Bay 12 Games Forum

...
Please try to stay on the topic of development.  These questions are kind of intrusive and irrelevant.  We aren't paparazzi.

So... me hiding out in Toady's closet with a camera right now is... inappropriate? 

Ah, yeah an alphabetizer does sound potentially useful.  It ought to be pretty straightforward to do - I'll try to incorporate that sometime. 

whoops - meant index.txt

modified my previous post with the correction

Alrighty, just uploaded a new version that ought to give some messages to let you know when it doesn't like something, and should provide some kind of output even when it fails. 

The problem stems from this program mainly being a raw structure issue diagnoser and formatter that I sorta shoehorned into a stonesense graphics converter, and as such there isn't much in the way of raw argument error processing.  DF complains about that stuff itself like 99% of the time, so I just never bothered with it.  As such, while the tag name (0th argument) does generate errors if it is unrecognized, the rest of the arguments weren't set up to.  The version I just uploaded should mostly fix that, though I'm sure there are creative ways still to confuse it. 

The first kind of error was coming from lines like this:
[MILITIA_CAPTAIN:1:4:AS_IS:DEFAULT]and is a result of a missing TILE_PAGE specifier.  Adding a TILE_PAGE (eg: PUTNAM_TROLLS) fixes this. 

The second kind of error stems from including a TILE_PAGE name, but not having that tile page defined in the file.  Simply copying the tile page specifier to the top of the file with the others should fix anything like this. 

Hope that helps

Getting errors I don't think are errors:

unrecognized tag:   WOOD_MAT   in/item_tool.txt   189
unrecognized tag:   FURNITURE   in/item_tool.txt   191
unrecognized tag:   INVERTED_TILE   in/item_tool.txt   193
unrecognized tag:   WOOD_MAT   in/item_tool.txt   202
unrecognized tag:   FURNITURE   in/item_tool.txt   204

The line numbers as reported by Notepad++ are 190, 192, etc..

Ah, thanks for noticing that - there were some missing tags for tools (there are probably lots of others as well).  The paramaters files can be modified to accept any arbitrary new tags, so if I end up slow to respond in the future you can do a direct fix.  I haven't bothered doing a really clean and complete job of tag definition compilation since there's been a new DF version in the ether for several months.  Once that rolls around I will probably spend some time getting everything new added in. 

Quote from: Heimdall86 on March 06, 2013, 02:52:46 am

Installed the java runtime but still having a hard time running the .jar files themselves unfortunately. Which specific java application should i be using to run these. (javaws,javaw,java,javacpl)
Sorry for being noobish. I keep getting "Server could not find or load main class raw ssense. Any chance you could explain this to me like I'm 10 how to convert masterwork mod files into stonesense.

I get the same error, what could it be?

Hmm, well it sounds like a java problem of some kind.  What operating system are you on and do you know what version of java you have installed?

The ssense converter doesn't work...

I need more information to be helpful here.  I will say that the converter is very basic, and doesn't handle every case completely.  If there is an errorlog produced by the converter, post a relevant-looking section here. 

I for one would like to welcome our new Carp Overlords!

vivant et rex impuneque seligit piscis

@CaptainArchmage: I think maybe you're not getting the full memory footprint reported by your OS; as far as I'm aware, the entire map's tiles are always loaded when you have a fort running.  If you have 139 z-levels  down beneath the surface, then neglecting any air above the fort that gives 139*(16*48)^2 = 81,985,536 tiles.  The minimum data stored per tile is ~12 bytes, meaning the absolute minimum amount of memory your map can possibly take up in main memory is close to 983,826,432 bytes, or just under a gig, before adding creatures, history, or items.  I would guess that you're getting reports about the active or possibly physical memory being used by the program, rather than the actual amount of allocated memory.  Since much of the map (and history and whatever else) probably isn't getting looked at very often, this number would be much lower.  On reload, you're getting to see the full number very briefly.  ~1.7 gigs sounds reasonable for the amount of data a 16x16 embark has loaded; and if you happened to settle on a location deeper than the one you did (I've frequently seen ~200 z-levels), or if you had a longer world history, then you'd be crashing.  (correct me if I'm mistaken about this interpretation)

In the future, perhaps it would be prudent for the game to simply not load regions of unexplored solid rock in order to save space in main memory as an optimization, and I'm certain there are other potential improvements to both the space and speed limitations that DF has, but like everything it's a matter of priority.  I would probably personally prefer optimizations to worldgen so that I can reasonably generate a 2000-year-world or so, or perhaps the ability to create player forts that straddle regions, but we'll see what gets done and when.