|What was memorable about this guy?|
I'd say for most players, one of the attractions of RPGs is the opportunity to play a character that does things you do not do in your normal life - sticking swords into monsters, throwing magic, or flying a spaceship. D&D 3E (and later) and Pathfinder are very good at letting us do that with tons of options and per-level multiclassing and point buy attributes - compared to the old "roll up your stats and figure out what to play" days it's a paradise for those who like to craft their character to match their specific vision. That's nothing but a positive. Well, almost...
The coal-filled stocking in this approach is that there are a lot of people who, once exposed to your vision of Abercrombie the Barbarian Prince will quickly point out all of the places it could be improved mechanically and all of the less-than-optimal choices you have made in creating him. "Why does he need a 16 Charisma - dump that to 8!" is among the kinder things you may hear. Communal min/maxing is just as annoying as individual min/maxing when it overwhelms the original concept. Much like MMO players, groups of players around the internet will quickly determine optimal approaches to specific classes and combinations whenever new rules or options come out and are often regarded as "experts".
If you're playing some kind of arena combat game against other players this may be useful.
If you're playing any other kind of game, it's of limited usefulness at best.
Sure, push your DPS way up there, that's really impressive. Hey, now we need an assist on this diplomacy check - oh, you have a negative? Ok, never mind. How about Stealth? Knowledge? Religion? Most optimized "builds' I have seen sacrifice a lot for increased efficiency in one particular area and that's not always the best answer to the things that go on in a campaign and they can get to be on the boring side when you're not slamming through one combat after another because that's a lot of what they focus on. Also it can warp the rest of the party. If we assume the tornado of steel barbarian can solo any monster in the game, the rest of the party may de-emphasize combat capability to try and shine in other phases of the game. Then the barbarian's player misses a session and suddenly combat goes horribly wrong .
Now you do get the opposite problem sometimes where someone takes a bard or a rogue type character and turns them into the jack of all trades and master of all trades too. That's not great for the rest of the party and rather than one player getting bored you have all but one player bored.
A lot of these overpowered builds rely on stitching together very specific abilities from across different classes and supplement books so one way to keep a lid on it is to limit options. Pathfinder has probably the biggest active universe for this kind of thing right now. Sure, the Technology Guide is awesome for the Iron Gods Adventure Path, but if I'm running Rise of the Runelords I probably don't need android PC's with chainsaws and laser pistols running around so the answer there at character creation is "no".
I've played and run a lot of games over the years but I have to say I've rarely seen the need for maxed out PC's. Right now the published adventures I am reading, mainly Pathfinder APs and the new D&D 5E adventures certainly do not demand apex character designs. So it's not pressure from adventure writers that drives optimized character designs.
I have found that campaigns are more enjoyable when people are playing a character they really like and that is often tied to designing it themselves. "Interesting" and "memorable" do not necessarily equal "efficient". Even when looking for power combos, if you find some combination of abilities that is particularly effective how much more satisfying is that than finding it out from some guy on the internet before the game ever starts? For Delve Night at the FLGS an internet-optimized hurricane of evocation may be fine but in an ongoing campaign it's different. Can you live with that character for a year? How about two?
Seeing it discussed online almost constantly I feel the occasional need to push back against the pressure to optimize everything. Ideally players find a balance between "fun/interesting to play" and "mechanically effective" that works for them and for the rest of the people in their group. Hopefully they take a little time and consider the non-mechanical aspects of the character to round things out.