Bug? GDSM squelched by 'bad'

[Raggy]

Just found a GDSM which was promptly squelched. I had body armour set to "good". I had to set it to "no squeclh" in order to see it again, and "bad" squelched it.

Surely it shouldnt be even squelched by "good" ?

[d_m]

Just found a GDSM which was promptly squelched. I had body armour set to "good". I had to set it to "no squeclh" in order to see it again, and "bad" squelched it.

Surely it shouldnt be even squelched by "good" ?

Do you have a savefile? If I create a set of "normal" GDSM (e.g. magical) then it only gets squelched by good or better.

[Raggy]

Do you have a savefile? If I create a set of "normal" GDSM (e.g. magical) then it only gets squelched by good or better.

Yes, here it is.

[Derakon]

DSM is simply another armor type, so if it has no ego or artifact qualities then "good" will squelch it. "bad" shouldn't squelch it unless it's generated with penalties, which I'm pretty sure can't happen (DSMs have an innate magical AC bonus of +10).

[Max Stats]

DSM is simply another armor type, so if it has no ego or artifact qualities then "good" will squelch it. "bad" shouldn't squelch it unless it's generated with penalties, which I'm pretty sure can't happen (DSMs have an innate magical AC bonus of +10).

But there is a disconnect in that squelching good body armor will squelch Red Dragon Scale Mail but not armor of Resist Fire (for example).

[Magnate]

But there is a disconnect in that squelching good body armor will squelch Red Dragon Scale Mail but not armor of Resist Fire (for example).

This should be fixed in 3.4 - either because of Eddie's revised squelch code coming in with rune-based ID, or simply with a new OF_GOOD flag for specific items like DSMs (ticket #1415).

Can anyone reproduce this bug or is it a one-off Windows-memory-leak freakshow of a DSM?

[d_m]

This should be fixed in 3.4 - either because of Eddie's revised squelch code coming in with rune-based ID, or simply with a new OF_GOOD flag for specific items like DSMs (ticket #1415).

Can anyone reproduce this bug or is it a one-off Windows-memory-leak freakshow of a DSM?

If you load that savefile I bet you'll see the bug also. I was able to.

[Magnate]

If you load that savefile I bet you'll see the bug also. I was able to.

Yes, but if you can't reproduce it with any other gold DSM, then the chances are that the object in that savefile is corrupt, rather than some dodginess in the code.

For example, was that savefile loaded in from a previous nightly? It's possible that it has some particular non-visible flag set or unset which is different from normal gold DSM.

I suspect there probably is a bug, but it will be much easier to fix if we can reproduce it with a second DSM.

[Magnate]

Yes, here it is.

What zip format is that?? Zip is telling me:

chrisc@junior:~/workspace/angband-github$ unzip -l ~/Desktop/Shelie_GDSM.zip 
Archive:  /home/chrisc/Desktop/Shelie_GDSM.zip
  End-of-central-directory signature not found.  Either this file is not
  a zipfile, or it constitutes one disk of a multi-part archive.  In the
  latter case the central directory and zipfile comment will be found on
  the last disk(s) of this archive.
unzip:  cannot find zipfile directory in one of /home/chrisc/Desktop/Shelie_GDSM.zip or
        /home/chrisc/Desktop/Shelie_GDSM.zip.zip, and cannot find /home/chrisc/Desktop/Shelie_GDSM.zip.ZIP, period.

Anyway. d_m tells me that the GDSM in question has only +6 AC. Since all DSMs are generated with a minimum of +10, this is being squelched as "bad" because it's below the minimum. This isn't actually a bug, according to the way the code is currently written (and it's Eddie's code, so you can be fairly sure it is doing what he intended).

There is a question to be considered though: should we change the default evaluation of "bad" from "less than the base object's minimum value" to "less than 0"? I am ambivalent about this, but it would seem to make sense in this case.

[jens]

I belive most people's intuition would use 'less than 0 is bad', and I can't think of any cases where it would not make sense...

[Magnate]

I belive most people's intuition would use 'less than 0 is bad', and I can't think of any cases where it would not make sense...

That's d_m's view too, so I'm happy to implement that in the next RC.

[Derakon]

I agree. Bad items are items that are worse than average, and average items are ones that don't have any magical bonuses.