II won't be the first person to compare WotC with Microsoft. IMO, they both owe as much of their success to the simple fact that nobody else has successfully marketed a superior system as they do to the inherent quality of their product. Here are the various FRPG's that I've played, along with my humble assessments of them.
Fantasy Hero--Combat's slow, but otherwise rewarding for warrior types. Spellcasters OTOH get shafted due to the fact that they have to pay character points (CP) for each spell individually, layering on loads of Power Limitations for diminishing returns in saved CP. A missile spell that's no more powerful than, say, a plain old crossbow (which a warrior doesn't pay CP to carry around) eats up a good 7 points or so, and is a lot more difficult and dangerous to wield. Moreover, the point cost issue makes the concept of high-level magic unfeasible. Also, spells are based on HERO-system powers, which are geared quite heavily towards combat application. Designing noncombat spells for utilitarian purposes such as communication and divination is very clumsy, based around two or three vaguely-detailed powers. All of this is exacerbated by a point-based character generation system replete with characteristics, skills, talents, perks, and disadvantages. After putting so much thought into a character, it's tough to treat him as disposable.
GURPS--Ah, there's nothing like GURPS' smaller-than-life take on fantasy to make you appreciate D&D's over-the-top spellpunk approach. Whereas D&D is too ridiculous and excessive to simulate traditional heroic fantasy, GURPS is too mundane and underwhelming. Conan wouldn't survive many violent encounters running around in that loincloth. Armor = Life. Encase your body in tempered steel. Then encase that encasement in even more steel. Convince your GM to give you one more point of PD by eliminating that visor from your helm; tell him you'll hire a seeing-eye beggar to guide you into the dungeon. Even magic is dull, broken down into an unspectacular college-based system, which means that as a wizard you have a small group of spells that you specialize in, and most of the colleges are just variations on the same theme (e.g. a "bolt" spell, a "detect" spell, and a "shield" spell). Don't expect the average wizard to have a Knock available to them.
Ars Magica--The various incarnations offer an elegant game that generates a variety of interesting characters. Magic is not something to become jaded with. The open-ended spell system is capable of just about any effect, and it rewards spontaneity and creativity. Combat has some interesting mechanics, such as a "Soak" roll to allow a tough/armored character to withstand more damage than a frail one. Most of the problems with this game are strictly how the designers chose to present it. A lot of players simply won't appreciate the stratified caste of player characters, with mages being the top dogs, "companions" serving beneath them, and down at the bottom the ever-pitiable "grogs".
Harn--This is a gritty system for a gritty world. Combat's fairly unforgiving of mistakes or plain ol' bad luck. Magic is college-based, so a spellcaster is more like a super-powered-mutant X-man than Merlin. Users a linear percentile-based system for determining success, which stinks. All-in-all, it's about as good as GURPS.
Warhammer FRPG---Fun game with a quick character-generation system, which is appropriate because combat is brutally lethal. Any attack that hits has a 1-in-6 chance of doing critical damage, making it almost foolhardy to become too attached to a character that can explode like ripe watermelon after a 10ft fall. Survival is less dependent upon an individual's ability or equipment, and more to a healthy reserve of Fate Points, which always seem to be in short supply. There are a lot of career paths to take, and you may find yourself eagerly awaiting your character's nigh-inevitable gory demise at the hands of some chaos thingy so that you can try out something different. Magic is designed to be powerful, with the trade-off being that you're supposed to trudge through a couple of careers as a worthless weakling before you get to cast that first fireball.
Runequest--Haven't played this system much, but it seems like a lot of work for a small return. Uses a percentile-based system, but if you're willing to do a little min-maxing you can get a skill over 100%. Plenty of ability scores, but most of them don't have significant impact on anything.
Rolemaster--I found the character generation system to be absurdly elaborate & random, basically a game unto itself (and not a particularly fun one). It took me an hour to create my first character, who died 30 minutes into the game. To a dog. Not a hellhound or shadow mastiff or some sort of dire canine. Just a regular ol' doggie that kept making his Dodge roll, and then managed to score a critical (criticals are the great equalizer in Rollmaster).
Arcanum--This is an old game from the 1980's that was, essentially, a third-party D&D enhancement that offered a good deal of flexibility (sort of a proto-d20 product). It had about 32 different character classes, most of which equate to what we now call prestige classes. Ahead of its time in many ways. It replaced the AD&D's saving throws with saves based directly on one of the 6 ability scores (mainly Dex, Con, & Wis). They had combat skills that granted various benefits relating to such things as mounted combat, archery, and weapon specialization. They had a character class called "warrior" which gained a new combat skill at 1st level, 2nd level, and every two levels afterward. Any of this sound familiar? Magic was broken up into 9 different schools, such as Astrology, Divine Magic, Black Magic, High Magic, Low Magic, etc, Each shared some basic universal types of spells (e.g. a 1st-level magic-missile type spell) but each also had its own particular area of strength. The various spellcaster classes were learned in one or two of the schools. Great game that offered a wealth of options for PC's without straying too far from archetypes. Don't know why it died on the vine. Poor marketing or poor market?