DBZ had a wealth of fun maps, with some that were incredibly well done. There are two in particular that stand out, one, I think its name was DBZ: Sagas, had an co-op arcade-y set-up and players trained between battles, going through every saga in Dragonball Z; the other was a sort of JRPG take on DBZ, filled with cutscenes, dialog, and specific roles for players, with many sitting it out when not required - both were tough, I don't think I ever finished them. DBZ maps were my bread and butter back then.
RP maps were also a lot of fun - lots of creativity by players and DMs. There was a really old one with Antioch in its name (Realms of Antioch or something) that was always my favourite, but the Volc's World maps had a great concept for multi-game campaigns, using similar units/stats across a variety of terrain. Many hours were spent in those games, with many worlds made.
LotR maps were everywhere back then, and they were a cool mixture of mass-attack game and overarching RPG/strategy, vaguely mimicking the events of LotR books. Other IPs had similar maps in concept, but I don't think any took off quite the same. Of particular note was LotR: Last Alliance, which always seemed like one of the best to me, with heroes that were strong, but not the basis for the entire match, with some cool secrets and strategies as well - it was really old, IIRC. There was a much newer one with base-capturing rewarding specialised units/perks, and some cool innovations in how armies were composed. Both had pretty good balance, I thought, which impressed me a lot considering how complicated they are.
But my hands down favourite maps? The Crash series, which are like the ultimate example of why Starcraft's UMS community was amazing. They were simple survival-RPGs where you levelled up a specified hero to new, powerful units, eventually becoming unstoppably strong at which point you levelled the entire map in search of four artifacts, each in a corner of the map. But what set it apart was how crazy and unrelated the sequels got. Crash 1 and 3 were made by the same person and were excellent; Crash 2 was made by some other dude, a real amatur, and was terrible; and Crash IV was made by someone else entirely and was really friggin' elaborate. Might be mixing up 3 and IV, but you get the idea - it's altogether a crazy legacy that every mapper respected.
Such a shame Crash never got replicated on WC3.
WC3 had less stand-out maps to me, and it took years for real amazing stuff to pop up. Of all of it, DotA was obviously the most noticeable, dominating the mapping scene and inspiring every clone imagineable; of them, Advent of the Zenith and Ninjas in Pyjamas seemed like the only really fun ones. But my favourite map was Dragonball Z: Tribute, a real innovation in how SC-style DBZ maps worked, utilising everything WC3 added to make this incredible, competitive-style RPG/arena brawler that vaguely followed the DBZ storyline; you'd have all the big bads and Z Fighters controlled by players, the respective groups allied with eachother, each with level-appropriate power-ups, giving specific points where one side or the other had an advantage, and each player - especially enemies - with radically different ways of powering up. WC3 produced less incredible maps overall, but the ones that were there ended up as remarkable classics.
Oh, and the custom-content community (like, skins and models and such) was amazing in WC3. There was a plethora of really good models and reskins, giving a lot of vareity to what was available. It was really fun to just browse what was out there while imagining the possibilities.
Nostalgia bombs everywhere.
It's a shame that hardly any of the popular game types that emerged from the UMS boom of Brood War and Warcraft 3 ever saw the light of day in SC2. I think the mapmaking community just said "fuck this" with SC2 early on, and that was that. I get so angry thinking about that sometimes. I almost wish SC2 had never been made for what it did to UMS. Killed it for both Brood War and SC2.Fuck.
SC2's map editor was the opposite of user friendly; it was really designed from the start for professionals and Arcade monetisation. This kneecapped any chance it had at taking off like SC or WC3. The ease of use yet flexibility was really a key factor in their respective succes, I think.
Plus, restricting the use of existing IPs completely killed big aspects of custom-mapping (as outlined above) which I think further hamstrung its ability to go anywhere, giving players at large nothing to really grab onto and enjoy.
