Jump to content
3DCoat Forums

Sakru

Member
  • Posts

    18
  • Joined

  • Last visited

Recent Profile Visitors

42,822 profile views

Sakru's Achievements

Neophyte

Neophyte (2/11)

  • Great Support Rare
  • Great Content Rare
  • Great Content Rare
  • Helpful Rare

Recent Badges

7

Reputation

  1. Sorry i was very busy last few days. I'm trying to understand what this report is about, but i really can find any clue.. i see you are trying to run /home/jfespinal/.local/share/flatpak/exports/share/mime/ but how is it related to 3D-Coat? You are trying to activate Linux version of 3D-Coat? If yes, then i would recommend to run it using Bottles, because as of right now Linux version is broken. We can continue in this thread - https://3dcoat.com/forum/index.php?/topic/30731-3dcoat-202406-crashes-on-recent-linux-distros/#comment-196389
  2. Here is what i found myself. The goal of this test was to analyze the stability and memory usage Steps taken: GDB: 1. Launched 3DCoat with GDB to capture the backtrace of any crashes. The program encountered a segmentation fault at xmlDictFree (dict=0x96e4540) in the librsvg module, specifically within the GTK icon loading system. <- i tried to open a project 2. GDB backtrace showed that the crash was triggered when handling SVG images, possibly during GTK theme loading. Valgrind Memory Analysis: Run 3DCoat under Valgrind’s memory checker (Memcheck) to detect memory errors. 1, Valgrind reported multiple warnings regarding unsupported compression types when reading debug symbols from system libraries like libgtk-3, libcairo, and others. This limits my ability to provide detailed memory analysis for these libraries. 2. Several errors related to conditional jumps based on uninitialized values were detected. One of these errors stemmed from the calculate_path function, which is involved in Python’s path calculation in the application. The program attempted to read invalid memory, causing segmentation faults. This was traced back to the calculate_path function and ultimately tied to the handling of Python configuration initialization. The tests were done hastily. I’m not sure if this will help anyone or provide the developers with new information (probably not), but I can send the logs if needed.
  3. If you're experiencing crashes, try testing the program using strace. strace ./3dcoat 2> log.txt
  4. When you run 3D-Coat using shell, you need to add the environment variables to your shell configuration. Like this: echo "export PYTHONHOME=/usr/bin/python3" >> ~/.bashrc echo "export PYTHONPATH=/usr/lib/python3" >> ~/.bashrc source ~/.bashrc This will fix HOME and PATH being not set. GTK errors might not always be related to GTK itself. They could stem from mismatched themes, system dialogs, or library conflicts in KDE. It’s recommended to test GTK apps in their native environments like GNOME or XFSE.
  5. If you have a second drive, you can create a separate bottle directory but keep in mind to mount the drive before use, or it will not work. If you use Ubuntu, you can do it automatically like this: sudo lsblk (or just lsblk) x----------------output-----------------x we will mount this drive -> sdc sdc 8:32 0 1.8T 0 disk <- disk └─sdc1 8:33 0 1.8T 0 part <- partition (if created) sdd 8:48 0 931.5G 0 disk └─sdd1 8:49 0 931.5G 0 part sde 8:64 0 1.8T 0 disk └─sde1 8:65 0 1.8T 0 part sudo blkid (will show IDs of drives) /dev/sdc1: LABEL="my_drive"512" UUID="264466014465D459" TYPE="ext4" PARTUUID="691e2ce5-01" | your drive ID | /dev/sdd1...... /dev/sde1... Create a holder for your drive -> sudo mkdir /home/user/3dcoat_on_bottles sudo nano /etc/fstab Add new line: UUID=264466014465D459 /home/user/3dcoat_on_bottles defaults 0 0 You can also remove bottles cache files. If you installed different dependencies and runners it can be huge in size - xx gigabytes. It is usually located here: /home/user/.var/app/com.usebottles.bottles/data/bottles/cache If you installed other runners you can remove them inside "runners" folder.
  6. Nope. It was more of a concept, so i did use some kit from the internet for the left arm. This one is made in 3DCoat (not done)
  7. Using the Wine version through Flatpak/Snap, you might encounter an issue where 3DCoat cannot see beyond the sandbox boundaries. To resolve this, you will need to specify the exact path to your working directory. Here’s how it’s done in Bottles: You go inside the bottle then -> Settings -> Manage Drives -> Create new drive -> Point it to your working directory. Then in 3DCoat, when you choose to open a project file, you go to System -> Drive -> "Your Drive" -> Working Directory
  8. Install Wine via Flatpak on Ubuntu, and it should resolve your problem. It's a straightforward process. You won’t notice any difference in performance when using Wine through Flatpak compared to other installation methods. I'm not sure if Wine is available as a Snap package, but if it is, you could try that as well. When you use Lutris or Bottles, they don't rely on the potentially outdated or broken Wine package from the Ubuntu repositories. This is similar to why Steam developed their own Proton runner rather than relying on Wine from distribution sources—it's done to ensure better compatibility and performance sudo apt update sudo apt install flatpak flatpak remote-add --if-not-exists flathub https://flathub.org/repo/flathub.flatpakrepo
  9. Strange.. It does work in my VM. Are you unable to install it or you have problem running it?
  10. CL (ClearLinux). (What do You want to say with the benchmark-pictures? That you can use runners, without any issues, and they will give you better performance since native 3D-Coat uses OpenGL.
  11. #Work in progress =) in total 6 Maelstrom gang models.
  12. Did some testing just to show the difference. Here is my testing rig: Top left: Crossover (their own Runner (Wine with DXVK)) Bottom left: Native Top right: Bottles (Soda-9.0-1 Runner with DXVK) Last one is ge-proton9-11 runner. I was too lazy to mount the Windows drive, but the last time I did, I remember my results being around 21k. Kron4ek offers slightly better performance, but I didn't had time to configure it for this test. My goal was simply to show that you can achieve better results than the native implementation by using custom Runners versus basic Wine or the native port. Proton has significantly improved the performance of 3D applications running through this compatibility layer. In some cases, it can even outperform native implementations, especially when those native implementations rely on older technologies like OpenGL. I wish 3D-Coat’s Linux version was not only properly ported but also ported to Vulkan instead of OpenGL, given how much more control it offers over your GPU compared to OpenGL—especially with multithreading in mind, which is a huge bottleneck for OpenGL. In short using Vulkan over OpenGL (native) generally provides better performance, or at least similar to what you get in Windows.
  13. Sure, it is necessary to get a working Linux-version. But using tools like Bottles with specific configurations (like Kron4ek runner with Wine-Staging + DXVK) can significantly improve compatibility and performance, allowing 3DCoat to run. And no, Wine Runners are not just simple "front end" of wine. Runners are pre-customized and pre-patched versions of Wine. CrossOver, develops and maintains their own custom version of Wine or a "runner." This allow CrossOver to run certain Windows software out of the box that might not work, with the native Wine version. Kron4ek runner do work on Ubuntu 24 without any issues. You might want to use latest Nvidia driver (if you use Nvidia). https://github.com/Kron4ek
×
×
  • Create New...