Not at all. In TA you would probably select all 50 units, point them in the right direction and wait for the dust to settle 15 seconds later.

In Kohan you select your company's formation. For example, if you are racing reinforcements to the frontline, or to a resupply zone, you might choose a column formation that sacrifices fighting efficiency for speed. When the troops are committed to battle, they are out of your hands. You can order them to retreat or rout, but while they are engaged the seven men of your company will fight and cast spells of their own accord until they are decimated or their morale drops so low that they rout themselves.

Not only is this just as spectacular to watch as any other RTS (7 soldiers times say 20 companies is 140 individual unit animations ... that's a whole lot of swords, arrows, and spell casting effects), it is a whole lot more strategically challenging. If you want your companies to gain experience you will need to retreat them from battle before they are wiped out. That means having reserves on hand to engage the enemy as your vanguard retreats to resupply. You need to constantly monitor the state of your companies and know where to apply the reinforcements, and when to turn tails and run. Elite troops don't grow on trees. Having a company wiped out will also frequently mean losing a Kohan.

The Kohan play a role in the game as hero characters. There is a set roster of Kohan, some of which you meet up with during the course of the campaign, others which you find in amulet form. Awakened Kohan which are on your side are available to be assigned as captain for any of the companies that you create.

Doing so means that the company will benefit from the Kohan character's unique spell casting or other special abilities. As the Kohan gains in experience, he or she will level up and become more powerful in combat. But if their company is destroyed and the Kohan killed in battle, they go back to amulet form and back to level one.

So it's worth trying to keep them alive.

There are a couple of interesting twists in defensive combat that are worth mentioning. Companies that are resting automatically entrench. This gives them substantial combat bonuses if they are attacked. Also, towns and forts are manned by automatically-created and replenishing militia. When enemy troops approach, the militia sally forth and engage. They can repel low level attacks but will be decimated by tougher troops, so militia really just buy you a few minutes while you force-march real soldiers to the action.

So the combat, in itself, is challenging and fun - but Kohan's success lies in its strategic focus. Modelling buildings, economy and units at a more strategic level frees you from the mundane minutae of empire management. When you don't have to worry about checking every two minutes to make sure your peasants aren't standing around dumbly because their tree is already chopped, you can worry about the big strategic questions.

You have to constantly balance the need for economic expansion against the need for military development to defend your economy. If you over-extend and new towns fall to the enemy, your empire can rapidly become unsustainable while the economic benefits that have been seized from you fuel your opponent's war effort. The loss of a key town can turn the tide of a battle very quickly. The strategic considerations flow to the way you conduct a war. As you capture towns, you might decide you cannot afford to spread to armies thinly to defend them. You might choose to raze the towns to the ground instead to prevent them being retaken. Or, if you have found yourself over-extended and want to pull back to a line of forts, maybe you will implement a scorched earth policy and raze your own towns to stop them falling into enemy hands and serving as re-supply stations for an advancing army.

These are the sorts of strategic considerations that gamers expect from a genuine strategy game like Civilisation, but that you do not generally find in an RTS.

On the technical front - sound and graphics are good, about what you would expect from a 2D RTS in this day and age. It is worth noting that the game does not support video resolutions below 1024x768, 16 bit colour, though.

The control interface is simple. Anyone who has ever played an RTS will have no trouble, as the game conforms to the general left-click select, right-click set destination standard. There are some tricky exceptions such as when you are setting a retreat point, which might cost you a few companies in a heated battle, but generally you won't really need to refer to the reasonably thorough manual and quick key sheet you will find in the box.

A big, big plus for Kohan is the excellent level/mission editor that comes with the game. You hardly even need instructions - using it is an absolute piece of cake. You just select a terrain type or object and paint it on the blank map with the mouse. Once you have created a level, you can even go so far as to script AI and events, link levels and even create your own campaign. You might need to refer to the editor manual that comes on the CD for that, though. If you can't be bothered, there are a number of user-created maps and missions available on the Web, as are a range of new AI scripts for fresh strategic challenges.

Between the multiplayer options, skirmish modes and level editor, there is as much gaming in Kohan as you could ever need.

But there are some annoying things about Kohan. To a very large extent these are the same things that are annoying about every RTS though. So, rather than describe these as flaws in the game I would characterise them as flaws in the genre. Kohan has successfully busted a few of the worse RTS conventions - perhaps the developers can take to these problems in a sequel.

For a start, the game doesn't quite shake the standard RTS level concept: survive the initial attack, build your economy, build your army, destroy the enemy. Sure, there are the obligatory few levels where you have a set number of troops, or time constraints or whatever - but frequently the basic level design is pretty thinly disguised.

Not that this is all that bad. The enemy will have constructed or captured a number of towns scattered around the map, all with economic value, and will employ pretty cunning AI in defence and attack with feints, counter-attacks, and sacrifices. So even the formulaic missions are not mind-numbing in the way many RTSs are.

The real problem with the formula levels is that you do find yourself drawn into a pressured opening clickfest. You sometimes get the impression that if you do not issue economic and unit building orders in exactly the right order in the first couple of minutes a level will become unwinnable. The harder levels seem to ramp up simply by applying more of this pressure in the opening minutes. As per the formula, once you have established adequate defences the rest becomes basically a matter of time unless you do something really stupid.

The game also suffers from a general lack of atmosphere. There is just nothing particularly engaging about the story, despite the solid if somewhat predictable premise and plot. It is mostly text driven, with adequate but uninspiring voice-overs. There are occasional rendered cut-scenes, but these lack any dialogue or continuity, and don't identifiably relate to the missions or the characters. They could be clips from any fantasy game selected at random for all the effect they have.

The main characters aren't really developed to the point where you really care who's briefing you, or who just joined or left your party. In fact some of the characters begin to become downright annoying. Mostly this a function of another terrible RTS convention, the 'unit acknowledgement'. While it is critical, in the heat of battle, to know that your mouse clicks have issued the order you intended, hearing a character utter the same cheesy and suggestive "What will you have of me?" time and time again really starts to grate. It's about time RTS games figured a way to use obvious visual unit acknowledgments, I reckon.

One character in particular is worth singling out for ridicule though - a sort of Kohan spirit-guide, who gives you mission instructions during the tutorial and early in the game, and turns up again at the end. The guy looks and sounds like Lorne Greene on narcotics. Kohan would also have benefitted from more attention to the small details that can help to create real atmosphere in a game. For example, the music is as uninspiring as you would expect from an RTS, and you will likely turn it off before finishing the first level. A less obtrustive, more varied score would have helped a lot.

Similarly the look and feel of the levels is all a bit samey. There are terrain variations, but these don't quite succeed by themselves in conjuring up a real sense of menace or anything else for that matter. Weather effects could have made a huge difference.

None of this is much different from any other RTS, or many other games for that matter. It's just that Kohan comes so close to actually taking you past the tired RTS conventions that when it doesn't try to break the mould it becomes quite noticeable.