Description: Runequest is a venerable old game that has been around for quite a few years and a number of different publishers. The original version, developed by Steve Perrin for play in Greg Stafford's world of Glorantha, was published by Chaosium in 1978, where it, and it's supplements, set benchmarks in quality that are still rarely met today. The game system, formally renamed Basic Role-Playing, formed the basis of a number of popular licences such as Stormbringer/Elric, Hawkmoon, Ringworld, Elfquest, and, most importantly, Call of Cthulhu.
It was later licensed to Avalon Hill, where it was supposed to be the flagship of the boardgame giant's new roleplaying game franchises (which included Powers & Perils and Lords of Creation). Unfortunately a decision to remove Glorantha from the game setting, and really terrible production values, made the title languish heavily. [Although Games Workshop, who was licensed to print their own version, did go on to enhance the franchise in the UK, at least until they decided to only support their own games.]
Business problems with Chaosium then created an interesting dynamic. Greg got the rights to Runequest and his world of Glorantha, whilst Chaosium kept the rights to Basic Role-Playing. Gloranthan development continued essentially in a different system (Hero Wars, later renamed Heroquest in honour of the Runequest expansion that never was ("coming in 1983!"). Greg (through his company Issaries), then licensed Runequest to Mongoose Games. Mongoose wanted to use it as the basis for their own house system of fantasy role-playing games, in addition to using it for Glorantha and a number of the old Chaosium licences thay had also acquired. Which definitely led to an interesting dilemma that they wanted produce "a generic system" where the actual ownership of the game system they were using was vested in another company (Chaosium, who had just produced the 4th edition of Basic Role-Playing). This may account for the slightly schizoid nature of the original Mongoose edition; it had to retain enough Gloranthan content to be Runequest but it also had to be open enough to be used in other products that had nothing to do with Glorantha.
Meanwhile, development of Glorantha continued apace at conventions and online, eventually resulting in the production of Heroquest II (which was actually supposed to be a generic system), and more importantly Sartar: Kingdom of Heroes which adapted the HQ2 system to the best-known piece of Sartar.
Runequest II is Mongoose's attempt to reconcile their game with the current theory of Glorantha, and incidentally, clean up all the mess they made of the first edition.
Setting: Runequest II is still written from the viewpoint of it being a generic system, although it does embody considerable Gloranthan sensibilities in it's portrayal of the magic system. The approach to cults (an important element of Runequest) has been redone, also generically, but also by presenting a tool set where it is comparatively easy to construct cults to your requirements, in or out of Glorantha. [One of the problems of the Mongoose 1st edition (MRQ I) was that it exported Gloranthic things such as "cults" and "rune magic" to non-Gloranthan worlds, such as Llankmar, Elric, and Hawkmoon, probably as a necessity to continue claiming it was actually RQ and not BRP.]
Character Generation: Essentially Basic-Roleplaying. Random characteristics, with percentage-based skills whose initial value is determined by the characteristics, and to which previous experience is added. One important change is that characteristics aren't tested directly. Instead there are various skills that serve as characteristic tests.
Mechanics: Definitely cleaned up from MRQ1. Opposed tests now use the higher quality of success wins (which means you want to roll low), or, if the same quality of success ensues, the higher roll wins (which means you want to roll high). This cures the old "hit!/parried!" problem of original RQ, but at the potential of making the game more deadly. If you or your opponent have skills over 100% then the amount the highest skill is over 100% is subtracted from all skills, which is a better mathematical approach than the previous "divide by 2" method.
The idea of actually needing to find physical runes and align with them (ala Swordbearer) in order to cast basic magic has been dropped. Instead the idea of runes can be readily seperated from the magic system. Runes may still be invested, but this is much more in the nature of a heroic ability than a default approach. This allows the magic system to drop all Gloranthan references with ease, and yet still allows the game to officially not be Basic Role-Playing.
The system proposed by Lawrence in MRQ1's Cults, Guilds & Factions (derived from the MRQ1 version of Elric) is now the default standard, and looks quite workable, especially in relation to Divine Magic (traditionally sourced through the cults).
The system remains, as was the original, intensely simulationist.
Thoughts: One has been a Runequest devotee since early 1979, when I first saw a set of rules with a lizard gnawing on the shield of a female greek hopalite. It was really several generations (depending on how one measures such things), ahead of the curve. For example it had a single simple resolution system that was used for everything, rather than the customary hodge-podge. It treated NPCs the same way as PCs, essentially allowing players to play "monsters," and even more importantly, encouraging the treatment of "monsters" as beings in their own right and with their own motivations. One enjoyed the world of Glorantha as well, although my campaign rapidly diverged from the official one, so much so that in it's latest incarnation names have been changed to encourage people that they aren't really in Glorantha anymore. Runequest always had problems with high skill levels, and this version seems to have hit a workable approach to the problem.
The production values on this edition are very good, and even better, they've managed to keep a consistent approach to the standard Gloranthan ideology. Some of the earlier MRQ1 expansions, such as Legends, weren't really suited to being part of the game's core identity (and more reminiscent of Mongoose's old D&D supplements), and had all the signs of being hurried into production in order to support the line. [Then again, I believe that Mongoose, like Avalon Hill before them, have effectively only licensed the name; only specific Gloranthan product is actually vetted.] Even better they are attempting to maintain fairly close compatibility with the Heroquest II development of Glorantha, albeit as a simulationist rather than narrativist game.
Unfortunately it is their second attempt, and the idea of investing again in a product line I'm not really going to use* is less than attractive (the next few planned books are essentially reprints with a couple of additional new things). I just wish that they had gotten it right the first time (considering they had an extensive history of different solutions to various problems to draw upon). Still, for people wanting to take up this game for the first time, I do actually recommend this version. It's very well done, much more approachable than Heroquest II (and without the vehemently fanatical supporters of that game that you will find online), and I hope they can keep up the quality an attract new people to the wonderful shared world that is the modern Glorantha.
Additionally there are a number of third party developers of Runequest II product, such as Clockwork & Chivalry (alchemy and clockwork "steampunk") in the English Civil war, and Deus Vult (a crusade era historical game). In this regard they've managed to create something on par with Chaosium's 4th edition of Basic Roleplaying, albeit one that is not as customizable (which is probably an advantage for an inexperienced gamemster).
Rating: Very Good. Much better than their first attempt anyway.
* I really didn't like Hero Wars when it first came out (it wasn't helped by the fact it made old White Wolf products masterpieces of copy-editing), and had picked up a game called Ironclaw. Unfortunately my regular players are rabidly anti-furry and so playing an anthropomorphic game was out of the question. However I liked the Ironclaw system and wanted to use it, so the obvious solution was to translate my old Dragon Isles (nee Glorantha/Dragon Pass) game into this system. And so was born my Runeclaw game which has been working extremely well ever since. [And it was interesting to watch the ideas behind Glorantha shift to the stuff I had adopted in this new game, so it still retains close compatiblity with the new Gloranthan ideology.]