Showing posts with label SciFi. Show all posts
Showing posts with label SciFi. Show all posts

Friday, July 19, 2019

Inspiring Images

With my move complete and unpacking as done as it will get until the last of my furniture arrives, I thought I'd post some pictures that are inspiring me today.






Monday, June 24, 2019

Automating Architect of Worlds Update

Lately I've been learning a lot about Python and how it handles objects through another side project, and I'm starting to see where I should go back and rework what I have so far for Architect of Worlds. This comes at a good time, since I've gotten up to generating everything but moons, and that step is looking like it will be a bit more intensive than the text suggests - in terms of programming, that is.

So I think my next step will be to go through the original code, clean it up as much as possible and implement a cleaner data management system. I also want to break out a couple of bits into their own functions so they can be reused whenever necessary...I'm looking at you, orbital placement! Once this is done, I'll proceed to do moon placement.

I also have a couple of questions about the model being presented, since it doesn't appear capable of producing the insane number of moonlets our gas giants have (79 for Jupiter and 62 for Saturn). Mind you, in terms of storytelling, most of those moons are inconsequential, but knowing their total number is helpful, and if I know that, I can place them automatically and develop them from there. So, because my application doesn't involve a human rolling dice and interpreting results, I don't see an issue with rolling up 120+ moons no one will ever care about visiting.

Also, When you roll things up randomly, sometimes you get interesting results that spark creativity. So I'd rather roll up a bazillion moons and risk getting a GM's creative juices flowing than say, "There are a lot of them, and they probably won't be interesting enough to justify the time you'll spend rolling dice to generate them." But that's because of the medium, not a fault in Jon's system by any means.

So that's my plan. I'll keep you updated on progress. Hopefully I"ll have enough put together to use soon!

Friday, May 31, 2019

Requires Concentrate - A GURPS Blog

I recently discovered Requires Concentrate, a GURPS blog on GMing primarily science fiction (hell yeah!) and fantasy games. And while I am still poking through it as I type this, I have to say this post on what happens to stuff in vacuum already has me in love with it!

I've said it before and I'll say it again: These kind of little details are what breathe life into exotic settings, like living in space. The idea of the glues that hold door signs to walls outgassing and losing efficacy when a ship decompresses is just cool and weird. That eggs won't be bothered by what can suck a human through a far-too-small hole in a very gory way is amazing. These little things will draw your players in. Use them!

So that's it for now. Go check out Requires Concentrate and tell me what you think of it in the comments below.

Monday, May 27, 2019

More Progress on Architect of Worlds Automation

I'm happy to report that I'm ever so closer to being up to date with what Jon Zeigler has already published. I've not completed Step Eleven (yikes! that was a mess) and finished off Step Twelve, as well. That just leaves a little more basic planetary information from Step Thirteen, and I can dig into the moon stuff.

If I am able to get this done before Jon drops the next portion of the system, I'll go back and try to clean up the mess I've made everywhere I making this work. I think I can get it much cleaner than it currently is. Then I'll start working on using the data generated to create a database I can further manipulate in a program like Access.

That or I'll get back to fleshing out mechanics and making setting content. Either one will work. But being able to generate systems quickly is key to making this game work at all. I want players to have the agency to just shoot off in a random direction or ask, "What's over here?" and I'm just a couple mouse clicks from some sort of answer.

Monday, May 6, 2019

Automating Architect of Worlds Update

I feel like I haven't made much progress on my automation of Architect of Worlds project. Step Eleven has proven far more troublesome than originally anticipated, largely owing to the fact that I didn't save enough information from Step Ten - specifically how far inward the dominant gas giant migrates. I had just stored the final position for later use. Oops.

Once this was rectified, I ended up reevaluating my approach to Step Eleven to put it more in line with what Jon wrote in the first place. See, originally, I was generating all of the orbits, stocking all of the orbits, and then deciding which orbits to delete, but I realized this was just taking extra time and wasting memory. So now the Step Eleven function will run through Step Eleven then decide whether or not to place another planet, and continue from there.

There is also the matter of how I will store all of the planets' data. I am still not 100% sure about how to set up the data structure for this project in the most efficient way possible. As it stands, I have a bunch of separate objects, but I'm starting to wonder if these should go in lists in a variable in the parent or not. So, frex, System has a variable called Stars that contains a list of Star objects, and each Star object has a variable called Planets that contains a list of planet objects, and so on.

I'm not sure I like this and haven't committed to it, yet. But it looks like I will need to pass a list of planets out of Step Eleven, and future steps will process this list, one planet-object at a time. I am really hoping someone leaves some suggestions in the comments, below, because this is definitely pushing my understanding of data handling.

Monday, April 8, 2019

Automating Architect of Worlds: Update

I recently reached Step 10 - Place Dominant Gas Giant in my quest to automate Jon Zeigler's Architect of Worlds star system generation system when I read his post on how reality is impinging on his development plans. While I agree with his approach, it got me thinking about how I want to deal with this particular issues and future updates to the program.

Monday, April 1, 2019

AI, Software, and Databases in GURPS

Obligatory Abstract "Cyberspace" Artwork
I recently found this post on computers and software over at Orbital Vagaries, and thought he hit on a few things that have always really gotten to me about how GURPS Ultra-Tech addresses computers:

  1. The in-game difference between different types of AI
  2. The cost of software
  3. The effect of databases

Types of AI

I've always liked the different categories of AI, but the functional dividing line between Non-Volitional AI and Dedicated AI was always blurry to me. Christian Blouin's description of how characters might interact with each really breathes life into these differences and codifies it in my mind. For me, that is critical, because if the GM can't visualize it, how can he describe it to the players?

The Cost of Software

The software pricing in GURPS Ultra-Tech is low. Very low. By it's measure, a modern copy of Windows or MS Office should be virtually free. Do you hear Satya Nadella laughing in Seattle? Yeah. And that doesn't even begin to address more niche software like ANSYS or MATLAB. And we still haven't gotten into highly-niche or classified military hardware that is exorbitantly expensive.

While Christian is right that Traveler prices are better, but I'd argue that there should be  modifiers for software that address competition, standards, etc. This is probably as simple as having set that addresses overall competition strength - e.g., Monopoly (x5 price), Minimal Competition (x2 price), Moderate Competition (x1 price), and Highly Competitive (half price) - and another set that covers the degree of robustness and rigor required by the end user - e.g., Extreme (x5 price), High (x2 price), Moderate (x1 price), Low (half price).

Thus, military software that is expected to be extremely robust and whose author has minimal competition due to the political nature of the contract awarding process would charge 10 times as much for the software as what a similar commercial product sold to civilians who can shop around and don't require extreme robustness might pay. And if there is only one company who can meet the needs for that military software, it would run a full 25 times cost!

I might also argue for a third multiplier for availability, but that is likely rolled into any rolls to obtain software illegally from black market sources. I think I would use legality class to set a penalty to the black market roll and let the margin of success determine how much the seller gouges the buyer.

Also, because the robustness/rigor modifier does represent how reliable the software is, this should provide a bonus to the integrity of that software, should it ever come up. Any time software might be required to make something akin to a Malfunction check or roll to resist being corrupted or hacked, it should get a +2 bonus for Extreme Robustness, +1 for High Robustness, and -1 for Low Robustness.

Effects of Databases

While he doesn't specifically spell out the mechanics of it, tying a skill cap provided by a database to a database gives them a nice mechanical benefit other than, "If you don't have one, you can't use Research." I also like that they go out of date as time passes, and if I'm reading the rule correctly, I love that it is logarithmic. I do think to should be logarithmic based on skill cap, though, and not skill type. The type should set a category for the timeliness of information, and the step down should be skill-based.

Overall Thoughts

I really like what's been done here and will take some time to digest it. I think there are some tweaks and additions I'd like to make, probably including rolling what's there into what Thomas Weigel presents in Thinking Machines (Pyramid #3/37 - Tech and Toys II).

Sunday, March 10, 2019

Naming Conventions for Stars, Planets, and Moons

As you all know, I have been working on automating Jon Zeigler's Architect of Worlds system, and I've come to realize recently that I need a nice, coherent naming convention for stars, planets, and moons.

As far as I am aware, the current convention is that individual stars get names A, B, C, etc. in order of mass from most massive to least massive, and planets append a lower-case letter on that based on the size of their obit, from innermost to outermost. In this convention, Earth would be designated Sol c. The second planet around the second most massive star in a binary would be [StarName] B b. I am not aware of any standing convention for naming moons around exoplanets.

I have a couple of issues with this convention. Firstly, it lacks a means by which to address moons. I need that because I will be generating moons. Furthermore, I dislike that both the star and planet are designated by letters. This just sounds weird when said aloud. It will also invariably become the butt of jokes at some point - there are just too many combinations that will elicit a giggles (Planet B-j?). So let's take a look at some alternatives.

Immediately, alternating between letters and numbers for unnamed planets comes to mind. So the third planet around the second star would either be 2-C or B-3. I kind of like keeping the stars designated by letter, so I'm going to opt for the second in this case. Also, that number might be designated by Roman numeral or by Arabic numeral. Using the previous example, we get either B-3 or B-III. Again, I like the latter - mostly because it looks cool.

So now I have a way to deal with stars and planets, but what about moons? Again, we want to avoid repeating what just preceded it because that looks weird and sounds weirder. So the current form is <letter>-<number>. This suggests the next should be a letter. Let's consider that moons are small, so let's make this a lower case letter. Taking the first moon about B-III, we get B-III-a. This is easy to read but in text, it's a little aesthetically lacking. So let's drop that last hyphen. Now we have B-IIIa. That looks good to me.

That covers moons around planets around single stars, but what about planets that orbit entire binaries...or trinaries...or quaternaries? I suggest the letter grouping for the cluster being orbited be used in the first spot. So a planet orbiting a binary would start with 'AB', and one orbiting the second binary in a quaternary would be 'CD'. If a planet orbits an entire trinary, it'd start with 'ABC'.

With this, we can describe any planet, moon, or star in a star system, and at a glance, know the basic nature of its orbit. I think that's a good start for now.

Friday, March 8, 2019

Automating Architect of Worlds

I've been using the Architect of Worlds system as a small side project to get some extra experience doing non-numerical-solving things in Python, and I am up to Step Eight of the Star System Creation bit. So far, its going amoothly, but I think I'm reaching a point where I need to sort out how to handle all the data being generated.

I'm currently thinking that I'd like some sort of parent-child structure so moons belong to their planets, and planets belong to their primaries, which in turn belong to their systems. I've never had to do this sort of data handling before, so this will be a good learning experience.

I'm also hoping to round out the sequence by generating as much of the non-gaming-information available on a typical Wikipedia page as possible - stuff like flattening, surface area, etc. Ultimately, I'd love for this to output a database I can keep in Access and use to generate pretty system reports for players with a couple of button clicks. I have no idea if I can pull that off, but I'll definitely learn a lot trying.

Wednesday, March 6, 2019

Architect of Worlds PDFs

Today I have a quick update on Architect of Worlds, Jon Zeigler's update to the star and planet generation system from GURPS Space. He has been posting individual steps for some time now and has covered how to create star systems and planets, thus far. He also has announced plans to post the remainder of the core system this year. But this isn't new stuff.

What's new is he has created a dedicated page with PDFs of the entire process that he plans on keeping up to date. This will make using the system much easier to use. So without further ado, here is a link to the new Architect of Worlds page.

Wednesday, February 20, 2019

Architect of Worlds

If you don't know about this, you should. Jon Zeigler, author of GURPS Space has been updating his system generation...um...system to conform with modern research. He hasn't posted all of it yet - it is still in the works - but for those of you who love science fiction and need to generate worlds, it is fantastic. What's even crazier is that he has posted it on his blog for free.

The entire Architect of Worlds system to date can be found here in PDF form:
Architect of Worlds

You can find the individual steps here:


So if you love science fiction gaming, world building (literally!) or generally want to see this amazing work come to fruition, make sure you go check out his blog. I know I'll be using this creation sequence the next time I run a science fiction game.