Pandemic 2 is an interesting study in contrasts. As a sandbox, it's great fun -- who amongst you hasn't wanted to try to give everyone in Western Europe bloody vomit with some horrible new disease? But as a game, it's a miserable failure -- if you try to play to win, you'll be bashing your head against its design limitations in no time.
I'm going to digress a moment and talk about the original Pandemic, which I played long before I saw Kongregate (indeed, it's not on Kongregate even now). It was an interesting little germ (ha ha) of an idea -- you controlled a disease with the goal of wiping out humanity. You gained points by infecting more people, which you could use to increase your transmission (for instance, by making yourself airborne) or increasing your lethality (for instance, by adding hemorrhaging as a symptom). It was a cute little game, but very simplistic, egregiously so in its geography (there's a total of 8 regions, including an "Eastern Europe" which stretches from India to Burma), and the optimal strategy becomes blindingly obvious early on: don't develop any lethal symptoms until you've infected everyone, and then once you do, bring the hammer down with as much lethality as you can and patiently wait for everyone to die. (That's right -- once you develop new symptoms, everyone who already has the disease also gets them. Don't ask me how that makes sense.) Anyway, it was a neat idea, but a little too simple to be much of a real game.
Pandemic 2 attempts to address many of these shortcomings and make the basic idea into a more substantial game. The result is a product which is vastly improved in some areas, retains some of the flaws of its predecessor, and takes a very large step back in one respect. Let me talk about the last of those items first. In the original Pandemic, there wasn't that much that happened in the very early stages (when your disease had only infected a handful of people) and the very late stages (when you have already infected everyone and were just patiently waiting for humanity to hurry up and die). That was OK, because you could just hammer the "next day" button until something did happen. Pandemic 2 has basically the same dynamic, except that the designers decided to make it a real-time game. This means that you can see the airplanes and ships niftily flitting about the globe. Unfortunately, it also means that you spend an inordinate amount of time (even with the game speed set to its fastest) just sitting around and waiting for something interesting to happen. This is not fun at all.
So, in Pandemic 2, you have a relatively detailed world map, with 21 different regions. Many regions have one or more airports and seaports (although, realistically speaking, every region should have an airport and seaport, but I suppose the designers took some artistic liberties), which can very quickly spread your disease from region to region. The game seems to overemphasize these modes of travel as opposed to overland travel, though -- is it really the case that a disease is more likely to make it from the US to Mexico via an airplane than a car? I doubt it, somehow. Your disease can be a virus, bacterium, or parasite, each with its own advantages and disadvantages; as time goes on, you acquire "evolution points", which you can use to buy new symptoms, means of transmission, and resistances. (Somewhat perplexingly, you gain evolution points very quickly early on, when your disease has only infected a few people, but the rate decreases sharply as time goes on, which can leave you just sitting around waiting for points to accumulate in the late game.) The symptoms are a double-edged sword: they increase your infectiousness and lethality, but they also make your disease more visible to authorities who can take countermeasures. Some symptoms are, on a net basis, more trouble to your disease than they're worth, and it actually costs more evolution points to get rid of these symptoms than to acquire them. Your disease can also have some traits, which are generally entirely random and affect your disease in various ways. There's also a wide variety of natural disasters which can hit countries, which may speed or slow the progress of your disease.
Now, for the two big flaws in Pandemic 2. First of all, the aforementioned authorities have a dizzying array of countermeasures they can take to combat your disease, ranging from handing out bottled water to declaring martial law. As far as I can tell, none of these countermeasures matters one whit once you've infected a region. They may slow the progress of your disease, but I have never seen a case where a disease has gotten a toehold in a region and then been stopped by these countermeasures (with one exception, which I'll discuss below). Conversely, countries can close their borders, seaports, and airports, and these measures are 100% effective at keeping your disease out if it hasn't yet gotten in, which is just as unrealistic. If you're trying to kill everyone in the world, then as soon as Madagascar closes its seaport (which is the only way in), you might as well pack up and go home.
The other flaw arises from the fact that, to avoid the easy win strategy for the original Pandemic, countries will crazily overreact. You can have a virus which is entirely asymptomatic and nonlethal, and yet as soon as it's infected a few hundred million people, countries will start shutting their borders, declaring martial law, and so forth. This is obviously quite unrealistic; from a gameplay perspective, it arises from two basic issues: first, that lethality and transmissibility are completely decoupled, and second, that when your virus acquires more lethal symptoms, that everyone who already has the virus will get these symptoms too. The obvious solution from both a gameplay and realism perspective is to eliminate these -- if you have a virus that infects 200 million people, some of them are going to die from it regardless of how harmless your virus is. The symptoms having good and bad aspects is already a step in the right direction; the game just needs to take it a bit further. And maybe make it so that your virus acquiring new symptoms isn't retroactive? It would change the game greatly, but I can't say it would be for the worse.
OK, last complaint. After a while, the humans will start working on a vaccine. Usually, you kill people too rapidly for them to finish the vaccine in time, but occasionally they'll finish and deploy it. When this happens, you get a coin flip! Sometimes it works and you lose. Sometimes it doesn't work and they'll try again (almost certainly they won't have enough time, though). And sometimes your virus mutates and becomes invincible to all vaccines. This is a hugely random swing (it can be affected by your acquiring drug resistance, but there's still, as far as I can tell, a coin flip underneath), and adds another element of annoyance to the game.
On to the presentation. The graphics are clear and crisp, though the automatic messages displayed in the information panel (which could really use a scrollback bar) occasionally are kind of nonsensical; a little editing would have helped. The background music is pretty good, and makes for a nice complement to the game. The game offers two different modes, "realistic" and "relaxed", which in practice aren't terribly different; your disease appears to be faster in relaxed mode, and a few things aren't included, but most of the gameplay is entirely the same.
Anyway, if you're just messing around with the game, it can be fun to see how much of humanity you can kill. But when you're trying to beat the game, you'll get very, very frustrated when you play 20 times, Madagascar closes off every time, and then finally on the 21st time (without you changing your strategy at all) you manage to get in before the close and win. This amount of dependence on luck results in a very frustrating game. While Pandemic 2 is definitely an improvement on its predecessor, it still has a way to go to really turn its potential into an enjoyable challenge.