Stars-Util
Stars-Util |
|
|---|
Finally, the tool *certain* people have been hassling me about for a while now has been releaesd :-).
mSplit is a small command-line tool to split a multi-year M file into it's consitutent 1-year parts. Since this doesn't involve decryptin anything at all, and the necessary information has in fact been available since the days of the "starsfile" tool, the source code is included in the ZIP (along with compiled windows and linux binaries).
As already announced on the Stars! Autohost forums, I've put together a page where you can customise the tech levels of the stars! battlesim test game, useful most notably for testing (you guessed it) the outcomes of various battles. The original battlesim already has all of the MT toys (for all but a few races, which missed out on the genesis device), as well as a wide selection of PRTs with various combinations of NRSE and Regenerating Shields, enabling most conceivable battle scenarios to be set up. The disadvantage used to be that the tech levels of 26 in all fields sometimes caused undesirable effects with stars! targetting algorithm - being able to select the techs for the simulation is useful to ensure that the miniaturisation benefits of having higher techs do not interfere with the battle results.
Also coming soon is, by popular harassment from m.a@stars, a little tool for splitting up M files. Look for more news in the next few days. (it's done, but I guess I should provide it with *some* instructions rather than just a bunch of assert()s when stuff isn't done just right ;P)
I've started to put together some of the cargo / fleet splitting examples that seem to me most critical, and attmpted to explain the problem a little more clearly; I'll expand on this in the next few days (can you tell I'm no huge fan of HTML-fiddling? ;P)
See here for the details so far.
If someone has tried to contact me in the last few... uhh... well, weeks mainly, but possibly months... at my sourceforge addrss (IE the one that's linked here), then I won't have answered; it seems my mail that goes via that address is being eaten by a grue :(. I'll update all my email links soon; in the meantime, feel free to contact me at das.wumpus@gmail.com.
Apologies for any inconvenience :(
Since the last update, a lot of work was done on the figures in the split_fleet_cargo.txt file. Thanks in paticular to sotek, Leit and EdorFaus for some great work. Unfortunately, we still haven't been able to work out exactly what the Jeffs had done here, and without it, it's difficult to add the finishing touches to the cheat detection tool; granted, one could allow a certain degree of error to account for the fact that we haven't worked out this detail yet, but surely there must be some simple "trick" to how the cargo splitting works that we've overlooked.
In that light, I again invite contributions and suggestions! I'll update the file and add try to organise the thoughts and progress made so far in the next few days. I realise it's more than a little confusing at the moment :)
A lot of quiet again, I know; work has been keeping me occupied, and the project remains significantly "secret" due to the cheating concerns that have been pointed out repeatedly.
There's actually been a lot of progress, paticularly in the last month and a bit:
I can finally load a HST file into what I hope is at least a reasonable structure in memory, rather than just dumping the inidividual entries to the screen and moving on. This is, of course, essential to do any reasonable work with the file. A similar approach could be taken loading M files alone, but this has somewhat shaky results so far, on the one hand because M files contain certain data that is never present in the HST file, and partially because the HST file's data is complete in a way which is by no means the case for M files.
In the process of this, the file loading code has been much more cleanly split from everything else, although I'm still not entirely satisfied with it. The whole code has become more "object orientedish", but there are also still quite a few modules which need cleaning up to the newer approach. Got to love code overhauls, they give you something to procrastinate.
The set of understood "news" items (those messages in the box,
like "Stinky Socks has built 69 mines this year") has also been
continually increasing; I think I've got almost everything which
comes up in most games now. No "class" of things which are
missing immediately comes to mind anymore - I've covered most of
the PRT and LRT specific things which came to mind, as well as a
whole lot of random events. My favourite message at the moment
is probably: Your starbase at %{planet} has built %{d} new
%{ship}[s] but due to a lack of 27B/6 Fleet ID Requisition forms
they have been merged with %{fleetName} which fortunately
happened to be nearby.
A "small" game_test module can now load an XY, HST
and X file together and attempts to validate whether the
commands in the X file are credible. It catches a lot of cheats
now that stars! itself does not catch, notably one involving
fleet movement that I've alluded to a few times in a few places.
It also tries to catch some interesting tricks that can be done
with cargo, but at the moment there are still some problems
there because I don't know how stars is supposed to
handle certain cargo actions.
Which brings me to a point where contributions are welcome once again. In split_fleet_cargo.txt I've listed a bunch of different fleet splits of a 2-privateer fleet carrying a bit of cargo. If someone has any thoughts on what rules Stars! might be using to perform these splits, please contact me! More details in the split_fleet_cargo.txt file.
Other progress has included a valuable contribution from Autohost's Ron, which I can't go into at this time. Thanks Ron!
All things considered, there is a glimmer of hope on the horizon that maybe, just maybe, something interesting can be released in the forseeable future. Don't hold your breath just yet, though :)