wip Real-Time Body Physics (DirectX 11)

[ArtEdit]

16 hours ago, CPVC said:

 

Try this build here. It might do the trick. Do a manual install. This build is similar to the last one, but makes more of the system DLL functions available through RTBP. I think I'm on the right track with that, but I'm not sure about this implementation.

 

RTBP v2.3~3 2026-06-29.zip 1.57 MB · 1 download

 

 

This issue has been fixed in the attached build I believe. It will be part of the next release.

It works now with this version when running in windowed fullscreen. However, if I go to the options and switch it to exclusive fullscreen, the game freezes without showing any error, either in the log or as an error message.

If I try to continue my save while using windowed fullscreen, it starts loading, but then suddenly closes without displaying any error in the log or any error message.

[CPVC]

5 hours ago, ArtEdit said:

It works now with this version when running in windowed fullscreen. However, if I go to the options and switch it to exclusive fullscreen, the game freezes without showing any error, either in the log or as an error message.

If I try to continue my save while using windowed fullscreen, it starts loading, but then suddenly closes without displaying any error in the log or any error message.

 

Hmmm, ok. I'll have access to a machine I can do a ton of extensive testing on later this week, with Windows 11 and an NVidia GPU. I'll try to get this issue sorted then.

[HentaiGai]

Latest update seems to have broken this mod, I even tried the latest version released yesterday, Better Exceptions says:

"The listed script mod failed to load and should be removed or updated. Failure: 'C:\Users\triba\Documents\Electronic Arts\The Sims 4\Mods\RTBP Mod\Scripts\Client' (Client) unexpected EOF while parsing (Client.py, line 19) (SyntaxError: unexpected EOF while parsing (Client.py, line 19))"

 

Hopefully, this information is useful!

 

 

EDIT:

I rolled back to the previous version I was using (RTBP v2.3 2026-05-19) and this version seems to be working fine

Edited July 1 by RealSiViX

[0Spirit0]

Also, your dll (I tryed all since version 2.1) don't work on latest nvidia drivers. Updated the drivers yesterday and the game don't run anymore.

[CPVC]

12 hours ago, RealSiViX said:

Latest update seems to have broken this mod, I even tried the latest version released yesterday, Better Exceptions says:

"The listed script mod failed to load and should be removed or updated. Failure: 'C:\Users\triba\Documents\Electronic Arts\The Sims 4\Mods\RTBP Mod\Scripts\Client' (Client) unexpected EOF while parsing (Client.py, line 19) (SyntaxError: unexpected EOF while parsing (Client.py, line 19))"

 

Hopefully, this information is useful!

 

 

EDIT:

I rolled back to the previous version I was using (RTBP v2.3 2026-05-19) and this version seems to be working fine

 

Woops, that's not supposed to be in there. That's an old file from an older version of the script mod. I've removed it and reuploaded. Thanks for bringing this to my attention.

 

4 hours ago, 0Spirit0 said:

Also, your dll (I tryed all since version 2.1) don't work on latest nvidia drivers. Updated the drivers yesterday and the game don't run anymore.

 

Yes, that appears to be the case. I'll get this fixed as soon as I can. Like I said earlier, I'm going to be doing some testing on a NVidia machine soon, so hopefully I'll find a fix.

[ArtEdit]

I didn't realize you had another version on the forum's main page, "RTBP v2.3~3 2026-07-01.zip".

I tried it, and this time I was able to load my save successfully in borderless windowed mode. The only issue is that the game crashes if I switch to fullscreen mode.

[CPVC]

On 7/9/2026 at 10:07 AM, ArtEdit said:

I didn't realize you had another version on the forum's main page, "RTBP v2.3~3 2026-07-01.zip".

I tried it, and this time I was able to load my save successfully in borderless windowed mode. The only issue is that the game crashes if I switch to fullscreen mode.

 

Still trying to fix that. I think I know what I need to do, though.

 

As for the other problems users have been facing, I've run into a little snag. Every single computer I've tested RTBP on it works out of the box, 100%, no problems whatsoever. However, I was wrong about the GPU in one of the machines. It was an AMD, not a NVidia, so I still need to test RTBP out on a machine with an NVidia GPU.

 

I'm thinking that I might be barking up the wrong tree a bit here, though. I have a version of the Sims 4 that is independent of all game launchers, so there is no Steam or EA App/whatever they're calling it now integration. It's possible that an overlay could be causing problems. I can confirm that it works with the Steam overlay, but I'm not certain about the EA one.

 

If any of you have been having problems with RTBP, try disabling the overlay if you can and let me know if it works.

[Nnuos]

First of all, thank you for your work. I noticed in the game that the shoulders of men tend to look like the illustration, which seems to be a problem with the breast.

[CPVC]

On 7/12/2026 at 6:44 AM, Nnuos said:

First of all, thank you for your work. I noticed in the game that the shoulders of men tend to look like the illustration, which seems to be a problem with the breast.

 

That is a known issue. Sometimes the male shoulders are erroneously detected as breasts. I'll work on it.

[evexec]

Just wanted to say thank you for the mod!! Also, you seem to be one of the best modders I've seen in regards to communicating with your users and just being overall active, and it's been super cool looking back and seeing some of the more technical details of your process (like the switch to angular physics)!!!

[CPVC]

On 7/17/2026 at 9:03 PM, evexec said:

Just wanted to say thank you for the mod!! Also, you seem to be one of the best modders I've seen in regards to communicating with your users and just being overall active, and it's been super cool looking back and seeing some of the more technical details of your process (like the switch to angular physics)!!!

 

Thanks for the kind worlds! I do what I can.

[jeno3]

Hi,

I hope you're doing well.

How is the physics coming along. Specifically, getting it to behave the way it should? I have the latest version now, but it all looks pretty chaotic to me. With the "angular" setting, the breasts sometimes pull upwards or to the side. This happens mostly during animations. The linear physics also seems quite wild and jittery for me, and I just can't seem to get it under control using the settings.

[CPVC]

Another experimental release here. If you haven't been able to get RTBP to work, please check this version out. If it doesn't work, please submit an error report. The information in your log files would be helpful. Thanks.

RTBP v2.3~8 2026-07-28.zip

 

Edit: If this build still doesn't work for you there is another thing you can try. Add this to the bottom of your Settings.ini file or append the second line if you already have a debug flags section:

 

Edit: Updated with fixed DLL.

 

[rtbp.debugFlags]
forwardFuncsNotImpl = true

 

This will make it so when the game requests a forwarded function, RTBP will report that it is not implemented. Some other DLL wrapper projects do this and it might work.

Edited July 29 by CPVC
Fixed DLL.

[jeno3]

Log.txt

[CPVC]

Whoops, I accidentally left some debug stuff in the v2.3~8 build. Please redownload and try again. Sorry about that.

RTBP v2.3~8 2026-07-28.zip

 

Edited July 29 by CPVC

[0Spirit0]

Still can't load the game

log sent to server

[CPVC]

Thanks for testing out the experimental releases. I'm working on fixing the "OpenAdapter10_2" bug if you've seen references to that at the end of your log file. I haven't figured out why it doesn't work yet and I haven't been able to reproduce the bug on my own, so thanks for for checking out the every increasing amount of experimental builds. This bug effects a lot of users and fixing it is my number one priority right now.

[skawars1983]

I have been trying to get the dx11 version to work with reshade, they both work separately but crash on startup when i use both. Am i missing something?

Log.txt

[jeno3]

So, the latest DLL from July 28th is working for me. I’ve attached a video because the movements look strange. Could it be that they’re operating in exactly the wrong direction, or am I mistaken? I hope you can fix it soon. I’d really love to help, but I don’t know how—I have absolutely no idea about the work you’re doing here. Anyway, please don’t give up; the angular physics look promising. Am I actually the only one having these problems?

weird_Movements.rar

[CPVC]

3 hours ago, jeno3 said:

So, the latest DLL from July 28th is working for me. I’ve attached a video because the movements look strange. Could it be that they’re operating in exactly the wrong direction, or am I mistaken? I hope you can fix it soon. I’d really love to help, but I don’t know how—I have absolutely no idea about the work you’re doing here. Anyway, please don’t give up; the angular physics look promising. Am I actually the only one having these problems?

weird_Movements.rar 98.43 MB · 1 download

 

Yeah, that doesn't look right. I must have tweaked something and forgot about it. I'll see what I can do.

 

Edit: Try this.

 

RTBP v2.3~8 2026-07-30.zip

Edited July 31 by CPVC
Fixed DLL.

[joezreezal]

This mod will be a must have for me as soon as OP adds belly jiggle 

[jeno3]

Hi, that looks a bit better, but with the angular physics, the breasts still disappear back into the body as soon as the upper body bends 90 degrees forward.

 

I’ve looked into the matter to see if I could help, but I can't reproduce the issue in Blender to give you coordinates you could incorporate into your calculations—assuming I’ve actually understood the problem correctly. But surely you need the X, Y, and Z values derived from a natural look for your calculation, right? Or am I mistaken? Another issue for me is that I’ve never moved bones in Blender before, so I’d have to familiarize myself with that first. When I angle the upper body 90 degrees using `b_spin0` in Blender, the breasts move along with it; so, I’d have to roughly replicate the calculation error that occurs in-game, then pull the breasts downward to determine the correct value. But do I understand correctly that you have to perform two calculations that cancel each other out, depending on the coordinate space they originate from? What you're doing is pretty complex—hats off to you.

 

So, for negative results, you’d need to determine—let's call it an "offset"—is that right?

[CPVC]

14 hours ago, jeno3 said:

Hi, that looks a bit better, but with the angular physics, the breasts still disappear back into the body as soon as the upper body bends 90 degrees forward.

 

I’ve looked into the matter to see if I could help, but I can't reproduce the issue in Blender to give you coordinates you could incorporate into your calculations—assuming I’ve actually understood the problem correctly. But surely you need the X, Y, and Z values derived from a natural look for your calculation, right? Or am I mistaken? Another issue for me is that I’ve never moved bones in Blender before, so I’d have to familiarize myself with that first. When I angle the upper body 90 degrees using `b_spin0` in Blender, the breasts move along with it; so, I’d have to roughly replicate the calculation error that occurs in-game, then pull the breasts downward to determine the correct value. But do I understand correctly that you have to perform two calculations that cancel each other out, depending on the coordinate space they originate from? What you're doing is pretty complex—hats off to you.

 

So, for negative results, you’d need to determine—let's call it an "offset"—is that right?

 

Sounds like you understand it well enough. Yes, an offset to the bind position could change the position the vertices rotate around, improving the appearance of unconventional meshes such as yours. RTBP rotates around the exact positions of the breast bones, but if an offset to that could be encoded into a mesh, that could be used in the calculations if present. I'll have to do some testing and see how meta data like that could be embedded into a mesh without changing its appearance. RTBP only sees the mesh data after it's uploaded to the GPU, so the meta data would need to be encoded in that.

[jeno3]

Hi cpvc,

I’ve been working on a test mesh in Blender 2.79 to help with the RTBP clipping issue when the Sim bends forward at a 90-degree angle.

To help you test the vertex-based metadata/offset approach, I have prepared the mesh as follows:

• Vertex Paint: I masked the breast area in pure Red (R=1.0, G=0.0, B=0.0), so your shader code knows exactly which vertices should be affected by the offset calculation.
• The Test Deform: I manually rotated the upper body in Edit Mode to simulate the 90-degree bend and repositioned the breasts to a hanging state where they no longer clip into the torso.

To give you a rough mathematical baseline for your shader experiments, I tracked the coordinates of the very tip of the left nipple (before and after the bend):

• Original Position (Upright): X: -0.091467, Y: -0.145079, Z: 1.33675
• Adjusted Position (90° Bent): X: -0.091467, Y: -0.327787, Z: 0.944211
• Calculated Delta Vector: X: 0.000000, Y: -0.182708, Z: -0.392539

Please note that this is just a first rough attempt to see if the logic works under the hood. I wanted to provide you with these numbers so you can test if your shader can successfully read the red vertex colors and apply a progressive offset based on these delta values as the spine rotates.

Let me know if these data points help you with your initial tests, or if you need the actual .blend file exported in a specific way!

 

Attached, you'll find the original mesh and the test for determining the position. No idea if that helps you.

Position_Test.rar

[jeno3]

Hi CPVC,

 

Following up on my previous message

 

The Physics Conflict:
Because my custom mesh slopes backward and attaches higher at the bottom, the actual Center of Mass is shifted significantly closer to the chest wall compared to the default Maxis geometry.

Since the current angular physics assumes the default Maxis mass distribution, it calculates an incorrect, overstated angular inertia. This is why the breasts clip into the body during overbending—the torque calculation is overcompensating because the physics engine "thinks" the lever arm is longer and heavier at the tip than it actually is on this custom shape.

Feature Request: Inertia/Tensor Slider for CC Support
To accommodate these realistic shapes, the Inertia Tensor/Moment of Inertia range (especially for the X-axis/Pitch) needs to be scalable downward by about 30% to 45%.

Would it be possible to implement a dedicated slider in the In-Game GUI for the Inertia/Tensor settings? Having an interactive slider would allow users and creators to easily fine-tune the physics properties for different CC breast meshes directly in the game. Lowering the inertia value via this slider would perfectly compensate for the shorter effective lever arm of adjusted mass distributions and eliminate the clipping issues entirely.

Let me know if these numbers help you with tweaking the boundaries and if a GUI slider for CC compatibility is something you could add!

 

Perhaps a good solution, and worth testing?