Lynx3d

knockback effects are actually a SMSG_MOVE_KNOCK_BACK, has nothing to do with this. Not really true, when using thunderstorm on mobs server does send trajectory movement, only difference is another flag that flips the orientation so "leap" becomes "knock back" effect. For players it's handled by the client of the affected player, server only sends command what momentum player got, and off you go... I experimented a while ago with a movement generator for knockback spells, but it's pretty rudimentary still (assumes flat terrain): http://github.com/Lynx3d/mangos/tree/knockback

So did anyone test it now? To me, Revils' code looks good, from all i could find the calculations seem correct now for 3.x. The only thing i can't really find any confirmation is the 0.02% part in hitChance = 95.0f + skillDiff * (skillDiff > 0 ? 0.02f : 0.04f); Wowwiki just states that formula like this, but at no other site or tool (like tankPoints, Rawr etc.) i can find this 0.02% part.

I just thought "wtf it still is as unreadable as before" but somehow, my browser cached it And yes, much better. With the light background it is no problem to read, and also doesn't look like someone posted console screenshots anymore ^^

whee syntax highlighting again Though the colors...maybe it's just me, but blue and purple on dark grey is almost unreadable here...

If you would just FINALLY look at the gcc documentation, we wouldn't have to do the reading for you: http://gcc.gnu.org/onlinedocs/gcc-4.5.0/gcc/Optimize-Options.html#Optimize-Options Besides, -fstrength-reduce does not even exist anymore, got removed from docs with gcc 4.2. And if you actually READ some of the google hits, you'd know -fno-strength-reduce was used waaaaaaaaaaaaaaay back in gcc 2.7 era to work around a compiler bug, because -fstrength-reduce is (err was) turned on with -O2. Note that -fno *disables* something... Just once more, adding random options you read somewhere is nonsense, most of them are either platform specific, compiler version specific, or simply useless. O2 includes pretty much everything that generally makes execution faster. O3 is a lot more aggressive, but may actually overtax your CPU caches, and hence even reduce performance. -march is the only "general" optimization left for x86-64 targets, and this assumes you will be running the binary on that very CPU *only*. Phenom would be amdfam10/barcelona but you need at least gcc 4.3 for that.

It's all documented at http://gcc.gnu.org/onlinedocs/ And if you don't even know what options do, don't use them. I've seen too many people using a wild mixture of redundant and unsafe options just because they read somewhere that it might improve performance a bit.

I bet they did read it, but you apparently have no clue how Cell actually works... "Normal" C/C++ code doesn't just "run on Cell" after throwing a compiler at it. Well it does, but not on the SPUs that make the vast majority of its transistors, but on the meager PPE that has the performance of maybe a Pentium3 at best. Using the whole Processor means writing new code that needs to be compiled with a special compiler and run in one thread for each SPU. But let's just be generous and still call that hypotetical rewrite "MaNGOS", would it perform well? Probably not... MaNGOS consists of two relevant parts really, a very branch heavy logic core, and some float heavy geometry interaction (vmaps mainly). The SPEs maybe could do the float parts (although i have yet to see a convincing raytracing demo on cell...) but the SPEs suck at branch-heavy code, and the the PPE is hardly capable to compare with even one core of a current Intel/AMD CPU when it comes to logic code...

ccache doesn't help there at all...the only thing it can do is work around the lousy file/environment change (mis-)detection of Make. But if you DO change the header, ccache has to recompile every source it includes it. Besides, i think he's on Windows anyway ^^

Just to make one thing clear, scaling everything with the same factor has no influence on threat mechanics at all. And since we are calculating with float, multiplying the internal values won't really do anything useful, so as mentioned SendThreatUpdate() would be the more logical place... Side note: I'm not sure whether the choice of patrick star as avatar influences the way i perceive your posts...

Or even better, don't copy ad.exe at all and use -i option to specify input path...

Apparently alternative was implemented in [10305]

So...pure SD2 problem now. I'll classify this one obsolete and move to rejected.

Moved to rejected then...

Just to clarify, the description of this board is "Discuss code, review existing code, submit patches, or write implementation drafts for new and missing features." Your posts "somewhat" lack a productive base (not to call it a blatant feature request), so you shouldn't be surprised of such precise answers as "write a script"...

Maybe you should read the SD2 changelog...

Any news on this? I had no luck finding that information elsewhere yet

Fix in [10461] Thank you!

Can't reproduce with [10457]. All mobs in Falthrien Academy work just fine. What does .gps give you for floorZ values? They should be (almost) equal to you character Z. If not, your Vmaps are not working.

Hm seems that kFreeBSD has a bit of a Frankenstein character...kernel is FreeBSD, rest is GNU (including glibc) so it's hard to say if you shall treat it as BSD or GNU/Linux when it comes to compiling, chances are neither works. But it seems that you should more treat it like Linux, that's how i interpreted the Debian guys, it is as GNU-Linux like as possible to make most Debian Linux packages compile with it. Still, the problem is that there doesn't seem to be an easy way to identify kFreeBSD via preprocessor macros Some suggested to test for "#ifdef __GLIBC__" but it wasn't quite clear to me if gcc predefines it, or if you need to include a special header first, apparently this was changed at some point, but i didn't find out from which to which. Maybe just try "#elif defined(__linux__) || defined(__GLIBC__)" in line 51...

Hm do i understand that correctly, you are running the precompiled ad.exe from the repo under Win XP withing VirtualBox on Linux, and extract from an OSX client install? Phew, hard to say what might be the problem... But just a suggestion, wouldn't it be easier for you to just build "ad" on linux? There are CMake files for it, just do "cmake ." in contrib/extractor and then "make". You might even have a chance to compile it on OSX that way and directly extract from there. Or as suggested, manually extract Spell.dbc with some MPQ tool, on Linux you can use mpqfuse to mount mpq files like a file system for example.

I guess you are talking about core dumps in order to obtain stack traces... you may want to read up on "ulimit -c" and "/proc/sys/kernel/core_pattern" then. Some distros may add their special sauce to it though...

So you want to select all spells with Atrributes & 0x40... Those are about 7500 spells btw, i'm not going to post them all

Okay this discussion is pointless. If you don't know what git is or what revision you are using, this can only mean you did not even obtain the source from our project repository. Read up the basics and come back if you can give some useful information about what you actually did so someone may actually be able to help...in case it still does not work.

I very much doubt that this is necessary to fix anything...it only wastes a lot of time and github bandwith... If anything was modified locally, "git status" would show. This can be easily undone with "git checkout -f". If you merged in any commits you want to get rid of, "git reset" is what you want, e.g. "git reset --hard origin/master". But we still don't even know what revision or branch you are talking about...

Yea there's not really a problem with make, it's just a pretty ancient tool IMHO. Timestamp based rebuild is so 1980... I've used scons for other projects and just loved it, it doesn't just blindly rebuild everything because you touched some header file (like, switching to some branch and back, or undoing a failed patch apply etc.) It does get quite extensive with scripting though, and also can get slow on larger projects. From what i understand, that's why waf was forked from scons but eventually was completely redone with somewhat different concept. Anyway, i guess i will make a new thread eventually, maybe it gets some new fans, but i expect MaNGOS to go with Cmake because of the windows people... While waf does work on windows, i bet 9 out of 10 people will scream "i need to install python? and use a command line?"