Skip to main content

Editing an M2 for Epsilon (M2i Method)

Guide by NORTE.m2 · Version 2.0


Required programs​

  • M2MOD (pre-configured version — includes an up-to-date listfile.csv) Download M2MOD

You'll need to choose one of the following Blender versions to work with:

Version 3.4​

This is a modified version built to maximize compatibility with WMOs. I recommend using this one.


Version 2.9​

This is the original version of the addon. However, I recommend using 3.4 for better WMO compatibility.


Other useful links (not required)

1. Setting up the programs​

  • Download Blender — just one of the two versions listed above.
  • Download M2MOD — the pre-configured version is recommended since it already includes an updated listfile.csv in the mappings folder.
  • Install the Blender Addon inside Blender.

Blender Preferences panel

Install button for adding the addon

Click [Install] and select the addon's .zip file.

Addon activated in the list

Enable the one called "Import-Export: WoW Tools".


2 — Preparing the files​

This method works by replacement. You'll need to choose a base armor piece to modify or replace.

In this example, we're going to work with these shoulder pads:

Example shoulder pad in-game 3D model of the shoulder pad

Go to a WoW file browser — in this case wago.tools. Search for your item — here it's shoulder_plate_dungeonpaladin_a_01:

Item search on wago.tools

Download:

  • The .m2 file (one per item; here two appear because "l" is the left shoulder and "r" is the right — we're only editing the left one).
  • All .skin files (only one in this case; if there were more, download them all).
  • Any .blp textures you want to modify.
Note

If you were modifying an NPC or Player model, you'd also need to download the .skel files.


2.2 — Converting to M2I​

Once the files are downloaded, put them all in a folder:

Folder with the downloaded files

Open M2MOD.exe. Under Source M2, select the item's .m2 file. The Target M2I path will be filled in automatically — that's where the intermediate M2I file for Blender will be created. (You can change this path if you prefer.)

Click Go! to generate it.

Optional bonus

If you want to see the texture in Blender, you'll need to convert the .blp to .png first: https://www.wowinterface.com/downloads/landing.php?s=734452651e00d9554b435e4acbc95c05&fileid=22128

Just drag the .blp onto the .exe and it'll create a .png in the same folder.


2.3 — Importing into Blender​

Import the .m2i file that was generated into Blender:

File > Import menu in Blender

Imported .m2i file in Blender

Optional bonus — Adding a texture

This only affects how you see the model in Blender — it won't change how it looks in-game.

Select the mesh and create a new material:

New material created in the materials panel

In the Shading tab, select the material:

Shading view with material selected

Add an Image Texture node using the .png you converted from the .blp:

Image Texture node in the shading editor

Connect the Color output to Base Color by dragging between the two dots. To make it look like it does in WoW: set Specular to 0 and Roughness to 1.

Connected nodes with Specular 0 and Roughness 1

Remember to enable material preview in Blender:

Material preview button in Blender viewport


3 — Editing in Blender​

In this example, we're replacing the shoulder pad with a custom model. You could also just edit the original model by deleting parts of it — the steps are the same, minus the part where you import a new model.

We're replacing the original shoulder pad with this one:

New custom shoulder pad in Blender

Export the new shoulder pad as FBX from its own Blender project:

Export as FBX from Blender

Go back to your shoulder pad project and import the new shoulder pad's FBX:

FBX of the new shoulder pad imported

Adjust its position so it lines up as closely as possible with the original:

New shoulder pad positioned next to the original

Make sure the new shoulder pad has a UV Map named Texture. If it doesn't, rename it:

UV Maps panel with the map renamed to Texture

Scale the original shoulder pad down until it's very small, then hide it inside the new model:

Original shoulder pad shrunk and hidden inside the new one

Once that's done, press CTRL + A and select "All Transforms". Do this for both the new and original shoulder pads to apply any pending scale, position, and rotation:

Apply menu with All Transforms selected


3.2 — Exporting the M2I from Blender​

Tip

Before continuing, save a backup of your Blender project at this point. You'll almost always need to make adjustments later and it's very helpful to be able to return here.

Select in this order: the new shoulder pad, then the old one, and press CTRL + J:

Merging both meshes with Ctrl+J

They'll now be merged into a single object, which should keep the name of the original WoW file — in this case Mesh0000:

Outliner showing the merged mesh named Mesh0000

What if my file has more than one Mesh?

Most modern models have multiple meshes: 0000, 0001, 0002, etc.

The extra ones can be deleted without issue, as long as at least one original remains. It doesn't matter which one — 0001, 0002, 0004 — as long as one original is still there.

It does not work to delete the original mesh, rename your new shoulder pad, and place it in the same spot. That will cause errors. One original mesh must always be kept — hidden inside the model.

Export your model as M2I again. You can overwrite the original M2I or give it a new name — it doesn't matter:

Export as M2I from Blender

Exported M2I file in the folder


4 — Converting to M2 for Epsilon​

Go back to M2MOD.exe and open the second tab.

  • Source M2 will already have the original .m2 selected by default.
  • Under Source M2i, select your newly exported .m2i file.

Click Preload and then Go!:

Second M2MOD tab with the fields filled in

You'll see this error — don't worry, it always appears. Click OK:

The usual M2MOD error when converting

Your files are now ready. An Export folder will have been created inside the folder where your .m2 was:

Export folder generated by M2MOD

Use the files inside to create your Epsilon patch.

Reminder

Include the .blp texture of the new shoulder pad in the patch, renamed to match the original shoulder pad's texture filename.

Patch structure with the .blp included


4.2 — Fixing issues after converting​

It's common to load your item in-game and find something looks off. Here's the workflow to fix it:

  1. Open the backup you saved before merging the two meshes. (Or in Blender, press Ctrl+Z until you're back to the step before merging.)
  2. Adjust the position again.
  3. Ctrl+A → All Transforms.
  4. Merge the two meshes.
  5. Export the M2I.
  6. In M2MOD: Preload → Go!
  7. Take the files from the Export folder and manually place them inside the Epsilon patch folder, confirming any replacements.

Disconnect your character in Epsilon and reconnect — the model will update to the new version.

TIP

I do this whole process without closing any of the 3 programs at any point, making small tweaks and reconnecting my character in Epsilon to preview the model until I get the result I want.


Bonus — Collections​

The process for a collection is the same, with one addition: it has multiple pieces, each one bound to a different bone.

The files​

A collection's files involve parts that cover the entire body. Example after converting:

Collection files after converting with M2MOD

How it looks in Blender​

Unlike a helmet or shoulder pad, you'll see many more meshes and bones (attaches). This is normal:

General view of a collection in Blender

Outliner detail showing all meshes and bones

How to edit it​

Follow the same steps as in the guide above. You can attach your new model to any of the existing meshes and delete the rest.

(Everything can go in one single piece, or different parts can go in different meshes. If a single mesh has too many faces it will cause errors, so you may need to split the model across multiple meshes.)

The key difference is that you'll need to specify which bone your new model is attached to.

The bones​

Click on any bone and press TAB to see the skeleton and its connections. The bone's name is displayed in the top-left corner:

Skeleton view with TAB enabled

Bone name shown in the top-left corner

Attaching the model to a bone​

As an example, we're attaching a 3D chest piece. We've identified that the chest bone is number 10:

Bone number 10 selected in Blender

Vertex Groups panel with the new bone added

Go to the Vertex Groups panel and add a new group named after the bone you want to attach to. You can have multiple bones — the order doesn't matter.

Once you've added it, switch the viewport to Weight Paint mode:

Weight Paint mode active

Paint the weights onto the bone:

Model painted in Weight Paint

  • Red → the area is strongly influenced by that bone.
  • Blue → weakly influenced.

Since this is a piece that only attaches to a single bone, paint the entire thing red. You can adjust the brush settings in the top panel:

Top panel with Weight Paint brush parameters


Some models attach to more than one bone. For example, a skirt would attach to each leg on its respective side, while the top part attaches to the hip bone.

For example, on Zekhan you can see his skirt has the following:

Vertex Groups of Zekhan's skirt with multiple bones

Weight Paint demonstration with multiple bones

Once the weights for each bone are set, the model gets merged with the corresponding MESH and exported normally.