Rebaked NPCs Pt. 2 (+ linux build!)
about 1 year ago
– Fri, May 09, 2025 at 01:35:35 PM
Hey there! Surprise bonus post 'cause the discussion about incorporating Kai's NPC Rebake last week unexpectedly popped off — y'all shared a lot of perspectives we both had and hadn't considered. Thank you to everyone for your input!
We all want the best possible Lancer Tactics and there are a lot of mutually incompatible ideas for what that means. This is complicated, and the discussion revealed some expectations & taut threads leading into the heart of this project. I'm glad this gave us the opportunity to unpack them. That means we have a long post ahead of us here!
First, just to get everyone on the same page about Kai's NPC Rebake: it's not a rules overhaul; it performs a relatively conservative rework of existing NPCs to address problems that have made themselves clear in the six years of play since Lancer was released. A lot of you were excited about LT fixing these issues by using the Rebake, and there were others with genuine concerns about deviating from a near-1:1 port.
We've had a week to chew through your commands, and I wanted to write this post to regurgitate and further digest the discussion [ew, olive, that's really the metaphor we picked?] and share our process. Each section starts with quotes we pulled that exemplify the topic, sourced (sans attributions) from last week's comment section, Discord, and itch.io.
The TL; DR: after reading everyone's feedback and talking it through with Carpenter and Josh, our plan is to implement most of the Rebake's changes for individual NPC classes + veterans/ultras, but skipping its alternate rules for NPC structure/stress + grunts. Additionally, we're leaving open the possibility ofpulling from other supplements and our own designs to further patch non-PC-option mechanics as-needed.
Let's go!
(oh also I got a Linux build up and running! Hot tip it works surprisingly well in desktop mode on the Steam Deck. I recommend also installing the itch.io launcher app for easy updates.)
Impact statement; Change is scary; Let's get on the same page
The biggest change from the Rebake is the blanket removal of tier 2 and 3 multiattacks. There's a reason no homebrew supplements in the community use these; they make high-level Lancer extremely rocket tag-y. More broadly, it removes or adds loopholes to things that inflict the ever-unpopular Stunned. It also reduces Accuracy inflation to make Evasion builds more viable & cover more impactful.
I wrote up an NPC-specific summary of the current version of the Rebake here; I was gonna put it inline here but even listing the classes made this too long. If you want detailed design notes for the mentioned changes, I encourage you to read Kai’s design notes inside the NPC Rebake itself.
Will it make Lancer Tactics better?
> "I think if the Rebakes make for a better game then that is the way to go even if it makes them not 1:1 with the core rules."
> "do whichever you feel is best for the game. Use the rebake, or don't. Or hell, cherry-pick the best bits of each if that's what you feel creates the best experience"
> '"the physical game it was based on" is a bad excuse for poor user experience - if including those updates makes for a better experience for players of your game, then you should go for it.'
The most common sentiment expressed was that if it makes the experience of playing Lancer Tactics better, go for it.
At a minimum, dropping tier 2 and 3 multi-attacks allows high level play to not be games of rocket tag. Needing to survive casually-structuring Quick Actions flattens the diversity of possible player builds.
Getting granular, almost all of the changes I see in the Rebake will smooth over both experiential and implementation difficulties with the core NPCs. To pick some examples:
- The core Breacher's Painmaker delays its turn until after a target. Nothing else in core plays with initiative ordering like that. The Rebake simplifies this to "the next attack the Breacher makes".
- The Sniper's Moving Target has always been an optional that drastically changes the threat it poses: the difference between firing its rifle every turn and every other turn. The Rebake fixes this by making it a cheap way to apply Sniper's Mark instead.
- The Rebake's Goliath's Towering Stride allows it to comfortably straddle uneven terrain; the core version's movement would be severely hampered by lack of available standing room on most maps.
- Linking different conditions is hard for the engine; the Scout's Marker Rifle now applies Shredded like a decent mech instead of the weird exploitable "as long as it's locked on" thing.
On the other hand, there are a few changes the Rebake makes that are designed for giving the GM or players more options that would be more difficult or unnecessary to implement:
- The Bombard gets different shell types that cause the cannon to need to be reloaded for specific tactical uses. The AI is just going to be worse at making those judgements.
- The Witch's Predatory Logic has a "can't brace against this" condition; joke's on you Witch, we've already changed Brace so it can't be used against most attacks.
Luckily, I'm not an automaton, and when I run into a situation where implementing a mechanic from the Rebake doesn't make sense I can just skip it and leave things as core (note that even without the Rebake this would apply, so there's no "purity" that's being lost). What these will end up being is impossible for me to predict ahead of time; what I can say is that NPC implementation will end up being a bit of a Lancer Tactics-specific patchwork.
The effects of human GM moderation; AI is bad at picking fun gameplay options
> "in-person DMs have the luxury of tilting the scales a bit to keep things feeling fun and fair - the AI in a videogame has much less provisioning for that sort of thing"
> "i feel that core lancer does have some flaws that require a human gm to moderate. With a gm playing full send optimally, it can be somewhat miserable. As such, a sidegrade might be a good idea."
> "Personally I don't particularly care about the NPC rebake and don't use it at my table, but I also know that what works for a group of friends with a live GM is not necessarily what works for a single-player experience run by a program, and what works on a physical (or virtual) tabletop is not necessarily the best thing for a computer game."
The core rulebook is designed for tabletop play with a GM and players. GMs can adapt to situations at-will, even bending or breaking rules if they need to. A computer opponent has no such judgement, and runs a greater risk of falling into nonsensical play.
An example: I was playtesting the Linux build on the Steam deck and a Demolisher spawned on the other side of the Riparian Zone rivers. It couldn't figure out how to cross, so it spent the entire holdout just handing out lockons and basic frag sigs. A human GM would have:
- 1. not spawned a demo on the other side of a river in the first place, and
- 2. been better equipped to find a creative solution for what to do with a demo in that situation (have it build a rock bridge? spawn an emergency Mirage? simply retreat and not take up an NPC activation?).
As quoted in last week's post, the Rebake would have helped here by giving the demo a few more interesting things to do (ie golf clubbing a piece of terrain) when it's at range, while keeping its core identity as a slow area-denial slugger.
Tabletop compatibility; New player onboarding
> "One phenomenon I've noticed in my D&D groups is the 5th edition rules are constantly being compared to Baldur's Gate 3, or worse even confused between the two. Its not a huge deal, but it creates some dissonance I've found can mess with new player expectations."
> "it's for sure going to be confusing when new players who've arrived at the game via the LT pipeline encounter the regular NPCs"
> "To be exact, what I'm particularly worried about is that the homebrew rules will 'replace' the original CRB rules, as then I can't really help out the people I'm introducing to the game."
Being a good neighbor to the tabletop and functioning as an introduction to Lancer for new players is the strongest argument I heard for sticking with core. Here's my one thought about this, plus two categories of quotes from other folks (none of them a definitive answer to this):
These changes are GM-facing, not player facing
> "This is from someone who has only played and not GM'd lancer, but I think that the stats of the NPCs aren't a huge impact on how the player interacts with the game. We would still have the same mechs available as players, with NPCs rebalanced for a single player experience. I could still do all the cool mech things I had wanted to do (like blasting someone into swiss cheese with a Raleigh)"
> "I’ve run games for a lot of (newish) players that would probably not be able to tell even if they meet two Assassins in the same encounter using Core vs Rebake other than that they have different abilities)"
> "While i feel the PC mechanics / mech features should be pretty faithful, I really think the specific NPC stats are not that integral to the identity (there are many and you interact with each individual NPC a bit rarely)"
Lancer Tactics is already off-core due to adaptations
> "The core rulebook is already not a reliable guideline for playing the video game just by virtue of existing changes necessary for a video game adaptation. And that’s the free version that more of the Lancer Tactics player base will have access to than the paid crb the npcs are in"
> "but I don't know how much LT aims to be an entry point to teach you how to play and run the game at your own table, you'd still have to learn the pdf and brush up on the rules if you wanna run Lancer anyway"
We've already had to make some unavoidable changes as part of the change in format that dwarf the Rebake in terms of players learning LT and then switching to tabletop:
- The distinction between quick and full actions, like Skirmish, Barrage, and Full Tech, have been dropped in favor of "action points". This has some knock-on effects like being able to move between firing weapons and the awkwardness of not being able to repeat weapon attacks but repeating Invades as much as you want.
- Flight in tabletop is granular: you can move up and down as many spaces as you want. This would be a nightmare to make UI for so we simplified this to a simple status that puts you 3 spaces above whatever ground you're over.
- Brace can normally trigger any time you take damage, but it is very rarely the correct decision. We obviously don't want to spam players, so we changed it to only proc on taking 50% of your health in one hit or getting structured/stressed (and I still think it's too much).
And that's not to mention our changes to line-of-sight to be more computer-friendly and completely dropping downtime + pilot combat rules. This is to say: the fact of whether or not the enemy Ace's missile launcher has Seeking is a less impactful concern when talking about what new players will be learning.
Expected a 1:1 port; Perceived authenticity
> "what I wanted was a translation of Lancer the tabletop game, warts and all"
> "i am simply of the opinion that authenticity to core Lancer is important in making what is likely going to be the only Lancer video game to exist"
> 'A lot of it is just that it feels a bit weird to go from "lancer with some changes to account for the medium" to "lancer with published homebrew and QoL changes to account for the medium"'
> "I also backed the project under the thought it was going to be the base rules and maybe some neat olive exclusives."
I take my Kickstarter promises very seriously (and have gotten increasingly conservative with them over my previous campaigns due to occasional trip-ups), and the idea that you might be getting something significantly different than what you were expecting is something that gets my attention.
We ran this campaign on "we're making a Lancer video game!" Specific mechanical adaptation requirements — the Flying status, dropping Barrage/Full Tech, curtailing Brace's trigger — have been well-received even though they're substantial and player-facing.
Conversely, the switch from 2D to 3D was not in original Kickstarter plans, but the change has been well-received and has made for a more polished & stronger game.
This underlines that not all changes in expectations are created equal, so it's interesting to see that the idea of us willfully changing any mechanics elicited a strong reaction for some.
So, even though I understand hesitation about implementing mechanical patches like the Rebake, I can't operate as a game developer without the elbow room to make changes. But I also want to make everybody happy (something that's both normal to want and possible to achieve). Kickstarter tries to make it clear that projects aren't pre-orders & changes happen, but leaning on that legalese leaves a bad taste in my mouth. I think if you absolutely need LT to be literally core rules, then that's not something we can accommodate & I'm happy to send you a refund — though I hope you can instead trust us that you're still going to get the Lancer video game experience that you were hoping for.
> "As an intellectual exercise, pure 1:1 core rules would be interesting to play. Given the choice between Core and Rebaked, I'd prefer the latter."
> "All I can do is say, fellow backers, we backed because we are excited about Lancer AND we believed in the ability for this team to bring it to reality. I ask that we trust those in the trenches on how to make a Lancer game, no matter how different it may be from expectation"
> "...what we'd be losing here is something that every GM of every game alters, willingly or not, every time they sit down at the table. NPCs never use the full range of their abilities, never act in the same way from table to table despite having the same statistics, and don't necessarily have the same statistics anyway because everyone tweaks stuff all the time, even if it's just accidental. Preserving their exact implementation in perpetuity represents misunderstanding their purpose; they're a tool for narrative and tactical tension, and they should be freely adjusted by any and every gamemaster to make them a better tool. That includes Olive."
Dev workload
> "But ultimately whatever makes Olive happier and more satisfied with the project wins here, and whatever doesn't make it stuck in development hell."
> "I think it's fine, just please don't go and add a huge amount of workload."
> "In the end I just want a finished product that won't become abandonware"
We'll do our best! I continue to be paranoid about overscoping. The biggest advantage to the Rebake is how much of a drop-in replacement it is.
Modding
> "If people really want the original core NPCs, they can be modded in 🤷♀️"
> "The Rebaked npcs sounds like a great mod/supplement to let people make after release, but i'd rather have the option for them down the line, instead of HAVING to use them :)"
We currently have no timeline for mod support. I dearly hope that we're able to get it working someday, but there are enough unknowns that I'm not comfortable making plans that rely on it. I mentioned this in the original post but saw this suggested enough I thought it was worth reiterating.
Just make a new set of custom-designed NPCs!
> "I personally think the best method is to tailor make an npc roster for the game itself, in a way that makes coding easy and make the game fun and interesting."
> "I don't see why it wouldn't be possible to just do quality of life things that are commonly discussed amongst the community like making Juggernaut more interactive. Seeing how those changes play out in the game, which this will be primarily for anyway and not for tabletop use, shouldn't make that much of a ruckus in terms of playtesting."
So: the biggest reason we've been able to move so fast with so few resources is because the game's already designed, i.e. balance and playtesting are done; we don't have to implement anything only to throw it out later because it no longer fit. Even putting our Kickstarter promises aside, inventing entirely fresh custom-built mechanics would be a relative tarpit. There's a reason the homebrew mech Xiong Xiaoli tier was limited to one person.
However, this brings up: having dipped our toes into opting into any improvements to core, we've opened up the door to the question "if you're making changes, why not fix X?" We've signaled that we're now making limited game design choices, which means we're no longer a 100% neutral translator of the one Holy Text. Uh oh! How do we start making those calls?
The soul of this: what are we making? Lancer MegaMek or Lancer Tactics? A virtual tabletop?
> "As absolutely killer of a CRPG as I think it'd make, this isn't a CRPG, and comparing it to CRPGs is a bit messy due to the narrative focus of that genre compared to the tactical wargamey focus of LT"
An admission: we've been having our cake and eating it too. Lancer Tactics has been riding the line between "a Lancer wargame simulator" and "a CRPG that tells a story". Here's two pages from our internal lookbook that Carpenter made at the beginning of the project to get everyone oriented:
We're trying to thread the needle of giving people action figures they can set up and smack against each other AND having a specific voice with stories that we think are worth telling & using Lancer as the vehicle to deliver them. The former is what people are going to be excited and show up for; the latter is how I sleep at night spending so much energy on making a video game in the face of the Horrors.
Battletech has an unofficial straight port from tabletop to video game called MegaMek. It's all toy box; an automated virtual tabletop for playing that one wargame. As satisfying as making cool robots go boom is, spending three+ years making a Lancer version of MegaMek sounds so so boring to me! That's not how I wanna spend my life!
So the plan has been to make a toy box that equips folks to tell their own stories in the Lancer system & universe (which is what y'all showed up for), and then use those same tools to make some campaigns that'll follow through with voices and characters and settings that I think are important. We stitch the whole thing together in a loose modular anthology structure and we've made a platform for people to tell their own stories while setting the political tone with our entries.
With that as the big picture, you might see why I'm not particularly biased towards being a strict Lancer literalist. I love playing Lancer (it's been my sicko hyperfixation for like five years now) but my lens for what I'm hoping to do with it is wider than making a totally neutral virtual tabletop (I spent so much space in the Kickstarter's description talking about the planet Viridian's colonial & ecological situation). This means I'm not afraid of improving player experience by making changes as needed, as long as they maintain the core fun of putting builds together and seeing the Raleigh go pop-pop-pop-pop-pop-pop-pop with its seven guns.
> "All in all, I think you have proven through each update and instance of the project that you are not only open and honest about these things, but you desire more than simply making a port. THAT is what I am most excited by now, and originally when I backed. It's not just LANCER, it is LANCER: Tactics."
So we're changing things now? Where does it end?
As discussed above, our biggest production advantage in this project is that we've known the shape of the game we're making from the start. I was able to go heads-down and work on the engine for a year before having anything playable. This would normally be a death sentence for a game's production, but since we already knew Lancer was fun we were able to get away with it.
With that in mind, it's only with extreme reluctance (note: it was my idea) that I'm opening the door to us explicitly taking on a limited role of game designers in choosing what Lancer patches to bring into LT. This whole discussion has generated a number of explicit constraints that I'm going to be looking for when considering divergences from the core rulebook:
-
We continue to do minimal design work. The change comes as a tested package that benefits player experience; I'm skeptical of one-off "hey I bet I know how to fix the balance on this specific thing" ideas without them being thoroughly vetted.
-
It does not increase scope. We're not adding completely new stuff. Having an associated LCP is also required since we're set up to consume those for names and descriptions.
-
It doesn't complicate compatibility with tabletop. The changes we're picking up from the Rebake are mostly GM-facing; NPCs with different systems are well within the core lancer experience from players' perspectives. Modifying a PC option is last-resort ONLY (e.g. Skirmisher II).
-
It's forward-compatibility-friendly. Mods are not on the roadmap, but I want to leave the door open to them. Any design updates we make should not change something that later work is likely to rely on being a certain way.
In this case, Kai's NPC Rebake checks all these boxes. Other sources we may look at to pull from are TK's Lancers in the Dark or the standardization of sitrep objectives as Payloads from Ralf's Enhanced Combat. I could also see someday finding an update for dangerous terrain & environmental effects that fit within these constraints.
This also means there won't ever be one definitive PDF/book source of truth for Lancer Tactics as a whole. That... seems fine? An NPC roster with the relevant info exists in-game in the places where it's needed. And someone who's played LT will continue to be better equipped to sit down at a Lancer table and play than someone who hasn't.
Outcome; so, what's the plan?
- We are going to prioritize player-experience compatibility with the tabletop rules. This means we're not going to use the Rebake's alternate structure/stress tables nor alternate grunt classes. I personally think they're good solutions in themselves, but their changes-to-rules/things-they-fix ratio isn't enough to justify picking them up for LT.
- We are going to pick up most of the rest of the Rebake's NPC class changes. Most of them reduce reliance on GM sandbagging, reduce my overall dev workload by letting me spend less time implementing abstruse minutiae, and are a light enough touch that I don't believe they compromise our Kickstarter goals of porting Lancer to a video game (especially in light of the larger player-facing changes we've already had to make). I'll skip the changes that don't make sense for LT on a case-by-case basis.
- We also reserve the right to make further non-PC-option patches as gameplay demands. For example, even before all of this I was going to ignore the tabletop ruling and allow Berserkers get a last Aggression hit in before dying. These changes will follow the criteria listed above.
- I don't think I'll get to starting the next batch of NPC implementations until like October, so there's plenty of time between now and then to sit on all this and let it ferment.
There has never been a flawless game of Lancer played; the quest for the holy 1:1 RAW grail is a fool's errand; tables out there right now are doing PvP (bleh) or have dropped the 1/round limit on overcharge (woah!); chaining your early access game to a beloved-but-unchanging set of rules means that it can't ever grow into something better.
Closing words
I love this process. I love getting the opportunity to hear folks voice their thoughts and feelings as we all orbit around this shared idea of what Lancer is and could be. Thank you again for spending your time sharing your perspectives, both in public and private. Our core team a lot of time talking this over and have done our best to chart a course through more political and charged waters than we were expecting. Even if you don't totally agree with a decision we made, don't think we made it lightly. I tried my best to listen and understand & I hope you felt heard.
Alright, now for real I gotta switch over to my non-LT work to pay some bills. See you in July after the break!
🌺 Olive
