Thursday, October 28, 2004

NBA Live 2005

I installed NBA Live 2005 (PC) yesterday.

Basketball games on both consoles and the PC have an erratic, inconsistent history. Basketball is a difficult game to simulate because so much movement is reactive and based on a player's proximity to other players. Finding and creating space is a core skill in basketball, and this kind of free-flowing motion has been very difficult for developers to recreate. To this day, I think my two favorite basketball games are TV Sports Basketball (Cinemaware, for the Amiga, 1989) and NBA Live 95 (Electronic Arts, for the PC, 1995). NBA Live 95, in particular, was a breakout game for EA, because it was amazingly fluid, the ball physics were quite good, and the sound effects were outstanding. That's very few outstanding titles over two decades, unfortunately.

The NBA Live series is a microcosm of the erratic history of basketball games. The games have always sold well, but in the last five years, Live was clearly the weakest of EA's sports lineup. Last year's revamp took major strides in improving the series, and I had high hopes for this game, given that it could be expected to add additional refinement to last year's effort.

Let me start with the positives. On the PC, the game looks fabulous. Player models and arenas just look jaw-dropping at the highest resolutions. Crowd sounds are excellent, and the announcing is very solid. The game also does an excellent job compressing breaks in play to keep the game moving.

All-Star Weekend, particularly the slam dunk contest, is remarkable. The slam dunk contest is a game in itself, and it's absolutely fantastic.

So, in a word, the game is sexy.

Now let's get to the problems, and I wish this section was shorter, but it's not.

What I'll always do first is watch the CPU play against itself and experiment with sliders over a 5-10 hour period. Even if you're playing the game, 90% of the players at any one time are CPU controlled, so if the CPU can't play against itself with a reasonable degree of fidelity, the gameplay experience is in trouble. So I've adjusted sliders and done the best I could to get to a reasonable CPU vs. CPU experience.

No matter the slider settings, here's a one-word summary of how the CPU plays basketball: lousy. It's very poor at creating space and even poorer at taking advantage when space is available. Player spacing on offense is horrendous. It's very slow to find the open man. The fast break is non-existent, which is tremendously frustrating. In almost all cases (breakouts after blocked shots and steals being the exception), CPU players will stop to receive passes instead of catching them on the run, and that kills the fast break. Even when the CPU has numbers, they will rarely take advantage, preferring to slow down instead. There's also far too much dribbling in response to pressure. Closely guarded players will dribble all over the place instead of passing the ball--in part, because teammates are very slow to offer help.

There are certainly visual anomalies--players 'skating' with each step, players passing without looking toward their target (and not in the Pete Maravich sense, either), phantom fouls, shots considered three-pointers when the shooter's feet are clearly on the line, etc.--but the biggest problem, by far, is the way that the CPU plays basketball. It's slow, it's ponderous, and it's boring.

Then there are the blocks. Gear up for 10-15 per team per game unless you want to edit player ratings--the sliders just can't fix the problem.

Toss in a few more problems--far too much 3/4 court pressure, odd ball flight on some passes--and the summary is messy, messy gameplay.

Now you may be thinking that today's NBA is a relatively poor simulation of good basketball, and I will freely admit the truth of that assertion, but it's poor in different ways. NBA Live violates so many of the fundamental tenets of modern basketball that while it it looks pretty and animates well, look more than six inches deep and you'll find air.

This is what drives me crazy about this game and EA Sports in general. If anyone who understands basketball sat down and just watched a CPU vs. CPU game for ten minutes, they would see these problems. How is that not part of the development process? Isn't watching the game and evaluating whether it looks like 'real' basketball an important step? This is true of ESPN games as well--these development teams need to spend far more time watching CPU vs. CPU games.

I know--you buy the game to play it, not watch it. But if you're playing single-player, 90% of the players are CPU controlled. In essence, you ARE watching a CPU vs CPU game with one human player inserted into the mix. So if CPU vs. CPU doesn't follow the correct fundamentals, the game isn't going to work.

I'm still hoping that someone can come up with some genius sliders to fix some of these issues, but right now, the game just isn't fun. And ESPN NBA was unimpressive as well this year.

Last year, the best basketball game, by far, was ESPN College Hoops. I've heard good things about this year's version and hope it comes through when it's released next month.

Site Meter