The first expansion of Arcforge (actually the second half of the original document, to my knowledge) clocks in at 76 pages, 1 page front cover, 1 page inside of front cover, 1 page editorial, 2 pages of introduction, 1 page ToC, 4 pages of SRD, 1 page advertisement, 1 page back cover, leaving us with 64 pages of content, so let’s take a look!
This book was moved up in my reviewing queue at the request of my patreon supporters.
This book requires the first Arcforge book, and for full use, you should also be using Ultimate Psionics and Akashic Mysteries.
Okay, while the book doesn’t start that way, let us begin as always, by taking a look at the class options presented within this book – these include two new archetypes, the first of which would be the biomech speaker druid, who adds Knowledge (engineering) to the class skills and gains a mech, as well as “biomech pilot” (should be capitalized) as a bonus feat, replacing wild empathy and nature bond.
This feat requires a bit of explanation: It nets you a partially biological mech, which unlocks a whole array of unique mech enhancements, which includes share spells, 25% to negate critical and precision damage that can be taken multiple times to upgrade to full-blown immunity, or treating integrated weapons as natural weapons. These benefits are potent, but considering the dual tax, and the per se plausibly minimum level requirements, they are valid. But back to the archetype: 4th level lets the druid 1/day as a standard action exchange one of the mech enhancements for another; 6th level and every 2 levels thereafter increases that by another daily use, and 8th and 12th level improve the action economy. The capstone delimits this daily-uses wise and replaces wild shape. Instead of resist nature’s lure, we have an untyped +4 bonus to saves versus Ex/Su, Sp and Psi-like abilities of biomechs, robots, and bio-engineered critters.
Before we go to the second archetype, we have to talk about the robot engine featured in this book, the modular robot engine: In addition to the robot subtype’s properties, the book champions them having Upgrade Points (UP) equal to twice the robot’s CR, minimum 1. Additional upgrades may be slotted on to a robot, granting +1 CR for every 2 UP beyond the normal amount, and if possible, these abilities have a saving throw DC of 10 + ½ the robot’s HD + the robot’s Intelligence modifier. For creators of robots, Improved Robotics nets +2 UP for any robot made (does that increase the robot’s CR for cap-purposes or not?). Most upgrades cost between 1 and 2 UP, with a few also offering a cost of 3 UP. These include additional modes of movement equal to base speed (including clumsy fly or burrow), an additional attack (both 1 UP), Dex to damage, all-around vision…there are a ton of cool ideas here, but the individual value of the upgrades oscillates rather drastically. Let’s take aforementioned movement mode upgrade: It places the same numeric value on climb or swim speed as on burrow speed, which is valid for a monster-customization engine. There is also one upgrade that is called “Autodestruct core”, which deals 1d10 times HD damage in a 30 ft.-radius as a full-round action, with half damage being fire, the other slashing, Ref halves. At higher levels, this also irradiates the area. Cool, but destroys the robot. Compare that to threat range +1 for all the robot’s attacks (which RAW even stacks with keen et al., which it really shouldn’t) or a threat multiplier increase by 1, which has no maximum cap. Both of these cost 1 UP. Notice something? Formatting is sometimes weird: “The robot gains burn (1d6) with attacks of a certain type.” – look no further than this to realize how important formatting is…making a melee attack work as energy-based is valued the same way as causing 1 Constitution damage on an established or maintained pin.
To make that abundantly clear: I *like* the robot customization engine! And heck, as a quick and painless GM-customization tool to make robots more potent and versatile it is absolute GOLD. The problem here is akin to others in the first Arcforge supplement– the almost obsessive system transparency between subsystems that are not, or no longer, balanced for parity. In many ways, the core issue of the robot upgrade system lies that it is not as finely balanced as a class option among its own options due to originally being a GM-facing tool. Okay, that’s something one can deal with. However, the system is opened to players, and to other subsystems, and that’s a really bad call. The mech engine, for example, has obviously not even cursorily been balanced against the robot upgrade system. You don’t have to be an experienced crunch wizard to see the disparities here at a glance.
One of the mech enhancements introduced herein, for example, nets you 1 Upgrade Point – this upgrade point can be spent on bonafide fly speeds, boosters, etc. – compare that to hover stabilizers or the feather fall-based modified aerodynamics enhancements in the first book. Need more? In the core engine, only mechs with the quadruped/threaded body type could gain a climb speed – well, now that limit’s been thoroughly squashed. In many ways, this small mech enhancement that allows for robot upgrades to be used for mechs is SUPER-broken and needs to die; it compromises the per se solid core engine for Arcforge mechs established in the first book. Thoroughly.
Unfortunately, this annoying lack of concern with system parity can be seen in some other instances. There’s e.g. the new feat chain that upgrades your astral constructs with upgrade points (because we all know that astral constructs really need a power upgrade */sarcasm*) or the option to add the aggregate template (more on that below) by adding AIs stored into the thus created astral robot. The *idea* here is AMAZING. Let me make that abundantly clear. I also love the world-building implications this has. Execution? Not so much. This book introduced the Biomech Construct psionic feat, which applies the biomech template to any astral constructs you create, and you can choose options from an enhancement menu or the metamorphosis powers of the same line, which is pretty potent as a whole. At one point, I am pretty sure that parity between power points and BP/corresponding abilities was considered, and the feat would be potent in that context; however, the final iteration of the Arcforge-systems has gone another way, which also destroys this assumption of parity.
The integrator aegis is a more complex archetype – this one gets rid of astral suit, but instead modifies their own body, which means they can wear armor. They get access to a limited array of aberrant customizations, and begins play with two robotic enhancements (see above) twice at 1st level, once at 2nd, and once at 12th level. Reconfigure is replaced with the inorganic property that provides a pretty darn impressive list of growing immunities. This theme is also emphasized by cannibalize suit being replaced with an option to ignore a whole array of negative conditions with limited daily uses. And yes, that’s flat-out “ignore” – not suspend or delay onset, ignore. And the character even gets hardness that scales, with all implications of hardness, and as a capstone, has a construct apotheosis. The aegis also features generally available customizations to gain weapon emulation, the ability to mimic psibertech, gain robotic upgrade points, or apply the astral suit’s bonus to touch AC. These general customizations should be taken with a grain of salt – I do not recommend any but the weapon emulation and the psibertech mimicry to be introduced – bingo, once more we have system-crossovers that pretty much go beyond what’s feasible. Robotics enhancement would, for example, net you an UP as a 2-point customization, which can be…well…overkill.
So yeah, the robot upgrade engine and how it pertains to robots on their own? Valid, fun and cool. How it interacts with other sub-systems? Broken. These crossover options need to die in a horrible nuclear fire, or the material needs desperately to be rebalanced to the power-levels and assumptions of the respective systems that they connect to.
The pdf then proceeds to provide specialties for technological daevics, the metahumans – essentially akin to passions, opening the flavor of the class and expanding it, which is a cool angle. Two are provided, the cannoneer (fixation) and espionage (vigilance) specializations. The cannoneer gets Perception, Stealth and Knowledge (geography) as skills, and the passion veil list includes the new tech-themed veils from Arcforge: Technology Expanded (with micro-missile gauntlet and nanite cloud delivering two veils usually exclusive to vizier and helmsman), as well as gorget of the wyrm, armory of the conqueror, courtesan’s cloak, sentinel’s helm, and lashing spinnerets, the latter usually being a guru/vizier-exclusive. The pdf also introduces a new veil, namely daevic aspect, which the specializations/passions both get – but interestingly, the effects differ by specialization. The core benefit for fixation would be a +1 insight bonus to attack rolls and AC, for vigilance, it’s be +1 (untyped, should probably be insight) to saving throws and skill checks against creatures you have identified. This differentiation also pertains to the chakra bind to blood. The cannoneer (fixation) gets a 20% constant miss chance, as per blur (spell reference not in italics), as well as a threat-range expansion of 1 with all ranged weapons used, explicitly applying after the Improved Critical feat or keen. Not a fan of threat range stacking per se, but not necessarily broken at 12th level; it should be noted that weapon and weapon property lack formatting in the pdf. Espionage (vigilance)’s blood chakra bind instead nets 30 ft. blindsight, +10 ft. per essence invested, as well as improved uncanny dodge as a rogue of the daevic’s level. Notice what I didn’t talk about? Yep, essence invested. The veil lacks its essence invested section, and e.g. Fixation provides no reason to actually invest essence in it beyond the base benefits outlined above. Pretty sure that’s a glitch.
Anyhow, back to cannoneers: 3rd level nets Precise Shot, 5th Deadly Aim, 8th level Improved Precise Shot or Pinpoint Targeting sans prerequisites. 6th level nets one of two aspects: Sharpshooter lets the character use a full-round action to make one shot, making a number of attacks they normally could execute, even with weapons that can shoot only a limited amount of times per round; for each hit, they deal damage and add it together, and if even a single attack was a critical threat, one confirmation roll at the highest BAB suffices. I am EXTREMELY torn on this one; on one hand, it is a good example of representing the one-shot-kill style of snipers; the damage this can accumulate, particularly on crits, is insane. Then again, that’s exactly what the ability is supposed to do. So yeah, it might not be for every game, but I actually like it, with reservations. If you need a nerf suggestion for a grittier game, make the 2nd iterative attack the only one that is consulted for critical confirmation; with a good build, this still will yield a reliable amount of critical hits, but not as much as the “one confirmation suffices”-angle.
Cannoneer requires the use of the armor penetration rules already previously mentioned in Arcforge: Technology Expanded (as an aside – they are one reason why I believe that this book and the first one were, at one point, one massive tome that was split); however, the rules actually are here, and the benefit increases armor penetration by 2, and all creatures adjacent to the projectile’s impact point or line of fire for automatic weapons are treated as if caught in a splash weapon (save based on class level and daevic’s Charisma modifier), with 12th level and 18th level increasing the radius of the splash by 5 ft., and the armor penetration by 2.
Let us briefly talk about armor penetration. That would be a kind of variant rule that is applied to weapons, with a proper table added: To explain the scale: A light crossbow or light pick has AP 1, an atom gun AP 14; muskets and revolvers, for comparison, clock in at AP 4. AP does pretty much what it says on the tin – it bypasses the respective value of armor. This value is enhanced by the enhancement bonus, if any – a +2 revolver, for example, would have an AP 6, value, the revolver’s base 4, plus 2 for the enhancement bonus. I like the idea behind AP per se – weak weapons like crossbows etc. definitely can use a power upgrade, and AP does deliver that, and the excessive-looking AP-values at higher level start making sense courtesy of a simple rule: It gets rid of that “attack touch AC”-caveat of firearms. Now, I do think that some of the higher AP values are a bit excessive, but having tested playing with the AP rules, I actually found myself liking them quite a bit. While my pretty conservative tastes would reduce the higher AP values and increase the lower ones for a more even playing field (that would also make tanking more viable and interesting), the notion behind this system is one I can definitely get behind. The Piercing Attack feat introduced herein increases any AP value of your weapons by 2 if you maintain your psionic focus, and by expending the focus, you can also apply the AP value to deflection, sacred or profane bonuses to AC.
The second specialization/passion for the daevic gets Perception and any two Knowledge skills, as well as HU.D. from Arcforge: Technology Expanded, as well as sentinel’s helm, courtesan’s cloak, dreamcatcher, collar of skilled instruction, essence of the succubus, cuirass of confidence and bloody shroud, as well as aforementioned, slightly problematic daevic aspect. 3rd level nets an untyped +2 to Knowledge skill checks and the ability to make them untrained, as well as the option to, as a swift action, make Perception and Knowledge checks. Rules syntax here is a bit ambiguous: Is that a swift action for both? Or a swift action for either? Additionally, we have a +1 DC-increase for veils used against identified creatures, which increases by a further +1 at 8th level and every 5 levels beyond. 6th level allows for the choice between optimizer and saboteur; Optimizers get the tactician’s strategy and one strategy, with an additional one unlocked every 6 levels thereafter; these use Charisma as governing ability score. Unfortunately, quite a few strategies don’t work for the daevic, as they are contingent on being a member of a tactician’s collective, which is a class feature the daevic does not have. No alternate means to determine eligible allies is provided either. Yep, another point for my assertion that Arcforge struggles when attempting to blend systems. Saboteurs gets 1d6 sneak attack and the unchained rogue’s “debilitating strike”, with 12th level and 18th level increasing the damage dice by +1d6 and the option to apply an additional debilitating strike effect whenever the ability is used. Why did I use quotation marks above? The class feature of the rogue is not called “debilitating strike” – it’s called “debilitating injury.”
The pdf also provides the psiborg racial variant for the noral race: +2 Constitution and Intelligence, -2 Wisdom, starts with a psibertech piece’s basic augmentation and treats their level as +1 for its purposes, and an increased implantation value of +1/2 character level (min 1) as well as a decrease of the Heal check to install cybertech by 10; the variant loses symbiotic resistance and surge for these. Androids can choose three new alternate racial traits for a similar start play with a basic augmentation, Small androids, and a bonus feat in place of nanite surge. Forgeborn can replace fearless with a piece of psibertech and its basic augmentation. There is more interacting with the eponymous psibertech: The Crystal Psiborg feat transforms a psicrystal into a piece of psibertech, granting you its base augmentation, but eliminates its autonomous ability to move; I don’t think doubling the personality based benefits in this context was the best call, though. Feat-wise, an Android or construct can choose the True Machine feat, which nets you full construct apotheosis, the robot subtype and 1 UP, or the clockwork subtype sans winding requirement. This feat may be taken at 1st level. Seriously, this feat is not a good idea – I’d rather recommend basing the like on a player race properly designed to account of the copious immunities of constructs. It’s not like we don’t have enough of those. Alternatively, if you wish to salvage the feat, I’d strongly recommend implementing a scaling mechanism that lets the player choose new immunities from the construct’s lists as the levels progress. Otherwise: Kill this with fire. It should also be noted that soulknives and zealots can use blade skills/convictions to tap into the psibertech engine so prominently featured above.
Which is also what we should talk about next, for 15.5 pages are devoted to psibertech. Cybertech is interesting, in that it blurs the line between item and class feature: A psibertech implant may be chosen in lieu of a power known, feat or selectable bonus feat; if the character does not have a manifester level, they use class levels in a class granting Wild Talent to determine manifester level-based benefits. A character must first choose the respective basic augmentation before choosing an advanced augmentation; if the character has 3 advanced augmentations, they get the ultimate augmentation at 20th level. HOWEVER, psibertech does occupy a slot and has an implantation value, and they have a weight. Unlike regular cybertech implants, these psibertech pieces can, as you can glean, not simply be bought…or can they? After all the information on individual psibertech benefits, there is a note that provides global pricing for psibertech crafting: 5K for the basic augmentation, +20K per advanced augmentation, and +50K for the ultimate augmentation; twice that price for being bought and installed. This latter section is one I’d be careful with, but do appreciate as a whole: Adding these costs to the class feature angle might be a way to keep the psibertech power in check for less high-powered campaigns. It should be noted, though, that in comparison to regular cybertech, psibertech is VERY low-priced. 1/day full-round action to replenish all power points is certainly worth more than 100K gold.
There is one aspect about psibertech that I consider extremely problematic and broken, regardless of campaign power level. You guessed it. Mechs can get psibertech if they have the proper body parts. Thing is, the pdf doesn’t really explain whether the pilot has to pay for in class features, whether the crafting is the only thing, or any limits – is psibertech a mech enhancement? Not that it’d matter. Psibertech as a sub-system has no semblance of power-parity with regular mech enhancements, and it’s very much obvious that this section was pretty much a very ill-conceived afterthought, as psibertech’s rules language never intersects properly with that of mechs, making cross-interaction very wonky at best. You guessed it: Kill it with fire. Scratch that, make it “Kill it with untyped damage”.
To give you an idea of what to expect: At the same implantation value of comparable cybertech, the arm of the augmented blade acts as a mind blade or call weaponry, which replaces your hand as though affected by graft weapon. Rules syntax isn’t 100% clear here in whether this means that only that hand is lost while the weapon is drawn, or whether the benefits of graft weapon also are assumed to apply. It could be read either way. The advanced Augmentations (header not properly bolded) include threat range increases of +1 (which do stack with keen, but NOT with static threat range increases, retain usefulness in areas where psionics don’t work, a critical multiplier increase of up to x6, or making it count as a DR-bypassing material chosen from adamantine, cold iron or silver. I think adamantine should have a minimum level here. The capstone ability nets auto-confirmations of critical threats. (Yes, there’s a lot of missing formatting here, unfortunately.)
The collectivist’s mental uplink requires a collective to do anything (and should probably specify this as a prerequisite), and allows you to add willing creatures as a move action, and expend psionic focus to add a willing creature as a free action. This is potentially very strong, but the “willing” caveat does prevent abuse via Unwilling participant etc. and a breaking of the offensive capabilities of the collective engine. The advanced augmentations here include making all collective members count as having your teamwork feats for the purpose of the psibertech augmented creature gaining their benefits, a shared awareness of creatures regarding concealment, and a very powerful one: manifesting powers through collective members – but that last one is actually properly kept behind 15th level and another augmentation as a prerequisite. Remote viewing through members is included, and the capstone doubles collective members.
A psionic tattoo-based one can be found, and there’s one that provides the ability to morph and infiltrate, including options to fortify their mind, deliver false readings, etc. There also is an athanatism-themed enhancement that could be thought of as somewhat themed in line with Death Stranding (which I btw. grew to absolutely love after hating it for the first 10 hours…) There also would be one that allows for psychometabolism powers to double duration (doesn’t stack with Extend Power); nomads gain mobility-enhancers (which include altering teleportation destination by up to half base speed – very cool!) – and yep, the nomad still has to have “line of site[sic!]” to the destination, but that’s at least just a typo. With mech pilot’s bond, you can teleport your bonded mech to you (again, this should have a prerequisite that, you know, it requires an actual bonded mech to do anything…), and a vocal enhancer can improve the mind-affecting abilities of the target. Interesting, btw.: There is a piece of psibertech that has construct-apotheosis as an advanced augmentation – only here, it’s locked behind a proper minimum-level-requirement. On a minor meta-level complaint, some bonuses here probably should be circumstance bonuses, as that’s usually the bonus type associated with regular cybertech. Some of these augmentations, just by the way, are pretty much game-changers. If you have plating of the psion-killer, you get the ability to choose an advanced augmentation to fire off a 30-ft. radius dispel psionics with ML equal class level (should be character level), usable every 5 rounds. This is very potent, but also pretty darn cool, and at 16 lbs. weight and implantation 3, it does have a cost. Flat-out power immunity for any power to which PR applies is locked behind 12th level, as another example for the augmentations provided here. I am not a fan of the one that nets you slowly replenishing technological charges and charges of psionic items, and replenishing the entire power point pool in 2 hours? Ouch.
An incorporeal phantom lite also ranks among the more potent pieces here. Fans of Path of War also can psibertech that interacts with Path of War Expanded’s sleeping goddess, which does, among other things, allow for the substitution of expending 2 readied maneuvers instead of psionic focus. This one is very powerful, but that’s what fans of Path of War know and expect. There also is an interesting piece of psibertech that eliminates the limitations of cybertech – only up to implantation value works at once, the rest becoming latent, and with an advanced augmentation, they can have two pieces of cybertech that take up the same slot, active at the same time. MOST of these are pretty well-balanced. Most. Not all. Temporal schemer’s interface, for example, has a cool base power – automatically notice delayed powers or those set up triggered. Cool! An advanced augmentation nets you, however, the option to make a move action to get as second swift action. That can be EXTREMELY powerful. Don’t do it. Seriously. Swift actions are extremely valuable. This should have, at LEAST, a 15th level minimum prerequisite. Readying a full round worth of actions is also insanely strong, and it introduces a whole can of worms. As a whole, I love a lot of the psibertech ideas – pretty much all of them. But their internal balancing, even without the subsystem-spanning issues, imho would have warranted further finetuning to ensure that the material is on a singular level.
The magic/tech item section includes a means to fire ranged touch, rays, cones and lines through weapons to add their enhancement bonus to the attack roll or save DC, doubled charges and ammo capacity for just the equivalent of +1 (x5 for +3 in the greater version), injecting weaponry, etc. – the armor penetration rules also are featured here, with AP-ignoring armor, laser-weapons increasing AP, etc. – some interesting ones here. Combined weaponry is interesting (but should specify that components lose e.g. finesse if not both of them have that property…); an modification that decreases charge uses, means to withstand psionics/tech-negating fields and 3 regular cybertech types are also included: One is essentially a template-based slavecollar, one decreases psychic enervation chance (with a limit). Realignment chips are interesting, if very low-priced– at 36,400 GP, they allow for the swift action regaining of psionic focus, though each subsequent use renders you first fatigued, then exhausted – and this DOES have a caveat that prevents abuse, bypassing immunity. Kudos for that! It’s in instances like this that the series shows how good it can be. A couple of neat items (balanced against the Armor Penetration-rules) are provided alongside two new artifacts, including the mighty mech/robot-crafting Arcforge.
Apart from the 4.5 pages of aforementioned robot customization engine, the last 30 pages of this module deal with creatures – first, with a whole array of new templates. These include sample statblocks, and feature the biomech template, data phantoms, a variant mindborn template, a template for modular constructs, one for the aforementioned slave-angle (shell), synthetic creatures, the anti-magic spellspurned and one for radioactive creatures. I did not reverse-engineer all of the sample creatures, but at a glance, the builds and templates generally are neat and interesting. The robot section ranges in CRs from 2 to 16 and includes terraformer bots (looking like birds with drill-beaks), and seriously ends the boot on a useful and versatile note.
Editing and formatting are, like in the core book, very much uneven – psibertech sometimes gets high-complexity stuff right, only to botch in the easy parts. Formatting is often very inconsistent, but less so than in the first book. This pertains to rules-language as well, which oscillates between “I’d allow that in a heartbeat” and “what were they thinking” – the latter applying primarily to anything regarding attempts to crossover between subsystems. The pdf sports a nice array of old and new full-color artwork. Bookmarks are annoying – the first part of the pdf is bookmarked with a few lines, but after the daevics, the entire template and robot and item section has no bookmarks, making navigation of these parts a pain.
Matt Daley & Michael Sayre’s Arcforge: Psibertech shows many of the same issue of the core book, but also does many things better: Within the individual frames of reference of the individual subsystems, the content tends to be cleaner than in Arcforge: Technology Expanded. While internal balancing isn’t often as tight as it should be, if you do exert some caution, you can get some seriously neat mileage out of this book. There’s this part of me that loves this book.
And then there is the part of me that is infuriated by the plethora of formal glitches and uneven balancing, and more so, by the absolutely broken links between subsystems.
These aggravating afterthought-like crossovers that compromise the systems they intersect with; both psibertech and the robot upgrade system have no business intersecting with the regular cybertech (pricing all off in comparison) or mech engines (balancing of UPs vs. mech enhancements all off); considering that the core book already had its issues in the core engines and how the engines of classes and sub-systems interacted with the mech-engine, adding these on top is asking for a colossal cluster-f*** – and not in the fun way, but in the “how the f*** does this line up” kind of way.
In short, when seen from solely a design/balance perspective, this is broken as all hell.
Dear lord, this is a rough beast of a book, and one that clearly shows that it desperately needed some serious playtesting, development, etc. – this, like its first book, could have been a milestone. It oozes cool ideas. But, and there’s not questioning that, it does fall short of its lofty ambitions.
And yet, while I consider this book DEEPLY flawed, I also can see it having its appeal: If you take care, eliminate the broken bits, and want to flex your design muscles a bit, it’s actually a book that’s surprisingly easy to redeem (provided you can handle the complexity of the subsystems): Rebalance a few components, limit psibertech and nerf a few parts, kill off the system-crossovers, and there you go – psibertech is a book you’ll get a TON of mileage out of. For me, as a private person, I can get a ton of fun out of this!
As a reviewer, however, I can only rate what’s here, not what I wish this was, or what *I* can make the book into for my table. I have to rate this book for what it is.
And it *is* a flawed book that shows glimpses of true greatness time and again, but still falters. Worse for the system-inherent context, this book compromises the mech-engine, which already was struggling under the none-too-great class option components in the first book, even further, at least if you are not careful and realize how broken those system-crossovers actually are. Considering that aspect, I should rate this lower than the first book; probably around the 2-star vicinity.
However, idea-wise, and within the systems presented, the book also does a lot right. Moreover, the extensive bestiary section is super useful for the GM, and, it covers almost half of the book, it needs to be weighed accordingly.
As such, I actually do consider this book slightly better than the first one, but it’s still an incredibly uneven book, one that makes the first Arcforge book more uneven as well; hence, my verdict will be 3 stars. If you are a very crunch-savvy GM (or simply not concerned about balance) then consider this to be a full-blown recommendation; if balance matters to you, do yourself a favor and bear my warnings in mind. I thought long and hard, and while I really wanted to rate this higher, round up, etc., but I just can’t justify doing so. The book has too many serious and pronounced issues to warrant doing so.
For me, as a person, it’s a good book that inspired me to flex my design muscles, but as a reviewer? As a reviewer, I can only recommend this with reservations and very pronounced caveats.
As an aside: One of the authors has expressed a desire to revisit this series at one point. I’d LOVE to see that.
Arcforge is one of those frustratingly-rough books that really shows potential, and this holds true for this second part here as well. Now, the first book and this one were probably part of an original document, split in production; the following installments were not, so I’m looking forward to seeing how/if they improved upon the lessons learned here and engines crafted.
You can get this book here on OBS.
Missed the first book? It’s also inspired, yet flawed – in different ways. You can find it here.
If you consider my reviews to be useful, please consider leaving a donation, or joining my patreon here.