Books/mythology/metaphysics discussions moved to: loopingworld.com

This means that the site here won’t (usually) be updated and I’ll eventually copy all of book-related posts over there. The rest of the stuff will stay here for as long the site stays up (not planning of pulling it down for the foreseeable future).

UPDATE: I’ll sporadically still post here, but it will be for writing about roguelike development, tracking my own (lack of) progress, or other quirky gaming things.

Posted in: Uncategorized | Tagged:

Minecraft Is Not What You Think, or:
How I Learned to Stop Worrying and Love Polybenzimidazole

This was a long time coming (and I wrote a slightly shorter version on some forum a while ago).

Thanks to Q23, I’ve known Minecraft since it’s pre-beginning. It’s all very vague in my memory, but I remember I tried an early tech demo with blocks when there wasn’t even a world generation, and then something resembling current Minecraft some time later. Interesting because it started as some Dwarf Fortress-like, a project only known in certain tiny circles. But for me, personally, Minecraft has always been disappointing.

The early tech demo didn’t convince me at all. The moment I broke a block and saw the remaining still floating in the air, with no physics, left me a sour taste. And when it became the real Minecraft it still was bad for me. The game is not intuitive. This was long before the recipe book. You had to look up things on some guide, or just trying random arrangements to find a recipe. Something that could be found immediately at random, could have required someone else half an hour. And then you had to memorize them. The survival game is shallow. Once you know the intended progress the survival aspects only last about half an hour: you get some tools, crafting table, start digging a hole until you find coal, coal leads to torches, so more subterranean expansion, looking for iron. But the survival itself is caused by the active pressure of monsters during the night, and food. If you don’t want to slaughter random animals, you can start a crop farm early on. Since the yields are always greater than the uses, food is infinite, problem solved. To skip the overworld monsters you need a bed. That means you need three sheep.

This wraps up Minecraft for me. Prioritize the bed before everything. Then settle, start a crop farm, dig down. Game over. Why? Because once you have infinite food, a magical object that makes you skip the night, and infinite torches and tools… The “survival” pressure is over. It’s not that Minecraft actually stops here, but the lack of clear goals (and “demands”) mean that every time I tried playing the game and reached this stage, I got bored. I don’t care about building pretty houses, it feels trivial and meaningless. The rest of the game seems shallow and there are tons of other that offer more depth.

(All I’m writing here is not “personal” ramblings, as always I analyze my own reactions because they eventually build a point. Game design.)

…But Minecraft is hugely popular, lots of kids play it. Everyone, at every age plays it. It is different why? Because most of those kids don’t come to Minecraft like I did. In the age of youtube and everything else, they probably have seen Minecraft being played tons of times before putting their hands on it. Or, they’ve seen their friends playing it, and playing together in multiplayer. The learning process, and recipe guessing, isn’t like mine, where you are alone in front of the game. You have friends guiding and helping you, so that you move at a brisk pace. Tutorials have no place here. Instead for me the recipe-guessing was like a leap back to the age of textual adventures, when you had to try commands over and over until you found something that worked and made you progress to the next room. Or dialogue in early RPGs where you had to guess the word that would trigger a new block of text and obtain some important info or progress. This wasn’t about having “more freedom”, it was about being tightly restricted and trying to guess what the designer was thinking. It’s not gameplay informed by knowledge, it’s informed by blindness. Groping in the dark. Making it frustrating.

That was it, and then it was not. About a year ago I got into Factorio. Even in this case, thanks to the forums I’ve known Factorio for a very long time. I installed it at least three times. Every time, I played for half an hour, or slightly more, got a “feel”, and then let it be. But this was different for Minecraft. I simply thought: “Cool, this is interesting, I’ll keep it for when I have more time to focus.” Last time, about a year ago, I played for slightly longer, two, three hours. I got halfway through the third tutorial. Something clicked. It was like a magic trick, i got addicted. Factorio became everything. But, for me, everything doesn’t mean I start playing the game obsessively, but that I start READING about the game obsessively. Forums, reddit and everything else, I started absorbing “information”.

For me, the AAA industry is almost irrelevant. I thrive in game communities and mod work. I thrive in the vision of players, that can see a game well past its original intention. I came from the Commodore 64, but I always wanted to tinker with games rather than playing them. When I had my first PC I got Doom. Doom for me wasn’t about playing the game. It was about hunting for levels made by players, trying all sort of editors. I never made ANYTHING even barely worthwhile. But I also knew everything, from obscure level editors, to Dehacked. The day I bought the first Quake, I installed an editor before I launched the game for the first time. Even today, I get so much more enthusiasm when I find some players’ project then the announce of a big new game. Battletech, for me, is nothing if there wasn’t Roguetech, Jagged Alliance 2 is nothing without 1.13, or AIMNAS (that sadly will never get even close to be completed), but there are tons of examples, from the classic Mount & Blade, to Starsector, X-Com (the old), all Paradox games, simulators like Falcon 4 and Silent Hunter 3, Freespace 2, Rimworld, Stalker (Anomaly), Terraria of course, but even Tarkov that walks a thin line between legal and not with its incredible single player add-on. Even Doom Eternal, that for me would be a much worse game without add-on support and some manual tweaks (I also got a constant crash that could only be solved by modifying code that could only be possible thanks to mod support). Every time there is some space, the community can make these games become SO MUCH MORE. To the point that the original game simply disappears. No reason to exist outside those mods.

The more the industry tries to wall itself off, the more it disappears for me. I’m not a passive consumer. I don’t give a shit about your movie-like stories. I can read books, I can watch movies that are made by real artists. Games for me are transformative. They are tools, languages meant to be played with. If your game isn’t open, for the community, then it’s already dead. That money will choke you to death. And that’s also why there’s that other space about roguelikes, and those communities and devs, open-source projects like CataclysmDDA. Let’s play, not watch passively.

Back to Factorio, I started to read obsessively about the game and found out that Factorio also had an extensive mod community. That was a door to a new world. I got to the third tutorial mission, and I never played “Factorio” again. It was over. What is Factorio for me, now? A collection of a few “packages”: Angel + Bobs + some MadClown and a couple of other minor components, Nullius, Space Exploration + Krastorio 2 + all Brevven meterals and some other stuff, and Pyanodon. That’s it. Four distinct collections of mods that represent the best versions of Factorio available. Games that are so expansive, addictive in all the right ways, that they become hobbies to cultivate for thousands of potential hours. And that’s the most FUN I’ve ever had, in all my gaming life.

I’m not analyzing further, there’s some stuff I wrote in the Factorio thread on Q23, but I love writing about this stuff. Factorio is great because its structure makes playing with mods straightforward. You don’t need to LEARN anything, strictly. You can just dive in the deepest end, it’s fine. That’s why I went from the third tutorial to the most complex Factorio had to offer. It’s not Minecraft, where you have to guess your goal, or randomly find a recipe. Factorio is like the code of the Matrix in front of you. It doesn’t require expertise, it’s not esoteric jargon. It just requires patience. You have a tech tree, all laid open. you have some limited tools. You just have to walk through it. Pyanodon is known as Factorio as its most complex, but when you start a game you soon find out you cannot research anything. There isn’t any conventional “path” to walk. And so it dawns on you: I can only build and play with what I already have. I can only put down the pieces, and connect them. And that’s like being a kid and playing with Legos. Minecraft is NOT Legos. Because everything’s hidden. In Factorio you see everything. You have all the pieces: make them sing. No one gives a shit if you make a disorganized spaghetti, you’ll always have time to reiterate, polish. Arrange the pieces temporarily to see how they operate, and when you figured it out you can clean the table and set down a better plan. Just… PLAY.

Factorio is addicting because you have infinite time, close to infinite progression and toys, and ALWAYS clear goals. Long term, short term. “One more turn”, in the sense there’s always something else that awaits. Small and big things that constantly present new problems and open new paths. It’s just PERFECT game design, distilled to its most pure. And what you build is… yours. Your own creation. It’s a bit the fun flavor of playing a Sim City, what you build is your creation, maybe to share with others with a screenshot. There’s always the fact that in Sim City the design constraints end up making every built city looking alike, but in Factorio there’s quite enough freedom to make your own misshaped spaghetti monsters, that you will love so much. Because you are their mom.

Factorio can deliver a similar exhilaration and fun of creativity you have when programming. The fun without the job, I suppose. But all games are.

So there was me, reading about the greatest, most complex mods for Factorio. Reading that a SINGLE Pyanodon run could take 1000 hours would excite me. Here I open another parenthesis… People usually get turned off by something like that. They hear: this takes 500 hours, and they NOPE out, thinking “this is not for me.” YOU ARE WRONG. Of course the number of hours in not indicative of quality time, but in these cases the prejudice is built on false premises. People who shut the door when they hear a game is too long is simply because they think they don’t have so much time. That’s the part where they are wrong. I came from a generation when games weren’t “meant” to be finished. From the penny arcades to the first home computers, you just played for a while with whatever you felt like playing. Everything changed, of course. But these types of games like Factorio are different from those linear AAA games that try to ape movies. A movie and a book have their meaning as a finite thing. But in a game, more content or a very long progression are just a guarantee of entertainment. It’s like you have a distributor of Fun Pills, but the can only contains 10 of them. Now imagine if they were INFINITE. The point is never “reaching the end”, but having fun WHILE you’re playing. Stop WORRYING about the end. And if you’re having fun, why do you want it to stop? It doesn’t matter if you have one hour or thousands, you simply get to the point you feel like getting to. You stop where you want, it’s fine. The game gets too complex? That’s also fine, you can stop. It’s not like only hearing half of a story makes it pointless without hearing the rest. If a game like Factorio has tons and tons of content through mods, that’s great. It’s like a huge bag of toys, that you explore at YOUR OWN pace, for as long as you want. You shouldn’t be scared that this bag is too big, because it’s always for you to employ however you like. It’s not a challenge impossible to match, it’s just a long, steady progression that presents new problems to solve. The game will be with you for as long as you want, and if you want to take a vacation it will still wait there for you when the desire comes up again. These are sandboxes, and you simply use as much sand you’re comfortable with. The fact the amount of sand is huge encourages you to play, because it means it will be there, ready, when you feel like stepping up the game. Scale to bigger things once you feel like you mastered your current level.

…So there was me, reading about the greatest, most complex mods for Factorio… and eventually I found mentions and comparisons of similar games. Satisfactory, Mindustry, Dyson Sphere Program, Oxygen Not Allowed. Until I found another obscure reference… GregTech.

Turns out GregTech is a Minecraft mod that makes the game more similar to Factorio. I started looking around and found some wiki pages. I learned there were multiple GregTech versions, and that they added block-machines to create some processes. It was interesting, but not especially mind blowing. The weirdest part for me was that I know pretty much everything on the field, at least superficially, but for some reason all this big section of Minecraft modding never surfaced. I knew there were mods for Minecraft, but they all seemed about people making big structures and all that decorative fluff.

Oh, I was wrong.

A door to a whole new world opened. And I spent the last 7 months reading about all of this, and a little bit of playing too.

Once I realized that the modding scene was way bigger then I thought, I took my usual approach: show me the deepest end. I was pointed to something called GT (GregTech) New Horizons (GTNH). In Minecraft, these “modpacks” dominate the scene. There are single mods that develop their own thing, but these individual mods are then organized together in what essentially are “total overhauls”. Part of this is because Minecraft modding is messy. It’s not like Factorio with an explicit mod support. The mods themselves have to be written in Java and are generally “dirty”, hacky work that require lots of expertise and skill. And because everything is quite messy, there will be bugs and problems, and solving them requires serious work. All this means that you need a good team to make a good modpack that can be well supported. You don’t just throw together some mods and play, you have to rely so much more on “curated” packages. The modpacks.

I launched GTNH.

And wow. This is the Minecraft I wanted, with what I hated removed. Not a survival exhausted in 20 minutes, but blazing onward to 600+ hours, of actual, “hard” content added to the game. And the most amazing, unexpected thing: …a questbook.

If my hugest issue when playing Minecraft was not having some incentive to push me onward, some clear goal that would enlighten the progression, this HUGE questbook, with more than 2000 quests, was at the same time a great guide to erase the problem of guessing recipes, and a clear path. It was finally Minecraft… with actual CONTENT to play with. With immediate activities, rewards, and long term goals.

I always thought Minecraft was just fluff, players spending hundreds of hours building pretty houses and castles, but functionally pointless. Because they were just abstractions. But now the whole thing takes a new light. I don’t care spending hours decorating something, for fluff, but if I’m going to spend hundreds of hours in a factory, then the way it’s functionally built, BECOMES IMPORTANT. You can plan and carefully design your places because the design has a function. You’re going to actually “live” in there. It’s not a backdrop, it’s your working environment. The same way you can spend hours setting up a programming IDE so that you are familiar with it. Things start making sense. You jump around like a bunny, you’ll get hungry sooner, need more food. If you build a path, then your speed is enhanced while walking on it. You want doors, windows, a place well lit. Stuff you build and use because it’s functionally active in that world. Stuff that has a purpose, a role to play. The survival aspects start making sense when they come together like this. You don’t just eat bread, because you need different types of food and nutrients. The world around starts to take a shape that is meaningful, where every “block” joins into a system. Onward.

If difficulty is a problem, it’s always because of accessibility. Having fun in a game is all about solving problems, but if those problems are too hard they become insurmountable walls. Even a small obstacle that doesn’t have a clear solution can become frustrating. The great thing here is that the functional complexity available in a huge modpack is accompanied with an equally huge questbook that works as a guide. Here’s your mountain of content, don’t be daunted, take my hand, I’ll guide you all the way. You just have to put the time, patience, and fun. But it’s all there because it is MEANT to be enjoyed. Your sandbox, with your guide. It is a huge mountain, but it is accessible. You just start to climb, setting your own pace and goals, step by step.

Now my Minecraft folder is… let’s see… currently at 52Gb.

I also intended to make a sort of guide here. So here’s what you need to dive into all this.

You need the Java version of Minecraft, avoid everything else. I don’t think the Java version is still available from Mojang, so you’ll have to deal with Microsoft. But once you got your account information, you don’t have to touch Microsoft stuff anymore.

You get this (instead):

(Note: community projects sadly often derail into lots of drama, and unless you’re directly involved, it’s drama that is very complex and hard to judge. I know very little of the present one, but MultiMC has also “forked”: https://polymc.org/news/moving-on/ …It can’t be too bad to have a more open platform, but otherwise, I have no clue on the matter.)

This is the program you use to download the game assets locally and the modpacks. Everything will be held within, without contaminating other directories, Windows main drive, profile and everything. The modpacks themselves, along with their options, configs and savegame data, will be put inside the “/instances/” directory. If you want to move all your stuff you simply have to backup your MultiMC folder and you won’t lose anything.

Beside MultiMC, and your user and password to access the main assets, you’ll need a Java version. Now… Modpacks depend on a certain Minecraft version, you don’t just download and play the latest. Most of the relevant modpacks are either 1.7.10, or 1.12.2. Those are the big “plateaus” where Minecraft modded has taken shape. But today there are also popular modpacks for 1.16.5, and a few things are trying to stay up to date (1.18.2, currently). Mods take time, questbooks too, good modpacks take even more time. That means that the more bleeding edge is the modpack, the more you lose in terms of mod integration and complexity. Many mods that define what modded Minecraft is haven’t moved past 1.12.2.

All this is also important because when you are PAST 1.16.5 you need Java 17. If you instead play a modpack based on 1.16.5 or earlier, then you need Java 8.

This is where I suggest grabbing Java:

Take a recent version if possible. But notice that the current one, doesn’t have a Java 8 package. So you need to scroll to a 21.3.1 or 21.2.0, for example “graalvm-ce-java8-windows-amd64-21.3.1.zip”, since you are looking for Java 8, and Windows (“amd64” indicates both AMD and Intel processors). Once you have the zip, you unpack it somewhere, and then configure MultiMC so that it will use that java version. Just look in the /bin directory for “javaw.exe”. Done.

It’s important you set the parameters, and every modpack has its own recommendations. This is what I use:
-XX:+EnableJVMCI -XX:+UseJVMCICompiler -XX:+EagerJVMCI -Djvmci.Compiler=graal -Dfml.readTimeout=120 -Dgraal.ShowConfiguration=info -XX:+UseG1GC -XX:+UnlockExperimentalVMOptions -Dsun.rmi.dgc.server.gcInterval=900000 -XX:+DisableExplicitGC -XX:G1NewSizePercent=20 -XX:G1ReservePercent=20 -XX:MaxGCPauseMillis=50 -XX:G1HeapRegionSize=32M

Also set the minimum and max memory allocation BOTH to 8192 MiB. All these settings aren’t perfect, and every pack has its quirks, but it’s a general default that should be working well. You can then individually tweak each instance while leaving the general settings alone.

That’s it. There’s an “add instance” button at the top left. You press it, go for example at the CurseForge tab and look for a modpack to install, the program will download the mods and create the instance. You launch it and play. These days I follow closely dev work directly on Discord channels, and often packs are available there as downloads. So you just import the zips.

At some point, for a reason or another, you’ll start tinkering. So it’s useful to know some general aspects. A modpack is generally built of three components: mod code, scripts, configuration files. Consider that MultiMC won’t automatically update a pack for you. You do it manually. It usually means creating a NEW instance, so a new subdirectory. You can then navigate to the old, and copy those files you need, to the new. Usually game options (options.txt), the “/saves/” directory, and some other stuff depending on mods, like minimap or tomb data. in any case, the instance directory will contain a “/minecraft/” directory, that’s where’s your stuff. The mod code is contained in .jar files, in the “/mods/” subdirectory. The scripts, that generally contain custom recipes, are inside “/scripts/”, and configuration files are under “/config/”. You can edit scripts and configuration files with just a text editor. When you manually move your data you don’t touch these, since they are set by the pack developers, and so might change when there’s an update. Other directories are usually generated at launch, so you generally don’t touch them.

Now, what do you PLAY?

Here’s a list. With a few notes, more or less relevant. I have less than 100 hours logged overall, when each one of these can take several hundreds. I really know nothing, but I read a lot.

My focus is of course on “industry”, complex processes and depth of gameplay in general. For reasons I’ll elaborate later, this generally means everything GregTech, but there will be other stuff listed.

– GT New Horizons [1.7.10] This is the big daddy of Minecraft modding. It’s one of those with the most active development, still to this day. It gets criticized because it’s “grindy”, but we’ll see later that grind can have a meaningful design purpose. Objectively, it has two strong positive aspects, that are almost unmatched in other modpacks: integration and progress. Integration means that this pack puts together the whole breadth of mods. It has many different biomes, types of food, monsters, along with magic and everything GregTech. These mods are integrated not only because they are in the same pack, but because the recipes are tweaked to depend on each other. Leading to an overall semi-linear progress, of a monumental “size” and scale, that is kept still accessible thanks to an expansive questbook. It’s essentially a flagship. The central pillar of this pack is GregTech 5, but after it was pushed by the community in a whole new directory, now known as GT5U (U-nofficial).

http://downloads.gtnewhorizons.com/Dev-Pack/Pack/Client/ The latest versions usually appear here. What you can find on curseforge is usually a bit outdated.
This is the main Github, the group of devs maintains a large number of 1.7.10 mods, at least those that can be messed with because set as open source.

Note: there are a couple of GTNH “forks”. One is GT Mega, but you need to hunt on discord for this one. It’s made by some embittered GTNH devs that left because of some drama and different vision for the modpack. The other is GT Impact, that you can find here. This one is set on “peaceful” mode, no monsters, only automation. Known to be very good, with interesting multiblocks and some other interesting choices. It has a significant problem though: no questbook. Because of that, you usually need to be already familiar with GTNH.

– Technlogical Journey [1.12.2] This is another unofficial GregTech. It’s based on GTCE (Community Edition), that is a port of GT5 for this new version of Minecraft (all official versions made by Greg didn’t move past 1.7.10). But this is not a well done and “full” version of GregTech, and the main reason why it was completely rewritten recently (but still not in this pack). The main show here is Gregicality, an extension of GregTech, only available here, that pushed the mid to late game toward future science technology. This is here because it represents another interesting extreme, in the complexity of the chemistry chains. It goes one step beyond GTNH itself at the late game, but it is much, much more stripped down in non-GregTech content. There isn’t much integration, and the questbook is more a general progress guide to follow than a complete tutorial. So it’s generally for those who have experience with modded Minecraft and GregTech especially.

https://www.curseforge.com/minecraft/modpacks/technologicaljourney This link is only useful to follow the Discord link. The updated version of the pack is distributed on the Discord link and you absolutely should avoid the Curseforge version. Even the questbook is incomplete in that version.

– Supersymmetry [1.12.2] Only for future reference, since there’s nothing to see here at the moment. It’s an ambitious modpack, science-based, built around the rewrite of GTCE (GTCE-unofficial) and a new version of Gregicality. But a new version of Gregicality (now called Gregicality-Science) is still rather far from release, and the pack, with the ambition to join Gregicality with NuclearCraft Overhaul and QMD, is even further away. Right now the pack is a meme, but it could become important.

https://github.com/Zalgo239/Supersymmetry (Zalgo, Tech22 and Pcm_Keywielder are literally doing God’s work in the proverbial sense, “work that is very important and necessary, especially that which receives little or no recognition or pay.”)

– Omnifactory/Nomifactory (STE) [1.12.2] Still the “lousy” GTCE, but the best introduction to GregTech if you want the focus on the factory rather than the whole modded world, as in GTNH. Since I also was not practical with GregTech, I needed some guide to then “graduate” to Technological Journey. My choice was Omnifactory STE. STE stands for the ominously named “Self-Torture Edition.” For the time being, it helped me. Omnifactory has now be renamed Nomifactory, so refer to that. Omnifactory STE still hasn’t transitioned to the new name. The difference, early game, is that it has much more expensive recipes and enables the steam age, that is instead skipped in the standard version of Nomifactory. Since I play to get familiar with GregTech, I do need that steam age, and playing through that first questbook page helped me starting to define the GregTech mental space. I’ll soon try to reproduce and match the progress into TJ. Omni/Nomi is mainly a GregTech pack, rather lean, and meant to be played without monsters. So it’s like a focused slice of Minecraft. All survival aspects are essentially removed. But there’s still plenty to keep you busy, and work hard.

Nomifactory dev (only use the dev version on github, the main description page has a link to the nightly builds and an info page)
Omnifactory STE Get the client on that page, I manually updated a few things myself, since it’s not as well maintained as the main Nomifactory, but it should be fine

Both STE and Standard have a port to the new GTCE-u, but it’s still in beta.
The STE version is currently only on the Nomi Discord channel.

I’ll add here that there are other GTCE-u modpacks in development. This includes Technological Journey 2. But I’d stay away until Gregicality comes out. I’ll link instead three other packs:
GregTech Community Pack, a lean pack with a tutorialized questbook. This is only GregTech base version + some small support mods. It’s meant to be an introduction to GregTech.
GregTech Expert 2
TerraFirmaGreg You may soon realize that GregTech seems to attract some niche Russian, Chinese and Japanese communities. This is a Russian guy who’s putting together GTCE-u with TerraFirmaCraft.

– FTB Ultimate [1.4.7] This is a jump back in time to a much, much earlier version of Minecraft. Ancient history. This is GregTech 2. But I played this to have an overall feel of the starting point in GregTech. It’s still greggy. The biggest difference is that ore generation is traditional. So you just start digging down and find plenty of minerals to use. Because ore generation is scattered all over the place, it means it’s more “exploratory”, and classic Minecraft experience with mods. There’s no questbook. This is a typical “kitchen sink” pack. Every mod does its own thing, there’s no “intended” progression, and you just take your own path. At the time there weren’t many mods, so players were familiar with everything available. They knew what to do and the game was more about creatively using those tools rather than structuring some intended progress. But it’s also the point where mods started to be aware each other, guiding toward a richer, complex experience. It’s fun to play. Use NEI, make a Pulverizer to double ores.

(Old, but it still lets players build… this)

From the MultiMC instance page, look at FTB Legacy, order by game version, scroll down close to the bottom for the correct version for 1.4.7. The pack version is 1.1.2… but there are a few caveats. You need to add a few things that are indispensable.
Block Helper
https://bdew.net/old-downloads/ (the one for Minecraft 1.4.7: “neiaddons-1.4.7-1.6.1.r8.jar”)
Mouse Tweaks

– Divine Journey 2 [1.12.2] Not GregTech. This is a good example of a “hard” progress based pack. Recipes, even of basic things, are modified. And the mods included are organized to provide a semi-linear progress. Since everything in it is modified, this gives something fresh to play for everyone


– Lost Era [1.7.10] This is another large pack for this early Minecraft version, despite being a recent one. It has GregTech 4, that was mainly a minecraft 1.6.4 mod, so this one here was an unofficial port. But GregTech is not the core here. It’s a much more simpler pack compared to GTNH, with meaningful but not “hard” progression. It’s especially great as an overall tutorial for mods. Featuring most of the main ones, including tech and magic. Stuff you learn here is always useful whether you move to newer or older versions. A great access point, with an excellent questbook.


– TerraFirma Rescue [1.7.10] This is the merge of TerraFirmaCraft and GregTech 6. It’s actually GT6 unofficial, it’s a bit outdated as a pack, not so well maintained. But TFC has some excellent gameplay aspects. I’ve only tried the early game, and the quests could use some polish, but once I gasped what I was meant to do it became quite satisfying and more “natural” than even standard Minecraft gameplay.


– Bears Den surviving Take 2 [1.7.10] Simple GT6 pack. Basically a ready test environment for GregTech 6. You can always drop some more mods in there, if you want.


– Test Pack Please Ignore [1.6.4] This is another reference point for GregTech, since it contains the last version of GT4 before the move to GT5. Despite its age this is a large pack, somewhat inheriting what FTB Ultimate was. In both these cases there is no questbook, so playing through it is not simple at all. you can probably look some older gameplay on youtube to get some ideas. I have a questbook for this version of GT, but based on a Russian hard pack with all the recipes changed, and if I backport the questbook, it crashes. But at least it can works as a general reference. It’s especially with one of the late GT3 versions, and then the move to GT4, that GregTech grew in ambition and also started “greggifying” recipes. This was a controversial design decision because if for example before you needed metal plates in recipes for GT machines, now metal plates are much more widely used, in order to justify building the plate-making machine, rather than crafting those plates manually with a hammer. Same as modifying some Minecraft canons, like 1 log only producing 2 planks, as long you don’t use a saw or a sawmill block. Leading to the general concept of “microcrafting”, also often considered as a bad thing.

This one is on Technic. Simply “add instance” in MultiMC, select the Technic tab and search for the pack name. I heavily tweaked my own version, with minimal testing, so it makes no sense to give more specifics.

– FTB Unhinged [1.5.2] Another GregTech reference point. I’m not sure if it’s really that interesting as a reference, but this, being late GT3, is right between the GT2 in FTB Ultimate, and the GT4 in Test Pack Please Ignore.

Use again the FTB Legacy tab in MultiMC.

– InfiTech 2 [1.7.10] To complete GregTech reference points. This is version 5 before it moved to the Unofficial version, and then GTNH. It is an important pack because it comes with a classic questbook. So it can be followed to see how GregTech was intended to work at this point in time. And how it spread to fundamentally change a lot of the early game.

https://www.curseforge.com/minecraft/modpacks/infitech-2 (yes, it says it’s outdated)

– Revolution 3 [1.7.10] Another generally ignored but important niche. This modpack includes Reika’s mods. Like Greg himself, Reika still maintains his mod suite to this day, but they are very rarely included in modpacks. The main reason is that Reika forbids modification and integration, because he thinks his stuff is precisely designed with a certain progression, but that goes against the intention of those who put together modpacks. These choices ended up in a sort of walled off world. This pack is a rare one that includes those mods, and also has a questbook. Reika’s mods come with their in-game documentation, but they still assume one is quite familiar with the modded games and tools.

This one’s on ATLauncher. So add instance, select the tab and search for “Revolution”. You want the “3” version. Don’t go to 4, it’s a different thing.

– Dragon Realm [1.7.10] This is a recent thing. Reika had his own server and modpack, where he played along the years and tested all his mods. He never released it because he didn’t want to maintain it, answer questions, deal with bug reports and all that. He recently decided to still release it. There’s TONS of information on the site, including specific install instructions that need to be followed. What’s missing is again… a questbook. So you either use Revolution 3 as a “map”, or you’re left figuring stuff in the game or watch some video.

https://dragonrealm.overminddl1.com/index.php Read everything here. Follow instructions carefully.

– FTB Academy + FTB University [1.12.2] These are “FTB” packs. These are usually well done, especially in this case. They are essentially big tutorials, making you familiar with all (most) of the mods. No GregTech, but you have a guided approach that shows you what’s the deal with Minecraft modding.


– Enigmatica 2 Expert (and Extended) [1.12.2] One of the most famous “hard” packs. It messes with some common recipes, like chests, but it is generally considered as one of the most accessible hard packs. Built to have a certain progression and integrating both magic and tech, similar to Divine Journey 2. No GregTech.

https://www.curseforge.com/minecraft/modpacks/enigmatica2expert/ Original
https://www.curseforge.com/minecraft/modpacks/enigmatica-2-expert-extended Extended

– Enigmatica 6 Expert [1.16.5] The first modpack I mention in this Minecraft version. A spiritual successor to Enigmatica 2 Expert, released days ago at the time of this writing. It’s actually way different from E2. Like, Enigmatica Exotic edition. It demands a lot more exploration to find useful random loot, and fighting enemies in generated structure. It heavily depends on magic. I suppose it’s less “industry” than usual, but will require automation. The pack is identical to the non-expert version. The only changes are “hard” recipes. Most of the questbook is also identical, but hardmode enables a page to show an ideal order of progression for the mods included. Whereas in non-expert those mods are relatively independent, in expert the recipes have many cross dependencies, so that you’ll need to progress in “x”, in other to unblock something in “y”.


– Create Above & Beyond [1.16.5] Hugely popular. Classified as hard pack, but nowhere close to the standard. It’s mostly about the popular mod Create, that is about automation but in a way that is conceptually opposite to GregTech. Create is about building Rube Goldberg contraptions. Think about “analogical” technology rather than the industry blocks of GT. But it’s fun for a lot of people and now in a pack with a defined progression. Most 1.16.5 packs are just “kitchen sink”, without a definite progression, and more “fluff” mods about the standard aspects of Minecraft, like exploration and combat. So there’s still a huge demand for something deeper, and on more recent Minecraft versions. This one is a decent mix of something that feels fresh, but also not shallow as most more recent things.


– Multiblock Madness [1.12.2] This is another tech-first modpack, oddly not GregTech. But it’s one of the more rarer packs that contains some of the most complex tech mods, like NuclearCraft Overhaul and QMD (particle physics). Another ambitious modpack is being planned that will include this, with GregTech (Supersymmetry), but it’s a long time away from a release. The only other modpack with a similar set of features, available at the moment, is Quanta. This one (Multiblock Madness) should have a decent questbook, that i think is important when the complexity scales up. I haven’t played it, so I don’t know if the questbook can be a good guide on its own, or if it’s better to build some experience first.


– SevTech: Ages [1.12.2] Another popular modpack next to Enigmatica 2. Rather unique because it has a technological progress that is gated through various ages. Rather than using a questbook, it goes for an enhanced achievements page, that can still be used as a general guide about what to do and how to progress.


– Compact Claustrophobia [1.12.2] I can’t remember if there’s a more up do date version out there. But this is interesting because you have a series of problems of increasing complexity, and you are stuck in tiny rooms. So the challenge is to both solve the problem, and use as less space as possible. The forced limited focus helps to concentrate on the problem only.


– FTB Oceanblock, Cuboid Outpost [1.16.5] Both recent packs I’ve seen played on Twitch.

As far as I know Oceanblock doesn’t have a dedicate page, for some absurd reason. But you can again “add instance” and look for it on the FTB section.

– FTB Infinity Evolved Expert [1.7.10] I haven’t touched this one, but it’s one of the classic ones, hugely popular.


Different, semi-incomplete sidetracks (pursuing complexity in different directions than GregTech, these are even more “niche”):

– TechNodeFirmaCraft https://www.curseforge.com/minecraft/modpacks/technodefirmacraft
– Ad Astra Per Nucleon https://www.curseforge.com/minecraft/modpacks/ad-astra-per-nucleon
– Per Fabrica Ad Astra https://atlauncher.com/pack/perfabricaadastra
– Fabrica Atlantica https://github.com/wormzjl/Fabrica-Atlantica
– Quanta https://www.curseforge.com/minecraft/modpacks/quanta

– Valhelsia 3 [1.16.5] Popular big “kitchen sink” pack. Quests were planned but never appeared. Just a collection of mods, supposedly well put together, and it lets you experiment with the best stuff available in this version of Minecraft. Minecolonies, for example. This pack essentially contains many of the popular big mods, but it doesn’t have a well planned progression, and it won’t guide you through that content. So it’s a mixed bag.


– Better Minecraft (Plus) [1.16.5] Another “exploratory” modpack. Extremely popular. I think the video looks quite amazing. There are versions for 1.18, both Fabric and Forge. But of course less meaningful mods, since the newer version are always weaker on content.


– Crucial 2 [1.16.5] This one is a well thought, minimal pack that voluntarily stays away from Minecraft modding complexities. It’s basically an improved vanilla Minecraft.


Misc 1.16.5 “questpacks”. All of these have lots of quests, and slightly different sets of mods. Sometimes they use the same quests. They are all kitchen sink without a really defined progression. Quality may be variable. Age of Fate seemed to lag for me. Despite the huge number of quests don’t expect them to cover everything. Some “branches” are more detailed than others, some could be missing entirely. In the end, despite the presence of quests you could feel like not knowing what to do:

– TNP Limitless 3 https://www.curseforge.com/minecraft/modpacks/tnp-limitless-3
– Monumental Experience https://www.curseforge.com/minecraft/modpacks/monumental-experience
– Dungeons, Dragons and Space Shuttles https://www.curseforge.com/minecraft/modpacks/dungeons-dragons-and-space-shuttles
– Age of Fate https://www.curseforge.com/minecraft/modpacks/age-of-fate

– Craft to Exile: Dissonance [1.15.2] Weird choice of Minecraft version. You try this yourself. I’m just pointing you to this image


– Chroma Technology 2 [1.16.5] Not on my radar, this one seems for the cool kids. Fighting dragons and collect magical weapons. It has the “Silent Gear” class of mods.


Unordered misc:

https://www.curseforge.com/minecraft/modpacks/telomerase (TerraFirmaCraft+ 1.7.10 pack)

It’s all stuff I bookmarked for a reason or another. There are MANY more modpacks that are better and more popular than these listed, like many of the FTB ones. I just directed my searchlight in some specific directions.

Some notions about GregTech.

The latest “offical” GregTech version is GT6, for minecraft 1.7.10. As I said above, there aren’t many modpacks for this GT version. Same as what happened to Reika’s Rotarycraft, another “complex” mod made by an engineer, with real physics simulated in the game (praise for Reika’s mods is more about the concepts involved, but much less for his coding skill and performance of those mods…). I’m currently trying to build a “mental map” of those versions of GT, to have a general idea of what changes from one to the other, and how they evolved in complexity and design. It will take me a while.

GregTech originally started (…okay. Remember that all you read here doesn’t come from a position of authority. I just gathered SOME knowledge. What I write is usually pertinent and documented, but it may be a generalization or a simplification) not as an individual mod, but as an “extension” to a much more popular mod at the time: IndustrialCraft 2 (by the way the FTB wiki can be quite useful, but there are two versions, with different content. Here’s one and the other). IC2 already provided a number of machines for ore processing and other conundrums. GregTech Intergalactical was meant to expand the scope, especially in the middle to late game with more advanced machines, more materials and processes. But as the mod became more popular and Greg kept working and adding to it, the scope of the mod increased.

You have to consider the modded game landscape at the time. There wasn’t much organization, and there were an handful of popular and well known mods that players had fully mastered. Those mods weren’t well coordinated with each other, there wasn’t an overall vision or progress. So the individual mods were extremely powerful tools that allowed players to conceive creative contraptions. But they were also kind of overpowered once you knew your way around. Rather than a balanced progress you had players bee lining right to end game in no time. This created a demand for “hard” modpacks, where the overpowered shortcuts were nerfed so that you felt like earning the better tools, and experience the technological progress within the game. If you have everything you want, on demand, the game gets kind of boring even if you have endless possibilities about what you build. Without some restraints, the lure of a reward becomes weak.

As the scope of GregTech widened and deepened, Greg started to “override” the landscape. He started to expand the game design, so that a certain progress was forced. He was nerfing certain recipes so that the use of machines was more rewarding, and also modified the recipes of the IC2 machines to require more complex and expensive materials. The motivation is obvious, he was transforming the technological progress so that mods were better integrated and so that crafting a new machine felt more rewarding. It’s a delicate balance, but required at the time when players had available overpowered machines from other mods that would immediately make a large portion of the content irrelevant. At the same time, when you start touching these aspects lots of players are disappointed, because you are adding “grind”, for example creating artificial dependencies just to justify the presence of a machine. If the metal plate machine in GT is only used for GT machines, then you could decide to ignore it, and in case just craft a few metal plates with a hammer. But if Greg modifies the recipe of EVERY other machine, now all requiring plates, then the plate machine becomes mandatory. And so more restrictive. Up to the point of creating a large drama, especially with the developer of another hugely popular mod (to this day): Tinkers’ Construct. With Greg going for the infamous choice of making the game CRASH if the Tinkers mod was present. Responsibility was bounced back and forth, it is not my role to be judge here, but the result was that the community started to get split, and Greg, not unlike Reika, grew more “radical” in his design choices, that ultimately pushed his work in its own niche and more insulated to the rest of the modding space.

This “greggification” of recipes started especially with GT3 and moved on to GT4. And from there the scope of the mod kept growing, moving away from a mere add-on for IC2. Today, GregTech 6 doesn’t depend on IC2 anymore, and its impact on the game is so wide that it can be considered a modpack on its own. With a specific design vision that affects the whole game.

Despite this, the official GregTech, again like Reika, never moved past 1.7.10. The scope of the mod was way too large for Greg to reshape around a radically different of Minecraft. But this didn’t stop Greg from radically redesign his mod, with GT6. As with happens with another hugely popular mod, Thaumcraft, versions start to differ so much in design, there there isn’t a “best” one. And that’s why players who liked GT5 decided to “fork” Greg work, leading to GT5-U, first, I think lead by some guy named BloodAsp, and then eventually handed over to the GTNH team that continues to this day, working on that branch of GT5, and expanding a lot more the complexity of the chemical processes.

When the modding community largely moved to the new standard Minecraft version (1.12.2), Greg stayed behind to do GT6 on 1.7.10, but some other devs/players tried porting GT to 1.12.2 (mainly a guy known as Archtech). Not GT6, but GT5/U. This version was GTCE, that never got a good reputation, both in quality of code and content. But it was the only version available, and when carefully worked in a custom pack like Omnifactory… it wasn’t bad. Another team started to work on extensions to GTCE, like “Shadows of Greg”, to add back and flash out what was missing, and then to expand further the mod, with “Gregicality”. Eventually, having to constantly deal with the limits of the code, they decided to do a radical rewrite of the whole of GTCE… leading to GTCE-U, released just this past December (which is impressive considering the redesign only started about 6 months before)… But still going through significant redesign phases as more interesting ideas from GT6 are backported, moving again the scope of the mod, from a loose port of GT5U, to the aim of making the “best” GT version, integrating the best ideas and practices (but it’s not an exact science, and there are always ambiguous debates on how “hard” the mod should be).

Today, there are a number of GregTech versions still maintained and relevant. Here’s a list:

GregTech 6 [1.7.10] The official one, still maintained by Greg. It’s considered “feature complete”, so Greg is mostly fixing bugs and compatibility.
GT5U [1.7.10] The flagship that is integral, inseparable from GTNH.
GTCEu [1.12.2] The recent rewrite by the Gregicality team, the most promising version, especially when the new Gregicality Science comes out and more modpacks are developed.
GregTech Intergalactical [1.18?] The only attept to port GT to the newest Minecraft. Despite you might expect lot of attention on this one, this seems a one-man work, and it develops at rare bursts because that main dev can’t dedicate time to it. I don’t really know why it’s much, much smaller than the 1.12 team, but… It is claimed to be almost feature complete, and still generally ignored for some reason.
GT4 Reimagined [1.18?] This is a port of GT4 that is using the same API/infrastructure of the GT version just here above. Test pack here
GregTech 4 [1.7.10] GT4 was for Minecraft 1.6.4, this version is a port to 1.7.10. Used for example in Lost Era (see above), it’s a smaller, simpler version compared to GT5, so preferred in this case if you don’t want GregTech to “monopolize” the design of the pack.
GregTech 6 Unoffical GT6 also has a “U” version. Used mainly in TerraFirma Rescue, but the mod is a bit outdated and not well maintained, so the main 6 is now ahead. Greg has “complained” that rather than fork GT6 it could have been a better idea to create it as an extension. So that the merge of new GT6 features could have been more automatic, and even offered the possibility to port the most liked features of GT6U in GT6 itself.
GregTech Experimental [1.12.2] This one is a port of GT3.
GregTech Classic [1.12.2] I don’t really know. As the description says it’s a mod to match a port of an earlier version of IC2, so it should be a modern port of how GT1 was originally.

Now I have my own plans, most (all?) of which won’t probably see the light. I’d like to write, or contribute, to a questbook. For example it seems there’s not a good, complete questbook for GT6, and my original idea was to make a large pack including both GT6 and Reika’s mods. Now the idea moved to try to use the lost Era modpack as a base, eject GT4, replace it with GT6 and Reikas’, and rework the GT quests from there, since the pack is already meant as a guide for all the mods it includes. But it is likely than when I start working on that it will evolve to become more like a “hard” pack with some changes to the progression. Another, very long term and probably impossible for my technical skills, would be to modify the code to TerraFirmaCraft, to make it compatible to general modding, and overhaul all the early game, convincing GTNH devs to embrace this vision. This is “unlikely”, because both players and devs are strongly against TFC, but I have my reasons and my own way to rearrange those mechanics. The actual hard part is the code, and that mans being highly unlikely, especially if I’m alone working on that…

…But, even writing a GT6 questbook is a very long project. I don’t know GT6. The path leading there would be watching Bear’s videos on GT6, reproduce all that stuff in game, and putting on the questbook everything I learn. right now, I’m only playing through Omnifactory STE, to get an idea of GT, and eventually move to Technological Journey to reproduce what I know over a “legacy” Gregicality pack. And that’s already a significant commitment…

The discovery, seemingly coming out of nothing, of modded Minecraft was for me an amazing experience. Like a kid walking into a gargantuan castle filled with new toys. Hard to believe. The “Recipe for Fun” is quite simple, and very similar to Factorio. The game throws at you a constant flux of tasks. To complete these tasks you need to fill a number of sub-tasks, and many of these will have cross dependencies. So that in order to do X, you also need A and B. But doing A also unblocks a new path, that might lead you to F and G, in a complete new direction, that eventually will lead back. This continuous flow of problems to solve, micro and macro, is extremely addicting because it signals clear objectives and then their completion enables new stuff to see and experiment with, to create new sets of problems, and new branching tasks. New machines to experiment with, new stuff produced, new places to explore…

The progress, especially in GT-themed packs, also follows a well tuned formula: you have to work hard to get to a specific goal. In order to get there you have to complete a number of different activities, it’s a journey. But once you reached the objective the reward goes in two different directions. It goes forward because you are unlocking a new branch of the tech tree, so new gameplay to explore, machines you can build, new things. But it also goes back, because sometimes you open new ways to get the same resources in much less time and effort. That’s what makes it rewarding. You aren’t simply leaving everything behind, making it all obsolete as it happens in many (not sandbox-y) games, to move to something else, and neither you have to repeat over and over the same process. You simply unlock backward-facing shortcuts that let you “optimize” what you’ve done up to that point. As in Factorio, you have sometime the meaningful choice of keeping your own “slow” line, or “refactor” and redesign it so it’s much more efficient once you’ve unlocked some “power-ups”. No one prods you annoyingly onward, you decide what to prioritize. The sandbox provides you the good tools, it’s up to you doing what you want with them.

I can even analyze my own experience to better understand what works so well, and what doesn’t. For example when playing Omnifactory STE I got to a point that felt like a significant stagnation for my own fun. It was as soon as I unlocked the new quest page after completing the first. This is not normal, because I got to the place where the pack actually becomes interesting (Omnifactory standard starts from this point, basically), so why it was the opposite for me? Because I didn’t have a clear objective. There are a bunch of new machines I can do, but I don’t have anything they need to be done, FOR. I can work to build those new machines, and activate some new processes, but I don’t see anything, at the moment, that I need these new things FOR. I have things to do, but not a clear goal. In the previous “page” of the questbook, instead it was different. I was struggling for scarcity everywhere. So I needed to do stuff, and had to work with important restrictions. I made a coke oven multiblock, that let me transform wood logs into charcoal and coke, but I couldn’t use it because it was getting immediately clogged in creosote oil. So I then built a big steam furnace powered with that creosote oil, and then went to create a giant, “undocumented” buffer steam to power a couple of macerators and start ore processing before it was intended to be reasonable. Basically I was doing one step while thinking at what that step would let me do. Planning ahead, always with another goal on my radar. As it usually happens, good game design here doesn’t even depend on the mod. It depends on… the questbook.

And that’s why I want to go there. I enjoy tinkering with game design, and in modded Minecraft the quality of the experience depends so much on how the different mods are used, how they are integrated, creating a system that lets the game come alive. The structure and organization. It’s also the real distinction between Minecraft and Factorio. In Factorio you’re always at the bird’s eye. Always planning. Factorio is relentless with its lure because there’s nothing in there to distract or slow you down. In Minecraft instead you live inside the factory and are part of it. You tinker all the times with wrenches and other tools. You manually take an empty tank, walk down to a river, place it on the sand, slowly fill it with a bucket in your hands, and then drag it back to put it in its proper place, and maybe in the process you get hungry, or a creeper explodes half your base. Incidental annoyances that depend on your previous choices, that form a complex system. You plan your factory as in Factorio, but you also walk inside it, manipulate it directly with your virtual hands. In Factorio you plan and design, and it’s a pure mental drug, in Minecraft you’re “doing” too, in a way that is fulfilling and also relaxing. A mix of micro and macro that create a varied, absorbing experience.

Why I wrote all this?!

Not for you. You, meaning a hypothetical reader. This blog/website has existed for a long time. The purpose is still the same. A bit inconveniently named, it was always meant as an archive for my own use, and incidentally whoever would find some value or interest in its content. I’ve always been on the forums, engaging directly in discussions. On the blog I was rallying knowledge.

I’d cut pieces of discussions and paste them here to give them emphasis. Set milestones in recurring debates. As if clippings from newspapers. It was necessary for the organization of my mental space, because if that space is well organized then it will eventually have an effect on the quality of my thoughts.

The title is a reference to a famous movie, and Polybenzimidazole is one of the most complex production chains in GTNH.

Tabletop RPGs, Computer Wargaming (part 7)

Season 1: Part 1Part 2Part 3Part 4Part 5Part 6 – Part 7

This wraps up the first forum thread. There’s a lot more, but I’m already tired of maintaining that part-index here above. So this closes the first section and what will come after will only be divided into parts in the title, and a separate category.

Here I am again for a brand new SIDETRACK! The loops ever getting wider and wider.

The whole point of the above whole, was to go dig in the lost tradition of RPGs, not back when they were simple and naive but when they were ambitious and convolutedly eccentric. That’s the real main root of RPGs that we collectively abandoned and forgot about: wargames.

So one of these sidetracks is retrieving some of those games I wanted to play for a very long time when I was a kid, but never ended up trying on a real Commodore 64.

The title I remember (reading reviews of) is “Halls of Montezuma.” As you can see there are a few, all using the same “Battlefront” engine or some variations of it. The comments I read about these games mention that most of them are only collections of scenarios, and only the Russia game had a proper campaign, so I picked that one.

I employed the well practiced strategy of poking at things to figure them out without a manual, and it’s all relatively straightforward. Some menus to toggle on/off some map features, some text screens to see reinforcements and various statistics, and so on.

The only problem is that I couldn’t find a command to actually MOVE stuff on the map, and I’ve tried for quite a while.

I suppose that menu is to set up ground and air support for that precise attack, and then there’s a menu with various options, like: ASSAULT – PROBE – DEFEND – RETREAT

And there seems to be some upper level “directive” that only switches from “resting” to “normal”.

Units that aren’t shown engaged in combat can only switch from “hold” to “deploy”.

That’s about it?

After trying for a while I went looking for some discussions online, and it does seem that you don’t push counters in this game… It should be some higher scale strategic thing, so you pretty much only set up the intensity of an attack, but it’s the AI that is in charge of the tactical game.

I guess this can help the game to be more balanced, since it’s all the AI doing those moves. With the player only playing from the backseat, the game probably holds up better when it comes to challenge, but it still feels quite limited… and dry.

You’d expect me to stop here, after having an idea of how the game works. But instead I’m looking for very specific things one doesn’t usually care about.

This time I started wondering how they put together that “tiled” hex map. Because after a quick analysis it doesn’t seem like they use the same “grid” that they used for text. So I started to wonder, are they using two separate graphic modes, shared on the same screen, one for the map and another for all the text?

Yet that tiled hex map defied my first attempts at deconstructing it. The vertical spacing makes sense, it’s horizontally that it does something weird.

To find a solution I had to remove that ugly CRT filter and resize to a native resolution. The final result is a grid of 8×10:

Horizontally the sequence is two tiles forming an unit, then one tile used for the hex grid. The hex grid goes all around the two unit tiles, so each unit tile has 10 grid tiles surrounding it. Those tiles can be empty, but they can always only contain the hex grid.

Now you can see that vertically the hex tiles form a column from the top to the bottom of the screen (horizontally with multiples of 3), whereas this doesn’t happen horizontally, because horizontally we have two unit tiles followed by four hex tiles. Producing that slanted offset…

Why should one waste time to figure out these relics of map representation? Because for my own game thing I was trying to figure out how to best represent an hex map having only available a square grid. So not only I need to find some graphical solution that looks nice, but also the “algorithm” that would build such thing.

It’s not as trivial as it might appear, and it was very interesting to go back and dissect this game.

This series was made by SSG, but the better known in the field is SSI. I haven’t dug as much here because I got sidetracked again, but SSI has its own way of building an hex grid, and its own solution produces an arguably prettier result, since these looks more like proper hexes compared to the hamburger-ized version by SSG.

(a matter of scale…)

…In the future I’ll try to code my very own solution. Because about two months ago I stumbled on this problem of “building hex grids with square tiles”, and after fiddling with the most common solutions for a while I eventually ended up with some weird hybrid that I’ve never seen used before. Only accidentally I now stumbled on these SSG wargames, and that solution has something in common with mine… But not quite the same.

So what’s the sidetrack that sidetracked me? That from SSG I moved to SSI, and the SSI catalogue is quite huge.

For most of everyone SSI is only synonym of the GoldBox games, but as it happens with the pen and paper RPGs the true root was in the wargame genre. SSI was primarily a wargame publisher.

My story is that, before getting watered down, pen and paper RPGs were after simulation and complexity.

Is it possible that this is true even on the computer side? Is it possible that the GoldBox games were watered down designs coming from more complex systems that preceded them and then got erased from our collective memory?


I introduce you to: Wizard’s Crown + The Eternal Dagger

In the SSI catalogue these are the only two fantasy games showing up as “intermediate”. Everything that belongs to fantasy is otherwise “introductory”. The system contemplates three tiers, but the “advanced” one is only reserved for the actual wargames.

The Digital Antiquarian reinforces the story:

At their best, though, the rules behind these games felt more consciously designed than the games in the bigger, more respected series — doubtless a legacy of SSI’s wargame roots. This quality is most notable in Wizard’s Crown. The most wargamey of all SSI’s CRPGs, Wizard’s Crown was not coincidentally also the first CRPG to be designed in-house by the company’s own small staff of developers, led by Paul Murray and Keith Brors, the two most devoted tabletop Dungeons & Dragons fans in the office. Built around a combat engine of enormous tactical depth in comparison to Ultima and The Bard’s Tale, it may not be a sustainedly fun game — the sheer quantity and detail of the fights gets exhausting well before the end, and the game has little else to offer — but it’s one of real importance in the history of both SSI and the CRPG.

And confirmed by the wikipedia:

Wizard’s Crown was the first RPG designed in-house by SSI, previously known as a wargame company. Its detailed tactical combat system came from Murray and Brors’s background in wargaming, and they brought the complexity of those games to Wizard’s Crown’s tactical combat. For instance, shields block attacks only from the front and left (shielded) side, and not from the rear and right (unshielded side). Spears can attack two squares away, flails ignore the defender’s shields, and axes have a chance of breaking shields. There is an option for “quick combat”, and regular combat can take as long as 40 minutes per encounter. This combat system influenced SSI’s later Gold Box series of RPGs, but it was streamlined and simplified.

I think The Eternal Dagger implements one idea I also had for my game (that I didn’t know existed already in this form): when you are in a dungeon the whole party is a “blob” represented by a single unit, but when you enter combat you instead deploy all the party units onto a separate tactical map. Actually Ultima does something along these lines already. The difference in my idea is that I wouldn’t use a separate map, but deploy directly the party on the same map, smoothing this transition somewhat.

The game system has a plenty of good ideas. No levels, skill based. Experience points are spent directly to improve skills. Every skill increased has a fixed cost, then the amount of the increase is rolled randomly, and goes 1 to 8 if the skill is below 100, then 1 to 4, 1 to 2, and just 1 when you go above 200, progressively slowing down.

On the excerpt above there’s interesting differentiation between weapon types (in the image you can see the considerable attack range when using a spear), but it goes further as damage is divided onto “thrust”, “cut” and “bash”, and of course this is matched by the armor types and their damage absorption values.

Another important feature is the facing and how it’s handled. If you aren’t moving you can switch the facing how many times you want, up to three movement steps you only have 1 facing change, and above three you just can’t change facing at all.

It becomes relevant because facing is used in combat calculations

There’s distinction between front, rear, shielded and unshielded side. Side and rear add a to-hit attack bonus, the shield adds to the defensive skill, since the to-hit roll is essentially attack skill versus defensive skill, plus some randomness added. The defensive skill is the weapon skill, but halved, so that’s why a shield is probably quite important since without one the mechanic favors massively the attacker.

But there are also a bunch of situational modifiers. Fatigue and morale, but also injuries. If the defender moved more than half his points, then his defensive skill is halved again, BUT only if it’s a melee attack, because if instead the attack is ranged then his defense skill is DOUBLED, because by moving he gets harder to hit.

And there’s more. Four types of melee attacks: normal, defensive, aimed and killing.

The difference between these four is how they modify to-hit chance and damage. “Aimed” attacks lower to-hit to increase damage, “killing” lower defense to increase damage, and “defensive” attacks lower to-hit to increase defense, with the manual suggesting to use this option when you are threatened by enemies on your sides and rear (since they’ll have their to-hit bonus due to their positional advantage).

In Wizard’s Crown aiming behaves differently, it lets you skip a turn to obtain a to-hit and damage bonus on the next.

Wizard’s Crown also has an option to stand or go prone, so that you can actually DIVE and avoid arrows, on the other hand making you more vulnerable to melee attacks. And a “shield bash” can be executed to SEND a character prone (but it doesn’t inflict damage).

Wizard’s Crown also seems way more complex when it comes to determine damage. The manual explains the process without giving much insight in the formulas. But it does mention the damage is made of both injury and bleeding, and that these are applied through a “multiple” that is determined by hit locations: chest and stomach causing more bleeding, limbs and head more injury. Bleeding is what causes death directly, injury instead can knock someone unconscious and also affect fighting skills.

Even here the horrible trend is showing, The Eternal Dagger, being the sequel, simplifies the rules instead of enriching them. All reviews seem to agree that Wizard’s Crown (the first) was the better one.

It might sound complicate to organize mentally, but the manual gives some good tips:

There are plenty of good mechanics everywhere:

The general idea I get out of these two games is that they had an excellent system waiting to be used in a broader game. On their own they also end up rather dry since there’s not much outside the handling of the combat. You have to camp frequently, eat and recover morale/fatigue. It’s essentially tactical hack and slash with a very bland setting and “flavor”.

The CRPG Addict of course played and completed both:

Wizard’s Crown (1985): P1 | P2 | P3 | Revisit

Eternal Dagger, The (1987): P1 | P2 | P3 | Won

And this is what he has to say:

“I can only say that I’m glad that they simplified it for the Gold Box games, because there are enough statistics and options in combat to give a migraine to Sun Tzu.”

My opinion is slightly different. A lot of what doesn’t work with this game depends on interface and controls. Moving around and switching the facing are extremely clunky, the command menu is a list of 20 letters you have to memorize, and not all mechanics are immediately clear even in the manual. But the system overall is elegantly designed and manages to make all the right choices: skill based progression, tactical movement and facing, weapon classes with perks, damage types, armor as absorption.

Once you have the overall picture the system is actually simple. Its basic matrix is well done and built on what matters. The problem is again that once put in practice the handling was rather clunky.

Tabletop RPGs, D&D and TACH0: final analysis final (part 6)

Part 1Part 2Part 3Part 4Part 5 – Part 6 – Part 7

Sub-part 3 (of 3) on TACH0

How about not even knowing if you hit, or if you did, if the hit did any damage?

It’s an interesting mechanic but I don’t think it applies all that well to D&D and similar.

Realistically you know when you swing your weapon if you’ve done any damage. It also goes against the most current trend. In general players roll their own actions, and rolling for damage is part of the satisfaction of “swinging the sword.” So in classic PnP it makes sense that you roll those dice personally.

There’s also this modern trend of letting the players roll ALL the dice. It comes in two flavors, either the NPCs and monsters have fixed stats, or every possible action is turned into a pro-active player action. For example in combat the players either roll their attacks, or their active defense. So when it’s the monsters’ turn it’s once again the players performing their defenses against incoming attacks.

Even in D&D you don’t usually know the exact damage because while Armor Class is usually known, the monsters’ HPs are instead generally rolled by the DM and kept hidden. So the system naturally mimics this aspect of realism. You know what kind of damage you’ve done, but you don’t know when combat is going to end.

It’s interesting to know there used to be referees in wargames too. It really does seem there’s not a single original bone at the roots of RPGs ;)
Part of this thread was about a “what if”, taking RPGs and removing one of their core parts, the DM. And so the idea of “solo boardgame” applied to a roguelike, dynamical world but with boardgame style mechanics. Just another hybrid (with no original bones, too).

Anyway, after the last pass above I think I might have essentially exhausted this THAC0 controversy. I’ve read a few more things but they only focus on aspects within the analysis above.

There’s an interesting one here, that deals with the mathematical translations more competently:

If you scroll down to the comments, the first one is from Mike Mearls, the lead designer of D&D 5e.

This discussion precedes 5e, so it’s interesting to know that in the end they decided to go with what I described above as the third method, the one where both the attack bonus and AC are ascending.

Now… that blog is well written and tries to promote this True20 style, that corresponds instead to the second method described previously. The problem is that it doesn’t take into consideration the practical use. That’s what someone points out in the comments:

I think that attempts such as this to redefine the THAC0 system into a ‘better’ formula fail for one simple reason.

I think that they fail on actual use the *in the game*. […]

A d20 system (and, in truth, the original THACo system) works better in actual play because each actor need only know their own part of the equation.

And, in my opinion, of those an ascending AC system is preferable because it operates through addition – and because it deals with the better ACs (those that would be negative in a descending system but are merely larger numbers in an ascending system) in a more intuitive way.

And another:

I think it comes down to the fact that you must consider the source of the of the variables, and how easy each is to obtain. The d20 usually comes first, whether playing or GMing the die is already rattling around in my hand when it comes time to make my attack. Level and mods for the most part come from my character or monster’s stats, likely on a piece of paper right in front of me. The AC, however, is external. I must query another human for that value. Therefore, I find it easiest to keep it on one side of the algorithm all on its own, so I can calculate one side all by myself, and then query the other human for the AC value to compare to my result.

The blog author responds:

It’s not a terrible point that the 3E system has all player-scores on the left and the DM-score on the right of the inequality. But that doesn’t mean that it’s easier.

In practice with the Target 20 system (which is indeed how I run all my D&D games), the player performs d20 + FtrLevel + Mods, and reports that to the DM. Then the DM adds the AC and compares that to 20.

I’m confident that this is still significantly easier than the 3E system.
(1) It’s easier for the DM to lookup, or even flat-out remember, a single-digit DAC than a double-digit AAC.
(2) I’m confident that it is indeed easier to add a small single-digit number and compare to 20, than to compare to any random, large, odd number.
(3) If the player reports a raw score of 20+ (actually, as long as the first word out of their mouth is “twenty…” anything), then I can just ignore the second digit, skip the last operation, ignore the AC entirely (!) and announce a “hit”.

But I don’t think (3) is correct, since descending AC can go negative. And I’d argue about (2) too.

I don’t think that taking a number like 12 (told by a player), adding 7 (hypothetical AC, known by GM), and comparing the result to 20 is actually faster than taking a number and comparing it to an AC you have under your eyes. Simply put, I think that addition + fast comparison is always consistently slower than one single “slower” comparison. Taking a number, mentally add another and taking the result requires more memory work and mental gymnastic… I can’t even imagine this being subjective.

Doing 12 + 7, already demands more than comparing 12 to 13. It’s a “miss”. The first case falls short of the “20” target, being 19, the second case 12 is obviously not equal or above 13.

Ultimately the best algorithm is the one that can handle well both cases: starting with AC hidden and then letting the players know the odds before rolling the die, after they figured out the mysterious AC through experience.

Let’s see the breakdown of all four methods in each of the two most relevant cases:

a) keeping AC hidden (requires knowing the roll)
b) knowing the exact odds (requires knowing AC)

1. THAC0
a) THAC0 – roll = the minimum AC you hit
b) THAC0 – AC = minimum roll you need to hit

2. True20
a) roll + attack bonus = DM adds AC, sees if 20 or above (but the player will mentally calculate 20 – the roll to eventually guess the AC of the monster)
b) (20 – attack bonus) – AC = minimum roll you need to hit (but not very smooth)

3. 5e
a) roll + attack bonus = the max (ascending) AC you hit, if the monster AC is higher, you miss
b) AC – attack bonus = minimum roll you need to hit

4. roll under
a) roll – attack bonus = the minimum AC you hit
b) attack bonus + AC = the number you need to roll under

THAC0 is not intuitive but works fine after you memorize the procedure and stick to it.
True20 is a bit gimmicky. It works superficially, but gets clunky if you want to grasp the mechanics.
5e is the most straightforward despite the larger numbers due to using 10 AC as the baseline.
The “roll under” works best in the second case, not so much in the first (and it strays away from the tradition of rolling high).

If you read 5e rules there’s no mention at all that the DM should keep monsters’ AC hidden. The formula given implies an explicit AC value, so no reference to option (a) here.

With both cases the “roll under” mirrors 5e, but since AC is descending the numbers are overall smaller. So without a preference for the two cases the “roll under” is one step ahead. And if we consider that players start with (a) and then would eventually settle with (b) the consequence is that (b) is more important in the longer term, and the fourth method is the one that makes it easier.

But “roll under” still has to deal with negative AC…

Tabletop RPGs, D&D and TACH0: final analysis continued (part 5)

Part 1Part 2Part 3Part 4 – Part 5 – Part 6Part 7

Sub-part 2 (of 3?) on TACH0

Still a couple of things to add to the above. After I wrote all that, I decided to search and read online if there were similar analysis of the THAC0, and I found a few.

There’s more historical insight about the wargaming origins of the system:

In the beginning, Gary Gygax played wargames. In wargames, you would have something like an Attack value and a Defense value. You would also have a table on the game’s rulebook: If attacker’s attack value is x, and the defender’s defense value is y, you roll a die and cross-reference the result against the chart (attack values on the x-axis, defense values on the y-axis) to see if you scored a hit.

Specifically, he played naval wargames. The term Armor Class refers to ships: how thick, and how well-covered the ship was in armor plates. An AC of 1 was very good: it meant first-class armor. AC 2 meant second-class, and so on, such that a higher numerical value for AC meant that the protection granted by the armor was worse, and so it was easier to score a damaging hit against the ship.

But it’s interesting because no one seems to analyze the same aspects I did above, like the split between knowing whether you hit or not, and the odds of the roll. Instead they make DIFFERENT observations that even lead to another set of formulas.

So again if I thought I exhausted the topic, I was wrong again. My perspective if wholly mechanical, or practical when it comes to “smooth” the formulas so that they are simpler to use. But their angle is instead completely different:

A player of the game ISN’T SUPPOSED TO KNOW the Armor Class of a monster. Because this adds mystery to those monsters. It increases uncertainty and tension in the game. The idea is that you are supposed to role-play, not play the rules. And that means that sometimes the mechanics are better hidden.

How this converts to the THAC0 if the system is about subtracting Armor Class from that THAC0? It’s simple, you have to move around the formula again:
1a. roll + attack bonus = THAC0 – AC
1c. AC = THAC0 – (roll + attack bonus)
1d. (simplified) AC = cumulative THAC0 – roll

Let’s say you got TACH0 16, with a +2 strength bonus trying to hit a monster with AC -1.

The standard formula says you get to a cumulative THAC0 of 14, then add the 1 for AC. So you need to roll 15 or more.

Under the new method first you roll the die, let’s say you roll a 12:
14 – 12 = “I hit Armor Class 2” (then the master will say if it’s enough, it’s not in this case)

Now I wonder… how those other formulas deal with this other approach of keeping AC unknown?

I’ve written down some examples and it seems the third method (ascending AC, roll over AC) is the one that works best by far:
It’s just roll + attack bonus, the result is the Armor Class you hit.

1d20 + attack bonus = AC being hit
You roll 12 and attack bonus is 7? You hit up to AC 19 (or 1 in AC descending)
You roll 8 and attack bonus is 3? You hit up to AC 11 (or 9 in AC descending)
You roll 19 and attack bonus is 5? You hit up to AC 24 (or -4 in AC descending)

If instead you wanted to know the odds, in those two examples, while knowing the exact AC:
AC – attack bonus = target number (to roll equal or above)
19 – 7 = you need to roll a 12.
11 – 3 = you need to roll a 8.
24 – 5 = you need to roll a 19.

Do we have a winner? Is there anything else to all this? The THAC0 keeps (mis)giving.

Tabletop RPGs, D&D and TACH0: final analysis (part 4)

Part 1Part 2Part 3 – Part 4 – Part 5Part 6Part 7

Sub-part 1 (of 3?) on TACH0

A note & update on what’s written below:
All this, including what came before and is coming after, is extremely convoluted and confusing to read, even for me. But the purpose wasn’t to create a well organized and easily readable analysis of some problem. The purpose instead was to chart as precisely as possible my own journey. To reproduce the order of every step, in the way they happened. To follow this path, along with all the winding sidetracks, and doubts, and switchbacks, and dead ends. It is extremely convoluted to read, but it is meant to preserve the fun of the curiosity & discovery.
It reflects my own preference: I don’t want to read an history book, I want to read about the personal journey of the historian. Traverse history by digging new paths.

I should probably splinter this to some specialized forum. There are a few but I have no idea where all this mechanical wanking might be more tolerated… While I’ll probably keep here some analysis about the older stuff because if I do that elsewhere and start making assumptions they’d probably skin me alive.

I was quite satisfied with what I wrote above and the turning around of the system I currently use, but as it often happens I read something else and suddenly all the doubts are back and I have to reconsider everything from the start. In that endless cycle.

But before going there, there’s something else to say about the THAC0 and its alternative(s).

Summarizing, the classic THAC0 is inelegant because it’s hard to remember intuitively all its aspects. You have a descending THAC0 value, when you level up, that depends on the class. Then descending Armor Class you have to subtract from that first value. But not always because AC can go below zero, and so in that case you have to add the number to your THAC0. It’s not that complex but it never gets “smooth”. You often bump into some mental hiccup when you try to straighten these ascending and descending scales.

This is generally all very obvious and plain but while writing this I realized a few more things. There’s a difference between knowing whether you hit or not, and knowing the probability of the roll.

In that standard THAC0 roll you can already “embed” some calculations. Strength bonus and eventual magical weapon bonus can already be subtracted from the total before combat starts. Whereas in the other method your attack bonus will have to be added every time to the roll, because the Target Number is fixed. So the first system is compressed to a subtraction + comparison, while the second system is still addition + addition + comparison. (one subtraction VS two additions, now it looks more balanced)

It doesn’t stop there. That’s to know whether you hit or not, but to know your basic chance you need to know the number you need to roll on the d20. In the first method you already know. The single subtraction gives you the raw TN to roll. Instead in the second method you need to add attack bonus to AC, THEN subtract the result from 20. That gives you the actual TN to roll on the die. Here it becomes one subtraction VS addition + subtraction.

Let’s see, you ultimately want to know the number you need to roll on the die, so that’s the variable to find:
1a. roll + attack bonus = THAC0 – AC
1b. (becomes) roll = cumulative THAC0 – AC

The THAC0 and attack bonus can be aggregated because they are both going to be static. The variable parts that cannot be pre-calculated are the die roll and the AC of the target.

In the second method you have, instead:
2a. roll + attack bonus + AC = 20
2b. roll = 20 – (attack bonus + AC)
2c. roll = (20 – attack bonus) – AC

You could aggregate the attack bonus to 20 and write it down, because they are both static. But it goes against the grain of this formula. So that you have two different ones. One to know if you hit, and another to know the odds (the roll).

I thought there wasn’t anything else to say in what I wrote above. Not only there was more to say, but it even lead to the opposite conclusion: the THAC0 might be conceptually clunky, but works better in practice for what matters! (knowing the odds)

Is that all? Nope, there’s a third method:
2c. roll = (20 – attack bonus) – AC
3a. roll = AC – attack bonus

This was possible because both AC and 20 are static (I’ll explain better below).

And so we distill the three methods to:
1b. roll = THAC0 – AC
2c. roll = TN – AC (going against the grain, but same as above)
3a. roll = AC – attack bonus

With the second system I didn’t consider the possibility of AC going LOWER than zero (because Dark Dungeons, where the second system comes from, still uses descending AC values). For example with magic armor. In the first system you just have to consider that negative AC is added to the total THAC0. So it’s now even simpler (as long you remember the rule). Just one addition. But with the second method you’d have to add the attack bonus to the die roll, then subtract AC… Only to realize it’s more immediate if you simply add the negative AC to the TN of 20. It becomes addition + comparison + addition (shifted to the TN). But to know the odds you’d have to subtract from an addition. It suddenly became even messier.

In this case it comes more naturally to consider die roll + attack bonus on one side, and compare it to 20 + AC (when it’s below zero). It’s an even simpler calculation, but it requires to keep in your mind these two-way resolution, where you add AC to the roll when AC is positive OR add AC to the 20 when AC is negative. It’s once again a bit “fiddly”.

Here comes our third option!

It might even be obvious: what if we turn around AC as well and make EVERYTHING ascending? If the crux of the problems with THAC0 was its descending nature on its improvement (as you go up in level, TACH0 comes down), then we can as well apply it to AC. Again with a mathematically identical mechanic but even more simple to handle. Maybe?

We have once again the die roll + attack bonus. Nothing changes here. But this time the roll needs to be on Armor Class instead of a fixed TN of 20. We essentially have the same addition, addition, comparison, with some slight shuffling. AC is now calculated by adding 10 to the AC you read in the monster or character description. And that’s the number you have to directly roll over.

(If plate is 3 in the classic AC system, it becomes 10 – 3 = 7 in this one, if leather is 7, it becomes 10 – 7 = 3. Negative AC of the classic system is added to 10 here. But once this is written down no more subtractions are needed. But see below, because that also can be compacted and simply use 20 – AC.)

The advantage should be obvious. The comparison isn’t against a fixed number, but it’s found automatically as soon you look up AC. Adding 10 to a number is the most simple operation, so that second addition in the formula is much simplified. Say for example you are dealing with magic AC of -2, in this system you’d read something like 12. To which you need to add 10. It’s immediately a 22, that you have to roll over (after you add the attack bonus to the roll). It would become more complex if you had AC under 0, in this system, but it won’t happen here because 0 is the value of someone wearing no armor at all. It’s the baseline.

Not only these three systems are mechanically equivalent, but it also means that conversions can be immediate.

Between the first and second method the only difference is in the class table that gives your progress as THAC0 or attack bonus, then it’s only the formula being different.

Between the second and third we only have to convert descending AC to ascending. And that’s quite immediate: 20 – current value.

This third system seems to blend the good parts of both, it’s easy to use for both deciding on a hit, and knowing the odds. You either add the attack bonus to the die to see if you hit, or subtract the attack bonus from Armor Class to know the odds.

…But then I wondered, isn’t there a way to convert that one subtraction to know the odds, to simply another addition?

We’re dealing with Armor Class and a fixed bonus, so why couldn’t they be organized to add both together and obtain a Target Number? But since my mathematical competence is rather poor, I couldn’t decide whether this was a real possibility.

It turns out it’s possible: you just have to turn around something else in that formula.

The key to make it work is that the roll itself needs to change. You don’t roll anymore above the value, but below. And AC, since it’s added to the attack bonus, needs to be descending, so that lower values lower the odds.

The only thing left to do is slightly shifting the numbers, because we are starting from 1 on the die, not a zero. And so that +1 needs to be factored either in the attack bonus or in the AC.

For example, let’s say you want to hit some guy in plate armor (3 classic AC descending) no Dex bonus, with a total attack bonus of 4. In the THAC0 system the attack bonus becomes a 16, from which you subtract the 3 AC, to obtain 13, the number you need to roll on the die to hit. Or 40% (success on 13-20, 5% each). In this fourth system, instead, you simply add the attack bonus to the AC. 4 + 3 = 7. That’s the number you need to roll under, but that needs to be shifted to 8 to account for that +1. The result is the same 40%. (you can then incorporate seamlessly the +1 in the attack bonus itself)

I wonder if there’s a D&D based system that uses this fourth method.

But anyway, all this is only part 1 of 3. Me trying to come to terms with mathematical skills I don’t have while trying to find solutions to problems solved decades ago.

DIGRESSION #1: Tables VS Formulas (this was where I wanted to go next, at the time… It didn’t happen)

Tabletop RPGs, dice wars: linear VS bell curve (part 3)

Part 1Part 2 – Part 3 – Part 4Part 5Part 6Part 7

This time I prepare for THE HOLY WAR.

Dice wars: linear VS bell curve.

I’m lost into many other sidetracks but I was now trying to wrap my head around some very basic mathematical problem and I could as well write it down since I stumble on this constantly.

This isn’t about the solo boardgame project and more about my other project on the very complex combat system, but I guess I’ll use this thread to write about all the things (I want to make an analysis about two of the lesser know systems, Chivalry & Sorcery and Ysgarth, next).

The theme here is about the basic mechanics of a to-hit roll.

If we consider on one side the classic D&D d20, and on the other side systems like RuneQuest or Harnmaster, that are based on d100, it doesn’t seem there’s a concrete mathematical difference…

For example if I have my sword skill at 50%, if I roll the d100 under 50 it’s a hit, if I roll over it’s a miss.

The difference from the d20 seems to be just about the granularity: the d20 progresses at 5% steps, in the exact same linear way. So let’s say my TACH0 is 16 and I’m hitting some guy with 5 Armor Class, I roll 1-10 it’s a miss, I roll 11-20 it’s a hit. We got the exact same result of the system used above.

TACH0 improves through leveling up, with 5% increments, in the same way a skill system does the same thing, with whatever step of increment you decide to use.

Are we dealing with mechanically identical systems, at a basic level?

What I was trying to figure out is how this (a linear distribution) compares to a system like GURPS (3d6), using instead a bell curve.

We miss some granularity because the results go only 3-18, so -4 steps available, but also due to the different distribution the most common results are in a narrower range. And basically the ten results between 6-15 are roughly 80% of all possible results. Compared to a d20 where that 80% is mapped across 16 results instead of 10.

What are the merits of a bell curve, then?

Because in the end, for a to-hit roll, the value is still a target number. If you use 3d6 and have to roll 11 or better to hit, then you still have the exact same mathematical probability of the systems used above.

If that’s true then the mechanical difference can only be when we move to consider the progress system:
+1.39 +2.51 +4.9 +6.94 +9.73 +11.57 +12.5 +12.5 and so on in reverse.

It’s essentially what’s being shown in the third column of this chart, from the compiled rules of Traveller:

It translates to: slow to both learn and to master. If you match this with a level system where you gain a +1 bonus every level, the result is that the character will always miss for the first few levels, then rapidly improving, and with the higher levels a progressively slower growth. (but not quite, GURPS makes the cost of the +1 variable on your Dex value…)

I’m trying to distill “meaning” out of all this.

Essentially, when you have a target number then bell or linear curve make no difference. You either hit or miss, it’s a binary result and the distribution doesn’t affect that result. For example, if we want to use a 3d6 bell curve, but want it to map on a d100 system, we can look at that image above, the 5th column. Those are the “steps” available with the 3d6 granularity. So, in a d100 system, you could have a skill at 16%, and that’s exactly the same as a “7” on a 3d6. By rolling 3d6 you continue to have 16% of rolling below. Simply to say that rolling a d100 and succeeding only on a <= 16 is mechanically equivalent as rolling a 3d6 and succeeding on a <= 7.

It affects instead the progress system, or the way that target number “moves”. But with a d100 I could mechanically copy the outcome by designing a progress system around the same values of the one based on the 3d6.

This, oddly enough, was triggered by reading about an OSR system (OSR = Old School Renaissance) called “Dark Dungeons”, that mimics the rules of the basic D&D set.

What was explained in the rules is that we all known the clunkiness of a system like TACH0 and there are always debates about what’s the best system.

Dark Dungeons replaces it by using a system that it claims to be mathematically identical, but that is much simpler to use: instead of a TACH0 value, you have simply an attack bonus. It’s given by your class and Strength bonus. So the to-hit roll is:
d20 + attack bonus + armor value of the enemy.
If the result is 20 or higher, you hit, if not, you miss.

Armor value in this case is descending but it works because you don’t have to subtract it from some arbitrary value, you simply add it to the roll. So if someone is without armor his value is 9, and that will obviously push the result closer to 20.

Ultimately it’s all about ease of use and nothing else, if the system is mathematically identical then there isn’t anything else to consider beside that ease of use.

While looking into this I found this page: http://www.darkshire.net/jhkim/rpg/systemdesign/dice-motive.html

It easily explains there’s a hierarchy of “ease of use”: comparison is easier than addition, which is easier than subtraction, which is easier than multiplication, which is easier than division. (for “comparison” I think it means “less than” or “more than”)

The classic D&D TACH0 is made of: dice + attack bonus, compared to armor value subtracted from a class-dependent target value.

Versus the simplified system in “Dark Dungeons” that is: dice + attack bonus + armor class, compared to target value (fixed at 20).

So: addition, subtraction, comparison VS addition, addition, comparison.

This is simplified in two ways, one addition replaces the subtraction and the comparison is always against a 20, so you don’t have to look it up. While both attack bonus and armor class have to be collected every time in both systems, since they are variable.

From that sidetrack the problem becomes: is there something that a bell curve offers and that can’t be “mapped” on a d100 system (or other linear systems)? Is there something that is intrinsically different?

In general, the distribution of the possible values does matter, for example the typical damage dice. Rolling 1d12 is different compared to rolling 2d6, because you’ll statistically obtain a 12 much, much more frequently by rolling a d12. But even in this case, if you are against a dragon with a million hit points, it doesn’t matter. Because over time those rolls will even out around the middle, and once again the two options become identical.

The general consensus (and math, we aren’t dealing with “opinions” here, the only opinion is the potential lack of insight) is that the bell curve is more “skill based”, and so less random. The distribution of those results is narrower, so the outcome is usually more predictable. Compared to a linear system where instead the outliers are just as frequent. In fact, the bell curve is also often matched with dice pools (Earthdawn being a good example), and especially with exploding dice. It means results are common (same-y), and when they get uncommon they have the potential to become truly extraordinary.

(But I’m not a fan of exploding dice. Even if they happen rarely there’s always this possibility the system goes completely out of control. It simply makes randomness king, eventually. And it is randomness that can happen at very illogical times. I understand the appeal, but I prefer something more structured.)

…And then I thought, what happens if I try to mix up everything?

My current system, at a basic level, has the to-hit roll based on the standard d100. You have a % skill, take a 1d100 and need to roll under your skill.

…What if I replace the d100 with a 2d50 (because being based on computer code I can afford something weird)?

The resulting bell curve is a bit more flat, but we obtain that “realistic” effect of having more predictable results. The whole range that goes from 2 to 23 is cumulatively just 10%. 26 is where the probability is 1%, so matching the linear system, and it grows to 51, where it’s 2%, and then goes back down to 1% at 76.

2-23 = 10.12%
24-75 = 76.88%
76-100 = 13%

There are two direct consequences of this experiment. The first is that I have to redo the criticals, since the range 2-10 is less than 2%. GURPS for criticals uses 3-4 and 17-18 (plus a few quirks, but let’s keep those out), so still less than 2%.

But more importantly, the whole idea of score percentages goes out of the window because your skill value isn’t mapped anymore 1:1 to the probability of that number. If anything, it becomes misleading.

So I though… Let’s go further and make more radical changes. I keep the 2d50, but with score percentages gone I can as well turn it upside down, so you don’t anymore roll UNDER, but over (as it happens above with Dark Dungeons). Let’s say that instead of rolling under a target number you have to roll over a fixed value = 100. So now we have the bell curve of the dice that gives us the baseline number. That’s random chance and it will follow the bell curve, so with the most typical results (80%) in the 25-75 range. To obtain a 100 I then would need some sort of skill value that goes from 75-25 (at minimum), of course.

In the end I would have mapped GURPS onto a 2-100 wider range, slightly flatter “bell”, then reverse it so you have to roll over, and using a fixed target number instead of a variable skill value. It’s not impressive, but it should work fine.

The bell curve is intrinsically counter intuitive. There’s no fast way to extrapolate the exact %. So that’s an aspect that is unfixable. Linear systems are immediately explicit, whereas bell curves are more opaque but simulate better a certain behavior.

At this point I started going a bit too far and wild with ideas. For example I imagined a system that came out of the desire to have a melee combat where every attack carries a momentum to the next. So instead of having a fixed skill value, it might work like first you roll the dice, let’s say you get a 30. So now you need at least 70 to hit 100. And you could have instead a skill “pool” that you can use. So in this case you have a tactical option. You either “push” the attack and spend 70 points (but maybe leaving you out open for a counterattack, if you spend too much on the attack and don’t have much left for defense), or you can instead not spend those points and wait the next turn in the hope for a more favorable roll. It even gets interesting tactically because you could try to “build up” an attack, but it would always be a gamble because if you got hit before you next attack you then would also lose all your momentum with it.

…So I’d have a system where every turn you get a “pool refresh”, maybe based on Dex, or Initiative, or both, and where you carry over the values from the previous turn, in one flow.

This gets too fiddly and with too much bookkeeping even for a computer system. But maybe I can replace those numbers with “tiers”, like every tier is +5 points. That way I can mask away the lack of explicit percentages inherent to the 2d50 (while retaining the benefits).

…But that would gave gone against ANOTHER idea I had. The problem with skill based systems is that the basic statistics become almost irrelevant. Once you get your skill value that’s all that matters, and the stats only give some weak bonuses here and there. The D&D d20 isn’t that different and in fact it’s not skill based but class based. So it’s actually the class that has the biggest impact on what your character can or cannot do.

My idea here was about making the progression of the skill directly based on the stats it is derived from. The basic stats that you generate during character creation would be your natural disposition. So for example a guy with very high Dexterity makes for a potentially amazing archer. But it’s not mandatory, you get good only if you practice enough. On the other side, if you have a poor talent you can practice as much as you want, but you will still struggle a lot compared to the other guy. How this translates to the game? You make the SKILL progress depend on the STATS. So your improvement with that skill, over time, directly depends on the stats that rule the skill. And you could still focus a lot and improve in a skill that doesn’t come naturally to you. You’d just observe a slower progress compared to someone with a natural talent. …But all this goes against the idea of fixed skill tiers to use (because the rate is directly variable in its granularity).

How to fix? Well, I thought of separating the systems. So you have “points” (like Dark Souls’ souls) that you spend to buy skill “tiers”. But the COST in points of the next tier depends on the stats. So a guy with high Dex will have to spend less points to buy a tier in the bow skill, compared to a different guy with lower Dex. And the tiers still end up fixed +5!

And it goes on and on. I shouldn’t waste time writing all it down, but I kind of need to, or face the risk of going again and again through the same ideas.

In any case, if you take it as a whole, all this brainstorming lead to a clunky (but potentially interesting, once massaged into a better form) system that looks VERY similar to the one used in The Riddle of the Steel. Also because once you use a variable system like this one you the realize you don’t really need anymore any target number. You just let the player decide the “intensity” of the attack and see what your opponent does with it. For example a weak attack that in the classic system would be an automatic “miss”, here could still land if the defender doesn’t defend, and still do a little bit of damage (because in my system the to-hit roll has influence on damage done with that attack). But maybe the defender went for that option to prepare a much bigger counterattack by not wasting his own points on that defense for that weak attack and pour them all on his own, stronger attack… It’s tactics bound to a nice random bell curve!

And this brings me to the last point: I still need to do more basic research.

This isn’t a wide argument just about linear systems versus bell curve. There’s a third option that is extremely popular, especially in all the more recent games. And it’s all about pools of dice rolled against a target number. The downside of these systems is that they are VERY OPAQUE, in the sense again that the probability of success isn’t immediately explicit, and calculations are non-trivial. Sometimes the rules manuals hand you a nice table, but quite often you are on your own. But on the other side it’s a system that embraces much more directly both the opposed rolls AND variable successes.

It’s not anymore a binary hit/miss, but how good was the hit and how bad the miss. In some elaborate systems, like those used by Fantasy Flight, you directly have “narrative” side effects:

That, too, can be mapped out of the narrative and in a more mechanical way, but I think it’s still too fiddly.

I certainly need to research more the basic advantages/disadvantages of a dice pool compared to the other two. Only when I know all the technical details I will be able to do some more fancy experiments.

Squaring a circle would be too trivial, so I’ll have to find some way to take this triangle and square it into this circle. Possibly with a very big hammer.

(All the while, reinventing wheels. If it wasn’t already painfully obvious.)

Destiny 2: minor lessons in game design

Old style posts! In 2021!

Today Bungie released a sort of manifesto detailing their ever moving plans for Destiny 2:

The main feature is the removal of “sunsetting”, a truly awful solution to a problem that the previous game director forcefully pushed, despite it was obvious from the very beginning that it would fail.

Since the explanation is terse and straightforward, why not commenting it.

Destiny is a loot-based game where weapons and armor you acquire can be leveled up to the current cap. Therefore every item is virtually always viable, as long you keep upgrading it.

This created a “power creep” because in order to make players chase new loot, devs had to make that loot more appealing by making it more powerful. Giving players incentive to leave their former stuff behind and adopt the newer stuff. Because new = better (more powerful).

The consequence of that is that something powerful arrived, and then had to be eventually nerfed, in order to fight that power creep and make the game once again balanced.

“Sunsetting” was meant to solve that situation and avoid the powercreep. It works by creating a smaller, curated loot pool, by giving all loot a fixed expiration date. It means that new loot that is added doesn’t need to directly compete with old loot, because old root is pulled out. New stuff in, old stuff out. Not being there, it doesn’t create competition with the newer stuff, so the newer stuff doesn’t need to be more powerful in order to be more appealing.

(This is what they want players to believe. The truth is that curating a small loot pool requires less work than a giant loot pool, and Bungie has serious production issues and is looking for ways to cut costs.)

Why did it fail, and it was obvious it would, without the need to test it?

Because it’s all smoke and mirrors without any substance. What the model DOES, instead of what bullshit story it tries to make you believe, is making players go through hamster wheels. Like all progression systems that aren’t built around content.

Look at this scheme:

– Before-sunsetting > the player goes through the hamster wheel because there’s a tasty reward at the end (a more powerful weapon).

– With sunsetting > the player is pushed through the hamster wheel, to PAY BACK A DEBT.

It is not surprising that the feedback loop where you pay a debt feels worse than the one where you are rewarded.

That’s really all there is.

Removing your current weapons, so that you are forced to obtain new ones, and so avoiding competition between old and new, doesn’t fix ANYTHING AT ALL. It’s a solution to the effect instead of the cause. It’s very clearly silly and misguided. A blind, partial point of view driven by convenience. Incomplete analysis leading to broken solutions. Trump’s way of being, to make an obvious example: saying things that are convenient, BELIEVING they are true. The true mark of an egomaniac (and that can always be verified, since they always lack motivation and proof, when you dig, like in this case).

As I said, all progression systems work like that. The difference is that they are usually a part of a bigger system. They are built around content. A progression system by itself is a pointless hamster wheel, even when it’s a well designed one. Destiny is a game with a severe lack of content, so the best solution is to maximize what is there, give it value. Bungie chose the opposite, employing the standard MMO technique of mudflation: removing the relevance of old content. Yet it works for MMOs when there’s content to offer, and it fails for Bungie because they cannot produce enough. In this scenario, “mudflating” the little content they have equals shooting themselves in the foot. Emphasizing hamster wheels, rather than content, and then pushing players through them by creating debts, only damages the ecosystem further. Until the game is left to bled out.

Game design isn’t politics. You cannot bullshit players through sleight of hand of a carefully worded a blog post. If your game design is full of rhetorical bullshit it just won’t pass the test of reality.

In the world we’re imagining, we’ll have space at the top end to create powerful Legendary weapons. Legendaries that are just better than other items in the classification. We’ll be able to do that, because the design space for weapons will expand and contract over time. Items will enter the ecosystem, be able to be infused for some number of Seasons and beyond that, their power won’t be able to be raised. Our hope is that instead of having to account for a weapon’s viability forever when we create one, it can be easier to let something powerful exist in the ecosystem. And those potent weapons entering the ecosystem mean there’s more fun items to pursue.

Legendaries that are just better than other items [because the other better items are gone].

Better not because they are better, but because there’s nothing to compare them to. Sleight of hand.