Sooo... I got Spikes Babylonians, was pixeling a bit of the scout... and quickly decided that I'd like to have a naked character
template first, without the bling and stuff. Being the most simple graphic, I wanted to start with a carrier. And since I really like
to pixel and did not want to start this project with a lot of grinding... yeah...



Not much, really, I only made new heads for the carriers for each available race and thought of a basic body design.

- the roman does not change much, but I made the beard* of the fat carrier more prominent.
- vikings have long blonde or red hair, light skin and blue or green eyes. Since they are in colder environment, they are allowed to
wear pants. I suppose their design could be split in two races (blonde vikings, red-haired celts)
- babylonians are based on the half-finished scout, with hair instead of bling. Darker shade of skin and a black beard.
- nubians are based on their soldiers and pretty dark. I decided to give them a necklace to improve readability.
- japanese get a traditional hairstyle, lighter skin and wear a kimono.


So far, I pixel for fun, not for a purpose. Being new to this project, I'm not sure what's the current idea on how to progress, and
what players actually want to see. As long as I have fun with something, I'll do it because I want to, but I already know that I won't
have fun copypasting endlessly for each carried object and micromanaging thousends of tiny graphics. Still, I could see myself grinding
through the stuff if that is what players actually want.

So my questions are these:
- Do players of RttR actually want change, or should everything stay as close to the original as possible?
- As far as I understand it, RttR only needs S2 for graphics. As such, if someone was to provide new textures, replaces each object and
was fine with only playing babylonians, could RttR be a game of it's own which practically uses S2 as a "DLC" for additional races and
alternative content? Is that in any way a goal of RttR, or would it be considered if graphics were available?
- How much alteration needs to be done so you would consider publication legal? (Eg. the romans don't change much, but even the tiny
change would need to be published with RttR as new graphics.)

I'm still in orientation phase. I tried to replace the jobs.bob and carrier.bob with my own to test my heads and failed miserably, so I
don't even know what I'm doing... don't expect too much of me for now. ;)

---

Those carriers are amazing, keep on going with your flow :)

Zitat:

With the addon system everyone theoretically can decide themselves how close they want to be to the original.

Zitat:

There has been some ideas/efforts made going in that direction, for example here: http://www.siedler25.org/index.php?com=forum&mod=forum&action=thread&id=881&start=1 Sadly Merri's graphics aren't shown anymore.

---

I would actually welcome a standalone RttR. However the work required is massive. It's not only nations, buildings, wares, people but also all the GUI stuff.
Not sure if that is really worth it.
So yeah I think an addon with nation specific figures would be great.

Note: You need to use the original color palette. Otherwise the import into the LSTs won't work.
About the bob files: I don't think we can pack them yet, as the format is quite complicated (some mapping from wares to bodies etc).
What might work: Using override folders. So create a folder in "<DATA>/LSTS/GAME" ("<DATA>" is the data folder for personal data. That is something like "~/.s25rttr" for linux or "Saved Games/Return To The Roots" for windows), call it "CARRIER.BOB" and put your bmps right in there name the same as the lstpacker outputs them. The loading system will load the original first, then places every entry from the override folder in the loaded archive. This way you don't need to pack stuff and can simply use bmps.

---
Github: https://github.com/Flamefire

---

It's really just a matter what I'm going for. If the end goal is a standalone game, I'd have to replace the wares settlers carry.
Otherwise, they can stay the same. Since it's already a massive undertaking (~25000 graphics?) I wouldn't want to do it twice, so I
want to plan ahead.

If I redo the wares, that opens new questions. Should/Could wares be race-dependent (eg. a coin with a square hole for japanese)? Could
wares have race-dependent names as well (eg. make goats instead of pigs with no functional difference)? Or would that rather be
realized as an addon of race-specific economy? (I guess functional difference in races is not possible yet)

---

This sounds great to me though it would be definietly a challenge to create THAT MUCH bmps for each
nation. In the first place - what I really would like to be improved/changed - is the skin colour for
African People as for now it looks ridiculous (black soldiers and white workers xD)... I don't really
know though how to implement this. All I do here is this small project with Celtic Nation
(graphics&soldiers) so I am noone :D

But still, if you would like to "fix" african people only than.. still -> huge amount of work :p

---

As of "could": We already have race-dependant shields and I want to push in that direction by using lua for game data (see the folder "gameData" where there is already terrains)
Currently I'm working on buildings and nations in general (especially for arivald) but that is a huge project already. When done I'll be going for fighters and then wares (at least graphics). Currently we have a different ware for each nations shield although it is actually just 1 ware but different graphics. So I want to unify that. When that is done we can have different ware images per nation. Then I'll do carriers/workers. But again this is a huge coding task and I have to see how I can make it work...

Functional differences could be possible (e.g. different costs for buildings) but not anytime soon as we want to stay as close to S2 as possible, so something like that is at the very back of the list.

My call: keep the wares the same except for the shield.

About redoing wares: You have to redo the carry animation anyway which includes the ware, right? IMO this is already a change big enough to keep at least the carried items the same as currently w/o copyright issues (the ware is just a tiny amount of the image and you painted it by hand, not Copy&Paste) but I don't know the legal side here.
As a complete replacement is unlikely due to the work involved I'd stick with the current wares unless you find a way to automate the replacing of the wares so we could have a separate set there.

---
Github: https://github.com/Flamefire

---

Zitat von Flamefire:

The real question is whether the ~150px a settler is made of (or an even smaller ware) is of artistic value. If it isn't, it's not
covered by copyright. If it is covered however, it does not matter how tiny it is compared to my own work. Especially if it isn't
insignificant. (German Law, "Schöpfungshöhe" and "Beiwerk")
But even so, it does not matter what the law says anyway. If Ubisoft calls to shut this down, who is going to go to court against them?
Not me... ;)

Anyway, I'll leave the wares as they are when I reach them. As I started with the Jobs.bob, it will be a while.

Zitat von Arivald:

The real challenge for me is handling the files. I spend as much time opening and saving each single bmp as I do pixelling, and that is
with a somewhat 'optimized' workflow (doing all races in parallel to reduce the the amount of file handling). It's silly.

Zitat von Arivald:

Let's make it a race, who's done first with his ridiculously oversized project? xD

---

Progress so far: Thin settler body and face...


I have a question: Does RTTR care about the unused space? Could I, instead of using the "original" size of the images, place everything
in a 32*32 px raster and use the same offset for everything? I think that would help a lot, especially since black and asians have
higher hair.

IF that is how it works - would it be possible to write a tool that "converts" images by adding/removing rows of empty pixels and
renaming the file appropriately? I think that would speed up the progress by at least 50%...

---

You could use an image tool like https://www.codeandweb.com/texturepacker to trim your sprite sheet and use any advanced file explorer to batch rename your files with correct offsets

---

No Rttr does not care about empty space. It uses the nx/ny to paint the image at the current position offset by nx/ny. It is very much possible to write a tool reducing the images automatically. The lstpacker uses libsiedler2 to load bmps and the classes used have methods to find the non-transparent bounding box. So using that, one can write a tool based on libsiedler2 only (similar to lstpacker) that loads a bmp, crops it and adjusts nx/ny appropriately.
Should not be to much work. Only thing to make sure is to adjust nx/ny in the correct direction. I keep confusing the exact meaning.

---
Github: https://github.com/Flamefire

---

Zitat von Jonas:

I doubt it. Looking through what that tool can do, there is no indication that it could add or remove whitespace based on instructions
given in the filename. It might be able to see the the sprites on my sheet as individual graphics, but there is no offset information.
Would be hard for a batch rename to guess them, especially since they are not the same as in the original graphics.

Zitat von Flamefire:

If RttR does not care about the empty space as long as the offset is correct, the cropping would be optional, as it would only save
some storage space... I'm going to assume that's not an issue in this day and age.

But could you write a tool that does the opposite - adding whitespace around the sprite based on a size and offset given?
Eg. I want 0.player.nx16.ny12, which is 32*13 pixels, to fit in a 32*32pixel graphic with offset nx16.ny26.
You'd need to subtract the original offset from the target offset and add that many empty pixel rows/columns to the left and top. (26-
12=14, so add 14 rows on top)
Then subtract the current size from the target size and add that many empty pixel rows/columns to the right and bottom. (original 13 +
14 added is 27; 32-27=5, so add 5 rows on the bottom)
Then, save that new, larger image with the same index, but offset nx16.ny26.

It seems simple enough, if you have the tools to manipulate the image in that way. I'm certain it could be done via
photoshop script, since all the required methods are available, but my skills are pretty much limited to a bit of self-educated VBA in
Excel. Also, sparrows, cannons, nuts and sledgehammer...

---