Magic is underpowered. And no, I'm not basing it only in the fact that the AI loses every single game in the same conditions. I get it. Magic is a different kind of faction an the AI just does not know how to play it.
Doing maths you can see that the power of the faction is just lower than, lets say Tech. (And tech is not overpowered by any means)
I could create an algorithm that calculates the perfect strategy that takes into account all factors from the faction plus some of the map and environment. In fact, i will do that and post the results in another post.
But for now, I would like to know if the project managers are open to balance changes in the tech tree.
If it is the case I could come up with a set of changes that can then be revised and approved or rejected independently.
I keep in mind that magic is a special kind of faction and that giving it more power doesn't mean converting it into mass producing unit factory.
Proposed changes:
1 - As an initial idea, I would increase the building speed of the initiate. This would give the faction the possibility to build new bases quickly. To avoid this messing with the initial base establishment, it could be applied through an upgrade (that would be original since there is no upgrade like that in the game, but the game is capable of doing this). It could be added into the energy sharpening upgrade.
Think of this. Because they build things magically, it makes sense that they are fast doing so.
2 - Also, magic initiates are the only worker units that can attack. The attack effect is cool and all of that, but in practice they are pretty useless. I would not buff them enough to make of them a useful offensive unit, but I would make the effect of the attack noticeable. A quick fix would be making the effects of the energy sharpening upgrade affect the initiate as well.
If the community and the mg team agrees with that, I can make a pull request to megaglest-data on github with the changes.