Talk:Oni engine patches (Mac PPC)/changelog: Difference between revisions

From OniGalore
Jump to navigation Jump to search
(env_ray_number and signatures)
(forgot something important about large textures)
Line 11: Line 11:
:I don't see how it could parse the command line without those signatures. [[User:Neo|Neo]]
:I don't see how it could parse the command line without those signatures. [[User:Neo|Neo]]
::Actually, it looks like the signature strings are only there to detect conflicts (if any) before calling the function. In that case the function isn't called at all and an error message is printed to the console. So that string is not strictly necessary, but if it's absent then the possibly invalid arguments will be passed to the function, which may not always handle them correctly. That said, it's also possible that some signature strings are incorrect: in this case I suppose the "security check" could have rather nasty consequences. --[[User:Geyser|geyser]] 02:34, 22 October 2008 (CEST)
::Actually, it looks like the signature strings are only there to detect conflicts (if any) before calling the function. In that case the function isn't called at all and an error message is printed to the console. So that string is not strictly necessary, but if it's absent then the possibly invalid arguments will be passed to the function, which may not always handle them correctly. That said, it's also possible that some signature strings are incorrect: in this case I suppose the "security check" could have rather nasty consequences. --[[User:Geyser|geyser]] 02:34, 22 October 2008 (CEST)
----
Question about the 1024x1024 patch. Previously we patched 0x1151C3 to accomodate 512x512 textures and it didn't work. Now we're patching 0x1220AF and 0x1220B3, and 1024x1024 works. Does it mean that 0x1151C3 was completely irrelevant and can be reverted? --[[User:Geyser|geyser]] 02:40, 22 October 2008 (CEST)

Revision as of 00:40, 22 October 2008

"at 0x16E834 Replaced chr_focus with chr_health"

Why would you do that? Chr_focus is very useful. Gumby 10:24, 20 October 2008 (CEST)

When it comes to replacing functions, you have to consider the length of the name and signature string. chr_health had a longer signature string "chr_index:int [hit_points:int | ]" than chr_focus "chr_index:int" So it will easier to find another function to replace chr_focus with the shorter signature string. EdT

Hey, why would you want env_ray_number anyway? You cannot set it higher than 20 and you don't want to set it lower than 20 either (unless your CPU isn't fast enough :) ). Neo

Neo, geyser asked me to test env_ray_number and to see if the call back was setup in the Mac engine EdT
That's right. And I'm glad I did: I had no idea that values larger that 20 had no effect ^_^ Live and learn... --geyser 02:34, 22 October 2008 (CEST)

Are the signature strings actually used by the engine? I assume there would be no problem if they were left blank. rossy 12:40, 21 October 2008 (CEST)

I don't see how it could parse the command line without those signatures. Neo
Actually, it looks like the signature strings are only there to detect conflicts (if any) before calling the function. In that case the function isn't called at all and an error message is printed to the console. So that string is not strictly necessary, but if it's absent then the possibly invalid arguments will be passed to the function, which may not always handle them correctly. That said, it's also possible that some signature strings are incorrect: in this case I suppose the "security check" could have rather nasty consequences. --geyser 02:34, 22 October 2008 (CEST)

Question about the 1024x1024 patch. Previously we patched 0x1151C3 to accomodate 512x512 textures and it didn't work. Now we're patching 0x1220AF and 0x1220B3, and 1024x1024 works. Does it mean that 0x1151C3 was completely irrelevant and can be reverted? --geyser 02:40, 22 October 2008 (CEST)