This is a pretty good talk by Matt Leacock about Pandemic's development, and game development in general.

The thing that struck the biggest nerve with me was the notion of finding the core(s) in your game and developing those being the most efficient route, otherwise you get caught up in minute iterations forever.

This comes across as very good advice to me, and has shaped my thoughts of playing/developing "Delivery/1955" in the past few days. I had already been thinking about how a "family" version of the game would work, and this helped to convince me to strip all the unnecessary "flavor" away, at least for the moment, and get a glimpse at how the core is working. Suddenly, I created a small prototype within a short amount of time, and the core of the game can probably be played in about an hour.

The stripped down version consists of the following reductions: no shares, no likes/dislikes, all countries beginning with the same amount of cash, the same minimum wages and maximum purchases, and much less cash differential (from $1 million to $200 million rather than $1,000 to $1 billion) (since the game will hopefully have exponential amounts of cash growth, this wouldn't be as big of a difference as it might seem at first glance).

And the core of the game appears to be setting your prices and wages. So playing around with that will be first order. But this can be worked on in a much shorter time than I was anticipating (I had been thinking of trying to start a prototype in June sometime, whereas this enabled me to begin with much less difficulty/time).

This doesn't mean the rest of it is expendable or useless fluff. In fact, I can see how the game is more boring with just the basics, everyone is doing similar things because there are less routes to go. I'm just finding it useful for getting some basics down and figuring out myself where the "simple" mechanics are running into frustrations (I'm having a mental image of land below a fault's locking depth moving and that fault occasionally rupturing as a result).


Last night, I played Reading Railroad for the first time. It worked differently than I was expecting/hoping, but that is not necessarily a bad thing.

There were some major issues with the factories which was just a detail issue to balance out - Seth was thinking of capping the number of letters you can possibly draw at 6, but this still seems like too much to me. If you come up with 2 5-letter words, you're still having an excess of letters, of which you're going to build up a continuously growing pile of unless you have an expectation value of 6 letter words (which in itself requires you to get the right letters).

Tile drawing was too random by my count. It could be really neat to have every letter sitting up somewhere, with their "Scrabble values" (common letters are worth 1, more difficult letters are worth more [Qs are worth 10, I believe]). You can purchase them for some amount relative to their "Scrabble value" (so, maybe a base amount of $5, -"Scrabble value" to a minimum of 0: $5 - $1=$4 for an E, $5 - $3=$2 for a P). This would likely inflate the values of the coins such that connection fees would cost more on the order of $3 - $6 per spot, but given how much cash we were earning off the letters, this doesn't seem to be problematic.

The biggest problem I had with the game was not feeling connected to the final letters. They had no real interaction with the game itself, they were just some sort of bonus points at the end of the game (like # of tracks built in Age of Steam, albeit worth much more). I didn't like the mechanism of picking them up, as the most efficient way of doing so is simply to build a lot of sets of 2 disparate from each other and then connect them up later on (connect up one set early on so you can build more than 2 factories) - you'll nab a lot of tiles before others get a chance at them, increasing your chances of getting "good" letters.

A suggestion might be similar to the above suggestion: purchase your letters. Perhaps you get some sort of cash bonus for connecting up different cities, perhaps depending on your network size. At the end, you spend this cash to spell out as many words of "The quick brown fox jumped over lazy dogs. Reading Railroad" as possible in some fashion.

Or perhaps letters you grab from the cities are "permanent" letters that you can always use (up to your largest network). This would reduce the function of the factories, but would increase the number of tiles you're having to process at any given time. I think I like this idea best, as it integrates two disparate parts of the game best.

Well, next time...