itch.io

Hey Dan,

That’s actually a really interesting question! In theory, a common SDF format would be awesome, but in practice it’s quite difficult. A lot of fundamental aspects would need to be standardized—like what blend/shape functions to use, how shapes are parameterized, or how hierarchies are handled.

For example, in Womp you can assign materials per shape, while in MagicaCSG and SDF Modeler it’s usually done per sub-model or layer. These differences make it tricky to create a “universal” format that would work well across different tools without losing information or behavior.

That said, it’s definitely something to think about for the future if more SDF tools emerge and interoperability becomes more relevant.

Cheers,
Sascha

Hi Dan, very cool — thanks a lot for your continuous support!

Cheers,
Sascha

Thanks Dan, I’ve actually already thought about adding an option to pick the mirroring axis – it’ll probably make its way in at some point. This also would add more flexibility to mirror on a single layer. Best regards

Hi, I’ll try to get my hands on a Mac in the near future and look into what kind of effort it is to port it. You're definitely not the first one to ask about this :D

Yes, that is expected—the OIDN interface changed after version 1.4.3 and the library even became more bloated with all the GPU support. I haven’t been able to reproduce the issue so far, but I might try reuploading the release to see if that helps. Thanks for trying it out!

If you have some time to spare, you could try replacing the library with the one from their GitHub page and let me know if the issue still persists.

Interesting! I’m not creating the links myself; they just come as part of Intel’s Open Image Denoise library. I also have a laptop with a recent Debian version for testing and haven’t run into this issue so far. But I’m still learning how to properly deploy software on Linux, so have mercy on me! :P I’ll try to look into it. Thank you for letting me know.

Best regards,
Sascha

Thanks! Glad you like it :) Open-sourcing is the most likely scenario at some point in the future.

Hi Dan,

Thank you so much for your kind words! I really appreciate your support. I’ll try my best to make it as cool and fun to use as possible, there is still plenty to do, but messages like yours are super motivating.

Wishing you all the best as well!

Cheers, Sascha

Hi,

Thank you for your feedback! I’m glad you’re enjoying the software. The ability to insert reference images into the viewport is indeed on my to-do list for a while. Currently, I’m focusing mostly on the big core features, but I plan to implement this functionality down the line, I promise. :)

Best regards,

Sascha

Hi Dan,

Thank you for the kind words! I’m really glad you’re enjoying the new features. The next update will include radial repetition as well as linear repetition in all three dimensions.

For now, my main focus is on the bigger core features, but I’ll continue to add smaller quality-of-life improvements along the way as well.

Best regards

Thanks for your comment! You're correct that it's currently not possible to bevel all edges directly, for example, on a cube. However, you might achieve a similar effect by using a negative shape with a chamfer blend. Let me know if that works for your use case!

Thanks for the kind words! It’s still pretty early, but I’m working on making everything more solid as I progress toward 1.0. You can already modify a cylinder by setting the cube’s bevel factor to 100% and adjusting the cone percentage. Material attributes like metalness, roughness and IOR can be controlled on the layer level, and you’ll see the results in the render preview.

Version 0.4 is back! Please note that using old 0.4 files is not recommended, as they might not work. Have fun!

Hi Dan, there were some issues with group transformations. I’ll be re-uploading a new version tomorrow, but my changes are partially incompatible with the earlier version. That’s mainly why I took it down—children of grouped strokes will likely end up misplaced. Cheers, Sascha

Sorry, I had to revert to v0.3.3 for now. I’ll re-upload v0.4 once a few issues are resolved.

Hi Dan,

Thanks for reaching out and sharing your thoughts. It’s great to hear that SDF Modeler reminded you of Organica.

The extrapolate feature you described sounds like it could be a really nice addition for creating more complex shapes. Right now, I’m focused on finalizing version 0.4, which will include a good set of new features and improvements, but I’ve also been considering adding something in the future that could create shapes along a curve, which sounds similar to what you’re describing.

Thanks again for the suggestion and for your interest in the tool.

Best regards
Sascha

Hi, thanks for the kind words and for your feedback!

For version 0.4, I’m planning to add Boolean subgroups as the main new feature, which will allow for more flexible object operations.

In the next smaller update, I’ll be introducing smooth blending for mirrored strokes. I’m also working on making the current Bézier curve more robust and giving users more control over the curve’s points.

Arrays are definitely on the way as well, as I’ve received a lot of requests for them.

A more powerful curve tool is something I plan to tackle in the future, once the core feature set is more developed.

Best regards!

I'm absolutely clueless when it comes to stuff like this, so you'll have to let me know if this works or not.

Here's the link -> https://discord.com/invite/42wGegU2

Alright, I’ll set something up and let you know once it’s ready. By the way, kudos for making the most of your available display resolution! :D

Okay, great to hear. The issue with the snapshot function seems like a GPU resource synchronization problem to me. It would probably be a good idea to have a channel where I can send you test builds, so you can let me know if the issue persists.

PS: My primary design goal for this software is to make it easy to use and fun, so no worries!

When working with the software, there are two things I really miss right now. First, I’d like to be able to rearrange the mirroring plane to move subparts away from the origin—this is something layer/subgroup transformations would allow. Second, especially with the newly added intersection operation, I really miss being able to do a cutout of the intersected shape with another additive shape underneath. Hope that makes sense. Cheers!

Hi, glad you like it! I plan to focus on adding more basic features first. Next, I’ll be reworking the object structure, adding boolean subgroups, and enabling group and layer transformations. This will open up a lot more possibilities. Best regards.

Hi, I tried to reproduce the issue on all my available GPUs, including two different AMD ones, but I wasn’t able to. This one will be tough to track down, so I’ll need more detailed information about when it happens and when it doesn’t. We should probably move communication to Discord or Twitter for easier troubleshooting. Also, you might want to check if your graphics drivers are up to date, just in case. Best regards

And this all started with the most recent version? In the latest update, I switched from fragment to compute shaders for rendering. I’m currently on vacation, but I’ll try to reproduce the issue once I’m back. I might need your help to test some possible solutions. I’ve also been thinking about opening a Discord server for easier communication.

Regarding your suggestion for better stroke/layer management, I plan to rework the internals to handle things like subgroups and combine the UI into a single tree view. I also want to add several quality-of-life features, so you might want to wait for what I have planned—though it will take some time. :)

Hi, thanks for the feedback. Does this issue occur only when creating PNG files, or does it also happen while using the software?

Hi, CTRL + LMB for selection still seems to work in v0.3. The issue might be that selecting multiple strokes from different layers isn’t possible yet. This will likely change when I combine the layer and stroke list views into a single tree view. Best regards!

In general, you can already use CTRL + C / CTRL + V to copy and paste strokes, even between layers. With CTRL + A, you can select all strokes and copy them to a new layer. Just in case you weren’t aware of that functionality, though a more convenient method would probably make sense too. :)

Update: I found a solution to display local/global translation along with rotation and scale in the universal gizmo.

Hi, thanks for the feedback! I plan to add more blending modes, including a chamfer blend, in the future. I’ll also add tool-tips to make the parameters easier to understand. The higher the bevel percentage, the more the shape transforms into a cylinder; the higher the round percentage, the more it becomes a sphere or capsule. Hopefully, that helps clarify things. Best regards!

Thank you :)

Hi, the global parameter only works with the 'Translate' gizmo, since you can't scale or rotate a stroke in global space. I should probably disable the UI for those options to avoid confusion. I'll also add a ring for uniform scaling and position the screen rotation ring more conveniently in my next smaller update. Best regards!

Hi again, you might want to check out the 0.3 release. The issue "should" be fixed. Best regards!

Hi, thanks for your feedback. I'll look into the issue.

There’s still quite a bit left on my to-do list, but I’ve added your suggestions as well. The lack of reference image support in most of these applications is probably due to how the SDF data is rendered—there aren’t any triangulated surfaces you can easily apply an image to. :)

Thank you for the boost in visibility! I really appreciate your support. I'd love to work on macOS/iOS versions in the future, but unfortunately, I currently don't have the necessary hardware to work on those ports :)

Ah cool. Thanks for your little introduction video. I really appreciate it :)

Thank you for reaching out and for your interest in the software! I designed SDFModeler with a minimum spec target of 2GB of VRAM to ensure it can run on a wide range of machines. Currently, there's no way to scale the internal resources significantly for higher-end systems.

If your question is about the memory usage statistic displayed at the bottom left of the screen, this number shows how much memory the SDF data uses within the 3D texture atlas. The purpose of this feature is to help users understand how much memory their creations are consuming. Additionally, the application dynamically adjusts the resolution of layers if the texture atlas nears full capacity. If you create many layers at maximum resolution, they may be down-scaled, and the memory usage statistic will explain when and why this happens.

I hope this clarifies things! If you have any further questions, feel free to ask.

Hi, sure go ahead. Glad you like it :)