An Experiment, Part III

After some writing and musing, it appears the third phase of this experiment is coming to a close. That doesn’t stop the experiment, merely the phase.

The experiment in question is this blog as a whole, and I’m here to discuss it’s trajectory with myself and determine my future action.

Firstly, as of writing this, absolutely no one has heard what I have to say. Not one. That doesn’t make this blog pointless, it merely makes it… lacking. I wish I had people to discuss things with! Aside from that, I quite like talking, and will probably continue to do so, but the fact of the matter currently is that this is an empty place when it comes to conversation. Sadly.

With all of that in mind, I will continue to manufacture my thoughts and ideas, but possibly more catered to myself and less to invisible viewers I don’t have. Time will tell.

I’d love your feedback. Ask me anything, really. Contradict me.

On Model Railroad Layout Design, and optimization.

Another topic I’ve been looking forward to talking about is Layout Design. It’s an interesting logical game, where you take set parameters (the size of the space you can build in) and somehow transform that into an optimized space to run trains in.

Now, there’s many styles and goals for a potential track plan: Some want maximum mainline length, some want the most operations-centric capacity, some just want a pretty scene to watch trains go by. Some mirror a prototype down to the last detail, some explore freelance environments of the possible and the fictional. Out of all these, few must be chosen to represent your idea of the “perfect” layout.

Like anyone else, I have some favorites in this motley bunch of options. I much prefer operational capacity to scenic or mainline capacity goals, and I love to explore the potential outside the real. For these goals, I have constructed some rules to help me optimize any potential layout plan I might have. If you’re in to operations or want to make a track plan that isn’t just copying a real location, I might have some tips for you!

But back to design. And more importantly, what makes the BEST design. A difficult question in any field, but this time I believe I’ve solved some cornerstone issues revolving around this very problem. First off, what makes a layout fun to operate? Beyond creature comforts like carpeting and good track conditions, what can I change to make my layout more fun and more engaging. The answer

Design for throughput to maximize operational interest.

Let me elaborate on what “Throughput” is. It’s not the number of trains that can fit on your layout per say, it’s more like the “total work” that can be accomplished in a limited amount of time.

Take for instance this helpful addition to almost any yard: A “lead” track. It serves to increase the number of cars a yard switcher and mainline trains can move by allowing them to flow past each other and not interfere. Thus, operational interest is increased due to more work being possible since the trains don’t have to wait for each other.

Here’s an example of something that decreases operational capacity: Switchbacks. A Switchback can potentially squeeze an industry into a smaller than average space, but it’s zig-zag pattern means trains must reverse up to 5 times (compared to only 1 time with a normal siding) to use it just once! Combine this with making the lead space on the switchback too short or adding an industry on it results in a double-whammy where trains can’t work the switchback in one pass, turning what would have been a simple 2-3 move operation into a 15 move nightmare. All this takes time, so while a model railroad that doesn’t have this switchback might lose some potential traffic, the railroad with a switchback will be a royal pain to operate!

Thus, a golden rule for any operating layout is formed:

Design to let cars flow on your layout as easily as possible!

The less moves to work the same number of cars, the better.

With this rule in mind, I can give you some hints and good practices for operational success:

Make sure that:

  • Yards have leads as long as the longest body track (if you’re running a switcher and other trains at the same time)
  • All passing sidings can support the largest allowed train to prevent interference.
  • All switch leads and stub allow trains to pass without splitting themselves up.

Avoid having if at all possible:

  • Areas where trains must serve the same industry twice due to lack of lead space.
  • No space for a switcher to manage their train outside of the industry tracks.
  • Positions that trains must use that block other trains from working.
  • Switchbacks, industries served by turntables, and industries that are hard to reach.
  • Inadequate space for pickups, setouts, and maneuvering in yards and industries.

If you’ve heard all this before, and still can’t seem to get it right, try this.

Assemble the longest possible train any given area could receive, and try to work it through the area. If you start seeing areas where you spend a lot of time working to get cars through a bottleneck, there’s your problem.

I hope this has been a helpful dive into my way of constructing operationally optimized layouts!

I’d love your feedback. Ask me anything, really. Contradict me.


Solving the Impossible: TF2 Layout Design

It’s been one of my many goals to find the “core” of a good TF2 map. I have so many questions, and very few clues. Here’s a list of questions and thoughts on the subject.

First off, if you’ve played TF2 for any length of time you’ve noticed that there are many, many maps. (over a hundred now, and more every update) They vary in quality, and certainly in age, but overall they have a distinct TF2-ish-ness to them. A proportion between areas, a flow of gameplay, a style unto itself. I aim to encapsulate that in formula, and demonstrate just what makes some maps good, and others great.

Often, TF2 mapping is shown to be an art. It’s complex, mysterious, and only grasped by the highest of the mapping community. It takes skill, creativity, experience, and feedback to make a good TF2 map. Sadly, this tends to be a blockade to “inquisition” involving the true nature of mapping.

I’m on a quest to find the golden rules of TF2 mapping that will point anyone on the right path to a structured, fun, and quality TF2 map. But first, we have many questions to answer:

What determines the layout of a TF2 map? What makes the areas flow from one to the next? How do you sculpt the traffic that any given area will receive? Can you even predict it? How much control do you have over where the player goes?

What makes a quality TF2 map? Merely how it looks? How it plays? Some combination? Which comes first, aesthetic or layout? Can a map be good on gameplay merits alone? Can a map be loved on aesthetics alone? How do you interpret feedback from players? How can you know what they like and don’t like?

Can a TF2 map be designed entirely from feedback? Does it need a plan, or can you just keep fixing things until it works? What is the core structure of a TF2 map, a simplified version? Or maybe it can’t be simplified, and every element, every vertex, every inch affects the overall product in a way undefinable?  How do you know what a change is going to affect? How can you make changes without knowing what you intend to accomplish?

This is a lot of questions, as you can see. A few I have answers to, but none I am sure of. Many also, I have no clue how I would solve. But here I will cover many potential “theories of the nature of TF2 maps”:

1: Design by proportionality. This theory states that TF2’s area design is made possible by the ratios of sizes between areas and between measurements, the way they connect to each other to produce an equalized flow to an objective.

2: Design by pressure. This theory gauges player “density” in areas and determines what classes like what areas. Then it attempts to mess with this “pressure” to allow all classes to shine in most, if not all area for their own reasons.

3: Design by routing. This theory tries to show players a path to their current “mini-objective”, like killing that player or getting this pickup. All these routes converge into a network where players navigate from goal to goal, whilst being able to avoid obstacles and traverse intuitively and easily.

Maybe it’s one of these, maybe it’s some other, maybe a combination, but on the whole I have no idea. TF2 mapping is a mystery to me, a mystery which I spend many hours thinking about, arguing about, and talking about. So I figured it’s perfect for this place, since all 3 I hope to occur here.

I’d love your feedback. Ask me anything, really. Contradict me.

Experiment Continued

But of what use is a “blog” if it’s not worth reading. Therefore I’m changing that, with questions as advertised, and some dry humor on top.

Firstly, the big question. Why make a blog? What do I gain as a being from typing my endless ideas into the vastness of space?

An interesting question for sure, but can that question be answered? Here’s a tangent on that very subject:

Godel’s Incompleteness Theorem. It’s a very complicated and heavy-weight mathematics proof showing something that shocked science, logic, and the academic world at the time. It seemingly comes up with a proof of the impossible: That logic itself can never be complete.

The core of this proof is as follows, without all the jargon and messy mathematics that would scare off the average reader. Godel proved that any system using Axioms (statements that are true no matter what, like X is equal to itself), will never be able to have enough axioms to prove everything. In other words, at some point you will need new rules of logic to solve new problems.

He did this in a very clever (albeit paradoxical and unconventional) way. First, he starts with any old system using Axioms of logic. Then, he poses a seemingly trick problem:

“This statement is NOT provable by the current axioms”

This question (because of logic and other reasons) can only have 2 answers: Either it’s a true statement, or it’s false. If it’s false, an immediate paradox arises: If it’s false, we had a proof for it being false. And if we had a proof from the axioms that it’s false, that also means it was TRUE! This is a contradiction, which means that it can’t possibly be false, right? This leaves us with only 1 other option. It must be true! However, this also leads to a problem: How did we ever prove it was true? If the statement is true, it’s also impossible to prove! Therefore, this statement is both TRUE and UN-PROVABLE.

But hold on a second, didn’t we just prove it’s correct? Doesn’t that mean that the statement must also be false? Not entirely. You see, we DID prove that this statement is correct, but we had to add a few new axioms to do it. We had to add the axiom that “this statement is NOT provable by the current axioms” to our list of axioms to make it work. It’s complicated, but without diving into the extremely complicated proof itself I can’t show you why this is the case.

So we solved it. This statement is unprovable, and also must be correct. We’re done, right? Here’s the genius of Godel’s proof: We can just ask the same question again! “This OTHER statement is NOT provable by the current axioms”, proving that there are an INFINITE number of unknown and unexplored axioms, meaning that logic will forever rest uncompleted.

Overall, this whole process might seem a little pointless. Why do we bother asking these questions about a certain statement being provable? Maybe it just applies to these odd, self-contradicting sentences. This, at least, was mathematician’s hope, until this too was shattered. A brilliant paper called the Paris – Harrington Theorem came along and proved, using a test model of a potential statement that needed to be proven, that certain problems in mathematics, science and logic could contain these dangerous “Godellian” sections that ultimately mean that they can’t and will never be proven by the axioms. Fear hung over everyone as debates raged over the possibility of very high-level problems being potentially unprovable, like the Twin Primes conjecture, Goldbach’s conjecture, and the Riemann Hypothesis.

Thankfully, we can expand the axioms a bit to help us with these particular examples. If we can manage to prove that these problems cannot be proven, then use our new axioms created by Godel like “statements that you can prove cannot be proven by the axioms are true”, we could potentially have a workaround to proving these great mysteries.

This is the first of hopefully many talks and ideas thrown around by me.

I’d love your feedback. Ask me anything, really. Contradict me.


An Experiment, as usual

But why do I say “As usual”. You didn’t know that. Did I? I’ve got so many questions for myself and others, I want to discuss them. This is an experiment to see if Blogging is a good way to do that.

I’ll be talking anything that’s logical. You want to talk aerodynamics? What about rocket science? The theory of everything? Social structure? Game design? These topics and many more I will be delving into on my quest for information.