Research the changes made in RSW version 2.6
rdw-software opened this issue · 5 comments
GAT 1.3, GND 1.6 and RSW 2.6, the waterinfo is still not correct for those new, they removed waterinfo in RSW 2.6 and moved it to GND 1.6 (they changed that somewhere between 2021-11-03 and 2021-11-16)
Source: Discord
Hi Duckwhale
I remember that the latest gnd version is 1.8, which is different from the 1.6 you mentioned, am I wrong?
As far as I can tell, all the modern GND files use version 1.7. The files from Arcturus/the iRO alpha use a different version, but since they don't have any headers I'm not certain which one that would be... presumably 1.5, since that's what the converted alpha maps (from the rathena forums) have been assigned? No idea how (or by whom) they were created, so I can't say more.
With regards to 1.8, I'd have to see if there is an updated version available on kRO. Last I checked 1.7 was the only version used.
I haven't seen any GNDs using 1.6, so maybe that's a mistake.
Update: It seems I didn't have the latest files from kRO. After updating, I've just encountered this error:
I'm willing to bet the original comment was referring to 1.8 and not 1.6 (typo or oversight?).
It looks like they changed the RSM(2) model actor definition; there's a lot more data in there now. Luckily I was able to find the known properties present in previous versions, but I've no idea what the rest is used for... just some random ideas/guesses:
- In a kRO video of the new snow areas I noticed there was a texture overlay that is clearly added on top of the terrain (see this video)
- I don't necessarily think it's related to the model actors, could be an effect... but it is local to the scene, and not part of the GND/GAT, so it might indeed be in the RSW somewhere?
- Strangely enough, it does appear close to the ice floes (a few seconds later in the video)
- Not sure what map is shown in the video, but
1@iwpis a similar one and has those floes as well
Here's the parts I identified crossed out, with only the unknown data remaining (1@iwp.rsw):
I'm guessing the first part has something to do with the animation, since it directly follows the animation type identifier. Perhaps a setting to repeat it/repetition count/delay before looping or some other way to modify it?
The latter part is even more troublesome; the most sensible interpretation is that of uint8s (some sort of coordinates or width/height perhaps?) and then there's something that may be a color/transparency or offsets of some kind?
Yet another, completely different idea, would be to look for a connection to the Bink video feature added to RSM2, though that seems likely to be part of the model itself and not likely to be a setting for the individual instances placed in the scene... 🤔
Overall pretty weird, but unless there's a way to test those maps in a real client I'm not sure if I can figure out more details.
Presumed structure so far (still need to check out some other 2.6 maps):
Update: I forgot about the node name and block type. It seems like the only change is the (possibly animation-related) byte after the block type.