ShootersForever.com Forum Index

GoldenEye 007 Nintendo 64 Community, GoldenEye X, Nintendo 64 Games Discussion
GoldenEye Cheats, GoldenEye X Codes, Tips, Help, Nintendo 64 Gaming Community


GoldenEye Setup Editor Feature Requests List/Bugs
Goto page Previous  1, 2, 3 ... 95, 96, 97 ... 105, 106, 107  Next
 
Post new topic   Reply to topic    ShootersForever.com Forum Index -> Q-Lab Hacking Department
View previous topic :: View next topic  
therektafire
Agent
Agent


Joined: 21 Aug 2018
Posts: 8
Location: none of your business ;)

 PostPosted: Mon Jun 06, 2022 10:33 am    Post subject: Feature request: "better" texture-to-face assignme Reply with quote Back to top

So I'm using Blender 3.0 to make a map and I've been using it's vertex color baking feature to bake lighting for my map, it's been working pretty well so far, but, to see the resulting vertex lighting I have to use a custom material that multiplies the vertex color by the texture color since being able to see both at once was seemingly removed in 2.8 Rolling Eyes This is a bit of a problem because there's a mixRGB node hooked up to the Base Color slot instead of an image, so the visual editor can't find the texture I want to use so my map is just flat shaded when imported into the editor, and in game too... I'd like to be able to see how my lighting looks in blender so I can see if I need to change the color or strength of some light or tweak the topology of some area so it lights better but I don't want to keep having to fiddle with materials hooking and unhooking stuff up to keep the editor happy :/ So I was hoping that maybe a feature could be added where you can change how textures are looked for in the mesh so instead of the material properties it looks at the name which would resolve the issues I'm having currently, it could be a toggle in the settings or something
 
View user's profile Send private message Send e-mail
SubDrag
Administrator
Administrator


Joined: 16 Aug 2006
Posts: 6118

 PostPosted: Mon Jun 06, 2022 2:16 pm    Post subject: Reply with quote Back to top

Sorry what? It sounds like a blender problem, but you'd have to describe the specific technical details that you are requesting. Are you talking about ge obj?
 
View user's profile Send private message
therektafire
Agent
Agent


Joined: 21 Aug 2018
Posts: 8
Location: none of your business ;)

 PostPosted: Mon Jun 06, 2022 3:17 pm    Post subject: Reply with quote Back to top

SubDrag wrote:
Sorry what? It sounds like a blender problem, but you'd have to describe the specific technical details that you are requesting. Are you talking about ge obj?


Pretty much yes. So basically, in blender you used to be able to view the color of the texture multiplied by the vertex color as a specific view mode, so you could see how your model will look like in game with the vertex colors applied. But in 2.8+ the only way to have that same view is by modifying the basic principled bsdf material to do the vertex color * texture color manually. The issue is that by doing that the texture doesn't really get set correctly on export, presumably because normally blender will only actually assign a texture to a face in the export if it's material has an image hooked directly up to the output in the bsdf. This means essentially that I either have to choose between being able to properly see how my map will look in game, and have none of the textures be assigned properly because the fbx importer can't detect them because blender didn't assign it in the material, or not be able to view the vertex colors in blender but be able to have the textures assigned in the editor correctly, but I can't have both :/ So I think that since the textures.txt file already exists looking for the texture in the material itself isn't that great of an idea for this reason and the importer should just use the material name to look up the texture instead, that way I can have my cake and eat it too Smile
 
View user's profile Send private message Send e-mail
SubDrag
Administrator
Administrator


Joined: 16 Aug 2006
Posts: 6118

 PostPosted: Tue Jun 07, 2022 2:03 am    Post subject: Reply with quote Back to top

Doubt it's this simple in practice. Names are often truncated during or mangled I conversion. You could export it once and convert to obj then reuse that material file instead later, but I doubt it'd be fully compatible with whatever blender doing for your shaded one. It's effectively the same.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Wed Jun 15, 2022 2:25 pm    Post subject: model import issues/mesh distortion Reply with quote Back to top

having a little bit of trouble with my model import

* Mesh distortion on import
* and shared vertices problem

I think theres a process of vertex sharing that is mucking with my model, and even changing weights of things.
i think the program moves all bones to 0,0,0, then applies those changes to the underlying mesh, causing massive distortion due to bone movement, then moves things into a predefined "H" pose, which makes things more or less okay.
This is messed up especially with Hips, that are unweighted mostly because thats how the others were done. TopJoint would be its natural weighting, but it only has weighting to 1 on the upper edge that meets the torso.

when animation is applied to this model, some strange things happen. The underlying mesh for the hips (only one that has this problem) is now altered, and it is unweighted by TopJoint, so it doesnt un-distort when bone stuff is applied to it.
This then gets plugged into the bone movements in the animation, and the other parts are animated around it, but the underlying mesh is all twisted.
I figured out that one reason i had alot of distortion of that part was because my character is 4 legged and rotated 90 degrees forward. If I rotate it 90 degrees backwards, import the model, apply the animatoin, its better, but still a little messed up.
The hips have some distortion along the top edge still. This is because its still doing the "take the model, move all bones to 0, move everything to a predefined H", which is a bit different from my bone structure. I think the hipbone is moved? but its not just that, the upper torso seems to be moved too.

I'm wondering if the key mistake i'm making is moving the bones positionally at all. The one tutorial I saw had a bone structure stretched onto an existing H posed model. Maybe I would not have much trouble if I did that as well.


https://drive.google.com/file/d/10JIqRg6A5mA1hHljtVaoGbG3jbqkRUuL/view?usp=sharing

2nd problem i have
One problem I'm having is here, where some parts of the upper left leg lose their weights:

https://drive.google.com/file/d/10Ixy31VPSlsLXwdZTaBYO4zjbVdvqqM5/view?usp=sharing

I figured out that it can be bad for 3 parts to share vertices, so I added slivers near the hips so that upper legs dont touch eachother or the torso, only hips.
The slice for the hips maybe just needs to be bigger, because i think vertex weights are clearly being shifted over to a different part.


Sorry if that thats long and confusing.

Tldr trouble importing with different bones, hips in particular, and shared vertex issue understanding needed pls :c
 
View user's profile Send private message
SubDrag
Administrator
Administrator


Joined: 16 Aug 2006
Posts: 6118

 PostPosted: Wed Jun 15, 2022 5:33 pm    Post subject: Reply with quote Back to top

I'm assuming some issues with modeling program - maybe others can help more if you describe what programs you're using and workflow. And they can let you know how to do it properly.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Wed Jun 15, 2022 6:34 pm    Post subject: Reply with quote Back to top

SubDrag wrote:
I'm assuming some issues with modeling program - maybe others can help more if you describe what programs you're using and workflow. And they can let you know how to do it properly.


Ah sorry this is all blender 3.1, and using better fbx exporter.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Wed Jun 15, 2022 6:35 pm    Post subject: Reply with quote Back to top

NightSkiesPony wrote:
SubDrag wrote:
I'm assuming some issues with modeling program - maybe others can help more if you describe what programs you're using and workflow. And they can let you know how to do it properly.


Ah sorry this is all blender 3.1, and using better fbx exporter.


Also its really that it gets mangled after it gets imported into PerfectEditor, and that makes the animations mangled too. I guess the thing I need to do is make it not get mangled by the import process somehow.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Thu Jun 16, 2022 1:31 pm    Post subject: Reply with quote Back to top

NightSkiesPony wrote:
NightSkiesPony wrote:
SubDrag wrote:
I'm assuming some issues with modeling program - maybe others can help more if you describe what programs you're using and workflow. And they can let you know how to do it properly.


Ah sorry this is all blender 3.1, and using better fbx exporter.


Also its really that it gets mangled after it gets imported into PerfectEditor, and that makes the animations mangled too. I guess the thing I need to do is make it not get mangled by the import process somehow.


I had some theories, i moved the bones in my imported model around to be very similar to the bones exported from existing characters.
Mainly I moved the hip bone #1
It made it a little better for model import, but torso and the rest is still moved downwardsa bit, makng it overlap the hips, but not distort it. So what I did was I distorted the hips in the opposite direction, and then imported. I exported from GE again and the hips looked like the original, but hips still moved down. This more or less worked.
Then I animated with the new base model (before import to GE), and applied it to an animation, then exported it again. The hips were distorted again in the same way as before all this.
You would think when you import an animation it would use bone data only, and not the model. This is really confusing.

Can you shed any light on what happens to the bones from imported models?

Im forming a theory that they need to be in the *exact* same positions as the character you're overwriting. Different bone rotations seem to be allowable, but maybe bone positions can't be. My character's joints are attached differently so it gives it shorter front "arms" (legs tbh).
 
View user's profile Send private message
SubDrag
Administrator
Administrator


Joined: 16 Aug 2006
Posts: 6118

 PostPosted: Mon Jun 20, 2022 3:15 pm    Post subject: Reply with quote Back to top

Didn't you get it working?
http://www.shootersforever.com/forums_message_boards/viewtopic.php?t=8145&highlight=blender

I can't really help Blender much, but probably is your skeleton. I think it recenters all joints to 0,0,0 on import. It imports the joints, and points are relative to joints, so it's not preserving it properly likely.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Wed Jun 22, 2022 1:31 am    Post subject: re: bone problem Reply with quote Back to top

I did make some progress.

Turns out the issue with the bones was topjoint not being at 0,0,0
Since the hips were mostly unweighted, they got distorted where they were weighted only, and stayed in the same place as original. The model meanwhile moved all around it relative to topjoint at 0,0,0. Also my model rotation was causing further distortion to the hips (4 legged creatures need to be sitting on butt for it to work right).
Solution was to ofc move the topjoint to 0,0,0, along with the hips. No more distortion.

Now i'm just trying to solve the problem of vertices apparently losing their weights. I notice 3 places this is happening. Between one upper leg and hips, and between upper legs and lower legs.
Upper leg - some points lose weighting to hips, and have no weight at all, but not every point.
Between both upper legs and lower legs, points seem to lose their weights, but get them back after some model distortion. Their position gets locked to something like the original model before import, and then they get their weights back. The bone is more rotated in the basic H model for some reason, so a seam opens up due to those vertices being moved somehow. I have a theory that the nearest vertices with a different weighting are just too close or something ... I'm really not sure about this one
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Tue Jun 28, 2022 12:31 am    Post subject: Re: re: bone problem Reply with quote Back to top

NightSkiesPony wrote:
I did make some progress.

Turns out the issue with the bones was topjoint not being at 0,0,0
Since the hips were mostly unweighted, they got distorted where they were weighted only, and stayed in the same place as original. The model meanwhile moved all around it relative to topjoint at 0,0,0. Also my model rotation was causing further distortion to the hips (4 legged creatures need to be sitting on butt for it to work right).
Solution was to ofc move the topjoint to 0,0,0, along with the hips. No more distortion.

Now i'm just trying to solve the problem of vertices apparently losing their weights. I notice 3 places this is happening. Between one upper leg and hips, and between upper legs and lower legs.
Upper leg - some points lose weighting to hips, and have no weight at all, but not every point.
Between both upper legs and lower legs, points seem to lose their weights, but get them back after some model distortion. Their position gets locked to something like the original model before import, and then they get their weights back. The bone is more rotated in the basic H model for some reason, so a seam opens up due to those vertices being moved somehow. I have a theory that the nearest vertices with a different weighting are just too close or something ... I'm really not sure about this one


I'm at a loss, I really need some support here

Vertices are absorbing weights from adjacent points from the same part maybe. Seemed like it was arbitrary points. I went over it with a 0-weight brush and now *all* of the points get nearby point's weights.

Before
https://lh3.googleusercontent.com/d/10KWxxy6U8RvAF6LQvMeBdLrmKGjwVveb

After
https://lh3.googleusercontent.com/d/10NHkFTcOZrrog4Ipk3un_jgIxV0HiIxe


I reimported the exported file to blender, just to ensure the better fbx export plugin isn't messing it up, and it's fine. Can you shed light on what is going on in here? I know weights move around if the points are close together part-wise, but I am not sure what this is all about.
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Sun Jul 03, 2022 3:22 am    Post subject: Re: re: bone problem Reply with quote Back to top

NightSkiesPony wrote:
NightSkiesPony wrote:
I did make some progress.

Turns out the issue with the bones was topjoint not being at 0,0,0
Since the hips were mostly unweighted, they got distorted where they were weighted only, and stayed in the same place as original. The model meanwhile moved all around it relative to topjoint at 0,0,0. Also my model rotation was causing further distortion to the hips (4 legged creatures need to be sitting on butt for it to work right).
Solution was to ofc move the topjoint to 0,0,0, along with the hips. No more distortion.

Now i'm just trying to solve the problem of vertices apparently losing their weights. I notice 3 places this is happening. Between one upper leg and hips, and between upper legs and lower legs.
Upper leg - some points lose weighting to hips, and have no weight at all, but not every point.
Between both upper legs and lower legs, points seem to lose their weights, but get them back after some model distortion. Their position gets locked to something like the original model before import, and then they get their weights back. The bone is more rotated in the basic H model for some reason, so a seam opens up due to those vertices being moved somehow. I have a theory that the nearest vertices with a different weighting are just too close or something ... I'm really not sure about this one


I'm at a loss, I really need some support here

Vertices are absorbing weights from adjacent points from the same part maybe. Seemed like it was arbitrary points. I went over it with a 0-weight brush and now *all* of the points get nearby point's weights.

Before
https://lh3.googleusercontent.com/d/10KWxxy6U8RvAF6LQvMeBdLrmKGjwVveb

After
https://lh3.googleusercontent.com/d/10NHkFTcOZrrog4Ipk3un_jgIxV0HiIxe


I reimported the exported file to blender, just to ensure the better fbx export plugin isn't messing it up, and it's fine. Can you shed light on what is going on in here? I know weights move around if the points are close together part-wise, but I am not sure what this is all about.



Was about to set this aside for good, but I still had a few ideas left, and I had time to waste.

Solved my own problem.

I noticed that the original models export and import just fine through blender.
I replaced the leg part that was problematic with one from that original model, and it was fine.


Then I noticed that when I redid the weights by hand after deleting the vertex group, all was well.
I made some random test weights to see if they would import the same, then I figured out that the vertices that suddenly got the wrong weights were ones that I had done by mistake and overwrote with 0 strength weights.

So I managed to fix this by deleting all the vertex groups, recreating them for each bone, and redoing all my weights on those parts, and never painting with a 0 strength brush (always 1/full strength), just using ctrl+z instead when I messed up. Now the weights are 100% perfect!


I am guessing what's happening here is ... program sees a non-"null" weight set on a vertex, it for some reason changes it to either a 0 or a 1. The program will warn you if its 0.123123 or something not 0 or 1, but it seems to convert it to a 1 no matter what it is? Just as long as that point has a weight attached for that bone number.

Adding this to my other post about model importing:
http://www.shootersforever.com/forums_message_boards/viewtopic.php?p=75649#75649
 
View user's profile Send private message
NightSkiesPony
Agent
Agent


Joined: 02 May 2022
Posts: 25

 PostPosted: Sun Jul 03, 2022 11:25 pm    Post subject: Re: re: bone problem Reply with quote Back to top

NightSkiesPony wrote:
NightSkiesPony wrote:
NightSkiesPony wrote:
I did make some progress.

Turns out the issue with the bones was topjoint not being at 0,0,0
Since the hips were mostly unweighted, they got distorted where they were weighted only, and stayed in the same place as original. The model meanwhile moved all around it relative to topjoint at 0,0,0. Also my model rotation was causing further distortion to the hips (4 legged creatures need to be sitting on butt for it to work right).
Solution was to ofc move the topjoint to 0,0,0, along with the hips. No more distortion.

Now i'm just trying to solve the problem of vertices apparently losing their weights. I notice 3 places this is happening. Between one upper leg and hips, and between upper legs and lower legs.
Upper leg - some points lose weighting to hips, and have no weight at all, but not every point.
Between both upper legs and lower legs, points seem to lose their weights, but get them back after some model distortion. Their position gets locked to something like the original model before import, and then they get their weights back. The bone is more rotated in the basic H model for some reason, so a seam opens up due to those vertices being moved somehow. I have a theory that the nearest vertices with a different weighting are just too close or something ... I'm really not sure about this one


I'm at a loss, I really need some support here

Vertices are absorbing weights from adjacent points from the same part maybe. Seemed like it was arbitrary points. I went over it with a 0-weight brush and now *all* of the points get nearby point's weights.

Before
https://lh3.googleusercontent.com/d/10KWxxy6U8RvAF6LQvMeBdLrmKGjwVveb

After
https://lh3.googleusercontent.com/d/10NHkFTcOZrrog4Ipk3un_jgIxV0HiIxe


I reimported the exported file to blender, just to ensure the better fbx export plugin isn't messing it up, and it's fine. Can you shed light on what is going on in here? I know weights move around if the points are close together part-wise, but I am not sure what this is all about.



Was about to set this aside for good, but I still had a few ideas left, and I had time to waste.

Solved my own problem.

I noticed that the original models export and import just fine through blender.
I replaced the leg part that was problematic with one from that original model, and it was fine.


Then I noticed that when I redid the weights by hand after deleting the vertex group, all was well.
I made some random test weights to see if they would import the same, then I figured out that the vertices that suddenly got the wrong weights were ones that I had done by mistake and overwrote with 0 strength weights.

So I managed to fix this by deleting all the vertex groups, recreating them for each bone, and redoing all my weights on those parts, and never painting with a 0 strength brush (always 1/full strength), just using ctrl+z instead when I messed up. Now the weights are 100% perfect!


I am guessing what's happening here is ... program sees a non-"null" weight set on a vertex, it for some reason changes it to either a 0 or a 1. The program will warn you if its 0.123123 or something not 0 or 1, but it seems to convert it to a 1 no matter what it is? Just as long as that point has a weight attached for that bone number.

Adding this to my other post about model importing:
http://www.shootersforever.com/forums_message_boards/viewtopic.php?p=75649#75649


Above ought to be fixed at some point. 0 weight should be 0 weight, not 1
There's a workaround for now though.
 
View user's profile Send private message
SubDrag
Administrator
Administrator


Joined: 16 Aug 2006
Posts: 6118

 PostPosted: Mon Jul 04, 2022 4:07 am    Post subject: Reply with quote Back to top

You'd have to post a file with weights that you think contain zero and non-zero. But there is no weighting in GoldenEye/PD, all weights should be 1 or the vertices shouldn't exist.
 
View user's profile Send private message
Display posts from previous:   
Post new topic   Reply to topic    ShootersForever.com Forum Index -> Q-Lab Hacking Department All times are GMT - 8 Hours
Goto page Previous  1, 2, 3 ... 95, 96, 97 ... 105, 106, 107  Next
Page 96 of 107

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

Cobalt 2.0 BB theme/template by Jakob Persson.
Copyright © 2002-2004 Jakob Persson


Powered by BB © 01, 02 BB Group

 


Please Visit My Other Sites: GoldenEyeForever.com | GrandTheftAutoForever.com

Got kids? Check out my Dora The Explorer site with games and coloring pages!

Our forums feature Nintendo 64 games, GoldenEye 007 N64 New Maps and Missions, GoldenEye Cheats, N64 Emulator, Gameshark, GoldenEye Multiplayer and more!

[ Privacy Policy ]