bit more to my mountain of debt at that stage. If someone is wondering why I didn't get fired from my company, that's because I had started the company with 100% ownership to give a slower start (no stock screen in this game).
In 1960 I used all the aid cash to start replacing all my tired old engines (mainly P8s) with the Norfold Western Class J. I only used steam in this game. The shiny new J Class was fast enough that I could run them as mixed consists and achieve 43mph with 8 cars on almost all runs, more than enough for the average express speed goal. From that point on I didn't buy any other engine type except for two Red Devil's that I used for a long run from Fukang to Yumen. The better fuel economy of the Red Devil is offset by the higher replacement cost. On that run I figured that it was too close to call, maybe the Red Devil still wasn't more efficient but it was close enough that it didn't bother me. Here I figured on replacement cost being at least 10% of purchase price, most replacements were done at 10 years, near the end of the game I had around 213 J Class engines worth $45M. So each year $4.5M of company profits were going towards this.
Because I never built any industry, some cargoes never showed up on the map. For example, there was never a single load of Paper on the map at any time because no Paper Mills ever showed up. In the TM enviroment and especially with dense farming communities it was a challenge to try to route to get consistent demand gradients.
Prior to 1971 all my track was double with stone bridges. This was necessary when using the P8 in the search for speed because the P8 is made for low value, high volume traffic. Upon switching to the J Class I about halved the engine roster, so there were far fewer potential train conflicts. I switched permantently to building single track because in 1971, bridge and track costs went up 70% everywhere except inner China (because the temporary event unwinds on the entire game instead of just inner China). $2M+ for a double-tracked stone bridge was too much. Track maintenance I calculated at approximately 8% of the book value attributed to track. Every 12.5 years, the company must pay track maintenance equal to the original cost of the track. As an illustration, a $2M bridge will cost $160k per year throughout the rest of the game. At the end of the game, annual track maintenance was up to $26M per year.
I did go after some connections based on info I got from the editor about what bonuses I would receive (better pulling power mainly). In 1971 I studied the events for a long time to work out why track costs had increased drastically. This comes from the temporary track cost reduction in inner China as part of the 1966 newspaper "Cultural Revolution!". Turns out that the -70% track cost reduction works great, however, the unwind event (Temporary Event appearing at the bottom of the event list of an in-progress game) the game automatically generates will apply to the whole game. I can see a similar thing has happened for other events as well. I can only assume that after so much gameplay this feature while not controlled has been accounted for in the design. Some fiddling with the triggers could likely "fix" this to a works-as-advertised status.
At first glance, the terrain on this map is extremely rough (not talking about number of mountains, just jagged edges) and I wasn't really looking forward to entering the mountains, because of the events situation and particularly the 1971 unwind, many of the mountain routes ended up being "cheap" because they were in inner China when compared to the cost of track elsewhere. As the years passed I did try to put some design into my mountain routes.
I don't know much about the history and I hope that my play with restrictions was within the spirit you intended for your game, otherwise I would have been swimming in cash and long ago the computer would have bogged me to a standstill. I had a good challenge and a few moments of doubt that I would ever make the Trainmaster target. Thanks! Also, for your patience before I got time to attempt such a monster map.
Some screenshots:
Notes: The first shot is from when I had just connected 230 cities, the second two after I let the game run to the end of the year. The -$138k industry profits is because I used some terrain levelers but only in the instance that building a station had badly terra-formed ground. Trying to avoid reloading a save, it worked sometimes.
PS. If you want to see what I did in more detail I can post the saved game but it's 86MB and zips to 19.7MB so I hesitated to do it.
While playing for this long I had some time to consider the game's mechanics and I am now pretty sure that the demands of the agricultural communities can never be satisfied (price goes red). A simple example of this is a regular RT3 Textile Mill. This mill is swamped with Cotton but has no supply of Wool. It is running at full production converting Cotton to Clothing. The price for Cotton is low (red), however the price for Wool is at the maximum. There is no way that the mill could porduce more even if it has a supply of Wool, but the game still calculates a full strength for that demand.
The root of this problem is that there is one figure that determines the maximum number of loads an industry can produce no matter how many conversions are used. In the case of the Textile Mill the maximum is 4 (or 8 with an upgrade). For the agri communities this figure is exactly 1. The Textile Mill only has two conversions, TM's agri communities have 12+ seperate conversions. The TM Supply Chain Guide claims that these are only demands, when in reality a look at the bca files reveals that they are actually all conversions. All these 12 conversions are limited to a combined output of 1 load per year. I believe that the majority of cargoes sitting at an agri community actually just rot away.