Tweet
When do you get ID status on your boots of slow descent? Currently, you only need to learn they have feather fall. If you have boots of stability, and learn they give nexus resistance, you do not learn the ego. That seems backwards. Nexus is what you need to know to differentiate between them.
It seems to me that the current system, giving ID when you know all positive attributes of an item, only makes sense if you assume that the char knows the number of attributes of the item.
Knowing the number of flags is powerful information. E.g., the only boot ego with two flags to learn is stability, so that ego is presented immediately after pseudo.
I've coded up the change whereby sensing is defined to show the number of flags to be learned. I believe this to be a prerequisite for rune-based ID, but perhaps it is not. In any case, it is self-contained and meant to be suitable for current V.
I addressed all things sensing. I changed some IDENT_FOO stuff to mean "you whether FOO applies" rather than the old "you know object has FOO". Some quiver issues are addressed in passing. I did a lot of code cleanup, better variable names and moving cloned coded into subroutines, etc. There is now a single place for tricky obvious stuff such as FA on gloves for mage casters, and would be the place for making SI not obvious to high-elves [ugh]. I also changed ego and flavor identification to be according to the edit files, rather than the current "learn one attack flag", so westernesse e.g. is obvious on wield.
This won't be suitable for 3.3 given the feature freeze. I'm sorry it took so long, but I have been busy. It's still not complete, in the sense that pseudo messages don't fit the new framework. I'm not aware of any bugs.
I think that it might make sense to incorporate the changes into V even if there are philosophical problems with the general thrust. I cleaned things up a lot, and it would be easy to put in a few changes to get different behavior. Using and presenting the number of flags, e.g., is controlled by a variable, so it should be trivial to turn off, but I haven't really tested with it off.
There is a lot of interrelated stuff. I do not think it would be practical to try to get pieces of code here and there.
Anyone capable of compiling their own should feel free to clone and test.
If you are interested in rune-based id, this is not that, but it gives a hint as to the power.
I started working off of 4/24 vanilla. I only incorporated one unrelated bugfix, to allow detect invisible to show invisible monsters.
The commit is https://github.com/eddiegrove/angban...9f6fcb733a1c75
I hope I did that OK. I've never used git before.
It seems to me that the current system, giving ID when you know all positive attributes of an item, only makes sense if you assume that the char knows the number of attributes of the item.
Knowing the number of flags is powerful information. E.g., the only boot ego with two flags to learn is stability, so that ego is presented immediately after pseudo.
I've coded up the change whereby sensing is defined to show the number of flags to be learned. I believe this to be a prerequisite for rune-based ID, but perhaps it is not. In any case, it is self-contained and meant to be suitable for current V.
I addressed all things sensing. I changed some IDENT_FOO stuff to mean "you whether FOO applies" rather than the old "you know object has FOO". Some quiver issues are addressed in passing. I did a lot of code cleanup, better variable names and moving cloned coded into subroutines, etc. There is now a single place for tricky obvious stuff such as FA on gloves for mage casters, and would be the place for making SI not obvious to high-elves [ugh]. I also changed ego and flavor identification to be according to the edit files, rather than the current "learn one attack flag", so westernesse e.g. is obvious on wield.
This won't be suitable for 3.3 given the feature freeze. I'm sorry it took so long, but I have been busy. It's still not complete, in the sense that pseudo messages don't fit the new framework. I'm not aware of any bugs.
I think that it might make sense to incorporate the changes into V even if there are philosophical problems with the general thrust. I cleaned things up a lot, and it would be easy to put in a few changes to get different behavior. Using and presenting the number of flags, e.g., is controlled by a variable, so it should be trivial to turn off, but I haven't really tested with it off.
There is a lot of interrelated stuff. I do not think it would be practical to try to get pieces of code here and there.
Anyone capable of compiling their own should feel free to clone and test.
If you are interested in rune-based id, this is not that, but it gives a hint as to the power.
I started working off of 4/24 vanilla. I only incorporated one unrelated bugfix, to allow detect invisible to show invisible monsters.
The commit is https://github.com/eddiegrove/angban...9f6fcb733a1c75
I hope I did that OK. I've never used git before.
Comment