mcbigmac

You'd have to make checks on NpcInfo send from server - and then if npc_id (or somehow else variate it's a NPC attached to a castle) and get the crest_id (crests are client stored - aren't they?) and send it. Which would require research\coding into l2server.exe. Which means you can't. Pay the creator of your Pack (Meaning, Vangath or AdvExt) to do it - or get someone like me to do it. Ether way - it's not just "simply" doing it.

FYI new blowfish key is readily availeble: "556C2BCC6C6C6C6C5BCB606BB190CE82" So adjust that to your bot client of choice - althought don't see how that should stop l2tower?

eh scratch that im re-tarded. Looks to be more simple thru Blowfish\Authport changing and other stuff - and simple virtual protecting of most important hook places. ...gotta remember not to judge until IDA gets opened.

Just uses Hint Antibot - can't quite remember what it does besides exchanging the blowfish keys and looking for specific programs and crcs. That and it bans on sight - which in my oppinion is a poor strategy for scaring people way rather than truly protecting yourself.

RageZone is C1 extended. PP irc is the pure c4 leak release.

Primary Key solves those issues - with the account_id and name. That's why starting to remove from the top - and then remove user_data , and then on to skill, item etc etc.

If your using Interlude - you'll be using work from the vangath\advext camp. ...both if i remember offer a scheme buffer based on each char - and a fixed set for you can change via scripts. If not - you could write one via AI - for a fixed set ala warrior, dagger, mage etc.

Because associating the unique remove factor - by account id is easier than weird inner joins with login dates from user_auth. Same with then having to remove data from user_item, and other bigger tables. start from top, and remove down heirachy wise :P

Has nothing to do with a 6 digit offer. Has something to do with he's a no one - and they do not want to risk renting to someone that doesn't atleast have a chance of success.

I didn't see any shortcut @ the end of createchar? (Was i blind?). Didn't you want to remove the current ones?

HOKAI SO. Upon CharCreate Success packet return from cacheD - 2 things happens User::AddNewCharID is called along with CDB::RequestCreateInitItems in L2Server. We're interested in the RequestCreateInit. It basicly sends usersocket, class id and char id to cacheD. Fast Forward to cacheD - and specificly to CreateInitItems Packet - thats the handler for aforementioned l2server packet. In this packet\function there's this little gem: 000000000051D3B9 call CUser__RegisterInitialShortcut This function several times uses the DB::Execute(In a wrapped form) function with a string found here: off_6A5760 Which happens to be : "{Call dbo.lin_SetShortCut (?, ?, ?, ?, ?, ?)}". So - you could hook the register initial shortcut and do your ownstuff db class and just set off the shortcuts yourself (Remember to learn IDs and types and stuff). OR Change the param values on stack for the executes inside the function. FYI This is for GF - but i assume it's 99,99999999999% same for C4.

It might be set by a non existing replyback from db on a char during CShortCut::Init.... But i don't recall seeing any sends that would indicate this. I'll take a looksy tonight - but it might be hardcoded into client somewhere. We talking GF or C4 i should investigate?

.... so it's a good idea to post something completely irrelevant just cause its a link? Because it's not like L2J official and other l2j forks have c4 documentation for packets or sniffers like phx have any packet documentation which would be needed to make anything work. Another theory could be you wanted to post something to seem smart - but that's just a theory. You gotz all the cracking and extender stuff to do so anyway - so i must be crazy.

Nice. ...oh wait - you posted packet structures and information in a thread regarding Raw AI stack information. You are one special piece of... special.

Id like to try fix it , but i'd need access to figure it out. I'm bad at geusstimating stuff to try without a full idea of what your doing wrong. Is it GF or C4?

GF will be impossible, even if removing all NPCS. The objects\npcs memory footprint is too large. You cannot use l2off without geodata for one - so removing is not an optin. And by testserver on C4 BASE - i mean loading server and then connecting perhaps locally with 0 npcs. It will lag as a bitch tho - since it will not be able to reserve ram (aka not use it, but just reserve) it and try to churn out routines faster. If you can operate under laggy conditions when switching windows and looking at stuff - while client is open then yes you can run C4. GF, not a fat chance. 8GB will also not help - you won't even finish loading l2server before itll most likely run out of memory footprints and shutdown itself.

I can't really see much wrong - but i haven't used l2comm much. Ill set it up locally and try myself. Possibly your SQL Scripts are bit screwy. Can you find the posts in the DB - with everything correctly set as you do from client?

google errors for more information on the error. Has nothing to do with MSSQL - more PHP of your scripts. So post errors, and your code instead of being vaue - takes longer to figure out where your problem is. If your using php 5.3 or above - see a thread in the help section regarding this issue of how PHP removed some of its MSSQL functionality.

Any changes can only be made to a character that is offline. So often - you select directly in DB by char_name, get char_id and use it kickuser and then use one of the modify packets. In one of your threads, i pastebinned the formats for all availeble packets in c4\gf. (There's more in GF, but they're mostly the same).

4 GB on a consumer OS - with tons of bloated sysinternals will be hard to run C4. Testserver sure - but bloated. GF will be impossible.

You get skype, now fake mexibarto.

So you understand - how can create what you want right?

Oh forgot about getdatetime completely. Then use it, and go make it - what's your issue? You can save the day(hour minute, if you want too) into i_ai vars and have them checked on npc_deleted + on_start. (And be sure to completely check what 1000\1001 does and how to handle it proper based on the event_args from the child classes)) NPCMAKER contain all loose vars like i_ai\i_quest like normal npc and gg objects. Each RR it will reset naturally - but else you can have a perfect 2 day spawn more or less. NPCPOS would just have respawn time as whatever and a instant spawn of the npc. Altho it will not solve your geodata problem - which you should go fix with HDGE instead of trying this imho ;)

I do not think you can fixed weekday spawn with current limitations of the spawn systems. You can get time of day hour wise - but not week by functions. (Atleast of what i can recall @ work - check all functions\vars yourself). You'd definately have to change the entire inheritance down to the base maker classes and create your own for Antharas\Valakas. I'll see if you can do a AI based way when i get home - but i can't remember any function\var way this would be possible. (Everything would be reset at each restart - so omsetimes you'd see longer than 2 day spawn even if you save variables on maker objects). You could do daily respawn by the gettime functions - and do something semi alike it. Let me think about it a bit :P What i don't understand is - why are you trying to fix a geo bug with a fixed spawn time and how will that stop the geo bug at all?

...you got the answer. If (php_ver > 5.3) go with microsoft sql driver for PHP, if not, make sure the extension for the php_mssql.dll is inside php.ini uncommented. Go google instead of ask for links - what the hell? plenty of easy packages like wamp and xampp and easyphp.