Overwrite->patch conversions

[Guest temujin.]

seriously, why does patching the world map make people so uncomfortable?

Too much info to deal with, I guess. Even more that with ARE files, though it may depend.

 

well, it may be too much to deal with for someone new to modding or for unskilled pretenders and street-kebab vendors like Baronius and Sikret (in which case, it's perfectly understable), but certainly not to the fixpack team who are fairly highly skilled.

 

which is why devSin's response is more surprising... and a little saddening. even if we accept that file patching wasn't 'invented' back then, there is still no reason why it shouldn't be done now.

[Salk]

even if we accept that file patching wasn't 'invented' back then, there is still no reason why it shouldn't be done now.

 

I fear that the reason lies mostly on the fact that we no longer have a Fixpack team, actively working on the Fixpack. It's sad but it seems so. CamDawg is gone and Nythrun as well. Several contributors have also moved ahead and DevSin doesn't do much modding anymore.

 

I think they're just trying to mantain the Fixpack without looking for ways to make it more modern and efficient.

 

There are some that from time to time contribute something (Wisp himself, aVenger, Mike to mention few) but it's individual work which sometimes doesn't even end up in the updated version.

[DavidW]

even if we accept that file patching wasn't 'invented' back then, there is still no reason why it shouldn't be done now.

 

Actually that depends on the reason for patching. In general, mods patch because they don't know what else might have been installed. This isn't really an issue for a Fixpack, because it can stipulate that it's supposed to be installed first. Even then, in general it makes sense to patch, because it makes the code easier to understand and maintain. But now there's a cost-benefit analysis, and if patching is too horrifically complicated, it's not worth it.

 

(I'm not speaking from any real experience with Fixpack, but IWD-in-BG2 works this way: it absolutely assumes a vanilla, unpatched install, but it's still largely patched because it's easier to do that way.)

[Ardanis]

Even then, in general it makes sense to patch, because it makes the code easier to understand and maintain.

I'm not sure about 'understand' part, the whole worldmap piece takes a couple of minutes just to read, nevermind to comprehend. Well, unless we already familiar with the subject.

 

I'd say that the point of code format being easier to maintain is that once we've got a working code stored in a lib, we can define several values in installer and tweak them at our pleasure at will.

[Jarno Mikkola]

But now there's a cost-benefit analysis, and if patching is too horrifically complicated, it's not worth it.

And then there is the fact that the worldmap has way too many states to be simply patched at any given state. I mean that as there is number of mods that all ovewrite the install times existing map... This has lead to the current situation, if you wish to have any two of the map modifying mods, you need to have the Worldmap mod too, which I believe does this: It first overwrites, and then patches itself up as it knows what to look for.

[CamDawg]

This is from memory, but as I recall there were essentially two major issues I went after with the fix:

 

1) Inconsistent travel times. Going one way was 12 hours and the return only 4. Replacement areas exacerbated many of these problems.

 

2) Lack of travel links that resulted in absurd travel times, i.e. chap 6 areas to Trademeet.

 

Now, there was one other thing I went after that is probably causing the huge comps between the original and fix maps--lots of redundant travel links, on every map edge. In the original you have a lot of areas that have (for example) two links to the north, one west, none east, and two more south, and few of them went to the same areas. So, I would get the links to match one another and put them on every edge, often resulting in an area which had previously 6 links now suddenly has 16 or 20. In retrospect this is (mostly) unnecessary and cosmetic, but in my defense I was still learning the wmp format and thought these were necessary.

 

As for why the overwrite vs. patch originally? The huge, huge comps. It was such a complete teardown by the time I was done that it simply wasn't worth it--basically the equivalent of deleting all abilities and effects from a spell and rebuilding them via code. At a certain point it simply becomes an exercise in wasting time.