Re: [glob2-devel] Minimap algorithm

[Stéphane Magnenat]

> Calm down Nct. If Brad wants to rewrite code, let him!! As long as the new
> system works as well or better (which it does not atm but will soon?) as
> the old, and of the same speed or faster, and of better coding (easier to
> add to) then who cares if it gets rewritten (its when its more complex to
> code, slower, or doesn't work the same that should get someone "somewhat
> angry, and frustrated"!! The game is supposed to be about speed and
> functionality, not showing off code you made. AFAIC, if the whole of
> Glob2's code got rewritten (removing all of your code and anyone elses), as
> long as it works, then fine. The player doesn't care who made the game, as
> long as it goes.

Usually, I'm not that much of a troller, but that is going too far.

The minimap code is fine and nice. It was somewhat slow 6 years ago, so that's 
the reason it was rendered progressively but nowadays I'm sure it can be done 
in one pass.

The problem is not the minimap, it is that in this case Bradley has rewritten 
from scratch a code that was well written, worked fine and produced nice 
result, where he could just have modified it. The result is an inferior 
version.

The problem is not about rewriting the game, it is about improving it. There 
is a huge list of things to do to improve the game, and rewriting the minimap 
is not part of it.

> And with the new code, developers will be able to add heaps more code and
> functions to the game. Its all good in the end.

That's a myth. The old minimap code is very readable and easy to work with.
The new brings no advantage on that side.

If you want to make more people add things to the game, implement the zip 
filesystem, add wild animals, do special buildings graphics and interesting 
campaigns, add a nice feature to download maps from the network within the 
game. This list can easily be extended for a long time.

At the end, it is about respect. Rewriting everything just because one think 
that he will by default produce far superior results, is pure arrogance and 
helps the project in no way. Specially when the result is inferior. 

I've no problem when Bradley rewrites code that is difficult to enter and to 
manage, such as the network. But that's not the case here.

Steph

-- 
http://stephane.magnenat.net