wungasaurus

That's not a WMO, that's a blender file.

Those are actually two completely different errors. Let me tell you about https://wowdev.wiki/Client_Error_Analysis. It includes information on how to get your error analysed. I can't promise actually doing it, but to analyse, one would need the client binary (wow.exe) the full error log in text form Even with that information, it likely is highly time consuming and not even guaranteed to yield results. It likely is a better idea to build a new patch without any file not needed for character selection and then search for errors in the changes made to the ones left.

Please report an issue with a fuller log (this one has multiple minutes not shown, and should probably at least show the few last lines before the crash) at https://bitbucket.org/berndloerwald/noggit3/issues/new instead of this forum. Also, to investigate this at all, one needs your custom file.

I don't know. All my knowledge is on https://wowdev.wiki/M2/.skel.

animations are in .skel for those models.

how did you get that text file? what is "checking"? what makes you think it is "complete"? what is "complete"?

How did you check that?

The lastest M2+anim files are no longer "fixed" by removing "the 8 bytes", which was a horribly stupid idea to begin with. These things exist because they have a meaning. Removing stuff is pretty much never going to fix shit. This is unlikely to result in animations freezing until a sequence change is triggered. Yet, weapons being on wrong positions during sheath is not a M2 issue but rather one from extended systems from newer client versions where there are offsets in addition to M2 based attachments based on item type etc, making the M2 based root attachment (i.e. 0,0,0 of the item being in hand and middle of back) no longer static meaning 0,0,0 is in hand, but not in middle of back. This afaik is "new tech" and can't really be done in 3.3.5a. I might be wrong though.

Attachments on the weapon are for enchant particles, not for how they attach to the character!

If all weapons on the back of the character are off-set: change that characters attachment points. If it is only the case for this weapon on all characters, — if the weapon is also off-set when held in hand, move all vertices or the root bone in the opposite direction. or if it is only off-set for this character — if the weapon is not off-set when held in hand, you’re out of luck. Ish.

The client hides geosets based on the items equipped. Merging them is likely to break stuff?

You may be interested in https://wowdev.wiki/Client_Error_Analysis, and since your error is not in the list, paragraph 3.

> multiple of 4 You mean power of two.

Please, do not use that listfile. It is incomplete and is missing 212206 in contrast to upstream. Please just use upstream. You can find the currently most complete public listfile at https://raw.githubusercontent.com/wowdev/wow-listfile/master/listfile.txt.

All FSeek()s have to know about that right additional bytes. They should be relative to MD20 start, not file.

Well, the first step would be to look at the .skin file and check if the claim is in fact true. Then to take the unconverted .skin file and see if it is the case there as well. If it is, think about how to handle that, what it means. Can it just be removed? If it is not the case, try fixing the .skin file to not point to bad data? Is it actually pointing to bad data? Is it corrupted or is it fine with a single bad field? To fix it, you have to look at the stuff and try to understand the error. Nobody can tell you "in this case, write a 1 there", since the solution is 100% data dependent. But well, I realize it is simpler to just yell out of the window that someone should fix something, without providing further information or actually directing the request to the person you rely on.

No need to update any converter, just fix it: https://wowdev.wiki/Client_Error_Analysis#0x00837A90 Oh, and if you want Adspartan to fix anything, you should report a bug to him, including the exact file and version converted. Otherwise, nothing will change.

no idea, sorry. try installing msvcp140 yourself instead of having those api-ms-*and msvcp* dlls in the noggit directory. where do those even come from?!

Suggestions on how to actually get help: name the exact issue, e.g. by providing a error log tell exactly what you did, step by step show people what the files actually are, e.g. upload the files detail on why you think you did everything right tell what you have checked already

The WotLK parser is very picky about the files it reads. It assumes that if there is a #MWMO chunk as the first chunk, the #MWID and #MODF chunks follow in that order. Otherwise, or after the three, it *will* be #MAOF, and after that it only uses the offsets in that chunk to parse #MAREs. Your crash is at the point where it tries to read a MARE based on MAOF, but that points into bogus. This is likely due to there being other chunks than (MWMO, MWID, MODF), MAOF, MARE. https://wowdev.wiki/Client_Error_Analysis#0x007CC41F_read_0x.2A.2A.2A.2A.2A.2A.2A.2A_.28bogus_value.29

since you say noggit is not starting, there shouldn't be a log? try deleting it and starting again. does a log appear? otherwise, you _do_ only have the process monitor log, so that should probably be the one to send.

Please, read the linked article and provide the info asked there. That way you can actually fix the issue, and make it easier for others..

Please see the introductionary paragraphs of https://wowdev.wiki/Client_Error_Analysis

What other logs do you have?